Home

main manual

1. f a m ATA A ppp A AR a a II a O Y E ES SS O O A O O O O A A ES A O A O SA A O IS A E 04A E E jo TT a f ft f f E wy fF O ff f f Tf jf fo Tt gf ft ips Re tr FF A f fT ft fj fw fT fF f a ft ff A tt fj fy fF f A f Tf f f tw fF O a ff OA jp A A ESS q AAA reg ip wy ff ro f ff Tf wy ft a a f ft fm cd f ff f gf Tp pA fT 1 tr jw O a f f ft jw _ ff ff ff f gf ff A tf E ee ee oe e H O 0 Fa AS IES O HA ES IA O IE 0 0 1 We 0 3 0 4 0 5 Fraction of sample rate You define the differentiator by the number of coefficients the frequency range of the
2. O al Ly O This option opens a dialog in which you can create a new data file from a time range and selected channels and frames of the current file Use the dialog to select the channels time range within the frames and the frames to be exported T rra Y You can choose all the channels individual a channels selected channels or enter a list of attimelel Lown 0 El channel numbers directly You can export all Endtimefsl lt High frames the current frame all tagged or un Frames fan frames tagged frames frames with a given state or you can enter a list of frames cos Once you have selected the channels frames and a time range click the Export button to write the data to the new data file This is the same as the Edit menu Copy As Text command but with the output sent to a text file and not the clipboard See the Copy As Text command for details of the dialogs This copies the screen area containing the window to a file Make sure that the window is completely on the screen and that 1t is not covered by any other window You should only use this option when the image you require is an exact copy of the screen If you need to scale the image or want to edit it a Windows Metafile copy is usually better This copies the window as a Windows Metafile This file format can be scaled without losing any resolution and is usually the preferred format for moving Signal images to drawin
3. Introduction Apply Digital filtering The Analysis menu Digital filters command is available when you have a data file open You can apply one of twelve stored digital filters to a set of waveform channels over a selection of frames though you can only preview one channel in the current frame at a time You can also create your own digital filter The program implements FIR filters Finite Impulse Response optimised to minimise the filter ripple in each filter band see page 17 5 for more technical information on the filters Digital filtering x Channel Sinewave 100 000H2 Filter Band pats 13 to 23 Hz Apply Comment la filter comment Close Help Show details The Filter field of the dialog box selects the filter to apply and the Channel field sets the waveform channel to preview The Comment field is for any purpose you wish there is one comment per filter The dialog shows the original waveform in grey and a filtered version in the waveform colour The Close button shuts the dialog and will ask if you want to save any changed filter and the Help button opens the on line Help at the digital filtering topic The Apply button opens a new dialog in which you set which channels and frames to filter with the Channels Apply Example low pass filter to and Frame list fields The Frame subset field can be Channels 14DCO 5kHa y used to further specify which frames are to be filtered Frame lst Tag
4. E A RR AS E E a i i i A HH LH i a p i l E i B a i i E i i i aa i AAA CE CECE Linear A A A A A i TLL LLL A A A A A A hh Y 7 LL IEE i E EE i i AR IE A ih An i if CO W IE un 1 TT A PCCP SS a a OA EE ee Ra 160 A a E E AP O E E A E A A A 0 HRS O ES IS O E E A Pl O E E SS ES E E IS ES O IS A O E FO A IS E E DA DO O O E A AA AAA AAA A HH HA a E E DS a a E a E E E E E E E E E E E oo oon A E oe oe ne ot oO anna a ERR O ee eee ee eee ERF O IO ES O E O O IO IO IO IO nn eee eee eee AA O SS O SO ES O E O O O O OS a O O O O A FS et te o 0 oot ee ee it ft tf fT ft ft fT fT 0 0 1 J 0 3 Fraction of sample rate E i g Hi i i E E i E E L i L g E i i a g i i L I i a a L i TT a Pa OA AA AA AAA T AAA A A A A CAC IO A AA AAA T A A A A MAC MA IO A A AAA AAA MEN T 01 mp AA A AAA T AAA NN AAA A 1 0 001 1111101018 1 0 M1 A A A AA AA E ts LA var prm 4 1 1 bands for differentiator var coef 10 10 coefficients needed band start band end slope weight prm 0 0 0 00 prm 1 0 0 45 prm 2 0 1 0 prm 3 0 1 0 FIRMake 2 prm coef 17 13 Signal for Windows version 3 Hilbert transformer 17 14 A Hilbe
5. function on a waveform channel to find the maximum and minimum values over a wide time range gives a slightly larger amplitude and a much closer agreement Valuate a line of text war lo hiMinmax 1 0 0 55 lo hi hilol 2 Messagel Amp sfindmp 2 2 lt f Power 2P 5 022 4 7138 ib 0 01 sece compile time 0 83 secs execution time gt gt a Eval You can use the Evaluate command in the Script menu if you want to try this It gives a slightly larger value for the amplitude and now the power calculated from the amplitude Amp 3 070068 Amp 2 2 4 712660 Power 4 713800 and the measured power differ by 0 025 For an explanation of the text in the Evaluate window see The Signal script language manual The duration of one cycle of the waveform the time between cursor 1 and cursor 3 is approximately 0 2214 seconds a frequency of 4 52 Hz Again this 1s in agreement with the displayed power spectrum This analysis creates a multi frame memory view by carrying out a leak subtraction analysis on the source data Leak subtraction is a specialised analysis used by voltage and patch clamp researchers The basic technique is to use a small stimulus one that does not cause the cell membrane ion channels to turn on and measure the current flow through the membrane impedance made up from resistive and capacitive components This leak measurement is then scaled to give the expected non ionic conduc
6. script language functions can be used to read and write the settings See the appendices for details of the individual auxiliary states hardware supported 7 8 New El Data Document Text Document Script Document Sequencer Document XY Document File menu The File menu is used for operations that are mainly associated with files opening closing and creating and with printing The first section of the menu is used for operations that create a new window The second section closes windows and saves the contents of a window and also includes commands for saving data under a new name and exporting data to a new data file or a different data format It can also be used to revert a text based view to its last saved state If you have Mail enabled on your system it will include a Send Mail command The third section holds a control for how changed data is written back to the data file The fourth section of the menu loads and saves the configuration files that control data sampling The fifth section controls printing Both the printer setup and the commands to print a window are found here The sixth section holds a list of the most recently used files The final section is the standard way out of any Windows application the Exit command This command creates a new Signal file This can be a sampling document an XY file or a new text based file You can activate this command with the Ct r1 N shortcut key or from the
7. to display The dialog also gives you the option of a j l Large tick spacing setting the x axis tick spacing Il Tick subdivisions 10 Logarithmic J Auto adjust units The Left and Right fields set the window start and end You can type in new positions or select values from a drop down list Each drop down list contains the initial field value cursor positions the minimum and maximum allowed values and the left and right edges of the window XLow and XHigh The Width field sets the window width if the box is checked Click the Draw button to apply changes without closing the dialog Show All expands the x axis to display all the data and closes the dialog Cancel undoes changes made with the dialog and closes it Close accepts any changes and closes the dialog Normally you let Signal organise the x axis style However when preparing data for publication you may wish to set the axis tick spacing If you prefer a scale bar to an axis you can select this in the Show Hide channel dialog You can control the Large tick spacing this also sets the scale bar size and the number of Tick subdivisions by ticking the boxes Your settings are ignored if they would produce an illegible axis Changes to these fields take effect immediately there is no need to use the Draw button The Auto adjust units option will cause the units displayed on the axis to switch to multiples of powers of 10 in order to keep the figures sensible when z
8. Modify channels Tag frame Analysis menu The operations available from the dialog include all of the frame buffer operations all of the channel data modification options plus tag and untag frames For operations that modify the frame buffer such as accumulating an average or summing frames the effect is straightforward For operations that modify file view data such as rectifying channels or subtracting buffer data from frames the changed data must be saved to disk if the action is to have an effect otherwise the changed file data will be discarded This is because Signal only holds one frame from a data file in memory at a time frames are loaded from disk as required and discarded when another frame is wanted Use the File menu Data update mode option to ensure that changed frame data is saved either unconditionally or by querying the user For memory views all of the document data is held in memory and changes to frame data are always saved This command provides a pop up menu specifying the data modifications that are available All the modifications operate on the selected waveform channels or on all visible waveform channels if none are selected If the frame buffer is being shown then they operate on frame buffer data In either case all data points in the channels are modified Most of the modifications are also available via the keyboard shortcuts shown in the menu As for the frame buffer operations changes to the frame
9. RAMP SZ SZINC RATE RATEW ANGLE WAITC RING RINCW PHASE OFFSET CLRC DELAY DBNZ BXX CALL CALLV RETURN JUMP HALT NOP ADD ADDI DIV RECIP MOV MOVI MUL MULI NEG SUB TABLD TABST TABINC CHAN POINTS REPORT MARK STATE SETS SWEEP WSWP TRIG TICKS BRAND MOVRND WAVE WAVEBR Write to digital output bits 15 8 Write to digital output bits 7 0 does nothing in 1401 plus Read digital input bits 7 0 copy to V56 test and branch Test last read digital inputs in V56 and branch Set DAC value for DACs 0 7 Increment DAC by a value Automatic DAC ramping to a target value Set the cosine output amplitude Change the cosine output amplitude Set the cosine angular increment per step AS RATE but waits for phase 0 Set cosine angle for the next step Branch until cosine phase 0 Change the cosine angle increment per step AS RINC but waits for phase 0 Defines what phase 0 means Offset for sinusoidal output Clear the new cycle flag Do nothing for a set number of steps Decrement a variable and branch if not zero Compare variables and branch xx GT GE EQ LE LT NE Branch to a label save return position Like CALL but load a variable with a value before call Branch to instruction after last CALL or CALLV Unconditional branch to a label Stops the sequencer and waits to be re routed This does nothing for one step No OPeration Add one variable to another Add a constant value to a varia
10. Signal port setup dialog from the Port Setup page in the sampling configuration are used You can set the channel name and units to anything you wish the port Zero value should be set to 0 and the Full scale value should be set to the value of a full scale signal with the MultiClamp amplifier gain set to unity It is very important that you enter the value for unity MultiClamp 700 gain not whatever gain you have in use at the moment The full scale value is the value corresponding to a 5 volt or 10 if appropriate signal on the 1401 input The checkbox for a 10 volt system should be set if you have a 1401 with 10 volt inputs otherwise it should be left clear The button marked Test at the bottom of the dialog can be used to test that Signal can communicate sucessfully with the Multiclamp Commander controlling software and will provide some diagnostic information if the test fails Of course you need to be running Multiclamp Commander on the same machine for Signal to be able to communicate with it 19 3 Signal for Windows version 3 19 4 MagStim auxiliary device support MagStim Safety notice Introduction Auxiliary states devices When sampling using multiple frame states Signal can control external devices such as stimulators in addition to switching the 1401 outputs This is achieved by using auxiliary states devices external equipment that 1s set up in different ways according to the sampling stat
11. Status bar 2 2 Stop band 17 6 17 7 Stop Process command 11 6 Stop sampling 3 18 3 19 SUB output instruction 6 22 Subtract DC level from data 11 19 Sum of data 12 5 Sweep Mode selection 3 4 Sweep mode Basic 3 4 Fixed interval 3 4 Outputs frame 3 4 Peri trigger 3 4 SWEEP output instruction 6 25 sxy file extension 8 1 Synchronisation of outputs 3 10 System resources 16 2 System software required 1 2 SZ output instruction 6 14 SZINC output instruction 6 15 TABDAT directive 6 9 TABLD output instruction 6 23 TABST output instruction 6 23 TABSZ directive 6 9 Tag frame 11 19 Telegraph configuration dialog 19 1 Telegraph output 19 1 Text output 9 1 Data specification 9 3 Format specification 9 2 From data view 9 2 From XY view 9 3 Threshold crossing cursor mode 12 2 TICKS output instruction 6 26 Tile windows Horizontally 15 1 Vertically 15 1 Time limit 3 15 Time shift 17 5 Time units in X axis dialog 2 5 Setting 9 5 Time zero 2 7 12 4 Times Specifying 2 5 Toolbar 10 1 Toolbar 2 2 Trend plots 11 7 TRIG output instruction 6 26 Trigger Analogue 3 7 Digital bit 3 8 Peri trigger modes 3 7 Pre trigger points 3 8 Sweep trigger 3 6 Waveform 3 7 Triggered start of sampling 3 18 Trough between cursors 12 6 Trough cursor mode 12 2 TTL compatible signals 3 2 Two band pass filter 17 4 Two band stop filter 17 4 TXT standard file extension 8 1 q Undo command
12. This command copies data views to the clipboard as text Text representations of Signal data can be very large and awkward to manipulate with the clipboard as an alternative you can write the text output to a file with the File menu Export As command The command leads to dialogs that set the text output format and the data to copy This dialog sets the text output format The first section sets an interpolation method for Waveform ae Interpolation method Linear channels This is because the Waveform channels will be output in columns with each row showing data sale a Waveform e wW m sampled at a particular time If the data is not sampled haka F F in burst mode then the data points in the file will not pi r all be sampled at the same time This will result in a idealised y a small error as data will appear shifted slightly to bring pecima places data time Go all the points into line This can be overcome by using Field width for all it jo interpolation to estimate where the waveform ona A MANIN a Wem given channel would have been at the time a point on Sting delimiter I another channel was sampled Linear and Cubic Separator Tab Spline interpolations are available wl Uo iho ee In the next section you can choose whether to output Cancel Headings Data values and Time values where appropriate for each of the Waveform Marker and Fitted data by checking the boxes All output is written
13. be prompted to save them before the application terminates 8 6 Undo Text output Bitmap output Metafile output CED binary format Paste el Delete Edit menu This menu holds the standard Edit menu functions that all Windows programs provide The majority of the menu is associated with commands that move data to and from the clipboard You can also use these commands to search for strings in a text window or to search for the currently selected text When the current window is a text based view the Edit menu operates in the standard manner you can cut and paste text between Signal text windows and other applications When the current window holds a Signal data document the behaviour is modified In a text based view this undoes edit operations In Signal data views you can undo display scaling operations for example when you drag a rectangle over a channel to zoom in or out or use the X or Y axis dialogs Changes made to an idealise trace formed from patch clamp data may also be undone A multiple level undo list is maintained You can cut selected portions of editable text to the clipboard from any position in Signal where the text pointer is visible This includes any text or numeric fields in dialogs and all text based views You cannot use this command in Signal data document windows You can copy selected portions of editable text to the clipboard If you use this command from a data document window the wind
14. channels to display in p Chan numbers Grid M Y Axis i x Axis your window A Signal data file can hold up to 100 channels so having the ability to choose the channels to display is important if you are to see any detail options M Numbers lv Title 4 Units M Small ticks T On right options M Numbers lv Title lw Units M Small ticks i Scroll bar Scale only Scale only The list on the left of the Alon aop Draw Close dialog holds all the channels that can be displayed You can also show or hide the axes grid and scroll bar in the window from this dialog and control the appearance of the x and y axes Check the boxes next to the items for display and click the Draw button to see the result The Scale only option draws axes as scale bars 5 Keyboard Marker 4 ADC 3 Waveform 3 ADC 2 Waveform 2 ADC 1 Waveform 1 ADC 0 Waveform Cursor measurements S Getting started Open the View menu Frame display list dialog Specify frames 1 4 as the frame list and click OK Use the View menu Overdraw frame list command to turn overdrawing on and off Experiment with selecting frames and with the colour cycling display mode The frame display list is a list of frames to show in addition to the current frame if o j ramez agged frames overdraw mode is enabled It is also possible to Frame subset Frame state xxx overdraw the buffer a special frame held in
15. count increments gt Ramping HALT task finished gt Done It is a property of signed integers that adding 1 to the maximum positive number yields the minimum negative number If you use ADDAC repeatedly with the same value eventually you will run off the end of the DAC range and come back in at the other end DAC units run from 32768 to 32767 In a 5 volt system with 16 bit DACs this is 5 0000 to 4 99985 volts The DAC unit value for 5 volts is 32768 but this number does not exist in 16 bit signed integers and wraps around to 32878 Users often want to set the DAC to full scale so for the DAC command used with expr not with Vn we change requests to set 32768 units to set 32767 units Unlike the digital outputs the DAC output changes when the instruction runs not at the next sequencer clock tick the changes may have a time jitter of a few microseconds This command starts a DAC ramping with updates every sequencer step If the DAC was generating a cosine the cosine output stops The DAC ramps from the current value until it reaches a target value when the DAC cycle flag sets You can use WAITC to test for the end of the ramp The RATE instruction stops a ramp before it reaches the target value RAMP n target Vn slope Vs Vstoff n DAC number in the range 0 7 available DACs depend on the 1401 type target This is the DAC value at which to end the ramp The units of the DAC values are those set in the outputs page of the
16. defined The frame list is dynamic in that if for example you request all tagged frames and subsequently tag a frame 1t will be added to the overdraw list Use colour cycling If you choose the special category Frame state xxx an extra field appears Selected frame state You should set this field to the state number that you want to display The Use colour cycling checkbox selects drawing of data from each frame in a different colour Signal contains a fixed table of 18 colours that it uses for this mode of operations If the checkbox is clear then all the display list frames are drawn using the same colour this colour can be set from the colour setup dialog These commands the two buttons at the lower left of data windows and the keyboard shortcuts Ctrl Right and Ctrl Left expand and contract the displayed x axis area The enlarge command zooms out by doubling the data region spanned by the x axis The reduce command zooms in by halving the data region The left hand edge of the screen is fixed unless the expand operation would display data beyond the end of the frame in which case the displayed area is moved backwards If the result of expanding would display more data than exists in the frame all the frame data is displayed This menu command double clicking the x axis of a data or XY view and the shortcut key Ctr1 X open Left s 0 01 y Show All the x range dialog which sets the region of the view der E H width poa
17. m be dragged into the display area to add a new pulse into the outputs Each iil A icon represents a different type of pulse i a square pulse without variations digital or DAC a square pulse with varying amplitude DACs only a square pulse with varying duration digital or DAC a train of square pulses without variations digital or DAC a ramp pulse with varying amplitude at start finish or both DACs only a sine wave without variations DACs only an arbitrary waveform on multiple DACs one only of these allowed Cel ajSjPA D This is the area at the bottom right of ve seamentaroto02 1d hz the dialog It holds the parameters ots 2 stes 01 defining the currently selected pulse Cente Wotsi Lenath s a1 The pulse can be changed by editing the Cycle fs 01 Startphase 50 parameter values The parameters shown vary according to the type of pulse the box title shows the type of pulse and the start and end times for the pulse again using the currently selected time units The details of the parameters are covered in the Editing the pulse parameters section below A number of operations in the pulse configuration dialog are carried out by dragging and dropping A major advantage of drag and drop is its graphical visual nature For this reason it is increasingly commonly found as a way of carrying out operations in Windows software To drag and drop a screen object place the mouse pointer on top of the obje
18. plug located on the front of the 1401p us and on the rear of the micro1401 Micro1401 mk II and Power1401 In addition to the data lines a TTL pulse is required on the digital inputs Data Available input to log a digital marker Digital input bit bit7 bit6 bitS bit4 bit3 bit2 bit1 bitO Gnd Digital input pin 5 18 6 19 7 20 8 21 13 Signal 1401 plus All others Data Available pin 24 pin 23 To log a digital marker apply a low going TTL pulse at least 1 us wide to the Data Available When the 1401 detects the falling edge of the input it latches the input data on the Power1401 micro1401 and Micro1401 mk II If you have a 1401p us you must keep the digital input data signals stable for 50 microseconds after the low going data available edge When Signal displays marker data from keyboard marker or digital marker channels it shows the code of the first of the four markers as well as the marker time Marker codes have values from 0 to 255 This is the same range of numbers that the ASCII character set uses and it is sometimes convenient to treat the codes as ASCII character codes for instance when dealing with keyboard markers At other times it is more convenient to deal with the codes as numbers Whenever Signal displays a marker code that has the same value as the ASCII code of a printable character it displays the code as a character otherwise it displays the marker code as a two digit hexadecimal number Hexadecimal base 16 numbers
19. 10 for a 10 volt 1401 Units The units for calibrated data This is a string from 1 to 6 characters long If you set the first character of the units as a space this allows future versions of Signal the option of automatically adjusting the units by replacing the space with a character representing a factor of a 1000 such as u or k Name The port name This is a string from 1 to 19 characters long it sets the title of the waveform data channel sampled from this port 3 8 Clamp configuration Outputs configuration Outputs type Outputs type None Timer period ms Outputs type Pulses Sampling data Options This is a string of 0 to 8 characters that holds online processing options for data from the port Characters corresponding to various processing options can be entered into this dialog Currently only one processing option is available enter an R character to cause online rectification of sampled data The Conditioner button opens the signal conditioner setup dialog if a signal conditioner has been found see the Programmable Signal Conditioners chapter The Telegraph button on the bottom left opens a dialog that allows you to configure amplifier telegraph support Amplifier telegraphs are signals usually analogue outputs from an amplifier that signal current amplifier settings such as gain and offsets By collecting and interpreting the amplifier telegraphs Signal can automatically adjust for change
20. 19 Run script from Script Bar 14 2 S s sequencer expression function 6 7 Sample Bar 13 1 Sample Bar List 13 1 Sample interval 3 1 Sample menu Output controls 13 2 Sample Bar 13 1 Sample Bar List 13 1 Sampling configuration 13 1 Signal conditioner setup 13 1 Sample rate For waveform data 3 1 Minimum 3 1 Sampling Aborting 3 18 Accept sweep 3 18 ADC ports 3 6 Artefact reject configuration 3 16 Automation configuration 3 15 Configuration contents 3 20 Configuration dialog 3 4 Continue with next sweep 3 18 3 19 Control of states 7 7 Controls during 13 2 Controls during 3 17 Digital outputs 6 1 File size limit 3 15 Finish button 3 18 3 19 Fixed interval sweeps 5 8 Index 5 Signal for Windows version 3 Fixed interval variation 5 8 General configuration 3 4 Keyboard marker entry 3 19 Maximum frame length 3 5 Maximum rate 3 5 Menu 13 1 Output during 6 1 Outputs configuration 3 9 Outputs frame sweeps 5 8 Overdraw frames 3 19 Pausing at sweep end 3 18 Peri trigger configuration 3 7 Peri trigger level adjust 3 19 Reject sweep 3 18 Restarting 3 18 Sample rate 3 5 Saving configuration 3 20 Saving new data 3 20 Setting configuration 3 20 Stopping sampling 3 19 Time limit 3 15 Triggered start 3 18 Write sweep automatically 3 17 Sampling configuration 13 1 Loading and saving 8 5 Save changed data 8 5 Preferences 9 6 Save file at sampling end 3 15 Sa
21. 1f at least one flag of the event matches the Include set and none of the flags of the event match the Exclude set The flags are An event where the amplitude is not an average of the raw data points Open time A period when the channel is open Closed time A period when the channel is closed First latency The first event in the idealised trace Truncated The last event in the idealised trace Assumed amp Bad data Event flagged as not suitable for analysis Level n Six flags one for each level of multiple level data The Level 1 flag 1s set for all closed times as well as the first open level The open closed amplitude histogram is almost identical to the amplitude histogram described already The differences are that the y axis will be a count of events rather than time spent at that amplitude and the ability to include and exclude events using the Include and Exclude flags as with the Open Closed time histogram The Maximum amplitude and Minimum amplitude fields set the amplitude range that is divided into bins and that sets the x axis range in the result This opens a setings dialog which defines the parameters for a histogram of event burst durations The Channel field selects a waveform channel from the source view This channel must have an idealised trace fitted in order for the histogram to be built The x axis of the result starts at zero so the Bin width and Maximum duration together define the number of bins in
22. 3 or 7 3 is 2 except in a 1401 plus where it is 3 MUL MULI Table access TABLD TABST TABINC Sequencer outputs during sampling MUL multiplies a variable by another variable then optionally adds a 32 bit integer constant and divides the result by a power of two MULI multiplies a variable by a 32 bit integer constant and divides the result by a power of 2 MUL Vas VO expr Shift y Va Va VDI texpr gt gt shirt MULI Va expr shift Va Va expr gt gt shift Va A variable to hold the result It can be the same as Vb Vb A variable used to calculate the result expr An expression that is evaluated as a 32 bit integer It 1s optional for MUL and required for MULT If this argument is omitted it is treated as 0 shift An optional argument in the range 0 to 31 set to 0 if omitted that sets the number of times to divide the result by 2 The following examples assume that V1 holds 10 and v3 holds 10 MULI ARO V1 100 10 10 MUL ML op O Y 992 T00 IA O 8 Tables are declared with the TABSZ directive and can be populated with data using the TABDAT directive Most access to tables is through the Vn toff method but there are also instructions for loading and storing a register in a table and for incrementing or decrementing a register used as a pointer into the table These two instructions load a register from the table and store a register into the table Many instructions can load ar
23. 95 that uses spare disk space for virtual memory management The user defined limits are 3 15 Signal for Windows version 3 Artefact rejection Creating a new document Frame zero Number of frames If this limit is enabled sampling stops when the set number of frames have been written to the data file Sampling duration If this limit is enabled sampling stops when the set time has passed since sampling was started File size Kbyte If this limit is enabled sampling stops automatically when the file size reaches or exceeds the set size The Artefact reject button opens a dialog that allows you to configure artefact rejection This provides mechanisms for automatically rejecting or tagging sampled frames if they contain an artefact normally caused by stimulation The artefact rejection dialog can be used to Mr configure Signal to automatically examine newly Artefact rejection mode None sampled data and if the data has reached the ADC limits to reject or tag the new frame Artefact level of ADC range jo Start time offset for search 2 o Artefact rejection is important when generatin J P 8 8 End time offset hor search s 0 2 averaged evoked responses particularly from the Allowed bad points percent 2 EEG where artefacts often occur and where the mathematical rigor of the averaging process will be affected by the presence of signals at the ADC limits Cancel The Artefact
24. DAC 02 Set slow pulse gt F for fast 0006 DELAY ms 99 1 Wait 99 ms gt F for fast 0007 DAC 0 0 Set output low gt F for fast 0008 DELAY ms 98 1 Wait 98 ms gt F for fast 0009 JUMP SL Continue gt F for fast Most of this sequence is similar to MyFirst pls The differences of interest are Steps 0 and 5 have a keyboard character defined as well as a label for looping The sequencer control panel will show buttons corresponding to each character defined pressing the button or the appropriate keyboard character will cause sequencer execution to jump to the relevant instruction A JUMP instruction is used instead of DBNZ so that the pulse output loops continue forever The remainder of this chapter is organised as a reference manual for the sequencer instructions You can find more information about the output sequencer in the Signal Training Course manual 6 5 Instructions Digital TTL compatible input and output see page 6 10 DAC waveform voltage outputs see page 6 12 Sinusoidal waveform output 6 6 see page 6 14 General control see page 6 19 Variable arithmetic see page 6 21 Table support see page 6 23 Access to data capture see page 6 24 Randomisation see page 6 26 Arbitrary waveform output see page 6 28 Signal for Windows version 3 These are the output sequencer instructions in Signal version 3 DIGOUT DIGLOW DIBNE DIBEO DISBNE DISBEQ DAC ADDAC
25. DIBEQ DIBNE or WAIT against a pattern DISBEQ branches on a match DISBNE branches if it does not match DISBNE pattern Vn Vn off LB DISBEO pattern Vn Vn off LB pattern This is 8 characters one for each input bit The characters can be 0 1 and meaning match 0 TTL low match 1 TTL high or match anything The 6 11 Signal for Windows version 3 DAC outputs DAC ADDAC bit order in the pattern is 76543210 You may omit the square brackets however the Format command inserts them Vn With a variable the result is input BAND Vn 7 0 BXOR Vn 15 8 A result of 0 is a match not zero is not a match LB The destination of the branch if the input was a match DISBEQ or not a match DISBNE This label must exist in the sequence This example shows a typical use of this instruction We want to run a different part of the sequence for three trial types signalled by external equipment that writes the trial type to digital input bits 1 and 0 00 means no trial 01 10 and 11 select trial types 1 2 and 3 TRWAIT W DIBEO igh da 00 TRWAIT Wait for trial gt Wait DISBEO Lucca OL TRITANI DISBEQ 10 TRIAL2 DISBEQ 11 TRIAL3 The output sequencer supports up to 8 DAC Digital to Analogue Converter outputs The 1401p us and Power1401 have four DACs and the Micro1401 has two However the Power1401 can be expanded to 8 DAC outputs The last value written to DACs 0 7 is
26. Frame points fiona m of the sampling parameters Click akis ceo ollet fo Pause at sweep end on a tab to display the corresponding section of the dialog The sections always ADC ports c available are General Port Cancel Runnow Hep setup Outputs and Automate There are other sections Peri trigger and States that hold extra information not relevant to all sampling configurations The Peri trigger tab appears only when the sampling mode selector in the General section is set to Peri trigger the States tab only appears when the General section Multiple frame states item is checked The General section holds a selector for the sweep mode the multiple frame states checkbox fields to define the waveform sampling rate and the frame width or points checkboxes to control the creation of marker channels and various general options plus a list of ADC ports to sample The Sweep mode selector defines how sweeps of data are taken and triggered and how sweeps relate to the outputs system It has six selections listed below Basic the trigger for a sweep of data is a TTL pulse at the start of the sweep and pulse outputs start and finish at the same time as a sampling sweep Peri trigger the trigger point can be before the start of the sweep at the start of the sweep or at any point within the sweep Pulse outputs start at the trigger point and finish at the end of the sampling sweep This mode allows a wide variety of triggers includ
27. Power Pulses This sets the main master stimulator power output as a percentage of the maximum available you can set any value from 0 to 100 This control is disabled for manually controlled states This sets the interval between the two pulses you can use any value from 0 to 999 milliseconds or 0 to 99 9 milliseconds for Hi res timing Setting a value of zero for a state will switch the BiStim into simultaneous pulse mode setting a value greater than zero switches it out of simultaneous mode While switching back and forth between modes during an experiment is possible it may increase the inter sweep delay This control is disabled if independent triggers are in use or for manually controlled states This sets the second slave stimulator power output as a percentage of the maximum available you can set any value from 0 to 100 This control is disabled for manually controlled states The Rapid is rather different from the other l a i TERE TEE two devices as it is capable of producing a Mitin cinco bici x train of pulses at high rates Rather than the Type of stimulator MaaStim Rapid simple case mounted manual controls of the Gam port for communications cow z ee other two units the Rapid has a separate F janore coil interlock switch control system that has to be disconnected to Te fiend sinele pulee mode gain access to the serial line control port Sower 10 10 pulses at 10 Hz limiting the usefulness o
28. S 4 9 Sinusoid gt Sine RATEW 0 0 stop at cycle end END WAITC 0 END wait for end gt Wait end RATE O20 stop now RAMP 0 0 0 07S Cl r ramp to 0 yolt cin 1 see RAMPDN WAITC 0 RAMPDN walt gt Ramp down HALT Each time the phase angle of a cosine passes through 0 a new cycle flag sets This flag is also set when a ramp terminates There is a separate flag for each DAC This flag 1s cleared by CLRC RATEW RINCW and when tested by WAITC WAITC n LB DAC T n DAC number in the range 0 7 available DACs depend on the 1401 type LB A label to branch to if the new cycle flag is not set If the flag is set the sequencer clears the flag and does not branch This instruction can produce a pulse at the start or at least a known time after the start of each waveform cycle The following sequence outputs 4 cycles of waveform at different rates on DAC 1 and changes the digital outputs for each cycle SZ Ta L0 make sure full size ANGLE ey Dae make sure we start at phase 0 RATE lO 1 degree per step to start with DIGOUT 00000001 so outside world knows RATEW Alege e 2 next cycle faster clear cycle flag wl WAITC 1 wl wait for cycle gt 1 degree cycle DIGOUT 00000010 announce another cycle RATEW lo A next cycle a bit faster w2 WAITC EZ wait for cycle gt 1 2 degree cycle DIGOUT 00000011 yet another one RATEW Locke 6 last cycle a bit faster w3 WAITC 1 w3 wait for cycle gt 1 4 degree cycle DIGOUT 00
29. Slope of best fit line 12 5 Sum 12 5 Underlying CFS data 3 1 Value at cursor 12 4 Waveform display Cubic Spline 2 6 Dots 2 6 10 5 Histogram 2 6 10 5 Line 2 6 10 5 Skyline 2 6 10 5 Waveform output 5 6 During sampling 6 1 Maximum rates 5 7 test from output sequencer 6 28 Waveforms Drawing mode 10 5 Web site 1 2 whole cell 4 1 Width of a memory view 2 10 Window for FFT 11 3 Window menu 15 1 Arrange icons 15 1 Cascade 15 1 Duplicate 15 1 Hide 15 1 Show 15 1 Tile Horizontally 15 1 Tile Vertically 15 1 Windows 15 2 Write to disk 3 17 WSWP output instruction 6 26 X axis control 2 5 X Range dialog 2 5 10 2 XY Draw Mode 10 6 XY view 2 12 Auto expand axes 10 4 Copy as text 9 3 Creating a new view 11 7 Delete channel 11 17 Draw mode 10 6 Example 2 12 Key 10 4 Line style 10 6 Options 10 4 Point style 10 6 Prompt to save unsaved view 9 4 ay _ Y axis control 2 6 Y Range dialog 10 3 Y zero 2 7 12 4 Zero data 11 19 Zero region 2 8 12 5 Zoom in button 2 3 Zoom out button 2 3 Index 7
30. Step both Step start or Step end This is an extremely versatile form of pulse for example by setting the start size to zero you can produce a ramp running from one level to another This item specifies a cosine wave output Of lt Ginewaveseomentatoitoo2 ld fixed duration amplitude and frequency for Sizetvolts 2 stars for output on a DAC For Signal version 3 sine Certe Motsifo Lenathis o1 wave output can be generated on all DACs In P e earlier versions only DACs 0 and 1 could be used The Size parameter sets the amplitude of the cosine wave the distance from the mid point to extreme The other parameters are Centre which sets the level about which the cosine oscillates Cycle s which sets the duration of one complete cycle and Start phase which sets the initial phase in degrees The Centre value is an absolute or relative voltage level as required Because the output is actually a cosine wave an initial phase of 90 degrees will start the output off at the centre level This item specifies arrays of data to be output to amar mavetorm ar 0E5to 0 3016 laf one or more DACs at a specified rate Output to Dal select Sta si 065 each DAC starts simultaneously and consists of PeelHa 5000_ Points f1258 the same number of points so the output also finishes synchronously The arrays of data can be changed in two ways by using the Signal script language functions and by copying and pasting Signal data using
31. XY view will be deleted first In addition to the Y axis optimisation control there is an extra checkbox for view X axis optimisation after processing This command opens a tabbed dialog joj from which you can fit mathematical aT f Fit settings Coefficients Results functions to channels in a file memory or XY view If you fit data to fac A 8 g A A 1 F 4 a channel in a file or memory view and error bars are displayed the fit Channel 1 Chan 0 Fit Exponential aa minimises the chi squared value o EZ cuota otherwise the fit minimises the sum of 7 Reference Cursor t a Imum iterations squares of the errors between the data al and the fitted curve for frames Current frame IW Make an initial guess at fit coefficients ignore current values In addition to best fit coefficients and M Showlit I fom Eus to an estimate of how much confidence to place in them you also get an estimate of how likely it is that the model you have fitted to your data can Dore Heb Cose explain the size of the chi squared value or sum of errors squared If your data does not have error bars these estimates are based on the assumption that all data points have the same normally distributed error statistics Fit settings Channel Fit Range Reference Maximum iterations for frames Analysis menu The dialog has three tabs Fit settings Set the fit type and range of data to fi
32. __ i __ i _ tf _ _ ff fy with slope of 2 0 HH HH HH HA HH AHH HH HH HA eH tt pp tw fl f toy Ty fT f fF f f ff ow ft a ft ff f f i tw 0 1 1 1 SSeS 0 a _ f fog it _ Ff O fT a A f O O E ta __ aka asa eS f f f fg ft tT fF Tf ff ff ff i fw E l ff f jf T fF ft a fw ff f f ff ff tf A AR O a a SI Be E A E E E EOI E O O O O O O O O E O O O O O O O O O O O O ee O O O O O O O E wm wy gy i 4 Soe ot eS O a a H EEE H HIII A Ff J mn jf O ft f f
33. a clear indication that the next sweep has started This is valuable because once the pulse output or sampling has started the Accept Reject button in the control panel cannot be used on the previous sweep of data The small floating window is the Sampling control panel It contains buttons and other controls to interact with the sampling Sampling will not start until you click Start in this control panel the Sample menu duplicates the panel controls If the Event 1 start box is checked sampling waits for an external signal after the start button is pressed This dialog controls which frames to include in the processing and how to update the memory view holding the analysis result see the Analysis menu chapter for a full description of this dialog The most common mode is All filed frames with an update every 0 frames every frame This dialog closes when you select either OK or Cancel You can recall it with the Process command in the Analysis menu The sampling control panel holds several buttons and checkboxes used to control and interact with data sampling M Sweep trigger When the control panel appears it looks like the picture to Eventi start the right Click Start to begin sampling or Abort to give up Reject Abort Restart immediately Once sampling has started the buttons change Sweep end but most checkboxes are always present and can be changed fe write to disk at sweep end Pause at sweep end at any time during sa
34. and Use Black And White Change Colours Application colours Channel primary colour Channel secondary colour Channel background colour XY channel colours View colours View menu If you have a colour monitor you can choose to display your data files in colour The Use Colour menu item switches from black and white data displays to colour If you change to colour the menu item changes to Use To Black And White You may prefer to work in monochrome if you have to print the end result in black and white You can choose the colours for almost Set colours everything in Signal If you open the dialog application colous with an active file memory or XY view the Data view background dialog has multiple pages Select a page with fwavefom as a line the drop down list at the top The pages are Waveform as dats Waveform as skyline Waveform as histogram Waveform histogram fil To change colours select one or more items Markers as dots in the list on the left and then click a colour Markers a lines in the palette on the right You can check the e Rate histogram result of your action with the Draw button Rate histogram fil a E Cancel removes the window and undoes any 2h Fesetal Draw Cursors amp cursor labels changes OK accepts changes and closes the IContrals met Cancel dialog The Reset All button returns the list and the palette to a standard set of colours You can change the f
35. and organisation You can run both version 2 and version 3 on the same system but not simultaneously as long as they are in different folders If you have version 1 on your system install version 3 to a different folder The installation program copies the Signal program plus help and example files It can also install 1401 support in Windows NT 4 If you use Windows 95 98 Me NT 2000 or XP there are instructions for loading the 1401 drivers using the Windows Device Manager If you are new to Signal please work through the Getting Started tutorial in the next chapter Where you go next depends on your requirements The Signal Training Course Manual is more descriptive than the other manuals which are organised as reference material However it covers all versions of Signal and you will occasionally need to refer to the other manuals for version 3 specific details The on line Help in Signal has a lot of information if in doubt use the F1 key for Help You can update your copy of Signal to the latest version 3 release free of charge from our Web site http www ced co uk in the updates section You can only update a correctly installed copy of Signal version 3 There are full instructions for downloading the update on the Web site You can also sign up for email notification of updates Once you have downloaded the Signal update you will find that the update process is very similar to the original installation process except that you must se
36. and start sampling this sequence would be used You can also set the output sequence file from the Sampling Configuration dialog This is the Help button It opens a window holding a list of the sequencer topics for which help is available You can copy and paste text from the help window into your sequence The name of the output sequence file to use during sampling is part of the sampling configuration The file name including the path to the folder containing it must be less than 250 characters long You set the file name either with the Current button as described above or in the Sampling Configuration dialog When you start sampling Signal searches for the output sequence file named in the sampling configuration and will generate an error message if it cannot be found Signal compiles output sequence files whenever you use them If a file contains errors you are warned and the file is ignored The sequencer runs instructions in order unless told to branch starting with the first instruction The sequence can either be restarted at the start of each sweep or it can be set to free run throughout data acquisition Sequencer execution can be re routed during data acquisition by associating an instruction with a key on the keyboard Each time the key is pressed or the associated sequencer control panel button is clicked or the script language SampleKey command is used the sequencer jumps to the associated instruction Signal records key
37. as seconds The two Line width items set the line width in points for drawing data and axes respectively These are relatively unimportant for displays on the screen where most reasonable settings will only select between lines one or two pixels wide and many different settings will produce the same display The line width controls are particularly valuable for printing where the lines drawn can get unsatisfactorily fine At CED we find that values of 0 75 pt for data and 1 0 pt for axes look pleasant The line widths also control various other drawing operations for example the axis width controls borders drawn around views and the data width sets the basic size of drawn dots and XY view lines and symbols You can change the order of data and memory view channels by clicking and dragging channel numbers The Standard display shows the lowest numbered channels at the top checkbox sets the order when you use the Standard display command or open a new window If you do not check the box lower numbered channels are at the bottom 9 5 Signal for Windows version 3 Do not use the flicker free drawing method Data Data export format Default data update Metafile scaling Version 3 04 implemented a new buffered drawing method that reduces screen flicker on updates However this may impact the display speed Check the box for the old method The Data tab controls the way data is stored and exported The data export format items
38. at the top right of the script window provides overall script language help The help is implemented using the standard Windows help system with contents indexes hypertext links keyword searches help history bookmarks and annotations If you are unsure about using Windows help use the Help menu Using help command to get detailed instructions This command is found in the Help menu It opens an information dialog that contains the serial number of your licensed copy of Signal plus your name and organisation Please quote the serial number if you call us for software assistance If there is a 1401 device driver installed the driver revision is displayed If the l l de da nal sspedla an Signal version 3 06 January 2006 cana DEAE ee Copyright Cambridge Electronic Design Ltd 1997 2006 will be warned Signal displays the type of 1401 and the monitor version if a Licensed to Simon Parker 1401 is connected and powered up A A Lazen Uie Seral Number 030123 1401 driver USB 2 02 Microl401 mk Il Monitor 09 If the monitor is not the most recent at the time this version of Signal was released an asterisk follows the version If it is so old that it compromises data sampling two asterisks follow the version Working set 400 to 4000 kB The Power1401 and Micro1401 mk II have firmware in flash memory Flash updates and instructions for applying them are available as downloads from the CED web site you can u
39. automatically re processed if appropriate For example if a memory view holds the average of all tagged frames and a frame in the source document is tagged or untagged then the memory view data will be automatically regenerated using the new frames The new window behaves like a file view containing one or more frames of data The simplest way to get a feel for this 1s to try it so Make the original file view of the data the current window by clicking on it You may find it easier if you close all the other windows first Use the Analysis menu New Memory View command to select a Waveform average The Settings dialog prompts for information to define the new window There are three fields that define the waveform average The Channels field rannels all Channels selects the channels to analyse You can select any as acannon s channel or list of channels that holds waveform data The channel list in the pop up menu only includes suitable channels for analysis Average axis starts at zero IM Display mean of data Error bars The Width of average field sets the width of the ei result in units set by the source data x axis units You can choose any width you like limited only by the width of the source frame Start offset 0 0 Fi The Start offset field sets the start point within the frame of the data that is included in the average This is specified as the offset from the start of each frame to the start of the da
40. being aborted then Signal saves the configuration as the file last sgc The saved configuration is used the next time data is sampled When Signal starts it searches for and loads the configuration file default sgc If this cannot be found it uses last sgc These files are kept in the directory from which Signal ran If sampling fails or is aborted the sampling configuration switches back to the configuration in use before you started sampling This describes a sequence of operations that build a new sampling configuration from scratch You will find that once you have built a few configurations it is simpler to load an existing configuration and change the sections that do not fit your requirements rather than re build entirely The steps are 1 Open the Sampling Configuration dialog using the Sampling menu or toolbar and set the sampling configuration limits and port information 2 Press Run Now from the configuration dialog or press OK and select the New command in the File menu and choose Data Document 3 Arrange the new file view as you require and add or remove duplicate windows Use the Analysis menu New Memory view command to add memory views as required and set their update mode and position on screen 5 You can use the File menu Save Configuration As command to save the configuration to disk at this point 6 Sample adjusting any positions display configurations and so forth as required 7 Once sampling has finished
41. can be used instead of the WSWP command to wait until a specific point in the sweep but based on points rather than time This instruction takes rather longer to execute than other sequencer instructions and may cause sequencer timing problems if used in circumstances when the 1401 is heavily loaded 6 24 REPORT MARK STATE SETS SWEEP Sequencer outputs during sampling The REPORT instruction records a digital marker if the digital marker channel is enabled as if there was an external pulse on the data available input see the Sampling data chapter for the input to use The MARK instruction does the same except it takes the argument as the value to record REPORT has no arguments REPORT OptLB MARK expr Vn Vnt off OptLB expr The argument should have a value in the range 0 to 255 If a variable or table is used the bottom 8 bits of the value are used OptLB If this optional label is present 1t sets the next instruction to run otherwise the next sequential instruction runs LB DIBNE AA Ly Waiting for Dit 0 REPORT Save a marker when this is set MARK LZ set code 12 as a digital marker This instruction sets a variable to the current sweep state code STATE VI ODE Vn Sweep state code Vn This variable is updated with the current sweep state code OptLb If present the instruction branches to this label This can be used in conjunction with the Signal multiple states mechanism in Static or Dynami
42. can enter the number of frames required The Clear memory view checkbox is ignored as the memory view 1s always cleared before processing The Frames between updates field sets how often the processing of filed frames occurs Set this to zero to process as often as possible If you are processing Sampled frames then this field is ignored and the memory view is updated for each frame This menu command opens the analysis settings dialog for the current memory or XY view This is the same dialog as the one used to define and create the new view except that the New button is now a Change button The Change button accepts the changed settings and rebuilds and clears the memory view New XY View Trend plot Analysis menu This command analogous to New Memory View is available when a file or memory view is selected It provides a pop up menu from which you can select an analysis type currently Trend plot analysis is the only one available Trend plots consist of sets of measurements taken from a source data document and plotted into an XY view Two measurements are taken from each frame one measurement generates the X part of the XY data the other generates the Y data Selecting Trend plot analysis leads into a Settings dialog where you define the analysis parameters and other information to construct the XY view This dialog is also available later on to change the analysis and view parameters Once you have set the required values in
43. control decisions on how waveform data is written to CFS data files The Save waveform data as field sets the preferred format for waveform data on disk Waveform data in CFS files can be stored either as floating point numbers of as scaled integers Scaled integers are the format of data sampled by the 1401 occupy less space on disk and are the only format that can be read by the DOS SIGAVG software but are less accurate as data are converted to 16 bit integer values and can overflow Floating point numbers are more bulky on disk but are more accurate and cannot overflow In most circumstances the format used for waveform data is set by the destination file or it is forced to scaled integer when sampling data using the 1401 but when creating a new CFS data file by other mechanisms saving a memory view or exporting to a CFS data file the format used is controlled by this field Select Real for maximum data accuracy Integer to produce smaller data files or for SIGAVG compatibility The Keep calibrated zero at zero volts checkbox controls how scaling factors for integer data are calculated Signal tries to use the same integer scaling factors as the previous values for the frame or the previous frame where appropriate but may need to recalculate the factors if the calibrated data becomes too large for the existing scaling or too small to represent accurately If this checkbox is set Signal will always calculate scaling factors that keep zero i
44. current view At the moment you can only override the view background colour The equivalent script command for this is ViewColour 10 7 Changing the Palette Signal for Windows version 3 E E Mi MAA A E MEA a EEE mi Custom colors CE A A i ie A Define Custom Colors gt gt Cancel Help To change a palette colour double click it to open the colour dialog and select a replacement colour The first seven colours form a grey scale from black to white and cannot be changed You can replace the palette colour with any standard colour or you can click the Define Custom Colours button to select an arbitrary colour Click OK or Cancel to exit The colour selection applies to all data files It is stored with the Signal preferences in signal sgp not in the data files When you restore a Signal data file the colours will be those that are currently active not those in use when you last saved it Keyboard display Windows software is usually orientated towards control by means of the mouse and menus but it is often convenient to use the keyboard instead For interactive adjustments of the data or display keyboard control can also be much faster With this in mind Signal includes keyboard shortcuts to handle most display manipulation requirements control Scroll data down up Cursor Down Cursor Up Decrease Y range increase range Ctrl Down Ctr1 Up Optimise Y range End Show all Y range Hom
45. data reached full amplitude or was closed New idealised trace Threshold Analysis menu Consecutive transitions in the same directions can either mean a sub conductance level or a pair of transitions have been missed If the Avoid sublevels box has been checked then by looking for turning points in the first derivative of the raw data the times of transitions can be deduced or their presence eliminated Any missed events are assumed to be of full amplitude or closed and are flagged as assumed amplitude Transitions having less than a certain amplitude are stripped out in a final pass At the fitting stage in the Event Details dialog amplitudes having the assumed amplitude flag set are held fixed then a second pass of fit is made freeing up assumed amplitudes above a certain duration The Advanced Parameters dialog can be accessed from the Event Details dialog by clicking the Parameters button Threshold crossing as a method for generating an idealised trace is needed if you have more than one Channel 1 Chan 1 channel in your patch or you want a simpler technique A Data start time Mirtime me and are not concerned with a very high time resolution Data endtime bMaxtme m of the result 0 ls Open above fi pmp The first three fields are exactly the same as for the OS a SCAN method The next two fields set the thresholds Ba 00 Fiam to use for detecting openings These are labelled Open Tonton Eon D pois above
46. each channel a horizontal line appears above or below the channel This horizontal line shows where the selected channel will be dropped Drag until you have a horizontal line below channel 1 and release the mouse button The Keyboard channel will now move to the bottom of the channel list Type Ctr1 Z or use the Edit menu Undo to remove your change You can move more than one channel at a time Signal moves all the channels that are selected when you start the drag operation For example hold down Ctr1 and click on the channel 3 number Keep ctr1 down and click and drag the channel 2 number When you release both channels will move The mouse pointer shows a tick when you are in a position where dropping will work The default channel order is with lowest numbered channels at the top of the display If you prefer the reverse order open the Edit menu Preferences and un check Standard Display shows lowest numbered channel at the top then use the View menu Standard Display command S 1 Click pn 2 Drag 3 Drop S 1 Move mouse 2 Shiftetelick 3 Drag A Shift Ctrl Memory views Getting started Click the 2 of channel 2 and drag it on top of Channel 1 and release 5 Keyboard 0 2 0 1 0 0 0 5 0 0 1 0 4 gt ADC 3 O pp a 0 1 Co LC Oe oS 0 5 Op 05 5 q 00 0 0 7 8 Nd 7 8 7 8 ms ms ms The channels now share the same space with the channel numbers stacked up next to the y axis The visi
47. field sets the maximum number of iterations to try before giving up It is possible to have the fit run automatically across several frames at once Choose the frames whose data you wish to fit here Signal for Windows version 3 11 10 Make an initial guess Show fit Coefficients Results The iterative fits need a starting point There are built in guessing functions that usually generate a starting point near enough to the solution that the fitting process can converge If you check this box these guessing functions are used each time you click the Do Fit button Otherwise each fit starts with the current values Check this box to display the current fit for the current channel If the From box is checked you can also choose the range over which to display the fitted data If this box is not checked the fit is displayed over the range that the data was fitted This page of the Fit Data dialog lets Fit data for example cfs Mal ES you set the starting values for iterative fits You can also use this page to hold Fit settings Coefficients Results some of the coefficients to fixed a Value Hold Lower limit Upper limit values and you can set the allowed EJE LS 000000 1900000 range of values for fitting 1 fo mooz0014878 FE o 1onoo00 2 fossi3e2zs2 D hooo fioooo00 If you know the value of one or more 3 fooussicesry Oo fo foo of the coefficients type the value in and check the Hold box next to it For gp 0 22e05
48. for other windows has no effect You can also change the colours used for syntax colouring and preview the effect Settings M Use syntax coloring Quotes Change Comments Change Default Text Change did Background Change The Show time as selector controls the time units used within Signal you can select Seconds Milliseconds or Microseconds The selected units will be used in the sampling configuration for all data files with a time based X axis including cursors and cursor windows and for all appropriate dialogs including the various process settings dialogs The main area of Signal not affected by this setting is the script language which will always see and use values in seconds though script programs can read the current settings and adjust their behaviour as required This setting does not affect any data stored by Signal just the way time is displayed to the user so you can switch settings without causing problems with data collected using another setting Some time values are saved by Signal as strings particularly the parameters for memory view and XY view processing online and active cursors and these may be misinterpreted after the time units are changed The frame start time shown in the status bar and in printouts may be set to be the time since sampling started in seconds the same value shown as hours minutes and seconds or the absolute time of day also shown as hours minutes
49. for resistance measurements must exist State zero always exists even if multiple states are not used Signal checks for incompatibilities between the requirements of clamping and other aspects of the sampling configuration at the start of sampling and will generate an error message 1f problems are detected When a sampling configuration with at least one Clamp controle x enabled clamping setup is used the clamping toolbar is created in addition to the standard Membrane HP mi 0 Signal sampling toolbars The clamping toolbar B 49 37 MOkme contains separate controls and information for each clamping setup that is enabled and can either be floating or docked to any edge of Signal The clamp toolbar contains three items controls for the holding potential a text display area and a button marked Membrane used to provide the membrane analysis dialog The holding potential or current for current clamp controls are used to adjust the baseline level of the stimulus DAC output The initial value of the holding potential is taken from the initial baseline level of the appropriate DAC in state zero Whenever the holding potential is changed the baseline level for the DAC is set to the new value in all states the holding potential is also set up in all states at the start of sampling Note that when editing the holding potential directly it is necessary to accept the new value by pressing tab or otherwise moving away from the edit
50. for the display of each output in the pulse configuration dialog See the Pulse outputs during sampling chapter for details of the digital outputs This option generates pulses and other outputs using a list of sequencer instructions that are executed inside the 1401 at a specified rate Each instruction carries out a simple function such as setting a DAC to a given value waiting for a specified time or looping Parameters General Port setup Outputs Automate DAC scaling and waveform pm A Full scale Sequence Step period ms i Zero value F Free run without restarts Waveform points i 000 Output rate Hz Jumps controlled by P Hz j 3 Keys and control parel WavefomDACs D PT Sequence file Browse C Signals MyFirst ple The sequencer includes 64 variables that can hold values and a table of data that can be quickly read and updated by scripts running in Signal Cancel Run now Help The details of the sequencer language and instructions are given in a separate chapter of this manual Sequencer outputs during sampling When Sequencer outputs are selected a number of sequencer controls are shown Signal for Windows version 3 Step period ms Free run without restarts Jumps controlled by Sequence file DAC scaling and waveform This item sets the clock interval for sequencer instruction execution and thus the rate at which sequencer instructions are execu
51. in fields that are either numeric or text A text field is a sequence of characters that may include spaces Text fields may hold numbers but numeric fields cannot hold text Signal outputs a separator between each field It can be set to Tab space or comma with the Separator selector The examples below use Tab as a separator You can set the number of decimal places to use for both data and time output This field sets the minimum width of each output field in characters You can mark the start and end of a text field with special characters usually so that a program reading the field can include spaces and punctuation within a field without confusion Use this field to set a one or two character delimiter or leave 1t blank The examples use as a delimiter If you check this box Signal outputs a header of the file name followed by the frame number before the text output This header is normally disabled as it may interfere with reading exported Signal data into spreadsheets No1Ssy cats Frame 4 All waveform data is output together with the first column holding the time values if enabled and then one column per waveform channel Waveform headings are a single line holding a title for each text column Following this are multiple lines with the waveform data times and values 9 2 Copy data selection Copy As Text XY view Find Find Again Edit menu W S W ADC 0 W ADC 1 W ADC 2 W ADC 3 W 0223675 1 0
52. in the sampling control panel enable and disable sweep triggers With sweep triggers enabled a sampling sweep will not occur until a trigger has been detected the sampling configuration determines what a trigger is With sweep triggers disabled a sampling sweep starts immediately For Outputs frame sweeps the sweep trigger starts the outputs rather than the sampling sweep For Basic or Outputs frame sweeps the sweep trigger is a TTL pulse that is applied to the 1401 and 1401p us event O inputs or to the micro1401 Micro1401 mk II or Power1401 Trigger input For Peri triggered sweeps the trigger can be any of a number of signals see page 3 7 There is a small 10 microseconds delay between the time of the sweep trigger and start of sampling This delay is affected by the outputs synchronisation controls in the Outputs configuration section When using Basic sweep mode it is possible to start sampling at exactly the time of the sweep trigger by providing the trigger pulse to both the 1401 EO and E4 inputs This mechanism is only available when the synchronised sampling option in the Outputs configuration is disabled For the micro1401 Micro1401 mk II and Powerl401 the trigger input is automatically routed to both EO and E4 3 6 Rising edge trigger Write sweep to disk Pause at sweep end Peri trigger configuration Trigger type Sampling data internally if appropriate thus guaranteeing a precise start of sampling relativ
53. increases the amplitude from zero to full scale we assume that the waveform is playing SZ 0 0 0 start with zero size MOVI Wey LOO proceed in 1 increments loop SZINC 020 0 a 1 increase DELAY ms 100 2 show some of the waveform at this size DBNZ WL GA LOO loop 100 times This sets the angle increment in degrees per step which sets the cosine frequency If the nominated DAC was ramping this cancels the ramp You can stop the cosine output with arate of 0 Any non zero value starts the cosine output RATE n expr Vn Vn off OptLB DAC n n DAC number in the range 0 7 available DACs depend on the 1401 type expr The angle increment per step in the range 0 000 up to 180 degrees The Hz function calculates the increment required for a frequency Vn For a variable the value 11930465 is an increment of 1 degree The VHz function can be used to set a variable value equivalent to an angle in degrees OptLB If this optional label is present it sets the next instruction to run This example starts cosine output at 10 Hz runs for 10 seconds and then stops it This is then repeated using a variable to produce the same effect C RATE Oy Hz ALO Store OULOUG ac 1U Hz DELAY om GOP Boel delay for 10 seconds gt Sine wave x gt RATE 0 0 Stop OULOUE HALT gt Stopped Y MOVI Vi VEZALO eset Vi equivalent or 10 Hz RATE UV start at 10 Hz DELAY S 10 1 X delay then goto exit gt Sine wave This instruction p
54. is the button previously labelled as Start or Pause at sweep end Continue Click it to resume sampling as if the Stop button had not been pressed If sampling stopped due to the frame count reaching a limit then pressing More resets the frame count and sampling runs until the count again reaches the limit otherwise the limit is disabled and sampling continues indefinitely Finish This is the button previously labelled as Stop If you click on this button the data capture is terminated and the sampling control panel disappears Click on the Finish button to end sampling If the sampling configuration has automatic saving to disk enabled the new data file will be saved at this point using either the automatically generated filename or a name entered by the user as appropriate You will also be prompted for a comment for the new file 1f this feature 1s enabled Once the sampling has actually shut down the sampling control panel is removed In the new data view frame zero of the data document disappears and the view changes to show frame 1 if it was previously showing frame zero If there are no saved frames the data document and view are destroyed giving the same effect as pressing Abort 3 19 Signal for Windows version 3 Saving new data Saving configurations Sequence of operations to set and save the configuration A sampling document that has stopped sampling is essentially the same as a document loaded from disk
55. is used to build sequencer data that is loaded into the 1401 and executed in exactly the same manner as Sequencer outputs The timing requirements and limits for these two forms of output are therefore identical The following table summarises the use of the various output methods in the different sweep modes More details can be found in the Pulse outputs during sampling and Sequencer outputs during sampling chapters Pulses Sweep and outputs triggered or free run together and are of same length Pulse output triggered when sweep trigger recognised same length as remaining sweep Pulse output triggered or free runs Pulse output starts sampling sweep at the required point Outputs triggered by internal timer starts sweep at specified time within outputs Single output set only same length as sampling sweep Sweep and outputs triggered or free run together Single output set only same length as sampling sweep Sweep and outputs triggered together by internal timer Sequence Sweep and sequencer triggered or free run together Sequencer triggered when sweep trigger recognised Sequencer triggered or free runs responsible for triggering the sampling sweep at the required point N A simulate with Outputs frame mode with free running sequence Sweep and sequencer triggered together N A simulate using Basic mode with free running script Free run sequence Sequencer starts immediately runs t
56. loss of power in the result A discussion of the relative merits of different window functions is beyond the scope of this manual We use a raised cosine window and compensate for the loss of power it causes If you sample a pure sine wave of amplitude 1 volt and take the power spectrum you will not get all the power in a single bin You will find data spread over three bins and the sum of the three bins will be 0 5 volts The factor of 2 in the power is because we give the result as RMS root mean square power This is illustrated by the example below where we have sampled a sine wave with amplitude 3 06 volts peak to peak amplitude 6 12 We have formed the power spectrum of the signal using a 256 point transform and zoomed in around the bins where the result lies If the sampled waveform was a perfect sine wave we would predict a RMS power of 4 6818 volts from this waveform 3 062 2 The cursor analysis of the power shows a total power of 4 7135 volts This is about 0 7 above the predicted result for a perfect waveform Leak Subtraction Signal for Windows version 3 Red Signal data3 cfs File Edit View Analysis Cursor Sample Script Window Help Waveform y Cursor regions for Power3 data3 Mel eA TEEPEE EEE The predicted result is slightly low because the waveform samples used for the cursor measurements are unlikely to lie at the exact peak and trough of any particular cycle Using the script language Minmax
57. menu or toolbar The command opens the New File dialog in pP which you select the type of document to x create You can create five types of document Signal data documents XY Test Document ES documents sequencer documents script Se ee Script Document documents and text documents Select the o Hep type of document click OK and Signal will open a new window holding an empty document of the specified type A sampling document window opens plus additional windows as set by the sampling configuration see the Sampling data chapter for details Sampling documents are not initially stored in memory like most new files but are kept on disk Until they are saved after sampling they are temporary files in the directory set in the Signal preferences The file name extension is cfs and the files hold CFS CED Filing System format files Text documents can be used to take notes build reports and to cut and paste text between other windows and applications The Log view is a specialised type of text document which is always present The file name extension is txt A script editor window opens in which a new script can be written run and debugged A script document is a specialised form of text document The file name extension is sgs A new window opens in which you can type edit and compile an output sequence see the Sequencer output during sampling chapter for details The file name extension is pile XY windows a
58. od f gs for state 1 on the front of the BiStim master unit you Ml Manual Power 1 20 can set the device into independent trigger Interval ms 100 Power B 30 mode referred to as IBT mode in the MagStim documentation When this mode _Copy to al _Copy above Tes is in use two separate triggers are used Cancel Help presumably signals from two 1401 digital outputs one trigger firing the main pulse and the other firing the second Power B pulse Because a BiStim behaves rather differently when in IBT mode it is important that you set this checkbox correctly to match the BiStim setup you will be using otherwise errors will be generated when you try to sample When this checkbox is clear the interval between the two pulses can be set to any value between 0 and 999 milliseconds the actual timing in the BiStim will be set with a resolution of 1 millisecond With this checkbox set the maximum interval is reduced to 99 9 milliseconds but the interval timing resolution is improved to 0 1 milliseconds This control is disabled if independent trigger mode 1s selected The BiStim settings available for the individual states are Set this checkbox for this state to be controlled by the manual controls and not by Signal clear it to give Signal control 20 3 Signal for Windows version 3 Power Interval Power B Rapid device Ignore coil interlock switch Use Rapid single pulse mode Manual
59. of 50 dB so the mean value is 70 50 70 3 or 63 33 dB We have two transition regions both the same width In the general case of transition regions of different sizes use the smallest transition region in the formula Plugging these values into the formula predicts 113 coefficients however only 111 are needed to achieve 70 dB var prall l2 3 bands for band pass var coef 111 111 coefficients needed band start band end function weight pra 0 110 0 00 pem 1 00 057 prm 2 0 2 0 0 prm 3 10 L0 05 prem 0 11 0 08 pea 1 11L lt 0 167 prm 2 11 3 L 07 prm 3 11 1 0 prm O 2 2 0 197 pea 1 12 50 502 prem 2 11 12 0 0p prm 31 12 L0 0 FIRMake 1 prm coef 0 a _ __ it tT tT O O O O O O CC a O O O O O O O E E A as HA aog ee REE EEE EEE EEE EEE EEE EEE EEE EEE EEE EEE ee EE EE EEE EERE EEE EERE BREE EERE EERE EERE REFER EEEEEEEEE A EERE EERE EERE EEE EERE EEE EERE EEE EH REE EEE EEE EE EEE EEE EEE EEE EEE EE EEE EEE EEE AR MAAAAAAAARARARARAAAAAARAAAAARAARA A II 20 APE TTT SHAH EEE TEEPE HII RITE a 1 i 0 ee ee ee I ee i ee ee O O 0 0 1 0 2 0 3 0 4 0 5 Fraction of sample rate 17 11 Signal for Windows version 3 Differentiators A differentiator filter has a gain that increases linearly with frequency over the frequency band for which it is defined There is also a phase change of 90 1 2 between the input and the output Ideal differentiator A EEE EEE
60. outputs duration and the trigger time within the outputs are not copied otherwise the target states are set to the length and trigger time of the current state Similarly checking the Fixed interval checkbox allows the fixed interval and variation values to be copied For the outputs selected all existing pulses in the destination states are deleted before the new data is copied in 5 4 Editing the pulse parameters Initial level Simple square pulse Varying amplitude pulse Pulse outputs during sampling In addition to changing the pulse start times by dragging and dropping all of the pulse parameters can be edited by using the values area of the dialog This area shows all the pulse parameters it is different for each type of pulse This specifies the state that the outputs are set to at the beginning of the pulse outputs frame this item is always present and cannot be deleted or teve Mots o moved The initial level has a single parameter PC o ie eee Level that sets the level that the DAC is initially set to The level entered is scaled before use as defined by the DAC settings in the output page Initial levels ld The three other parameters at the bottom of the values area Step change Repeats and Steps can be used to define a built in variation in initial level Built in variations are described under Pulses with variations below For digital outputs only the initial level parameter is needed S
61. pattern This determines the new output state The pattern is 8 characters long one for each bit with bit 7 at the left and bit O at the right The characters can be 0 1 i or standing for clear set invert or leave alone You may omit the square brackets however the Format command will insert them DIGLOW 001i Clear bits 3 and 2 set 1 invert 0 DEGROW Lera i invert 0 again to produce a pulse DIGLOW V10 use variable V10 to set the pattern Vn With a variable the new output is old output BAND Vn 7 0 BXOR Vn 15 8 The variable equivalent of 001i is 241 256 3 and of TE i is 255 256 1 If you use a table value set the same value in the table that you would use for a variable You can use the varValue script in the Scripts folder to calculate variable or table values OptLB If this optional label is present it sets the next instruction to run This example produces ten 1 millisecond pulses 100 milliseconds apart These instructions test digital input bits 7 0 against a pattern see the Sampling data chapter for connections These are the same inputs that will be used for digital markers in the future DIBEQ branches on a match DIBNE branches on a non match Both instructions copy digital input bits 7 0 to v56 VDigIn for use by DISBEQ and DISBNE DIBNE pattern Vn Vn off LB DIBEQ pattern Vn Vn off LB pattern This is 8 characters one for each input bit The character
62. pulse outputs to be used for membrane analysis Clamp setups define the data file channels that hold the current and voltage data so that analysis can find the correct data and checks of the units for these channels so that current and voltage values a e Help can be scaled correctly to amps and volts In addition the setup defines the DAC used to control the membrane current or voltage so that the holding potential and pulses can be manipulated Parameters C 4 fork Signala winD ebugiLAST 5GC General Fort setup Clamp Outputs States Automate Clamp setup 1 Type Whole cell voltage clamp Stimulus channel n Control DAC nm o Response channel n Clamp setup 2 Type Not clamping ki E Eontro DAE fo Stimulus channel Response channel State for resistance measurements 4 1 Each clamp setup specifies the experiment type the data file channels holding the stimulus and response signals in voltage clamp the stimulus is the voltage and the response is current in current clamp experiments this is reversed and the DAC used to control the stimulus The Experiment type item can be set to Not clamping to disable use of that clamp setup or to Whole cell or Single channel each either voltage clamp or current clamp If both setups are set to Not clamping then all of the online clamping support is disabled so you can carry out non clamping experiments The Stimulus channel
63. script recording is in progress the rightmost indicator in the Signal status bar will display the text REC as a reminder When you turn recording off a new script window opens that holds the saved script commands If you then compile and run this script the actions that you performed while recording was on will be repeated You can use this mechanism to record a sequence of actions that you wish to rerun at some later date to find out what script commands correspond to a given menu command or user action or to record a sequence of actions that can be copied into another script or edited to produce a complete scripted application 14 1 14 2 Debug bar Script Bar Script List Signal for Windows version 3 You can show and hide the debug bar from this menu when the current view 1s a script You can also show and hide the debug bar by clicking the right mouse button on any Signal toolbar or on the Signal background You can show and hide the Script Bar and manage the Script Bar contents from the Script menu You can also show and hide the Script Bar by clicking the right mouse button on any Signal toolbar or on the Signal background The Script Bar is a dockable toolbar with up to 20 user defined buttons Each button is linked to a Signal script file When you click a button the associated script is loaded and run There is also a user defined comment associated with each button which appears as a tool tip when the mous
64. tO F Logarthmic I Auto adjust units Close cursor positions The Show All button expands the time axis to display all the data The Draw button updates the display to show the time range set by the Left Right and Width fields In addition to typing times or selecting a time from the drop down list you can type in expressions using the maths symbols add subtract multiply and divide You can also use round brackets For example to display from 1 second before cursor 1 to one second past cursor 1 set Left to Cursor 1 1 and Right to Cursor 1 1 The Draw button is disabled if you type an invalid expression or if the Right value is less than or equal to the Left value or if the new range is the same as the current range The Large tick spacing and Tick subdivisions fields let you customise the axis Values that would produce an illegible axis are ignored Changes to these fields cause the axis to change immediately you do not need to click Draw The Auto adjust units option will cause the units displayed on the axis to switch to multiples of powers of 10 in order to keep the figures sensible when zoomed well in or well out This option affects only the axis the units used by the cursors etc will still be the same Checking the Logarithmic option will switch the axis from linear to logarithmic modifying the displayed range only if it included negative values In logarithmic mode another check box Show powers will appear Thi
65. text 9 3 Cut text 9 1 Delete selection 9 1 File comment 9 4 Find text 9 3 Frame comment 9 4 Paste 9 1 Preferences 9 4 Select All 9 4 Show event details 9 9 Undo 9 1 Edit text Editor settings 9 4 Email support 8 5 Enlarge view 10 2 Error bars 11 1 Errors Output sequencer compiler 6 28 Evaluate 14 1 Event 1 sampling trigger 3 2 3 18 Exchange data with another computer 8 5 Exit 8 6 Export data As bitmap file 8 4 As CFS file 8 4 As text file 8 4 As Windows Metafile 8 4 To clipboard 9 3 External convert sampling 3 5 External digital states 7 5 FFT Fast Fourier Transform 11 3 File comment 9 4 File comment at sampling end 9 7 File format converters 8 3 File icons 1 3 File menu 8 1 Close 8 3 Close All 8 3 Data update mode 8 5 Exit 8 6 Export As 8 4 Import 8 3 Load configuration 8 5 New File 8 1 Open 8 2 Page Setup 8 5 Print 8 6 Print preview 8 6 Print screen 8 6 Print selection 8 6 Print visible 8 6 Revert To Saved 8 4 Save and Save As 8 3 Save configuration 8 5 Send Mail 8 5 File name extensions 8 1 File name generation 3 15 File size limit 3 15 File view 2 1 2 9 Filtbank cfb filter bank file 17 3 Filter bank 17 3 Finding a pulse 5 4 Finish sampling 3 18 3 19 FIR filter 17 1 Attenuation and ripple 17 7 Coefficients 17 5 Differentiator example 17 12 Frequencies 17 5 Frequency bands 17 6 Maximum useful attenuation 1
66. the Settings dialog the new XY view is created and the standard Process dialogs can be used to control the analysis An XY view holds lists Sarees of XY points one list Plot Channel Channel1 y ae aon per channel The trend om ECE mee plot analysis can create data points using a wide variety of types of measurement for both the X and Y values Up Time mz MasTime r Time mz Cursor r to 32 channels of XY Width 0 000000 me Width 0 000000 me data can be created When the command is Points in plot f0 New Cancel used the Trend plot settings dialog is provided to allow you to enter the trend plot settings Y measurements Type Value at point Source channel 1 chan 0 1000 Hz s measurements Type Value at point Source channel 1 chan 0 1000 Hz The dialog consists of four regions At the top is a control to select a channel and buttons to add and delete channels you cannot delete the last channel In the middle are two similar regions setting the parameters for the X and Y measurements and at the bottom are the standard buttons plus a control for the points per channel You can set the channel title in the Plot Channel box and create additional channels with the Add Channel button The Delete Channel button deletes the current channel you cannot delete the last channel The two measurements sections are the same Both hold a selector for the channel to take measu
67. the sinusoid To have the output rising through 0 set the phase to 90 Each time Theta passes through zero a new cycle flag sets The RAMP RATEW RINCW WAITC and CLRC instructions clear the flag This instruction sets the waveform amplitude If a wave is playing the amplitude changes at the next sequencer step The amplitude is set to 1 0 when sampling starts SZ n expr Vna Vn off OptLB DAC n n DAC number in the range 0 7 available DACs depend on the 1401 type expr The cosine amplitude in the range 0 to 1 A cosine with amplitude 1 0 uses the full DAC range Vn Variable values 0 to 32768 correspond to amplitudes of 0 0 to 1 0 values outside the range 0 to 32768 cause undefined results OptLB If this optional label is present it sets the next instruction to run 6 14 SZINC RATE RATEW Sequencer outputs during sampling This instruction changes the waveform amplitude The change is added to the current amplitude If the result exceeds 1 0 it is set to 1 0 If it is less than 0 the result 1s 0 SZINC n expr Vn Vn off OptLB DAC n n DAC number in the range 0 7 available DACs depend on the 1401 type expr The change in the waveform scale in the range 1 to 1 Vn A variable value of 32768 is a scale change of 1 0 16384 is 0 5 and so on OptLB If this optional label is present it sets the next instruction to run You can gradually increase or decrease the wave amplitude For example the following
68. there are limits to the number of coefficients it is worth having apart from increasing the time it takes to calculate the filter and filter the data Although the process used to calculate coefficients uses double precision floating point numbers there are rounding errors and the larger the number of coefficients the larger the numerical noise in the result Because the waveform channels are stored in 16 bit integers there 1s no point designing filters that attenuate any more than 96 dB as this is a factor of 32768 2 Attenuations greater than this would reduce any input to less than 1 bit If you are targeting data stored in real numbers this restriction may not apply FIRMake filter types Multiband filters Digital filtering It is important that you leave gaps between your bands The smaller the gap the larger the ripple in the bands 0 1 0 3 0 4 0 1 0 3 0 4 This is illustrated by these two graphs They show the linear frequency response of two low pass filters both designed with 18 coefficients we have used so few coefficients so the ripple is obvious Both have a pass band of 0 to 0 2 but the first has a gap between the pass band and the stop band of 0 1 and the second has a gap of 0 05 We have also given equal weighting to both the pass and the stop bands so you can see that the ripple around the desired value is the
69. three items holding the time for the DAC to stay at 0 the DAC amplitude and the time to stay at the amplitude Some example table data is given but the data could also be set with the SampleSeqTable script command 6 23 Signal for Windows version 3 Access to data capture CHAN POINTS TABSZ TZ 4 sets of 3 items TABDAT ms 50 2 VDAC32 1 ms 50 3 TABDAT ms 100 2 VDAC32 1 3 ms 70 3 TABDAT ms 200 Leo ms 90 3 TABDAT ms 400 1 9 ms 12 0 3 G MOVI VIO use Vl as the table pointer LOOP DAC O30 strt with the DAC low DELAY V1 walt for first period gt Low DAC 0 V1 1 get the DAC value DELAY Vier wait for second period gt High TABINC V1 3 LOOP DAC 0 0 strdy up the dac Most activities in the sequencer are independent of the sampling process However there are times when you need to know the value of a channel to decide what to do next The CHAN command gives you the latest waveform value or number of events on a channel The TICKS command tells you the current time in terms of the sampling clock ticks The sequencer can also send information in the other direction This instruction gives the output sequencer access to sampled data on a waveform channel You can also use this command to get the most recent value written to the DAC outputs The variable value is set to O if the channel is not being sampled This instruction is not available for the 1401plus CHAN Vn chn Vn ChanData chn e
70. use the standard digits 0 to 9 but also use a to f for decimal 10 to 15 Thus 00 to 09 hexadecimal is equivalent to 0 to 9 decimal 0a to Of is equivalent to 10 to 15 decimal 10 to 1f hexadecimal is 16 to 31 decimal 20 to 2f 1s 32 to 47 decimal and so on The printable characters as far as Signal is concerned span the hexadecimal range 20 to 7e 32 to 126 decimal and are as shown in the table To find the hexadecimal code of a printable character add the number above the character to the number to the left of the character For example the code for A is 41 To convert a code to a character look up the first digit in the left column and the second in the top row For example 3 codes to the intersection of the row for 30 and the column for f 3 3 Signal for Windows version 3 Sampling configuration General configuration Sweep mode Before you start to sample data eee hans with Signal you must set the sampling configuration This is done through the Sample menu eee ae T i r i r BVO Sampling Configuration dialog Sweep mode Basic El Ke General Fort setup Outputs Automate which is also available by using a f Mutiple frame states aa toolbar button The sampling B UOL esise sari see ae configuration dialog is a tabbed Sample rate Hz 1000 E gt dialog it contains a number of Frame length ms 1000 T Rising edge trigger tabs for selecting separate sections
71. variables V1 to v64 that can be read and set by on line scripts e It supports a user defined table of values for fast information transfer from a script e Itcan read the latest value from a waveform channel and using this information provide real time fractions of a millisecond responses to input data changes e It can set or get the sweep state code get the start time of the sweep wait until a specified time within the sweep or trigger a sweep e Itcan control and monitor the arbitrary waveform output You write sequences with a text sequence editor each text line generates one instruction While sampling is in progress you can A a interact with sequencer execution using Step 0001 G to start the sequencer control panel The control panel displays the next 5 Fesetouputandstop Reset output and stop i A A G Pulse outputs sequence step and any display string associated with it You can use display strings to prompt the user for an action or to tell the user what the sequence is doing If there are any keyboard controlled entry Step 0001 G to start 3 4 _ _ _ _ points in the sequence these are also shown in the control panel You can show and hide the control panel with the Sequencer Controls option in the Sample menu or by right clicking on any toolbar to activate the context menu You can also dock 1t on any edge of the Signal application window When un
72. waveform channel which is the last port in the ADC ports field as this is the trigger channel For Analogue and Analogue trigger modes the cursors show the trigger level and the trigger level minus or plus the hysteresis this latter level is shown as the Arm level For Analogue trigger mode the two separate trigger levels are shown During sampling you can adjust the trigger levels used by moving the cursors they cannot be moved to a different channel as the trigger channel is fixed If you use the keyboard command Ctr1 PgUp Signal will switch to displaying the last frame that was saved to disk You can tag data frames as they are sampled by pressing Ctr1 T Most standard display manipulation mechanisms work in exactly the same manner online but frame overdraw mode works differently on frame zero it never erases old data but just redraws it in grey and the frame display list is ignored This provides a very nice storage oscilloscope style display but if any part of the view 1s redrawn the previous traces are lost You can use the Edit menu Clear command to erase all the previous traces Sampling can be stopped by clicking on the Stop button or by the sampling limits being reached When sampling is More Sweep tigger stopped Signal is in between sampling and finishing sampling The sampling control panel is still present but the decept Abort restar buttons have changed Sweep end write to disk at sweep end More This
73. x axis range zoom in The left hand edge of the display remains fixed You can zoom in until the ratio between the total length of the frame and the width of one screen pixel reaches about 2 billion In practice this means you can zoom in as far as you like This button corresponds to the View menu Reduce View command Ctrl Left Click this button to double the displayed x axis range zoom out The left hand edge of the window does not change unless the start plus the new width exceeds the length of the frame in which case the left edge moves back If the new width would exceed the total length of the frame the entire frame is displayed This button corresponds to the View menu Enlarge View command Ctr1 Right Click this button to add a cursor to the display Up to 10 cursors can be present in a window A cursor is a vertical dashed line used to mark positions You can remove cursors by using the Cursor menu Delete command You can add a cursor in two ways 1 To add a new cursor in the centre of the window click on the button 2 Use the Cursor menu New cursor command or its shortcut Ctr1 2 3 Signal for Windows version 3 Controlling the display gje AG S Click the cursor button so that at least one cursor is visible Drag the cursor and observe how the mouse pointer changes Use the Cursor menu Label mode command There are four labelling styles for the cursor no label position position and cursor numbe
74. you should null it out by changing the offset in the Channel parameters dialog Signal normally expects the first channel of signal conditioning to be connected to 1401 ADC port 0 the second to port 1 and so on Connect the conditioner output BNC labelled Amp Out on the 1902 and OUT on the CyberAmp to the relevant 1401 input This arrangement can be adjusted to start with another port but in all cases the conditioner channel number must match the ADC port to which it is connected Most signal conditioners connect to the computer with a serial line You will have received a suitable cable with the unit Basic communication and connection information is stored in the file CEDCOND INT in the system folder of your computer This file holds General Port COM1 The Port value sets the communications port to use We would recommend you use the Edit menu Preferences to change the port If this file is missing COM1 is used Preferences can also set a diagnostic option enabled in the file by Dump 1 If this entry is included Signal writes a log file called CEDCOND LOG to the current directory when it initialises the conditioner As mentioned above signal conditioners normally start at channel zero and go up to the last channel The conditioners are searched for assuming this arrangement the search starts at channel 0 and continues up until a missing channel is encountered This search process can be adjusted by two more lines that ca
75. 000 y asoa y Number of bins 200 Data end time Bin size New Cancel keep the amplitude range constant Finally if the Number of bins is changed then the Bin size changes again keeping the amplitude range constant Power Spectrum Windowing of data Power spectrum of a sine wave Analysis menu This analysis creates a memory view that holds the power spectrum of a section or sections of data If multiple sections are processed the result is an Channel Al Channels El averaged power spectrum The result of the analysis FFT size 256 hd 1s scaled to RMS power so 1t can be converted to energy by multiplying by the time over which the transform was done There are three fields to set in New Cancel the dialog the waveform channels to analyse the number of points in the Fast Fourier Transform FFT used to convert the waveform data into a power spectrum and the start offset within the frame for the data The channels and offset fields behave the same as their equivalents in the waveform average dialog Start offset 0 0 ls The FFT is a mathematical device that transforms data between a waveform and an equivalent representation as a set of cosine waves each with an amplitude and relative phase angle The version of the FFT that we use limits the size of the blocks to be transformed to a power of 2 points in the range 16 to 4096 You set the FFT block size from a pop up menu The way the maths works out the result
76. 000100 last cycle number w4 WAITE 1 w4 wait for end gt 1 6 degree cycle RATE Ly Uet stop waveform 6 17 Signal for Windows version 3 RINC RINCW CLRC These instructions behave like RATE and RATEW except that they change the output rate angle increment per step by their argument rather than set it RINCW clears the new cycle flag RINC n expriVn Va off OptLB DAC n RINCW n expr Val Vnt off OptLB DAC n n DAC number in the range 0 7 available DACs depend on the 1401 type expr The change in the angle increment per step You can use the built in Hz function to express the change as a frequency Vn For a variable the value 11930465 is a change of 1 degree You can use the VarValue script in the Scripts folder to calculate variable values OptLB If this optional label is present it sets the next instruction to run This example starts cosine output at 10 Hz and lets you adjust it from the keyboard RATE des AED FStart G tpu t at 17042 WES JUMP wt HALT stops all output gt P 1Hz M 1Hz P RINC 1 Hz 1 wt 1 Hz faster M RINC 1 Hz 1 wt 1 Hz slower These instructions can be used to produce waveforms that change gradually in frequency The following code generates a linear speed increase every two steps on DAC 1 SZ Ll0 make sure full size ANGLE LOS O make sure we start at phase 0 RATE ire ee 1 degree per step to start with MOVI VE 900 PL IVU SEEDS O Enek Loop CRINC 0 01 ive d M
77. 1 n DAC number in the range 0 7 available DACs depend on the 1401 type expr The phase angle to set in the range 360 up to 360 Vn For a variable the value 11930465 is a phase of 1 degree to be precise 4294967296 360 is a phase of 1 degree You can use the VAngle function to convert degrees into a suitable value for a variable OptLB If this optional label is present it sets the next instruction to run This example sets the phase angle to 90 degrees directly and by using a variable There is no need to use the VAngle function we could have set V1 to 1073741824 However VAngle 90 is much easier to understand ANGLE T oO set the DAC 1 cosine angle directly MOVI V1 VAngle 90 ANGLE Tevi set using a variable This changes the relative phase of the cosine output for the next cosine output A common use is to change the output from a cosine maximum value at phase zero to sine rising through zero at phase zero PHASE n expr Vn Vn off OptLB DAC n n DAC number in the range 0 7 available DACs depend on the 1401 type expr The relative phase angle to set in the range 360 up to 360 The relative phase is set to 0 when sampling starts Set 90 for sinusoidal output Vn For a variable the value 11930465 is a phase of 1 degree to be precise 4294967296 360 is a phase of 1 degree You can use the VAngle function to convert degrees into a suitable value for a variable OptLB If this optional label is p
78. 1074 A 6 1670 0 46875 Oe 23430 0 23700 L37451 4 601182 0 446078 0 27832 Ow 33300 1 69922 0 10799 0 43945 0 21729 Each Marker channel is output separately The output starts with the channel headings if these are enabled The data follows in two columns a time followed by a character gn Keyboard 20 20608 a 21 24544 uy Fitted data is output with two lines of headings if enabled followed by the data Single exponential fit on channel 1 Amplitude 1 Time constant 1 Offset ee O61 U0 S 75 208 Once you have defined the output format you have to select the data to copy This is done using the eme rra Y same dialog that is used to select the data to be art time 2 Mintimel exported to a CFS or text file You can specify the channels to use the time range for the data within E time each frame and the frames When you are satisfied Frames Tagged frames with the selection click on Cancel to quit or Export to start the copy process cares This menu item is available in XY views It copies the XY points for all visible channels to the clipboard The first line of output for a channel holds Channel cc nn where cc is the channel number and nn 1s the number of data points The data points are output one per line as the X value followed by the Y value separated by a tab character These commands move the text cursor E to the next occurrence of a string in a o o text base
79. 1s carried out in such a way that the buffer holds the average of frames accumulated If the buffer contains data before averaging starts this will be included as the first frame of the average There is an alternative form of this command available only through the Ctr1 Enter shortcut and the Multiple frames dialog which removes the current frame data from an average in the buffer this will not work correctly if the frame was not accumulated into the buffer average in the first place If you mix the buffer averaging commands with the normal addition and subtraction operations you will find that normal addition and subtraction on the buffer resets the average by setting the count of sweeps so far to one The actual addition and subtraction act as you would expect This command keyboard shortcut Ctrl M e provides a dialog that can be used to carry out numerous operations on multiple frames in the Peerten Scale selected channels al document The dialog contains a selector for Scale factor to apply 1 25 the operation to be carried out a field to enter any operation data required this is hidden if j F Tagged frames the operation does not need it plus a standard Guz Tagg set of controls to specify frames in the data Frame subset Frame state xxe document to be used The dialog can be used Selected frame state po repeatedly by pressing the Apply button it doesn t disappear until Close is clicked Close 11 18
80. 2 3 Display all horizontal 12 4 Display Y Values 12 4 Move To 12 3 Move To Level 12 4 New cursor 12 1 New horizontal cursor 12 3 Renumber cursors 12 3 Renumber horizontal cursors 12 4 Cursors Active 12 1 Add horizontal 12 3 Adding 2 3 12 1 Button for new 2 3 Delete cursor 12 3 Delete horizontal 12 3 Display all 12 3 Display all horizontal 12 4 Horizontal label style 12 4 Label style 12 3 Mode 12 1 Move window to centre a cursor 12 3 Offset channel to centre horizontal cursor 12 4 Regions 12 5 Renumber 12 3 Renumber horizontal 12 4 Value at 12 4 Values between 2 8 12 5 Curve fitting 11 8 Testing the fit 11 12 Curve Fitting 11 8 Cut text 9 1 CWAIT output instruction 6 17 CWCLR output instruction 6 18 CyberAmp Interactive support 18 1 _P DAC connections for Power1401 6 12 DAC output during sampling 6 1 DAC output instruction 6 12 DAC output range 3 14 DAC outputs 3 14 DAC outputs for state 7 6 DAC scaling and units 3 10 DANGLE output instruction 6 16 Data Exporting 8 4 Incoming sampled data 3 16 Saving 8 3 Data channels 2 1 Data export CFS data format 9 6 Data update mode 8 5 Data view Copy as Text 9 2 dB scale 17 6 DBNZ output instruction 6 19 DC measurement area for subtract 11 19 Decibel scale 17 6 DEFAULT S2C configuration 10 8 DEFAULT SGC default configuration file 8 5 DELAY output instruction 6 19 Delete channel 11 17 Delete fra
81. 5 1 Close document 8 3 CLRC output instruction 6 18 Coefficients of filters 17 5 COFF output instruction 6 17 Colour Enabling 10 7 Index 1 Signal for Windows version 3 Colour dialog 10 7 Channel colours 10 7 Comment Output sequencer 6 7 Comment file at sampling end 9 7 Compile output sequence 6 2 Conditional averaging state 11 5 Conditioner CEDCOND INI settings file 18 4 CEDCOND LOG file 9 8 Sample menu 13 1 Serial port 9 8 Configuration files 8 2 Contents 3 20 Load and run from Sample Bar 13 l Load and save 8 5 Connections Power1401 DACs 2 and 3 6 12 Signal conditioner serial 9 8 18 4 Waveform output 6 12 Continue sampling 3 18 3 19 Convert foreign file format 8 3 Copy Cursor values 12 4 12 6 Data view as text 9 2 File and Memory views as pictures 9 1 Text 9 1 XY view as text 9 3 Copy data As binary numbers 9 1 As bitmap file 9 1 As picture 9 1 As text file 9 1 Cosine wave output 6 14 Count of markers 2 8 CPHASE output instruction 6 16 CRATE output instruction 6 15 CRATEW output instruction 6 15 Creating a new document 3 16 CRINC output instruction 6 18 CRINCW output instruction 6 18 CSZ output instruction 6 14 CSZINC output instruction 6 15 Cubic Spline draw mode for waveforms 2 6 Cursor labelling styles 2 4 12 3 Cursor menu 12 1 Cursor mode 12 1 Cursor regions 12 5 Delete 12 3 Delete horizontal 12 3 Display all 1
82. 7 Signal for Windows version 3 Variables VAR directive Script access to variables VDAC16 expr Converts expr user DAC units so that the full DAC range spans the full range of a 16 bit integer Use with variables for DAC and ADDAC VDAC32 expr Converts expr user DAC units into a 32 bit integer value such that the full DAC range spans the 32 bit integer range Use this to load variables for use with the DAC and ADDAC instructions Used with care the built in functions allow you to write sequences that operate in the same way regardless of the sequencer step time or DAC scaling values You can use the 64 variables V1 to v64 in place of fixed values in many instructions In the sequencer command descriptions vn indicates the use of a variable Where a variable is an alternative to a fixed value expression we use expr vn Variables hold 32 bit integer numbers that you can set and read with the SampleSeqVar script command Some variables have specific uses Variables v57 through v64 hold the last value written by the sequencer to DACs 0 through 7 and v56 holds the last bit pattern read from the digital inputs with the DIBxx or DIGIN instructions You can assign each variable a name and an initial value with the VAR directive Names must be assigned before they are used usually at the start of the sequence The syntax 1s VAR Vn name expr comment VAR does not generate any instructions It makes the symbol name equivalent
83. 7 Copy from buffer 11 18 Copy to buffer 11 18 Delete channel 11 17 Delete frame 11 17 Exchange buffer 11 18 Fit data 11 8 Keyboard alternatives 11 20 Modify channels 11 19 Multiple frames 11 18 New idealised trace 11 13 New Memory View 11 1 New XY View 11 7 Open Closed amplitude histogram 11 16 Open Closed time histogram 11 16 Open Closed times 11 13 Process settings 11 6 Subtract buffer 11 18 Tag frame 11 19 ANGLE output instruction 6 16 Append frame 11 17 Append frame copy 11 17 Area between cursors 12 5 under curve between cursors 12 5 Area under curve 2 8 12 5 Arrange icons 15 1 Artefact rejection 3 16 Artefact rejection dialog 3 16 ASCII values table 3 3 Assume Power1401 hardware 9 7 Auto Average of waveform 11 2 Automate Directory for file 3 15 File naming 3 15 File saving 3 15 File size limit 3 15 Frames limit 3 15 Time limit 3 15 Auxiliary states devices 20 1 Average see Mean Average of waveform 11 1 Number of sweeps 10 4 Axis controls 2 5 10 2 Show and Hide 10 3 B Band pass filter 17 3 Band stop filter 17 3 Bitmap output 8 4 9 1 Black and white displays 10 7 BRAND output instruction 6 26 BRANDV output instruction 6 26 Breaking out of processing 11 6 Buffer Add frame data into buffer 11 18 Add to current frame 11 18 Average into 11 18 Clear buffer 11 17 Copy frame data into buffer 11 18 Copy to frame data 11 18 Description 11 17
84. 7 6 Multiband example 17 11 Number of coefficients 17 9 Nyquist frequency 17 10 Pink noise from white noise 17 8 Ripple in bands 17 7 Slope for differentiator 17 8 Transition region 17 6 Weighting 17 6 FIR filters 17 5 FIRMake filter types 17 7 FIRMake Discussion 17 5 First frame 10 1 Fit data 11 8 Fitting 11 8 Fixed interval period 5 8 Fixed interval sweeps 3 4 Fixed interval variation 5 8 Font for new data documents 9 5 for text views 9 5 Font selection 10 6 Footer 8 5 Format of Signal data files 1 3 Frame buffer See Buffer Frame comment 9 4 Frame zero 3 16 Frames Channels 2 2 Comment 2 2 Description 2 2 Display list 10 1 Extra data 2 2 Flags 2 3 Goto frame 10 1 Next 10 1 Overdrawing 10 1 Overdrawing online 3 19 Previous 10 1 Specifying 2 3 State 2 2 Status bar information 2 2 Tag 2 3 Tagging and untagging 11 19 User variables 2 3 Variation 2 2 Frames limit 3 15 oS Gradient of line 2 8 12 5 Grid Set colour 10 7 Grid show and hide 10 3 Group channels 10 3 H HALT output instruction 6 20 Hardware required for Signal 1 2 Header 8 5 Help 2 1 16 1 Help desk 16 2 Hexadecimal marker codes 3 3 Hide window 15 1 High pass filter 17 3 High pass filter example 17 10 Hilbert transformer 17 8 17 14 Histogram draw mode for waveforms 2 6 Horizontal cursor labelling styles 12 4 Hz sequencer expression function 6 7
85. 755 A 000000 example in an exponential fit you may know that the final coefficient aa Clear fit the offset IZEL Exponential fit from 0 00844749 to 00144894 30 points 28 iterations The limit values are applied after each ha EE iteration The fit may have to follow a convoluted path before 1t converges on a solution so do not set the fit limits too close to an expected solution as this may prevent convergence The Estimate values button can be used to guess initial values for fitting based on the raw data The Clear fit button removes the fit from the channel The results page of the Fit Data dialog holds information about the last successful fit done with the dialog The page has three regions coefficient values at the top a message area at the bottom and a plot of the residuals differences between the fit and the data in the middle The residuals are displayed immediately after a fit but will not be displayed 1f you close the dialog and reopen it olx rit data for Test fit Fit settings Coefficients Resuts Double exponential fit channel 1 Oto 50 50 points Sum of errors squared 0 08763 estimated probability 0 8333 Coefficient values Residuals Message area Analysis menu The Value column holds the fitted value that minimised the chi squared or sum of squares error for the fit The Sigma column is an estimate of how the errors between the fitted curve and the original data translates into u
86. 9 1 Un magnify mouse pointer 2 4 Update mode for changed data 8 5 Updating Signal 1 2 us sequencer expression function 6 7 V Value at cursor 12 4 Value between cursors 12 5 Index 6 Index VAngle sequencer expression function 6 7 VAR directive in output sequencer 6 8 Variables for output sequencer 6 8 VarValue script 6 8 6 10 6 11 6 18 VDACO 7 sequencer variables 6 8 VDAC16 Sequencer expression function 6 8 VDAC32 Sequencer expression function 6 8 VDigIn sequencer variable 6 8 VHz sequencer expression function 6 7 View handle Interactive access to 15 2 View menu 10 1 Add frame to list 10 1 Channel Information 10 4 Colour commands 10 7 Draw mode 10 4 Edit bar 10 1 Enlarge and reduce 10 2 Font 10 6 Frame display list 10 2 Goto frame 10 1 Info 10 4 Keyboard alternatives 10 8 Next frame 10 1 Overdraw frame list 10 1 Previous frame 10 1 Show buffer 10 1 Show Hide channel 10 3 Standard display 10 3 Toolbar 10 1 X Axis Range 10 2 Y Axis Range 10 3 Voltage clamp 4 1 Voltage limits TTL inputs 3 2 W WAITC output instruction 6 17 WAVE output instruction 6 28 WAVEBR output instruction 6 28 Waveform average 2 10 Waveform data Amplitude histogram 11 2 Area 12 5 Area over zero 12 5 Average 11 1 Connections 3 1 3 2 Copy and Export format 9 2 Leak subtraction 11 4 Mean value 12 5 Modulus 12 5 Multiple averages 11 2 Power spectrum 11 3
87. 90 result windows before I run out of resources as long as Signal is the only open application If system resources is a problem consider changing to Windows NT 2000 or XP where this is not an issue This command provides a dialog with a small piece of information in a Did you know form Further details can then be requested This dialog can also be set to appear when Signal is first run If you have an Internet browser installed in your system this command will launch it and attempt to connect to the CED web site www ced co uk The site contains down loadable scripts updates to Signal and information about CED products If you are having trouble using Signal please do the following before contacting the CED Software Help Desk 1 Read about the topic in the manual Use the Index to search for keywords related to the topic 2 Try the help system for more information Use the Search facility to find related topics 3 If none of the above helps FAX email or call the CED Software Help Desk numbers and addresses are to be found at the front of this manual and in the Contacting CED help page to be found near the start of the help contents Please include a description of the problem the Signal serial number and program version number and a description of the circumstances leading to the problem It would also help us to know the type of computer you use how much memory it has and which version of Windows you are running
88. A IED NN CA ee ere DE TTT TT oe A A A CA T TT TT TT TTT T ET ere DA ALLELE A Pi EE eee o uy o uw o c Ln ao I I I rm I dP 2 lt 2 S O o Effect of odd and even 0 5 0 3 0 4 O 2 Fraction of sample rate is 0 Another way of looking at this is to consider that a filter with an even number of coefficients produces half a sample delay The output halfway between points that are alternately a and a must be 0 You can use the formula given for the low pass filter to estimate the number of The graph shows the results of this high pass filter design with 69 coefficients which coefficients required but you must round the result up to the next odd number gives a good result and with 68 coefficients which does not In fact if we had not given a factor of 10 weight 20 dB to the stop band the filter with 68 coefficients would not have achieved any cut in the stop band at all of half the sample rate each cycle will contribute two samples The samples will be 180 out of phase so if one sample has amplitude a the next will have amplitude a the next a and so on The filter coefficients are mirror symmetrical about the centre point for a band pass filter so with an even number of coefficients the result when the input waveform is The reason for this unexpected result is that we have specified a non zero response at the Nyquist frequency half the sampling rate If you imagine a sine wave
89. A Z Labels are not case sensitive Labels may not be the same as instruction codes or variable names key In this optional field key is one alphanumeric a z A Z 0 9 character When this character is recorded as a keyboard marker during data capture the sequencer jumps to this instruction Each key can occur once Upper and lower case are distinct The key appears in the sequencer control panel code This field defines the instruction to be executed It is not case sensitive argl Instructions need up to 4 arguments and are separated by commas or spaces These are described with the instructions If an argument can be represented in different ways they are separated by vertical bars read as or for example expr Vn Vn off In this case the argument can be an expression a variable or a table reference comment The text after the semicolon is to remind you of the reason for the instruction If a key is set this comment also appears in the sequencer control panel gt display When a sequence runs text following a gt in a comment is displayed in the sequencer control panel to indicate the current instruction Many instructions allow the use of an expression in place of a constant value indicated by expr An expression is formed from numbers round brackets and the operators and and sequencer expression functions It cannot include a variable The operators and multiply and divide have higher priorit
90. DELAY DBNZ CALL CALLV RETURN Sequencer outputs during sampling These instructions do not change any outputs or read data from any inputs They provide the framework of loops branches and delays used by the other instructions The DELAY instruction occupies one clock tick plus the number of extra ticks set by the argument It produces simple delays of 1 to more than 4 000 000 000 sequencer steps DELAY expr Vn Vnt off OptLB expr The extra sequencer clock ticks to delay in the range 0 to 4294967295 The s ms and us built in functions convert a delay in seconds milliseconds or microseconds into sequencer steps Vn Variable or table index from which to read the number of extra clock ticks OptLB If this optional label is present it sets the next instruction to run This example uses display messages to tell the user what the sequence is doing SET 1 00 1 0 run with 1 millisecond clock ticks DELAY 2999 wait 2999 11 milliseconds gt 3 second delay DELAY OM GOP mall 3 seconds 1 tick delay gt 3 second delay DELAY Vi LB wait V1 1 ms branch gt variable delay DELAY V1 9 V1 9 is table index gt table delay DBNZ Decrement and Branch if Not Zero subtracts 1 from a variable and branches to a label unless the variable is zero It is used for building loops DBNZ Vn LB Vn The variable to decrement and test for zero LB Instruction to go to next if the result of the decrement is not zero DBNZ 1s often used w
91. E all be shown and hidden by using the popup menu generated by clicking the right mouse button on an unused part of the Signal window the blank parts of the toolbar area are suitable and always visible during sampling This command is only available on the toolbar It is equivalent to selecting New in the File menu then choosing Data Document That is to say it prepares Signal to start sampling with the current sampling configuration This command hides or shows the sequencer control panel that is available during sampling 1f the output sequencer is in use 13 2 Compile Script E Run Script Evaluate Turn Recording On Off Script menu The Script menu gives you access to the scripting system From it you can compile a script run a loaded script evaluate a script command for immediate execution and record your actions as a script You can find details of the script language and a description of the script window in the separate manual The Signal script language and in the on line help The script menu commands are This command is enabled when the current view holds a script It is equivalent to the Compile button in the script window Signal checks the syntax of the script and if it is correct 1t generates a compiled version of the script ready to run This command pops up a list of all the scripts that ERR have been loaded so that you can select a script to _ Compile Script run Signal compiles the select
92. E E tae mcactaastass 8 6 EE S PARTSI E A AEE A A teed NE aetoo 8 6 Sid TE AT AEA E EAA EE EA ET E E EE E A A ETT 9 1 A SE ATAA A AEE E E EA SE T EAA ETA N 9 1 Gi Ren eRe A O II A A 9 1 A A ta alot ats 9 1 A O II O II A E Oa ee 9 1 B AA E A ACE A 9 1 A A a had site Rael 9 2 SU ANS Text as cate eared nea eee eee 9 2 COPAS ORE OY VIEW sine aa clea ee ee i ered ete ee aes 9 3 Find Find Again and Find Sole eae 9 3 Sele A N E E 9 4 Repic sg i E N e La a 9 4 Tie conin ibane a a oe ee ese iD enrol re ere 9 4 Pe TAMNG COMIMCMU teen A ii heel coor et a a eat ins Matias nit eluate 9 4 dE Gh e eine ee Ee er Ean 9 4 SOW CV CIC AIG rence nn sane N 9 9 o ecc 10 1 Toolbarand TUS DI Na 10 1 Previous frame and Next AMS a a tie 10 1 GOO AMS E SE A saa ai 10 1 SOW o AA Sn A A AAA AE away Bearers 10 1 COV ELA AW Mame ISU ii sch A E ais 10 1 PROG drome LOSS sit A A AA 10 1 Frame display AA Ee OE Re AO AE 10 2 Finlarce view and Reduce VIC W sitet intl eum eel Mees 10 2 KAS VAMOS ib A chia ee e ate chin Se nteentactehie 10 2 LA A aha acne ttre natty ac a ein eaten eae ase 10 3 Sanda la td on eR eee OR oe me ae er 10 3 CUSTOMISE split a a a 10 3 Chamelintormaton mia lada 10 4 PUSO NAO ri 10 4 ME y A er ee area ee were 10 4 DA aC A 10 4 Signal for Windows version 3 Analysis menu Cursor menu A RT 10 6 A On Ren PE I OEE OP L PEI RA AE eer AIE E See ORCI Senet eee E Teen ae Seen 10 6 Use Colour and Use Black And Wiiites i255 aha
93. EX PlCSSIOMS IES 6 7 Wid DIGS A eSiasal sancti disadbalsat a a a eam dda e a 6 8 A O O a a akeedeld weaned 6 9 Sequencer instructor TIL CMCC dos 6 10 Dista ide 6 10 DACIU PS EA AE haa a 6 12 Cosine output control ANSHUCUONS aa 6 14 General COMO gt ra E Bucur A A EE O A 6 19 Variable ACUSA A A dcaainudslecunnnuouaes 6 21 Taole SAC COSS A A dee noite keane tacit 6 23 ACCES O Cala Cipi ece a a ines aseeregnss LES IAE 6 24 Randomis ato and 6 26 A MA a a a eae ees 6 28 Sequencer COMPLI CILOL mess aro marines 6 28 Sampling with multiple States concoccccnconcccnconcccccnnconcccncnnronncnnrnnncnncnnrnnnrnncnnrnanenarnanos 7 1 Wiha t Can multiple states dOl dede 7 1 Defining multiple States A ii 7 1 Dynamic outputs States si 7 2 NN A 7 4 File menu Edit menu View menu Table of Contents ii A G eae cle et aad ere oe 7 5 Slate outis SIES oa icone ee es 7 6 Controlling multiple states ii 7 7 E ETE EE PEA E EA A A sada naneaaradageeseatadceaeatacsmanecasagnreys 8 1 NO A A E E E E S 8 1 ODE age ae E E oe A E a T eee eee 8 2 pOr dN Tana e ern A gt i E A re 8 3 port OD C A AN 8 3 A E ET A E A E EAE E E AE E EE E oe gee eee 8 3 COTA A O NN 8 3 DAVES anda AS io 8 3 EXPOR AS iio 8 4 al A A II II Nee 8 4 A O A TN 8 5 WALA CALC Mode tddi 8 5 Load and Save ConfieuratiON ie a a 8 5 Pao a a A A ee ee R 8 5 PrE DIEVI e W aaee an nese eee Bee Ney E A A T ee ee 8 6 POE VISIDIC Print ANd Print sele Mrs ai 8 6 A
94. Exchange with frame data 11 18 Multiple frame operations 11 18 Remove frame from average 11 18 Show buffer 10 1 Subtract current frame 11 18 Subtract from current frame 11 18 Bxx output instructions 6 21 CALL and CALLV output instructions 6 19 CANGLE output instruction 6 16 Cascade windows 15 1 Case sensitivity In searches 9 3 9 4 Output sequencer 6 7 CED Software help desk 16 2 CEDCOND INI 18 4 CEDCOND LOG file 9 8 cfb file extension 8 2 CFS Library 1 3 Standard file extension 8 1 cfs file extension 8 1 CHAN output instruction 6 24 Change colours 10 7 Change process settings 11 6 Channel display Colour override 10 7 Draw mode colour 10 7 Order 9 4 Channel Information 10 4 Channel selection 2 1 Channels 3 point smooth 11 19 5 point smooth 11 19 Differentiate 11 19 Integrate 11 19 Marker channels 3 2 Modify data 11 19 Negate 11 19 Offset data 11 19 Rectify 11 19 Scale data 11 19 Set DC measurement area 11 19 Shift data 11 19 Show and hide 10 3 specifying 2 6 Subtract DC level 11 19 Types of channel 3 1 Waveform channels 3 1 Zero 11 19 Clamping configuration 4 1 Clamping experiment 4 3 Clamping support 4 1 Clear data 11 19 Clear text or memory view 9 2 Clipboard Copy cursor values 12 4 12 6 Copy data view as picture 9 1 Copy data view as text 9 2 Copy to 9 1 Copy XY view as text 9 3 Cut text to 9 1 Paste text 9 1 Close all associated windows 8 3 1
95. GO W eena r E paces ie etna eas Sunes Nasa aeatted oune siamese Senes taeeenaedt 15 1 O A rad acces P E ce as ea oe E as E E E ES 15 1 Sn O a eer ee eee ere 15 1 Tie otitis 15 1 He Nec id 15 1 Castilla 15 1 ARA AS mii ere en one nee cee ene eR ee ee ee ee eee ere ee ee ee 15 1 0 LOTA O ON Or O E O 15 2 A a ea aie arse ne i teria clan ie eee cna ise peered cae Sunt aanensennc on oenuedacln ss 15 2 Nel MENU scsi cect aces pete she ets ensd deteaietedsneeehcieaad ete sete teeta 16 1 Usine NG lPi ack fade aa tesoesatacew as lett totes eeu ee ee 16 1 A a T sieese enna 16 1 Ty ECC AY E A nu EAS 16 2 View WED dei iii ii 16 2 Other SOULCES OF Ml O 16 2 Digital e 17 1 Ji AN E AA E A E IA 17 1 ADD eee A E a A AA A erie e MOU 17 1 il AAA a O 17 2 A A no EAS og co ee 17 3 NT ey De NE AEE TE A E AE annie wnat E AE toan aes E oa 17 3 ER A II E O IA 17 5 FIRM ake lo secon tee teadreca see E e AT EEA 17 7 FIR filter examples AMP en ne ee eee 17 8 ADE TAO eee eee ee eee 17 14 Programmable Signal Conditioners ooccocccocccnonocanocnnonncnanonanonanonnnonnnnanonarenarenaronaroannnnns 18 1 What a signal conditioner does cccceeeeeeeeseeesseeeseseeseesseseessssssssssssessssssssssaasasaaaaas 18 1 A A A eae eee 18 1 OTOL paneles 18 2 Setting the channel Sain and Offset siccsccccadeacsesvesiccvieedgaseoiassenvives iniesGersesnloed nsopiestexvieer tors 18 3 Conditioner COMME CULO an s 18 4 viii Amplifier Telegraph Signal for Wind
96. However unless you are using automatic saving the data has not yet been saved in a permanent disk file though it is stored on disk To keep the data you must save the new data using the File menu Save command If you try to close the document view without saving the data Signal will check that you really want to do this Data documents are always stored on disk Other document types are kept in memory until you save them We keep data documents on disk because they can be very large When you use the File menu New command Signal creates a temporary file in the directory specified in the Edit menu Preferences dialog If you do not specify a directory the location of the temporary file is system dependent When you save a new data document after sampling Signal moves it to the directory you specify To avoid setting the sampling analysis and screen configuration each time you sample you can save and load sampling configurations from the File menu The configuration includes The sampling parameters in full including port information for all ports The position of all windows associated with the new file including duplicated windows and the sampling control panel The displayed channels and display modes of the channels in the windows The outputs to be generated during sampling The multiple states information including the protocols The processing parameters and update modes of all memory views If sampling ends without failing or
97. I V Curves 11 7 Icons for files 1 3 Icons arrange 15 1 Idealised trace Drawing mode 10 5 Idealised trace editing 9 9 Idealised traces 11 13 import folder 8 3 Import foreign data file 8 3 Impulse response 17 5 Inflection cursor mode 2 2 Installing Signal 1 2 Instructions for output sequencer 6 6 Integrate data 11 19 Index 3 Signal for Windows version 3 J JUMP output instruction 6 20 Kk Key for XY view 10 4 Keyboard control of display 10 8 Keyboard control of sequencer 6 1 6 7 Keyboard driven analysis 11 20 Keyboard markers 3 2 Enable 3 6 Entering 3 19 Label for cursor 2 4 12 3 Label for horizontal cursor 12 4 Last frame 10 1 LAST SGC last sampling configuration 8 5 Leak subtraction 11 4 Least squares fitting 12 5 Licence information 1 4 Line draw mode for markers 10 5 Line draw mode for waveforms 2 6 Line style in XY views 10 6 Line thickness for printing 9 4 Line width for displays 9 5 Lock y axes 10 3 Log view 2 1 Logarithmic scale 17 6 Low pass differentiator filter 17 4 Low pass filter 17 3 Low pass filter example 17 8 M Magnify pointer 2 4 MagStim auxiliary states device 20 1 MARK output instruction 6 25 Marker count 2 8 Marker display Dots 10 5 Lines 10 5 Rate 10 5 Markers Drawing mode 10 5 Maximum Sampled frame length 3 5 Total sampling rate 3 5 Maximum and Minimum cursor modes 12 1 Maximum be
98. OO der EES O DBNZ V1 loop 10 degrees per step The next example produces 90 cycles each increasing by 0 1 degrees per step per cycle SZ taal tec make sure full size ANGLE IO make sure we start at phase 0 RATE Li ag 1 degree per step to start with MOVI Vis 90 pan 90 Steps OL vas Loop RINCW LOL e aL NO degrees Tosi walt WAITC 1 wait Bice au Welt TOr Iep SNH se eg DBNZ V1 loop 10 degrees per step This instruction clears the cosine output new cycle flag If you have been running for several cycles and you want to stop the next time phase 0 is crossed use this instruction immediately before using WAITC CLRC N OptLB DAC n n DAC number in the range 0 7 available DACs depend on the 1401 type OptLB If this optional label is present it sets the next instruction to run This example starts a sinusoid and stops it at the next phase 0 crossing after a user stop requests Because the sinusoid passes phase 0 in the WAITC instruction and does another step in the RATE 1 0 instruction we offset the phase by 2 steps However this would cause the start of the sinusoid to be 2 steps wrong so we change the start angle to match G PHASE 1 2 Hz 2 ANGLE 1 2 HZ2 10 compenstate for ending so we start in correct place RATE LA 2 H2 Output HERE JUMP HERE Output Ls running Runnang S CARE 1 Stop output WT WAITC 1 WT wait for cycle end gt Waiting RATE 1 0 HERE stop and then idle 6 18 General control
99. Settings frame list directly You can also select the current frame all frames tagged or untagged frames or frames with a given state code If you choose the state code option the dialog displays a field into which you can enter the state code to use 11 5 Breaking out of Process Process command with a new file Process settings Signal for Windows version 3 The frame list values are evaluated when the Process button is used the frames are processed and the results added into the memory view data The dialog window will remain on screen until removed with Cancel This means that you can set the frames to Tagged then adjust the frame tagging in the source data document and click the Process button to analyse the selected frames Some sets of frames have subsets which can be used to give a more specific list of frames to use If you check the Clear memory view before process checkbox the memory view is cleared before the results of the processing are added The Reprocess if source data changes checkbox enables automatic re processing Automatic re processing is optimised to try to prevent unnecessary work but can still slow Signal significantly on occasion particularly with large files If you check the Optimise Y axis after process checkbox the memory view y axes will be re scaled after the new results are added into the memory view to best display the data If the Settings button is pressed the process dialog is
100. Signal for Windows Version 3 Copyright Cambridge Electronic Design Limited 1997 2007 Neither the whole nor any part of the information contained in or the product described in this guide may be adapted or reproduced in any material form except with the prior written approval of Cambridge Electronic Design Limited Version 3 00 Aug 2004 Version 3 01 Dec 2004 Version 3 02 Jan 2005 Version 3 03 Apr 2005 Version 3 04 Sep 2005 Version 3 06 Jan 2006 Version 3 07 Apr 2006 Version 3 08 Nov 2006 Version 3 09 Feb 2007 Published by Cambridge Electronic Design Limited Science Park Milton Road Cambridge CB4 OFE UK Telephone Cambridge 01223 420186 International 44 1223 420186 Fax Cambridge 01223 420488 International 44 1223 420488 Email info ced co uk Home page www ced co uk Curve fitting procedures are based on routines in Numerical Recipes The Art of Scientific Computing published by Cambridge University Press and are used by permission Trademarks and Tradenames used in this guide are acknowledged to be the Trademarks and Tradenames of their respective Companies and Corporations Signal for Windows Getting started Sampling data Table of Contents aaa 1 1 DCEO UICC Olt AA E E 1 1 New Acatlres TA VECINO as 1 1 HANA do 1 2 rala e a nase Yo ltodasta scandent bade eericasieaaes 1 2 Update Sinai ir Cie 1 2 RO dll noes 1 2 O 1 3 Usine tas mati 1 3 Direct access t
101. Spike2 for Macintosh files as well as data from several third party vendors Signal searches the import folder in the Signal installation folder for CED File Converter DLLs If you need to translate a file format that is not covered please contact us and describe your requirements The script language command to convert files is FileConvert Signal can import idealised traces from the CED Patch software These are files of the type res or r generated by the Pulsed Analysis or Continuous Analysis programs Once imported the idealised trace will be stored in the resource file associated with the data file i e the sgr file and can be processed in the Analysis menu see the Analysis menu chapter for more details This command is used to close the active window It is equivalent to double clicking the control menu icon at the top left of the window in windows that have one or pressing the right hand top corner X button in Windows 95 and NT 4 0 etc If you use this command on a new memory view a newly sampled data document or a text based view with text that has not been saved a dialog will ask 1f you wish to save the text before closing the window This command closes the current window and all windows associated with the file In addition to saving the state of the data file in a sgr resource file Signal offers to save the state and contents of any memory view windows that belong to the data file Next time you open the data file
102. UT 001i clear bits 3 and 2 set 1 invert 0 DLGOU T Fresku i invert 0 again to produce a pulse DIGOUT V10 use variable V10 to set the pattern Vn With a variable the new output is old output BAND Vn 7 0 BXOR Vn 15 8 The variable equivalent of 001i is 241 256 3 and of PA i is 255 256 1 If you use a table value set the same value in the table that you would use for a variable You can use the VarValue script in the Scripts folder to calculate variable or table values Opt LB If this optional label is present it sets the next instruction to run This example produces ten 1 millisecond pulses 100 milliseconds apart MOVI lo V1 holds the number of pulses LOOP DEGOUL Altar a DI FA 0 A gt Pulsing DIGOUT Pate Bake Te O zbit 0 Low gt Pulsing DELAY ms 100 4 4 inst in the loop gt Pulsing DBNZ Vi LOOP count down gt Pulsing HALT finished gt Done The DIGLOW instruction changes the state of digital output bits 7 0 of the Power1401 and Microl401 see the Sampling data chapter for the connections It has no effect on a standard 1401 or 1401p us Unlike DIGOUT the output changes occur immediately they do not wait for the next sequencer clock tick You can take advantage of this to change all 16 digital outputs almost simultaneously within a few microseconds by using DIGOUT followed by DIGLOW DIBEQ DIBNE DISBEQ DISBNE Sequencer outputs during sampling DIGLOW pattern Vn Vn off OptLB
103. Value script in the Scripts folder calculates variable values for the digital and the cosine instructions Scripts can set and read the variable values with the SampleSeqVar script command See The Signal script language manual for details You can set initial values from the script as long as you set the values after you create the new data file but before you start sampling Values set in this way take precedence over values set by the VAR directive 6 8 Table of values Declaring the table Setting table data Accessing table data Long data sequences Sequencer outputs during sampling The Signal sequencer supports a table of 32 bit values when used with the Micro1401 and Power1401 The 1401p us does not support this feature You declare that a table exists with the TABSZ directive which normally occurs at the start of your sequence TABSZ expr Where expr is an expression that sets the number of items in the table The table size must evaluate to a number in the range 1 to 1000000 Each table item is a 32 bit integer and uses 4 bytes of 1401 memory The maximum size in a 1401 with IMB of memory and assuming that there is no arbitrary waveform output is around 150000 items The first table item has an index number of 0 the second item is index 1 and so on From the script language you can move data between an integer array and the table with the SampleSeqTable function You can also preset table data from the se
104. a a ee ias 10 7 Chanee Colour sacri is Bie a behead diets thie ened 38 ce 10 7 Keyboard display contol icin a a sad dusethat ia 10 8 sh stuf E EN O Ec 11 1 New Memory VIEW ait coche este E E E ER 11 1 Wavclorm o A E TOEN 11 1 AMO AVIDE n a a et A EE 11 2 AUIS Aston E OR aE EO REEN 11 2 PRO WEl S pect direne T wenigontvndaddes mewaus vuadecheud tawneandeal 11 3 Eak SUD IFA cOn dal 11 4 SS teen O O e WOTCR Im YO Ae WOE este e EY 11 5 Process command with a new file oooccconcccnnnccnonccnonoccnnnorononocnonocnonoccnnnrcnonicnonos 11 6 A II O et eer mene a gy ee 11 6 ING Ne NM Wy oss A A tne sioete 11 7 A ne ree meron eet emery artes me Creer ery ee ety ee ea me ree renee 11 7 A O NIN TE 11 8 UN ee eer eee ne ae eee eee ene ne ee eer 11 13 New 1dealised trace SCAN nidad 11 13 New idealised trace Threshold cccccccccccccceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeees 11 15 Open Closed time DIO ra 11 16 Open Closed amplitude hiStograM ooooooncnnonauanananonononononnnnnnononnnnnnnnonnnnnnnnn nono nnnnnnnnnnonnos 11 16 B rstduration DIO As 11 16 Append AMAS a AAA 11 17 Append AMECA E O AE 11 17 Dee NS aa a a r a ater aae ects hace ahaa 11 17 DA is ETETE E E E E A E EEEE A A EA eae oats E A 11 17 AEDE ERE E E E ET E E E A AEA E A E E E E A E E ess 11 17 CEO e aea O A 11 17 EP a me nts a a rar AAE 11 18 Copy Com DIe onee ene ere a Hen ne eee ee roe ee ee ene ee 11 18 Exehange DOr Trenari nn eee ene dl ee
105. a percentage of the maximum available you can set any value from 0 to 100 This control is disabled for manually controlled states The buttons labelled Copy to all and Copy above can be used to set up many states quickly Copy to all copies the currently selected state to all other states while Copy above copies the current state s settings to all higher state numbers These buttons are available for all types of MagStim The button labelled Test tests if communications with the MagStim can be established It checks that you have the correct COM port that the serial line is connected and that the MagStim is communicating correctly with Signal These buttons are available for all types of MagStim The BiStim is more complex than the 200 o gt f Magstim configuration in essence it is two 200 units with one aaa aes eae slaved to the other When a BiStim device M Magstim BiStim is selected two controls governing overall om port for communications COM 2 i behaviour are shown TF BiStin in independent trigger mode F Use Hires timing mode This sets how Signal will expect the BiStim state manual control ree State 1 ower ZU power b SO after 100 ms to be configured rather than how it will be State 2 power 30 power b 30 after 100 ms set up by Signal Normally a BiStim generates two output pulses one at the time of the trigger and the second at a preset interval after the trigger Using the controls an
106. ace view Press Find Next to move the 9555s Wih var text cursor to the next occurrence MT Match Case Direction Replace Al of the search string Replace to mM si Cancel replace the search string found E Down with the new text and Replace All to find and replace all occurrences of the search text The search is normally insensitive to the case of characters if you want to Match Case in a search check the box The search starts at the cursor position and stops at the end of the document unless the Wrap around search box is checked in which case the search will continue from the start of the document after reaching the end I Wrap around search This command is File Comment for example cfs available with file and memory views it allows you to see and sk Cancel edit the file comment The file comment is a single line of text attached to the data file it can be entered at the end of sampling or by using this command Sampled after applying treatment bb This command is available with file and memory views 1t allows you to see and edit the frame comment The frame comment is a single line of text attached to each frame in the data file that 1s available for any purpose The Edit menu Preferences command opens a dialog in which you set items relating to the screen appearance metafile output hardware connections and where to store data during sampling Most of the preferences are stored in the Windows regist
107. ae aha A A A AAA A a 20 0 20 oo oo oe oo oo eo tO EERE EERE EER Ey Gap 0 08 EE TA SE ASE ES ELE E CEE OE A Gap 0 04 40 0 15 ee ee ee ea ta ttt tt tt tS 60 SR a A A O A A a a 0 10 80 0 05 100 l A U SS 88eeees __ A 120 0 00 AI gt 140 ae 0 1 0 2 0 Fraction of sample rate The frequency response is shown in dB which is a logarithmic scale A ratio r is represented by 20 log p r dB A change of 20 dB is a factor of 10 in amplitude 6 dB is approximately a factor of 2 in amplitude The graph shows that a frequency in the stop band is attenuated by over 110 dB a factor of 300 000 in amplitude with respect to the signal before it was filtered 5 Because we didn t specify what happened between a frequency of 0 08 and 0 16 of the sampling rate the optimisation pays no attention to this region You might ask what happens if we make this transition gap smaller The lighter line in the graph shows the result of halving the width of the gap by making the stop band run from 0 12 to 0 5 The filter 1s now much sharper However you don t get something for nothing The attenuation in the stop band is reduced from 110 dB to around 70 dB Although you cannot see it from the graph the ripple in the pass band also increases by the same proportion from 1 part in 30 000 to 1 part in 300 We can restore the attenuation in the stop band by increasing the number of coefficients to around 120 However
108. agStim documentation if you require a different arrangement In the case of a Rapid stimulator the 26 way connector is also used by the integrated control system and this will need to be disconnected before the control cable can be connected Magstim PC serial 26 Way D Type 9 Way D Type high density plug socket CABLE SCREEN Trigger BNC plug to 1401 digital out Magstim PC serial 9 Way D Type 9 Way D Type y ket api Serial cable or 0 o o o O Magstim Trigger 15 Way D Type BNC plug plug to 1401 digital out Trigger cable If you are not using the external trigger input to trigger the MagStim from the 1401 you will need a different cabling arrangement for example to take the MagStim low going trigger out signal on pin 8 of the 26 way connector and use that as an input to the 1401 to trigger the 1401 sweep Consult your MagStim documentation for more information 20 6 Notes on use Auxiliary states devices A straightforward arrangement to use a MagStim with Signal would be as follows l Connect the serial port of your computer to the MagStim serial line input using the appropriate serial line cable Connect the trigger BNC plug to the 1401 digital output port 0 BNC socket found on the front of all modern types of 1401 If you are using a 1401plus the digital output pulse is available from the 25 way digital output socket on the front of the 1401 2 In the outputs page of the sampling config
109. age is shown at the top of the window Initialise loop counter Just less than 0 25 s gt Waiting Set output high Wait 9 steps Set output low a Wait 8 steps Loop required times gt Done Stop outputs 6 2 Format Format with step numbers Current fa Help 2 Loading sequence files for sampling Getting started Simple sequence restarted each sweep Sequencer outputs during sampling This aligns the labels key RIEIS Biel ES codes instructions and Formatting Done EEs 2 any arguments output text Initialise loop counter Just less than 0 25 s gt Waiting and comments and Set output high Wait 9 steps removes step numbers oe ee a Undefined labels are not a Loop required times flagged but there must be Stop outputs gt Done no other errors This does the same job as the Format button and also starts each line with Initialise loop counter Just less than 0 25 s gt Waiting the step number Step Set output high n b b f 1 Wait 9 steps umoers Can De userul as Set output low i b b Wait 8 steps they give an indication Pegs Aen E when you are running out Stop outputs gt Done of space and can pinpoint the line where your sequence is not behaving as you expect This compiles the sequence and if the sequence is correct saves it and makes this the current sequence for use during sampling If you were to open a new data file
110. ake script command Signal takes care of all the frequency scaling and recalculation for you The remainder of this description is to help users of the FIRMake script command but the general principles apply to all the digital filtering commands in Signal Users of the FIRMake script command must specify frequencies in terms of fractions of the sample rate from 0 to 0 5 For example if you were sampling at 10 kHz and you wanted to refer to a frequency of 500 Hz you would call this 500 10000 or 0 05 17 5 Example filter Coefficients and frequency response for low pass filters 17 6 Signal for Windows version 3 The heavy lines in the next diagrams show the results obtained by FIRMake when it designed a low pass filter with 80 coefficients with the specification that the frequency band from 0 to 0 08 should have no attenuation and that the band from 0 16 to 0 5 should be removed We can specify the relative weight to give to the ripple in each band In this case we said that 1t was 10 times more important that the stop band 0 16 to 0 5 should pass no signal than the pass band should be completely flat We have shown the coefficients as a waveform for interest as well as the frequency response of the filter The shape shown below is typical for a band pass filter One way of understanding the action of the FIR filter is to think of the output as the correlation of the waveform and the filter coefficients pS f Mo R
111. al switch on the CED 3304 You specify a current range when you define the stimulus settings in Signal Signal will not let you sample if the wrong range has been set on the CED 3304 The timing of the 3304 current pulses is controlled by a TTL signal connected to the front panel Trigger input Signal can control the timing of the current pulses with one of the 1401 digital outputs in Dynamic outputs mode or external equipment can control the pulsing in Static outputs mode For safety reasons the CED 3304 will switch off a pulse after a preset limit is reached see the 3304 documentation for details of how to control this limit Signal monitors the stimulator health waiting until it is ready for use before allowing a sampling sweep to proceed and terminating sampling if a hardware problem develops To aid data analysis Signal saves the stimulation current in the data section variables of the sampled file This value in uA is placed in the first user frame variable in the sampled data file The saved values can be retrieved using the script language later versions of Signal may make use of them directly To configure the CED 3304 in Signal open the sampling configuration dialog and make sure that Multiple states is enabled in the General tab Open the States tab and make sure that you are in Dynamic outputs or Static outputs mode If CED 3304 support is present there will be a button labelled CED 3304 If this button is not present reinstal
112. all that the corner frequency for a low pass analogue filter 1s usually stated to be the frequency at which the filter response fell by 3 dB which is a factor of 2 in amplitude when the response falls to 0 707 of the unfiltered amplitude If we use the analogue filter definition of corner frequency we see that we have produced a filter that passes from 0 to 0 115 of the sampling rate and we wanted from 0 to 0 1 so we can move the corner frequency back This will increase the attenuation in the stop band and reduce the filter ripple as it widens the gap between the pass band and the ot A stop band If we move it back to 0 085 A TT the attenuation in the stop band increases to 84 dB Alternatively we could move both edges back keeping the width of the gap constant This leaves the stop band attenuation more or less unchanged but means that the start of the stop band is moved lower in frequency Filter response 17 9 weight alU prm 3 0 prm SJ lL ii Ai OB A nd lt 07 0 1 function sL prnl 0 User pra LIS 0 Fr Resp 0 ONF 0 5 7 binss 1 fr band end coer l pen 1 0 prm 1 1 coef 1000 0 00 Usas prm band start A high pass filter is the same idea as a low pass except that the first frequency band is a stop band and the second band is a pass band All the discussion for a low pass
113. all the data view windows and their contents will be restored If you open saved memory view data then that will be opened as a file view and restored to its previous state as well Save will save the current document under its current name unless it is unnamed in which case you are prompted for a name before it is saved Save As is used to save the document with a different name leaving the original file intact The Save command is not available for a data document unless the data has been changed since the last save Data files are kept on disk not in memory as they can be very large Changes made to a data file are permanent as they are made on disk When you save newly sampled data you are giving the data file on disk a name replacing a temporary name If you save it to a different drive from that set in the Preferences Signal copies the file to the new drive and deletes the original If the file is large this operation can take a noticeable time When you are working with a file view the current frame for the view is held in memory this frame will be discarded when the view switches to a different frame Any changes made to the frame data while it is held in memory must be saved before the new frame is loaded or the changes will be lost You can write the changed data back into the file using the Save command The Preferences dialog allows you to select what happens if the frame is changed while data is unsaved the default action is to
114. also be normally distributed with twice the variance of the errors We use this to estimate the standard deviation of the data and then we apply the probability test We label this as estimated probability The same comments about likely values apply as for the Chi square fits except that very small values may just mean that our estimation process fails for your data The coefficient Sigma values are calculated on the assumption that the model fits the data that all the original points have the same standard deviation and that the standard deviation of the original data can be deduced from the residual sum of squares errors This command analogous to New Memory View is available when a file or memory view is selected It provides a pop up menu from which you can select an analysis type The analysis types are intended for use by patch clampers doing single channel experiments Currently New idealised trace SCAN New idealised trace Threshold Open Closed time histogram Open Closed amplitude histogram and Busrst duration histogram analyses are available Selecting these leads to a settings dialog An idealised trace needs to be fitted to a data trace before histograms can be built If you set up histograms without an idealised trace fitted they will still be created but will not contain any data If you are working on line it 1s worth noting that the idealised trace will be generated first so that other analyses that depend on it will still
115. alues section below You can select a pulse by clicking on 1t with the mouse Alternatively you can toggle the selection through the various pulses on an output using buttons in the Controls section You can change the start time of a pulse by dragging it around the display area Experiment with clicking on pulses to get a feel for how pulse selection behaves Note how the display of pulse parameters changes as you select different pulses This is the central area of the 000s 1 00 dialog which contains a lt gt De Com ster 715 babel Fos JA pa number of controls used to interact with the dialog and displays providing information At the top of the controls area at the left hand side is text such as 0 00 s indicating the time for the start of the pulse output frame The right hand side holds similar text showing the time for the end of the pulse frame When a pulse is being dragged about the display the current pulse position is shown level with these numbers All of these times are shown in the currently selected time units Below these time indicators are a number of controls from the left these are These buttons are used to toggle the selected pulse through the pulses on the current output trace in forwards or reverse time order This button is used to delete the currently selected pulse This button is only enabled when multiple states controlling dynamic outputs are in use see the Sampling with multiple states chapt
116. alysis or to customise Signal behaviour in some way The icon on the right represents a Signal sequencer file These files store sequences of output pulses for use during sampling The first section of this manual introduces you to Signal by suggesting a few tasks you might undertake to familiarise yourself with the system We have supplied an example data file for you to experiment with so there is no need to have your own data available at this time A second chapter introduces sampling new data Subsequent chapters describe using pulse outputs during sampling and multiple frame states The second section describes the individual menu commands We suggest that you work through as much or as little of the familiarisation section as you feel you need then dip into the menu details section as required for more detailed information Once you are familiar with the program you may wish to investigate the script language so you can automate your data capture and analysis This is described in the separate manual The Signal script language The online help system duplicates all the information in the manuals and is often the fastest way to look up a topic This manual covers version 3 04 of the Signal software Later versions of the software will have further enhancements which will be talked about in the revision history found by selecting the Index from the Help menu Some users may wish to write their own applications that manipulate Signal data
117. ameter sets the length of the pulse Both of these fields will use the current time units If the No return checkbox is checked the pulse does not return back to the initial level when 1t ends but just stays at the pulse level giving us a single step change In this circumstance the length parameter has no effect This specifies a square pulse whose amplitude Square pulse varying height at 0 60 to 0 70 Id varies as it is used The Size Start and Length Size mv ET Start s jos M Noretum parameters are exactly the same as for the Length s o1 simple pulse In addition there are three Stechange 0 Repeats fe Stes fe parameters controlling the built in variation This type of pulse is not available for digital outputs The behaviour of the built in variation is described under Pulses with variations below 5 9 Signal for Windows version 3 Varying duration pulse Square pulse train Ramp with varying amplitudes Sine wave Arbitrary waveform This specifies a pulse whose amplitude IS Square pulse varying length at 0 75 to Id e constant but the pulse length changes as it is Size votts fi Start s 0 75 used This is the only pulse with a built in Length s 01 Push back Step change o Repeats fi Steps o variation that is available for the digital outputs The Size Start and Length parameters are exactly the same as for the simple pulse The other parameters control the variation with the ex
118. and Close below for outward currents or Open Baseline track jo points below and Close above for inward currents Having interpolation None F two thresholds in this way provides some protection F Multiple level o a against false events being detected which are actually noise Hew Cancel Transition points can be set to provide a dead time after a transition before another can be detected This allows for settling of data which has been filtered The Base level is used in multiple level analysis The thresholds for subsequent levels are calculated by first averaging the two thresholds then doubling the difference between this average and the base level The result 1s assumed to be the current per channel opening for subsequent levels and so subsequent thresholds are calculated by adding multiples of this value to the original thresholds By default an event will start at the first data point found to be across a threshold and have an amplitude which is the average amplitude of all the data point within the event period The Interpolation field can be used so that each transition has a start time calculated by assuming that the transition occurred between sample points at a time calculated by extrapolating between the points Currently only linear interpolation is available This assumes a straight line represents the data between sample points Select Multiple level if you have more than one channel in your patch Outward current me
119. and Copy above can be used to set up many states quickly Copy to All copies the currently selected state to all other states while Copy Above copies the current state s settings to all higher state numbers This button tests if communications with the CED 3304 can be established It checks that you have the correct COM port that the serial line is connected and that the CED 3304 1s operating correctly The Help Cancel and OK buttons have their usual meanings If you want to use the CED 3304 from Signal very simply with a single current setting you must still enable multiple states Set one extra state so states are enabled and set the desired current for state zero and then sample using state zero only This gives you access to the configuration dialog to set the basic CED 3304 parameters and enables checking of CED 3304 behaviour during sampling and the recording of the stimulation settings 20 9 Signal for Windows version 3 CED 3304 Connections and cabling 3304 serial cable Notes on use The CED 3304 uses a USB A B cable or a 9 pin serial line cable both are supplied with the unit for control Use one cable or the other If you plug in both the USB connection will take precedence See the Operation chapter of the CED 3304 Owners Handbook for more information about the control cables The front panel Trigger input should be connected to the 1401 digital outputs or other TTL pulse source using a standard BNC cab
120. and Response channel items set the data file channels that will hold relevant data For voltage clamp experiments the stimulus channel is voltage and the response channel is current for current clamp it is the other way around The channel items can be set to any valid channel number from to 80 If the channel specified does not exist in the sampling configuration then the text for the incorrect item will be shown in red If the channel does exist the channel calibration will be checked for valid units The units for a voltage channel must contain the character V and start with either V or one of M U N P or F lower case characters are also allowed implying milli micro nano pico and femto volts respectively Plausible legal voltage units would include V mVolts or uV The initial character is used to scale the measured values to actual volts during online analysis Similarly the units for a current channel must contain the character A and start with either A or one of M U N P or F nA and pAmp are obvious examples The Control DAC item sets the DAC in the pulse outputs that is used to control the stimulus The DAC specified must be from O to 7 be in use and the units for the DAC specified must be appropriate for voltage or current according to the experiment type or the DAC text will be shown in red The State for resistance measurements
121. ans that a channel opening produces a more positive current This is a convention An inward current would produce a more negative current when a channel opens New will take you to the standard process dialog described above but will not generate a result view The results are drawn instead in the file view on top of the original waveform 11 15 Open Closed time histogram Open Closed amplitude histogram Burst duration histogram Signal for Windows version 3 This opens a settings dialog which defines the parameters for a histogram of event durations The Channel field selects a waveform channel from the source view This channel must have an idealised trace fitted in order for the histogram to be built The x axis of the result starts at zero so the Bin width and Maximum duration together define the number of bins in the histogram The two regions labelled Include and Exclude represent the flags associated with Settings for OpClHistifexample Ea Channel 201 ADC 0 Idealized Bin width 0 001 5 Maximum duration o ES Exclude I Opentime Level P Open time Closedtime Level Closed time T Firstlatency Level 3 Firstlatency Level 3 T Truncated Level4 Truncated T Level 4 Assumed amp Level 5 M Bad data T Level 6 Include Level T Level 2 Assumed amp Level 5 T Bad data T Level 6 Cancel each event An event is included in the histogram
122. art of a sampling sweep Signal writes values to the 1401 DACs states I Individual repeats Cycles before idle State 3 data n E Repeats d digital tput and digita outputs EEN DACO according to the current lnver ee state of the experiment 000 These outputs could select the stimulus or have other 1234 effects as required O Digital output bits 7 In addition to defining the values to be output Static _ Cancel _Fnnow Help outputs states requires that you also set up how Signal will sequence the states This adds quite a number of extra controls to the dialog for details of these see the section on States sequencing above The main box in the centre of the dialog now defines the values to be written to the digital outputs for each sweep Because these are now digital outputs the enables for the bits to be used are in the Outputs page but otherwise this and the adjacent buttons behave in the same way as for External digital mode The actual digital outputs and connector pins used are the same as for pulse outputs see the Pulse outputs during sampling chapter State 3 data Repeats i DAC 0 2 DAC The State data box to the right of digital output bits sets the DAC outputs for a selected state You can select the state for which values are shown by clicking on the digital outputs line for that state The DAC outputs used 0 to 3 only are available are enabled and disabled in the Ou
123. ary by incrementing their amplitude or duration by fixed amounts This provides a straightforward way of generating a repeating set of pulses from one definition If you are using Outputs frame or Fixed interval sweeps this dialog is used to set and control the length of the pulse outputs frame the start of the sampling sweep within the outputs frame and the fixed repeat interval Pulses dialog layout The dialog is divided into four sections Display This occupies the upper part of the dialog and shows the currently defined outputs as graphical traces At the top of the display is a solid line that shows the portion of the pulse output frame period that is covered by JT 1L M the sampling sweep If you are using Outputs frame or Fixed interval sweeps this line is thin where no sampling is taking place 5 1 Signal for Windows version 3 Se Controls State 1 Ia Label Fast The outputs themselves are displayed starting with DAC zero at the top and finishing with the digital outputs at the bottom with the higher numbered digital outputs first Only traces for enabled outputs are displayed The output traces are labelled with the DAC number or digital output bit number The display also includes a dotted rectangle drawn around one of the pulses or around the initial level of an output This indicates the currently selected pulse whose numerical parameters are displayed at the bottom of the dialog see the V
124. ata BGT V1 4000 START 1f above upper limit wait BLT V1 0 START if too low wait IN MOVI VAS IA timeout 4 instructions loop INLOOP CHAN Vs to check if still inside BGT V1 4000 START if above upper limit wait BLT vV1 0 START if too low wait DBNZ V2 INLOOP see if done yet REWARD DIGOUT Medion pes 1 Task done OK DELAY S 1 leave bit set for 1 second clear done Dit DIGOUT LF cesta se did O PENE next task We want the data to be in range for one second There are 4 instructions in the loop that tests this so we set to the loop to run for the number of steps in a second divided by 4 For this to work correctly the sequencer must be running fast enough so that 4 steps are no longer than the sample interval for the waveform channel This instruction moves an integer constant into a variable The syntax is MOVI Vn expr OptLB Vna expr Vn A variable to hold the value of expr expr An expression that is evaluated as a 32 bit integer OptLB If this optional label is present it sets the next instruction to run MOVI is not the same as the VAR directive The VAR directive sets the value of a variable when the sequence is copied to the 1401 and does not occupy a step The MovI instruction 1s part of the sequence and set the value of the variable each time the instruction is used The MOV instruction sets a variable to the value of another with the option of adding a 32 bit number and dividing by a power of tw
125. ata chapter the pulses configuration dialog Pulse outputs while sampling and dynamic outputs multiple states Sampling with multiple states In addition to the clamp specific analysis mechanisms the documentation on the standard analysis menu should be useful as all of the general analysis features in Signal can be used on clamp data The trend plot generation and curve fitting sections of the analysis menu may be of particular interest Specialised data acquisition and analysis can be carried out by using the Signal script language The online clamp support within Signal consists of a separate page within the sampling configuration where the clamp setup can be configured plus mechanisms for online analysis and experiment manipulation that make use of this information to provide automatic resistance measurements holding potential or current controls and a membrane analysis display Because Signal is a very general purpose program that can be configured and used in many ways there are a number of limitations to the types of experiments that can be used with the clamping support A significant amount of the clamping support consists of checking for incompatibilities between the clamp setup and other aspects of Signal sampling rather than simply configuring the clamping The clamp page in the sampling configuration is used to configure Signal online clamping support This page allows two clamp setups to be defined along with a state set of
126. atal Data2 and so on or a name produced from the file name template if one is available You can customise the view by adjusting the x axis y axes channels and other aspects of the view A sampling document is different from other types of data documents because it starts at frame zero while all other types start at frame 1 Frame zero is a special frame that holds the transitory data for the sweep currently being sampled frames 1 onwards hold data that has been written to the new file Sampling is a cyclical process of collecting a new 3 16 Online analysis Sampling control panel Sampling data sweep of data into frame 0 deciding if it is to be written to disk and writing it if necessary clearing frame 0 and then starting off the next sweep This process continues until enough frames have been written to disk or until sampling is stopped Data is shown in frame zero for as long as it is the most recent data Frame zero is cleared when pulse outputs for the next frame starts or when sampling of the next frame is triggered For Basic and Peri trigger sweeps this means that data is displayed right up until data from the next sweep starts to be drawn For Outputs frame Fixed interval or Fast fixed int sweeps the frame zero data will be cleared when the pulse output for the next frame starts and the new data starts to be drawn when the sampling is triggered There may be a period when no data is shown which provides the user with
127. ate sequencing control are not present and o x Comcel __ Pun now Hop no difference in Signal s behaviour for the different states A a E mn EEE A E mn ml pl EEE CORBERA OOCK EE EE The number of extra states is set as normal while the rest of the states page is used to defines the mapping from digital input values to the frame state The Digital inputs enable checkboxes at the bottom of the dialog control which inputs are used for this unchecked inputs are ignored The main area in the centre of the dialog defines for the enabled digital input bits only the bit patterns that correspond to the various states The bits 0 to 7 are shown in a row with O on the left A blank location corresponds to a zero bit low or 0 volts a checked location gives a set bit high or 5 volts When Signal is sampling it reads the digital inputs at the end of each sampling sweep The bits read are then checked against the bits for each of the states starting with state 1 and the frame state is set to the first one that matches If no match is found then the frame state is set to zero The bits for state zero are ignored See the Sampling data chapter for details of the digital input connections Note however that if an input is not connected then it will read as high It is therefore important to disable any unused inputs OOO EREE KOO EREE DKO EREE OOK EREE OODUKEEE E MMT O Digital inputs enable 7 The buttons to the left of
128. band to differentiate and the slope The example above has a slope of 2 Within each band normally only 1 band is set the program optimises the filter so that the amplitude of the ripple error is proportional to the response amplitude A differentiator 1s normally defined to operate over a frequency band from zero up to some frequency f If is 0 5 or close to it you must use an even number of coefficients or the result is very poor You can estimate the number of coefficients required with the following function dB the proportional ripple expressed in dB e E the highest frequency set in the band even Non zero if you want an even number of coefficients func NCoefDiff dB f evens if f lt 0 or f gt 0 5 then return 0 endif f 0 5 f var n5 if even then ne dBt43 837 f 35 547 7 0 22495429 312 f ns n 1 band 2 next even number else if 0 0 then return 0 endif n dB 29 33 f ns n bor 1 next odd number endif return no end For an even number of coefficients this is unreliable when f is close to 0 5 For an odd number no value of n works if is close to 0 5 These equations were obtained by curve fitting and should only be used as a guide To make a differentiator that uses a small number of coefficients use an even number of coefficients and don t try to span the entire frequency range If you cannot tolerate the half point shift produced by using an even number of coeffic
129. be created if the Create toolbar button for protocol checkbox in the protocol definition is set and there are not too many protocol buttons the limit is 20 Pressing one of these buttons causes the relevant protocol to be executed immediately Execution of a protocol is begun by the user pressing the Run button with that protocol selected or the button for an individual protocol or by another protocol chaining to it or by means of the script language Execution of a protocol is stopped when the protocol finishes executing by the user pressing the Idle or Manual buttons in the states control bar or by another protocol being started by whatever means In addition to the standard multiple states facilities that control 1401 outputs it is possible to install support for auxiliary states hardware with Signal Auxiliary states hardware is separate external hardware most often a type of stimulator that cannot be adequately controlled using the 1401 outputs that is set up in a different way for each State Auxiliary states hardware support is provided by means of a DLL that is copied to the Signal installation directory a separate DLL is normally supplied for each type of hardware When auxiliary states hardware is installed an extra button labelled with the external hardware name is provided in the States page to allow the settings for the hardware for each state to be defined In addition the SampleAuxStateParam and SampleAuxStateValue
130. be used see the Cursor menu chapter Value ratio the product of the channel value at the time specified and the value at the reference time both using the specified width Value product the ratio of the channel value at the time specified and the value at the reference time both using the specified width Value above baseline the difference between the channel value at the time specified and the value at the reference time Only the reference time uses the specified width The channel selector and time entry fields are as standard for Signal and should be easy to use Don t forget that in addition to entering a time value directly or selecting an item such as Cursor 2 or XLow you can apply an offset to these selected value allowing you to enter Cursor 2 0 1 or XLow 1 The Points in plot field at the bottom allows you to specify the number of points this channel can contain before old points are deleted to make way for new Set this field to zero to allow all points to accumulate The New button or Change if this is used from the Process Settings command closes the dialog creates the new XY view and opens the Process dialog described above Processing for trend plots is very similar to standard memory view processing the frames specified are used to generate measurements which are added to the view data If the Clear XY view data before processing checkbox is checked all of the data points in the
131. ble Division and reciprocal of variables Copy one variable to another Move a constant value into a variable Multiply two variables multiply by a constant Move minus the value of a variable to another Subtract one variable from another Load a register from the table and store a register to the table Increment a register and branch while within the table Get the latest waveform value from a channel Get the number of points sampled into the current sweep Forces a digital marker to be sampled Records a specified digital marker Get the current sweep state code Set the state code for the current sweep Load a variable with the sweep start time Wait until a given time within the sweep Trigger the start of a sampling sweep Load a variable with the time in Signal time units Random branch with a probability Load a variable with a random number Start arbitrary waveform output Test arbitrary waveform output and branch on the result Instruction format Expressions Sequencer expression functions Sequencer outputs during sampling Blank text lines and lines with a semicolon as the first non blank character are ignored Instructions are not case sensitive Each instruction has the format num lab key code argly argZ Comment gt display num An optional step number in the range 0 to 1022 for information only Lab An optional label up to 8 characters long followed by a colon The first character must be alphabetic
132. ble y axis 1s for the top channel number in the stack To move a stacked channel to the top double click the channel number Stacked channels keep their own y axes and scaling To remove a channel drag the channel number to a new position When you drag channels and at least one of the selected channels has a y axis you can drop the channels with a y axis on top of another channel with a y axis As you drag a hollow rectangle appears around suitable dropping zones You can also drop between channels when a horizontal line appears Merged channels are drawn such that the channel with the visible y axis 1s drawn last If you have a channel that fills in areas such as a marker channel drawn as rate mode put 1t at the bottom of the stack as it will mask channels below it in the stack Hold down the shift key and move the mouse over the data area for channel 2 Hold the shift key down and click Drag up and down and release the mouse 5 Keyboard on 0 2 4 B gt 01 i a 0 0 N a ott DI ADC 1 y o A ADC 0 Y o o o La When you click with shift down the mouse jumps to the nearest channel boundary and you can change the boundary position by dragging With shift down you can move the edge up and down as far as the next channel edge You can undo changes or use Standard Display to restore normal sizes Add Ctr1 to scale all channels with a y axis If there are no channels with a y axis then all channels scale You can forc
133. c outputs mode only to produce a sequence that behaves in a different fashion according to the sweep state This instruction sets the state code for the current sweep SETS expr Vn Vnt off OptLB Sweep state expr expr This is the state code from 0 to 255 normally Vn When a variable or table entry is used to set the state the value sets the sweep state OptLb If present the instruction branches to this label This can be used instead of the normal Signal multiple states mechanisms to generate data files with separate state codes attached to the data sweeps This instruction should not be used with a sampling configuration which uses multiple states as the two mechanisms will be in conflict use it in a sampling configuration with multiple states turned off To match the built in multiple states mechanism you should restrict yourself to state codes from 0 to 255 However you can use other state code values if you wish This sets a variable to the time of the current sweep start plus an optional expression SWEEP Vn expr Vn Sweep start time expr Vn This variable is updated with the start time of the current sweep expr This optional expression will be added to the time This can be used to find the start time of the current sweep or a time within the current sweep If there is no sweep in progress the start time of the previous sweep is used 6 25 Signal for Windows version 3 WSWP TRIG TICKS Randomisati
134. c states sequencing including protocols When Signal begins sampling with the CED 3304 support enabled it checks for a correctly functioning device If a CED 3304 is found Signal will check the range switch setting initialize the device and enable the trigger While sampling is in progress Signal sets the output current using the state data before each sweep At the start and end of each sweep the CED 3304 health is checked to make sure it is OK When sampling finishes the CED 3304 trigger is disabled Because of the extra overhead in setting up current levels for each sweep and checking that the device is operating correctly you may find that the maximum sweep rate with the CED 3304 is reduced 20 10 Index 1401 Device driver 16 1 Monitor revision 16 1 1902 AC coupling 18 3 Interactive support 18 1 Trigger inputs 18 3 3 3304 auxiliary states device 20 8 3 point smooth data 11 19 5 5 point smooth data 11 19 Abort sampling 3 18 Active cursors 12 1 ADD output instruction 6 22 ADDAC output instruction 6 12 ADDI output instruction 6 22 Adding a pulse 5 3 All pass filter 17 3 All stop filter 17 3 Amplifier telegraphs 19 1 Amplitude histogram 11 2 Analysing data 2 10 Analysis menu Fit data 11 8 Analysis menu 11 1 Add to buffer 11 18 Append frame 11 17 Append frame copy 11 17 Average into buffer 11 18 Burst duration histogram 11 16 Clear buffer 11 1
135. can display it in dB if you wish The Frequency field is empty as there is only one band and it extends from 0 Hz to half the sampling rate Gain 0 5 10 15 20 25 30 35 40 45 5 Frequency Hz 17 4 FIR filters FIR filtering Frequencies Digital filtering The FIRMake FIRQuick and FiltCalc script commands and the Analysis menu Digital filters dialog generate FIR Finite Impulse Response filter coefficients suitable for a variety of filtering applications The generated filters are optimal in the sense that they have the minimum ripple in each defined band These filter coefficients are used to modify a sampled waveform usually to remove unwanted frequency components The algorithmic heart of the filter coefficient generation is based on the well known FORTRAN program written by Jim McClellan of Rice University in 1973 that implements the Remez exchange algorithm to optimise the filter The theory of FIR filters is beyond the scope of this document Readers who are interested in learning more about the subject should consult a suitable text book for example Theory and Application of Digital Signal Processing by Rabiner and Gold published by Prentice Hall ISBN 0 13 914101 e 8 e 8 8 e e x o o o 0 Waveform 0 8 o 1 O O O O O 2 o 0 O Output gt Coefficients This diagram shows the general principle of the FIR filter The hollow circles represent the filter coefficients and th
136. ccepted The Hysteresis units are y axis units per second Data This mode finds the first point in the search range where the slope changes sign Put another way it finds a localised peak or trough The Width for slope measurement field sets the data range to calculate the slope The picture shows this method used to find the top of a sharp rise where Maximum mode would get the wrong place To use this you would probably set a cursor on the peak slope and start the search from that point This method finds the start and end of a fast up or down stroke in a waveform The Width for slope measurement field sets the time width used to calculate the slope The Slope field sets the percentage of the slope at the search start to find To use this mode set a cursor at the peak slope then use it as the start point and search for the required Slope percentage 15 usually works well Data 12 2 Repolarisation Expression Delete Delete Cursor 1 0 0130226 Cursor 2 0 00124294 Cursor 3 0 00564972 All Cursors Fetch Move To Display All Label Mode Label Mode Position Number Position and Humber Mone Renumber New Horizontal Cursor Delete Horizontal Delete Horizontal gt 1 0 336456 on chan 1 2 0 476711 on chan 2 All HCursors Fetch Horizontal Cursor menu This mode finds the point at Which a osa 0 waveform returns a given percentage of the 100 distance to a baseline in
137. ce of user units as defined by the Channel parameters dialog The number at the bottom right 1s the mean level of the signal in the area marked above the number Signal conditioners differ in their capabilities Not all the controls listed below may appear for all conditioners This example is the CyberAmp control panel with the right hand half omitted The controls are This is the 1401 ADC port number whose conditioner settings are being adjusted Only ports for which a conditioner was found are shown If your signal conditioner has a choice of input options you can select the input to use with this field The choice of input may also affect the ranges of the other options This dialog field sets the gain to apply to the signal selected by the Input field Signal tracks changes of gain and offset and will change the channel scaling factors in the ports configuration to preserve the y axis scale To make the best use of the accuracy of the 1401 family you should adjust the gain of your signal so that the maximum input signal does not exceed the limits of the data displayed on the right of the control panel Some signals are biased away from zero and must be offset back to zero before they can be amplified If you are not interested in the mean level of your signal only in the fluctuations you may find it much simpler to AC couple 1902 or high pass filter CyberAmp the signal and leave the offset at zero A low pass filter reduces hi
138. ception of the Push back checkbox If this is checked increases in the pulse duration delay the start of following pulses by the same amount Decreases in the pulse duration move the following pulses earlier in time If the checkbox is clear changes in the pulse duration do not affect the time of following pulses This specifies a series of non varying square 9 pychanaogseos Md pulses This type of pulse is available for DACs Size vots 3 Start s 0 34 and for the digital outputs Pulses 10 Lenath s foo Gap s 0 01 The Size Start and Length parameters are exactly the same as for the simple pulse The extra parameters are Pulses which sets the number of pulses in the train and Gap s which sets the interval between the end of one pulse and the start of the next This item specifies a pulse with different start voltage ramp pulse art 1to0a iW and end amplitudes so that the top of the pulse starpvoks fo Start s 01 T Noretum can be sloping The variation in amplitude if Endo fi Length s 02 Sten both y Step change fo Repeats fi Steps o this is used can be applied to either the pulse start or end or both The Start and Length parameters are exactly the same as for the simple pulse the size parameter is also called Start but shows the DAC units to avoid confusion The extra parameters are End which sets the pulse level at the end of the pulse and a selector for the variation which can be set to
139. ch frame This mode of operation is called External digital states perhaps you go on to make the experiment more complicated You want four separate stimuli now maybe two different tones for each ear and rather than using a fixed stimulus order you want to randomise them Rather than design hardware complex enough to do this you can use Signal to generate outputs which control the stimulus record the outputs used with each frame as the frame state and randomise the state order The analysis is essentially unchanged apart from now producing four averages but Signal is now in control of states sequencing and much more complex patterns of stimuli are possible This mode of operation 1s called Static outputs states Finally rather than just using Signal to select tones from the external equipment you now want to generate the tones directly from within Signal using the sine wave outputs available from the pulse outputs system Signal now has multiple sets of pulse outputs one per state and as it switches state it switches between the sets of pulses to generate different outputs This mode of operation 1s called Dynamic outputs states These three scenarios briefly describe the three ways that Signal multiple states can be used and touch on some of the ways that states can be sequenced during the experiment There are many other things that can be done using multiple states but they all share a common theme different conditions in the experiment
140. changed data back into the file using the File menu Save command or you can use the File Update mode dialog to select what happens if the frame 1s changed while data is unsaved Changes made to non channel data such as the frame state or flags are always saved You will often need to specify the frame or frames to be used for an operation You can select the current frame in the view or you can enter a single frame number directly To specify more than one frame you can enter a frame list such as 1 50 60 61 70 80 or you can select options such as All frames Tagged frames Untagged frames or Frames with state n with a separate field for entering the value of n giving a wide range of possibilities These mechanisms are also available in the Signal script language The bottom edge of the data window holds five buttons and a scroll bar Try them If you resize the window the same data is redrawn to fit the window The scroll bar controls movement through the data within the current frame while the buttons allow stepping from frame to frame changing the x axis width and adding a cursor Click these buttons to move to the previous or next frame in the file CFS files contain frames which hold similar data you can use these buttons to move from one frame to another These buttons correspond to the View menu Previous frame and Next frame commands PgUp and PgDn keys there is also a Goto frame command Click this button to halve the displayed
141. ck to the operating system this may interfere with MagStim communications and cause the unit to disarm 20 7 Signal for Windows version 3 CED 3304 auxiliary device support CED 3304 Safety notice Introduction CED 3304 support configuration The CED 3304 current stimulator can be used to generate controlled constant current stimulations of up to 10 milliamps and is fully controlled by Signal allowing you to set a different level of current for each state Signal controls the CED 3304 programmable constant current stimulator using a serial line or USB port to interact with the hardware The CED 3304 current stimulator can generate outputs of up to 90 volts which lies within the potentially lethal range Before using the device you must read the CED 3304 Owners Handbook paying particular attention to the warnings and precautions section In particular you are reminded that the CED 3304 is not qualified for human use It is your responsibility to ensure that Signal s control of a CED 3304 is set up in an appropriate and safe fashion for the intended use This documentation only describes the CED 3304 controls available from within Signal for a complete understanding of the effect these have upon the behaviour of the stimulator you should consult the CED 3304 Owners Handbook The CED 3304 produces current pulses in one of four current ranges 10 and 100 microamps and 1 and 10 milliamps The current range is selected by a manu
142. cked linear if not checked The Show axes check box controls the axes of the raw and filtered data display The frequency response display shows the ideal filter as solid lines for each defined band linked by dotted lines which mark each Transition gap between the bands All transition gaps have the same width The calculated frequency response is drawn as solid lines and is greyed when the filter specification has been changed and the response has not been calculated to match The circles can be dragged sideways to make the edges of the bands steeper or less steep or you can edit the band edges as numbers in the Frequency panel on the right You can also drag the bands sideways and the band gaps The mouse pointer changes to an appropriate symbol to indicate the feature you are dragging If you edit the numbers in the Frequency field the Set button is enabled so you can force a recalculation of the filter The filters produced by the program are not defined in terms of 3dB corner frequencies and n dB per octave as is often the case for traditional analogue filters The 3dB point column is present to help users who are more comfortable describing filter band edges in terms of the 3 dB point If you check the Continuous update box the filter is updated while you drag the filter features around If you have a slow computer and this feels ponderous you can clear the check box in which case the filter is not recalculated until you stop changing fea
143. com 1 cor 2 Type of stimulator Com port for communications Com port for second unit State 1 0 dual tigger State 2 power 30 power 30 dual tigger Settings for state 1 Manual Power 5 20 Power 2 7 30 Copy to all Copy above Test Cancel Help ower 20 power 2 3 the maximum available you can set any value from 0 to 100 This control is disabled for manually controlled states This control sets the second stimulator power output as a percentage of the maximum available you can set any value from 0 to 100 This control is disabled for manually controlled states 20 5 Signal for Windows version 3 MagStim Connections and cabling Cabling for 26 way combined serial and trigger connector Cables for separate 9 pin serial and 15 way trigger connectors There are two connector arrangements provided by MagStim for serial line control and external triggering of the unit Older 200 and BiStim units have an arrangement with a 9 way Serial line and separate 15 way trigger connector while newer 200 and BiStim units and all Rapids have a high density 26 way combined serial line and trigger connector In both cases two separate connections need to be made serial line control from a PC and a TTL trigger signal from the 1401 The standard high level trigger input suggested by MagStim works well with Signal and the 1401 and is used in the diagrams shown below consult the M
144. ct then press and hold down the left mouse button The mouse pointer may change to indicate the start of a drag operation and an icon may be attached to the pointer to indicate that the object has been grabbed Still holding the mouse button down move the mouse pointer to drag the object to its required destination The mouse pointer may change while you do this a common effect is for the pointer to change to a no entry sign a circle with a diagonal line through it to indicate that you cannot drop the object at this point When the object is at the required destination release the mouse button to drop it in place To add a new pulse into the outputs select the pulse required from the pulses area of the dialog Drag the new pulse from the pulses area into the display the mouse pointer changes to a a hand with a plus sign to indicate that you are adding a new pulse As you move the mouse pointer about the display a vertical line indicates where the new pulse will go and the drop time is displayed in the control area Once the pulse is correctly positioned drop it into place by releasing the mouse button If you cannot get the display area to accept the new pulse the mouse pointer is always No Entry this could be because of the following reasons e You are trying to drag a DAC only pulse into the digital outputs e You are trying to drag an arbitrary waveform item into a set of outputs that already contains an arbitrary wavefo
145. d set the slowest sequencer step rate that is fast enough for your purposes If you overload the system so much that the output sequencer cannot keep up Signal will warn you of this when sampling finishes If the 1401 is extremely overloaded by the combination of sampling and sequencer output sampling will fail with an appropriate error message The clock interval that you set is rounded to the nearest 4 microseconds to generate the actual sequencer tick interval For modern 1401s the Power1401 and Micro1401 mk II intervals less than 5 milliseconds are rounded to the nearest 0 1 microsecond All subsequent timing is based upon the rounded interval so that full accuracy is maintained The output sequence is stored as a text file with the extension PLS You can open existing Signal output sequence files or create new ones with File menu New There are five buttons at the top of the window Formatting Done MOVI 1 10 Initialise loop counter DELAY 246 Just less than 0 25 s gt Waiting LP DAC 0 2 Set output high DELAY 8 Wait 9 steps DAC 0 0 Set output low DELAY 7 Wait 8 steps Loop required times Stop outputs gt Done DBNZ 1 LP HALT This checks the sequence to make sure that it is correct with no labels missing or duplicated and no duplicated key codes The picture shows sequence with a simple error the LB in the seventh line should be LP The line in error is marked and an explanatory mess
146. d view The search is normally insensitive to the case of characters if I Match Case Direction Mark All E Up you want to Match Case in a search I Regular expression P Cancel check the box The search starts at the M Wrap around search cursor position and stops at the end of the document unless the Wrap around search box is checked in which case the search continues at the start of the document after reaching the end Searches are line based you cannot search for text that spans more than one line If you check the Regular expressions box you can include pattern matching characters in your string The simplest pattern matching characters are Start of line marker If you include this character it must be at the start of the search text The following search text is matched only if it is found at the start of a line End of line marker If you include this character it must come at the end of the search text The preceding text will only be matched if 1t is found at the end of a line Matches any character To treat these special characters as normal characters with regular expressions enabled put a backslash before them A search for would find all lines with a as the first character anything as a second character and a period as a third character To search for a character list enclose it in square brackets for example aeiou will find any vowel For a character range use a hyphen to link t
147. data will be saved or not according to the file data update mode The behaviour of the modifications themselves are mostly straightforward Subtract DC measures the mean value of the channel data then subtracts this DC offset value from all data points Normally the DC level is measured over the visible frame area the X range for the DC measurement can be set using the Area of DC item Differentiation replaces each data point with the difference between that point and the previous point and divides the result by the sample interval the first data point is set to zero Integration replaces each data point with the sum of all data points up to and including that point multiplied by the sample interval 3 point and 5 point smoothing replace each point with the average of the 3 or 5 points centred on that point The scale and offset data options provide a dialog in which a numeric value can be entered Scaling the data multiplies each data point by the number entered offsetting adds the number entered to each data point The shift data option rotates data points by shifting data from one time to another within the frame Note that this operation rotates data points that fall off one side of the frame are shifted back in on the other side so the operation can be reversed without loss of data There are special shortcuts Shiftt lt and Shift gt to shift left and right by one point The last four options are for inter channel arithmetic A channe
148. defining which state is used when Number of extra states 4 The simplest way to do this IY Individual repeats Cycles before idle E is to control the state State 3 data manually which can be Repeats fi done using a control bar but often we want some form of automatic states sequencing Signal provides essentially two separate forms of states sequencing numeric or protocol In the numeric ESA Bunion Help modes there are three of these simple numeric or randomised sequencing 1s provided while in protocol mode a precisely defined complex sequence of states can be generated In non protocol states sequencing modes the user is limited to specifying how many of each state are to be used and how they are to be randomised if at all The Individual repeats checkbox selects if each state should be repeated a different number of times or if all states are repeated the same number of times If it 1s clear the Repeats item is used to set how many times each state is repeated If the checkbox is set separate controls are used to set the repeats wanted for each state see below The Cycles before idle item sets how many times the full set of states states repeats is repeated before states sequencing is automatically stopped and Signal switches automatically to state 0 Set this 1tem to the number of sequencing cycles you want or to 0 if you want states sequencing to repeat forever until stopped 1 2 Individual state rep
149. distribution about the correct value To allow us to calculate likelihoods we assume that this probability distribution is a normal Gaussian distribution In the real world or course only some of this may apply You may have no a priori knowledge of the distribution of errors in your original data and this distribution may be anything but normal In the ideal case where you know the standard deviations of each data point the fitting minimises the chi squared value which is the sum of the squares of the differences between the model and the data points divided by the standard deviation of data point values Given a chi squared value and the number of points it was measured from we can calculate that probability of getting a chi squared value at least this large due to random variations in the data This is the value given in the Results tab Ideally you would like to see a value around 0 5 meaning that you were equally likely to get a larger value as a smaller one Values very close to 1 mean that given the errors in each point the data is too close to the model Either the error estimates are too large or the data has been improved Although you can hope for probabilities in the range 0 1 to 0 9 values down to 0 01 may occur for acceptable fits and even smaller values can occur 1f your error distribution 1s not as normal as you thought Very low fit probabilities will occur if e E A e3 your data contains variations that are ia a a sig
150. do not need to make the connection to follow this description Click the Run now button in the sampling configuration and then click the Start button The sequencer control panel is now visible Ster 000 Glosa It displays Step 0000 in the top left window and a blank area to E po the right If the control panel is not docked there is also an area for comment display currently clear If you start sampling you will see that the control panel displays the text after the gt when the relevant steps are being executed if you are sampling the output you will see the ten pulses being generated As this is our first sequence we will explain it in detail Click the Format Initialise loop counter Just less than 0 25 s gt Waiting and add step numbers i Set output high Wait 9 button at the top of the a 1 sequencer window All of eee eae i S Loop required times the lines get a step number Stop outputs starting at zero this is the step number displayed in the control panel You can remove step numbers with the Format button Step 0 is executed at time zero in each sweep The MOVI V1 10 instruction on this line sets the value of variable 1 v1 to 10 This variable is used as a loop counter in the sequence so this statement sets the number of times that execution will go round the loop The remainder of the line is a comment The next line step 1 holds the instruction DELAY 248 This causes sequencer execu
151. docked the control panel displays sequence entry points as the key that activates them and a descriptive comment When docked the keys are displayed as buttons and the comment is hidden to save space Move the mouse pointer over a key to see the comment as pop up text Click the mouse on a key and the sequencer will jump to the instruction associated with the key The key is also stored as a keyboard marker This is equivalent to pressing the same key in the time window or using the script language SampleKey routine The control panel is always displayed if you start sampling using Sequencer outputs from a Signal menu command or from a dialog If the sample command comes from the script language the control panel visible state does not change The control panel position is saved in the sampling configuration However the position is restored only if the control is currently invisible or floating and the saved position was floating if the control panel is docked we assume it is positioned where you want it The sequencer clock starts within a microsecond of recording time zero the time at which the Start button is pressed and is time locked to the 1401 sweep timing and waveform channel recording Each clock tick books an interrupt to run the next sequencer instruction and updates digital output bits 15 8 if they were changed by the previous instruction 6 1 Signal for Windows version 3 The sequence editor Compile An in
152. due to the Pr aa gt om A sae wrong model being used becomes lc a AAA much larger than the errors due to uncertainty in the data values and you get a residual plot like this one This area displays a summary of the fit information that you can select with the mouse and copy to the clipboard The first line holds the type of the fit the channel number the ordinate range and the number of points in this range For example Double exponential fit channel 1 0 to 50 50 points The contents of the second line depend on the source of the data If you are fitting a result view channel that has error information displayed the second line displays the chi squared error value for the fit and the probability that you would get a chi squared value of at least that size if the function fits the data and the errors are normally distributed For example Chi square value 58 6 probability 0 5867 In all other cases the second line displays the sum of the squares of the errors between the data and the fitted function and an estimate of the probability that you would get a sum of squares of errors of at least this size based on the assumptions that the errors in the original data had a normal distribution that was the same for all points For example Sum or errors squared 1 22 estimated probability 0 8553 If the probability value is very low or very high there are extra lines of information warning that the fitted function plus normally distrib
153. e Signal auxiliary states device support provides a mechanism for controlling arbitrary equipment and setting it up in different ways according to the sampling state in use In addition to controlling equipment settings by state Signal is also able to provide device specific configuration dialogs to check the equipment health and readiness while sampling to save the stimulation parameters used in the new data file and to save and load auxiliary states setup information to and from sampling configurations Script language functions to access auxiliary device settings are also provided Signal currently supports two types of auxiliary states device the MagStim range of transcranial magnetic stimulators and the CED 3304 programmable constant current stimulator You can select which of these devices you wish to use or none during the Signal installation process if you wish to change the type of auxiliary states device in use the simplest way to do this is to re install Signal If you need auxiliary states device support for other equipment please contact CED The MagStim range of transcranial magnetic stimulators are widely used to provide non invasive neuronal and muscular stimulation Signal provides support all three types of MagStim the 200 BiStim and Rapid as the software uses a serial line for control of the stimulator only the modern MagStims that provide serial line control these have a suffix to the model name can be used In add
154. e Y axis dialog Ctrl Y Scroll left right Cursor Left Right Decrease X range increase range Ctrl Left Ctrl Right Show all X range Ctrl Home X axis range dialog Ctrl1 X Next frame previous frame PgUp PgDn First frame last frame Zoom un zoom channel Hide selected channels Customise display Ctrl PgDn Ctrl1 PgUp Double click Del Ctrl Del Some of these shortcuts are documented with the appropriate menu command others do not have an equivalent command All display shortcuts are listed here for convenience There are more shortcuts provided for data manipulation see the Analysis menu chapter 10 8 New Memory View Ha Waveform Average Analysis menu The Analysis menu creates memory or XY views that hold analysed data from data channels from other data documents and provides access to other analysis functions The menu is divided into seven regions The first region holds commands that operate on existing memory or XY views The second is used to create new memory or XY views holding data created by built in Signal analysis mechanisms It also contains the curve fitting option The third region holds commands to append new data frames to documents and to delete appended frames The fourth region holds commands that use the frame buffer It also includes the multiple frames dialog which carries out operations on many frames The fifth region holds channel data modification commands The sixth region co
155. e 10 1 Button 2 3 NOP output instruction 6 20 _ Q Offset data 11 19 OFFSET output instruction 6 17 One and a half high pass filter 17 4 One and a half low pass filter 17 4 Online clamping support 4 1 Online data processing 11 6 Online pulses control 5 9 Online update of memory view 11 6 Open Closed times 11 13 Opening Configuration file 8 5 New document 8 1 Old document 8 2 Optimise Y axis 10 3 Options for XY view 10 4 Order of channels 9 4 Output sequencer 6 1 Access to data capture 6 24 Add constant to variable 6 22 Arbitrary waveform output control 6 28 Calculate variable values 6 8 6 10 6 11 6 18 Compare variable 6 21 Compile sequence 6 2 Compiler errors 6 28 Control panel 6 1 Copy variable 6 21 DAC outputs 6 12 Disable interactive jumps 3 12 Divide variable 6 22 Example 6 3 Expressions 6 7 Format text 6 3 Format with step numbers 6 3 Get current sample points 6 24 Get current sample time 6 26 Get current sweep state 6 25 Get sweep start time 6 25 Instruction format 6 7 Instructions 6 6 Multiply variables 6 23 Negate variable 6 21 Randomisation 6 26 Reciprocal of variable 6 22 Set file to use 6 3 Set state of current sweep 6 25 Set variable value 6 21 TABDAT directive 6 9 Table of values 6 9 TABSZ directive 6 9 Index 4 Index Trigger sweep 6 26 Variable sum and difference 6 22 Variables 6 8 Wait till time in sweep 6 26 Outputs Absolu
156. e described in the Script menu and Sample menu chapters The debug bar is Soke described in the script manual Sample bar These commands the buttons at the bottom left of the data window and the shortcut keys PgUp and PgDn change the current frame to the next or previous frames in the data document If the current frame is the first or last frame in the document then the corresponding menu item and button are greyed out The Ctr1 PgUp and Ctrl PgDn shortcuts change to the last or first frame in the document when sampling Ctr1 PgUp switches to showing the last frame filed until the frame is manually changed This command shortcut Ctr1 G opens a dialog that gets the frame number to move to This command shortcut Ctr1 B toggles between showing the frame buffer and the current frame When the buffer is shown the menu item is shown checked and the view title is modified to show that the buffer is visible The frame buffer is a separate frame of data behind each open CFS file which is maintained by Signal See the Analysis menu chapter for more details of the frame buffer This command shortcut Ctr1 D switches the display between normal mode displaying the current frame and overdraw mode which also displays all frames in the frame display list When overdraw mode is enabled the menu item is shown checked The frame display list holds frames to display in addition to the current frame These additional frames use different colou
157. e all channels to scale by lifting your finger off the Shift key leaving Ctr1 down after you start to drag the boundary So far you have been looking at windows holding data read from a disk file We call these File views There is another type of data window called a Memory view which holds data created by the Signal program that is held in memory This data is usually the result of some sort of analysis When a memory view 1s saved to disk and then re loaded 2 9 Signal for Windows version 3 it has then become a file view the two types of view are very similar A simple way to create a memory view is by analysing file view data There are two steps in the analysis 1 You set the type of analysis the channels to analyse the width or number of bins of the analysis result and any other parameters required This creates a new empty memory view with the appropriate frame width and channels 2 You define the frames from the file view that are to be analysed and Signal carries out the analysis and adds the result into the memory view You may repeat step two as many times as required to accumulate results from different sets of frames of data You can use the Analysis menu to add additional frames to the memory view Each frame can hold the result of analysis of different frames from the original file One way of using this would be to separate averages for each frame state in the source file Processed memory views will be
158. e new values corresponding to the amplifier settings In all cases you should note that for example data drawn at 3 mV would continue to be drawn at 3 mV It is only the axis limits that may change Programmable signal conditioners see the Programmable signal conditioners chapter are controlled through communication serial ports Check the Dump errors box to write diagnostic messages to CEDCOND LOG in the current folder The lower half of the dialog 1s used to display the current status of the conditioner This tab contains a couple of check boxes regarding the script editor Save modified scripts when run can be checked so that any changes to a script will be automatically committed to disk each time the script is run If Enter the debugger when a script has a runtime error is set then if a runtime error in a script happens then the script view will be displayed with the debugger running This allows the user to check on the values stored in the variables as well as looking at the call stack This tab limits the processor time consumed by the Signal user thread while sampling and idling with a script running If Signal takes too much time the system feels unresponsive It does not affect the time critical thread that writes sampled data to disk You can read more about threads below Signal runs a Cancel Apply Help background routine when it has idle time and also periodically on a timer The routine handles the fo
159. e option from the context menu Lock axes and Group offset are visible when the current channel shares a y axis with other channels They are enabled when the channel is the first in the group If you check Lock axes the grouped channels not only share the same space they also share the y axis of the first channel in the group The Group offset field sets a per channel vertical display offset for each locked channel to space out channels with the same mean level Draw applies changes to the Top Bottom Lock axes and Group offset fields Cancel undoes any changes and closes the dialog Close accepts changes and closes the dialog The remaining options are all identical to those already described under X Axis Range see above Standard Display This command sets the current data memory or XY view to a standard state In file and memory views 1t turns on all channels in a standard display mode and size and ordered as set in the Edit menu Preferences with x and y axes on and grids off In an XY view all channels are made visible the point display mode is set to dot at the standard size the points are joined and the x and y axis range 1s set to span the range of the data Customise display This command opens a dialog N w that sets the channels to display M Chan numbers I Grid MY Axis MX Axis Ve in a data or XY view and the AE keyboard Marker e Onn pega Y Numbers Y Numbers display of x and y axes g
160. e pointer lingers over a button F TOOLMAKE Bindump This command opens the Script CAsignaliSCRIPT 5 calib sqs Delete List dialog from where you can TOOLMAEE C siqgnal SCRIPT SS TOOLMAKE q3 contr ol th e contents o f the Scri t Bindurnp CAsignal CRIPT S46 indump sa Clear All p Bar or The Add buttons opens a file dialog in which you can choose one or more Signal script files Cancel SGS to add to the bar If the first line of a script starts with a Label TOOLMAKE Conment Wite a oobar diven serpi skeleton i SS ae C signalsSCRIPTS TOOLMAKE sas interpreted as a label and a comment otherwise the first 8 characters of the file name form the label and the comment is blank The label is separated from the comment by a vertical bar The label can be up to 8 characters long and the comment up to 80 characters A typical first line might be Pl SToolMake Write a toolbar driven script skeleton You can select an item in the list and edit the label and comment This does not change the contents of the script file You can re order buttons in the bar by dragging items in the list The Delete button removes the selected item Clear All removes all items from the list The list of files in the Script Bar is saved in the registry when Signal closes and is loaded when Signal opens Each different logon to Windows has a different configuration in the registry so 1f your system has three different users each has their
161. e solid circles are the input and output waveforms Each output point is generated by multiplying the waveform by the coefficients and summing the result The coefficients are then moved one step to the right and the process repeats From this description you can see that the filter coefficients from right to left are the impulse response of the filter The impulse response is the output of a filter when the input signal is all zero except for one sample of unit amplitude In the example above with 7 coefficients there is no time shift caused by the filter With an even number of coefficients there is a time shift in the output of half a sample period The Analysis menu Digital filters command deals with frequencies in Hz as this is comfortable for us to work with However if you calculate a FIR filter for one sampling rate and apply the same coefficients to a waveform sampled at another rate all the frequency properties of the filter are scaled by the relative sampling rates That is the frequency properties of an FIR filter are invariant when expressed as fractions of the sampling rate not when expressed in Hz It 1s usually more convenient when dealing with real signals to describe filters in terms of Hz but this means that each time a filter is applied to a waveform the sampling rate must be checked If the rate 1s different from the rate for which the filter was last used the coefficients must be recalculated Unless you use the FIRM
162. e the Browse button to select the sequence file directly This section contains controls that define how DAC outputs are calibrated for sequencer and arbitrary waveform output It also sets the arbitrary waveform output rate length and DACs used Both the sequencer and arbitrary waveform output assume that all the DACs have the same scale factor and zero setting Full scale This defines the full scale output level of the DACs in the units that you wish to use corresponding to a full scale output from the DAC This item along with the Zero value is used to convert from the user units entered into the sequence into actual DAC values Zero value This defines the value in your preferred units corresponding to a zero volt output from the DACs This value is usually 0 Waveform points This defines the length of the arbitrary waveform storage area in points Values from 2 to 10 million can be entered Output rate Hz This defines the rate at which the arbitrary waveform is played out through the DACs In conjunction with Waveform points this sets the maximum duration of waveform replay Values from 1 to 1 million can be entered The maximum achievable rate depends on the type of 1401 and the number of DACs used Waveform DACs These checkboxes set which DACs will be used for waveform output If one DAC 1s selected the waveform data consists of a list of values for more than one DAC the data for the DACs is interleaved Only DACs 0 to 3 can b
163. e to the trigger Sweep triggers are normally on a falling edge of a TTL pulse Check this box to make them on the rising edge This checkbox sets the initial state of the Write to disk at sweep end checkbox in the sampling control panel When this is set sampled sweeps are automatically written to disk when the sweep finishes This checkbox sets the initial state of the Pause at sweep end checkbox in the sampling control panel When this is set Signal waits at the end of a sweep instead of immediately starting the next sampling sweep The Peri trigger section holds peme information that is specific to the Peri trigger sampling mode It is only available when Peri triggered sweeps are selected in the General section General Pert trigger Fort setup Outputs Automate General Trigger type Analogue Pre trig time 3 E Trigger values for ADC port 1 Threshold level o Hysteresis 25 At the top of the dialog 1s a selector for the type of trigger and a field for the pre trigger points Below this is a section holding details of the trigger parameters The contents of this section changes with the type of trigger the individual fields will be described along with the various types of trigger Cancel Run now Help This can be set to one of Analogue Analogue Analogue Digital or Event The three analogue types monitor the last ADC port in the sampled ADC ports list for a trigger The trigger lev
164. e to the current sampling time in sequencer clock ticks and adds an expression or 0 if expr is omitted The s ms and us expression functions can be used to make the sequence independent of the clock rate ELCKS Vn expr gt Va Srgnal Eume F expr Vn This variable is updated with the current time expr This optional expression will be added to the time This can be used with the CHAN command and variable related branches to check the timing of external pulses The sequencer runs under interrupt and competes for time with other interrupt driven processes in the 1401 interface This causes some Jitter in the timing The jitter for a Microl401 or Power1401 is typically only a few microseconds For a 140 1plus it can be a few tens of microseconds depending on other 1401 activity These functions use a pseudo random number generator The generator is seeded by a number that is based on the length of time that the 1401 has been switched on BRAND branches with a probability set by the argument or by a variable This could be used when several different stimuli are required but in a random sequence BRAND LB expri Vn Vn off LB Where to go 1f the branch is taken expr This is the probability of branching in the range 0 up to but not including 1 6 26 Poisson process Scripts and variables Sequencer outputs during sampling Vn When a variable or table entry is used for a branch the value is treated as a 32 bit u
165. e used for arbitrary waveform output 3 12 Sampling data Pulses or sequencer Pulses output is easier to get started with and supports multiple states directly it will be Basic Peri trigger Outputs frame Fixed interval Fast triggers Fast fixed interval suitable for most users of Signal You should consider using the sequencer if you are unable to achieve the effect you require with Pulses output The table below summarises the main differences between these two forms of output Graphical sequence Text sequence Edited with Built in graphical editor Built in text editor Visualise output Yes No Stored as Part of the sampling configuration Output sequence PLS files Implemented by Drag and drop editing Machine code like language Ease of use Very easy to learn and use Takes time to learn Flexibility Uses pre set building blocks All features available User interaction Script interaction Arbitrary waveform Sweeps Timing Buttons trigger jumps Variables and table data Data loaded by script Can be sweep independent One instruction per text line Pulse editor while sampling Add delete and modify pulses Data in sampling configuration Locked to sampling sweeps Several instructions per item Though the way in which the required outputs are defined in Pulses and Sequencer outputs are very different the actual generation of outputs is carried out identically When you use Pulses outputs the pulses information
166. eat counts Sampling with multiple states The Ordering selector to the top right of the states page selects the type of sequencing to be used this can be set to Numeric In this mode the extra states are used in numerical order with each state being used a specified number of times as set by the overall or individual repeat counts First state 1 is used n times then state 2 and so forth Once the last state has been done one sequencing cycle has been completed and the sequencing either stops or restarts with state 1 as set by the Cycles before idle control Random In this mode one cycle of the sequencing again uses each state the number of times specified by the repeats but the order of the states is randomised So with two states and two repeats each cycle lasts for four frames two of which are certain to be state 1 and two state 2 but in a random order When the sequencing starts another cycle the states order is re randomised Semi random This is a slight modification of Random mode where the repeats are not all randomised together across a cycle If there are n states then the first n frames will contain one of each state in a random order as will the next n and so forth So with 2 extra states and two repeats the first two frames will contain one each of states 1 and 2 in random order and the next two frames will again contain one each of state 1 and 2 again randomised So really this achieves the same thing as setti
167. ecked and you want to process only un tagged frames see below then Signal will scan the data file for enough un tagged frames to form each average If it is checked then tagged frames are not included in the average but are included when working out which frames to add Thus some of the averages would be formed from less frames Averages formed from no frames are set to zero If you check the Error bars box extra information is saved with the result so that you can display the standard deviation and standard error of the mean of the resulting data The Waveform Draw Mode dialog controls the display of the error information The New button or Change if from the Process Settings command closes the dialog creates the new memory view and opens the standard Process dialog described below This analysis creates a memory view that is a x histogram of the number of data points found at a 1 ADCO y Channel particular amplitude range of a channel in the file A time range 1s given in which the data will be analysed The Maximum and Minimum amplitudes determine the start and end points of the x axis in the memory view These fields of the dialog are linked to the Bin size and Number of bins If the amplitude range is changed the Bin size changes to fill the new range and keep the Number of bins constant If the Bin size is changed then the Number of bins changes to Data start time Maximum amplitude 5 000000 Y Minimum amplitude 5 000
168. econds and S 1 is the built in function that tells us how many steps there are per second If you would rather express this in terms of a rate then prob 1 0 rate S 1 where rate is the expected rate in Hz TENHZ BRAND TENHZ 1 0 10 S 1 10 Hz mean rate DEGOUL Tainaa 1 set output high DIGOUT PRA O TENHZ set output low goto TENHZ This example generates a digital output that pulses to produce an approximation to a Poisson distributed pulse train with a mean frequency of 10 Hz The approximation improves the shorter the step time The mean interval between pulses is 100 milliseconds plus the time for 2 steps and the shortest gap between pulses 1s 3 sequencer steps From a script you can set sequencer variables as 32 bit signed integers For the range 2147483648 to 4294967295 we must use negative numbers This script example shows you how to convert a probability into a variable value and pass it to the sequencer Proc GetBrandVar prob vo prob 1s probability ve 18 variable prob 4294967296 0 range 0 4294967296 is 0 to 1 0 if prob gt 4294961235900 then prob gt 42949672 95 0 endif if prob gt 2147483647 then prob 4294967296 0 endif if prob lt 2147483647 then prob 2147483647 endif SampleSeqVar v prob end 6 27 Signal for Windows version 3 Sequencer compiler MOVRND Arbitrary waveform output WAVE WAVEBR error messages This instruction generates a random number
169. ed script and if there dll lalo are no errors runs the script If you run a script Tum Recording On Bindump twice in succession Signal only compiles it for the Debugbar ar first run saving the compilation time If a script Es stops with a run time error the script window is Sept List brought to the front and the offending line is highlighted Peakaverage You can also select the Load and run option from which you can select a script to run The script is hidden and run immediately unless a syntax error is found in it This command and the Ctrl L keyboard shortcut open the Evaluate dialog where you can type a line of script commands for immediate execution The window remembers the last ten lines of script entered which are shown in the drop down list You can cycle round the saved lines using the lt lt and gt gt buttons The Execute button executes the line entered Eval adjusts the line to include an Eval on the last statement so you can see the result You can execute any script that can be typed in one line which can include variable declarations Evaluate a line of text EvallChanMeasure 1 1 O 00078 0 4986 a lt lt Cancel Eval Execute You can record your actions into a script that will produce equivalent actions Use this command to turn recording on and off When you turn recording on Signal begins to save script commands corresponding to your actions While
170. eee 11 18 Add to DUI Sessa sa a cade 11 18 Subitact DU tai da a o adds 11 18 Average alo DUE dd 11 18 Mia e OS 11 18 Modit CAIN Sd dni 11 19 A ni 11 19 Distal Ms a oa 11 20 Keyboard analysis Control ooooonnnnnnnnninininnnnnnnnnonononannononononnnn nono nono nono nono nono nana n nana n RR 11 20 O 12 1 SO A E EIEI O 12 1 O 12 1 DA a Rte PN one ROUTED E OCR eee PN ee dis 12 3 Pe ien ta oy ac soca cea ee cold ee essa each se cold coat ated AA 12 3 Move MON soos oot caress ci ence cae ane eee 12 3 Dispara A e Ste eae 12 3 DS E arua AA 12 3 ys A A E EEEE DN 12 3 New Horizontal CUTS O a A AER E E 12 3 Delete horizontal ade 12 3 ELC NOTI ZONA atada 12 3 MOVE LOCC Fonar aE T E E T E 12 4 Displaralhorizontalas dd 12 4 Honzontal labe MOS A da 12 4 yi Table of Contents Reonmber dE here tae 12 4 Display VALS o E e de Cele 12 4 Cron Ten ON S ie e a a a cence Sauna a a anaes 12 5 Sample MEN Uers 13 1 So A II E OA 13 1 Sample Bar Sa Eos 13 1 Ss eee eo ee 13 1 Show Sainpline controls aii 13 2 SHOW PUSS CONG reson OD ER SO N E E T R 13 2 SAPIENS ES 13 2 SHOW SEQUENCER COMO lic 13 2 104 9188 9 4 1 1 9 0 COPEC o einem eek man ee er mee mene mee nee ete eae eee eee eee eee eee 14 1 COMPA a ee eee ae oe eee eee eee ee eee 14 1 A ee en A eee EE 14 1 E es ee A ee ee 14 1 TT RECO ASA A A A NATA 14 1 DEDO or OY ee EME te et a Oe pee ee ea 14 2 o E alsa dares Gaal N E aac 14 2 A 14 2 WINdOW Mel iriiria aaa aAa a a 15 1 TUT CALS WIN
171. els are shown with the sampled data as a pair of cursors which can be moved without stopping the sampling to alter the levels The Digital trigger waits for a specified state on a bit in the 1401 digital inputs while the Event trigger is a TTL pulse just as for the Basic sample mode triggers Each form of trigger has different parameters Analogue Trigger on a positive going level transition The parameters are Threshold level and Hysteresis both in units set by the channel calibration The trigger process first waits for the sampled data to go below Threshold Hysteresis and then triggers when the sampled data value rises above Threshold The hysteresis acts to prevent false triggering by noise as the sampled data passes downwards through the threshold level triggering can only occur after the sampled data has clearly been below the threshold If you find that you are having problems with false triggers due to noise increase the Hysteresis value Analogue Trigger on a negative going level transition This is identical to Analogue but in the opposite direction The trigger process first waits for the sampled data to go above Threshold Hysteresis and then triggers when the sampled data value falls below Threshold Analogue Trigger on signal moving outside a pair of levels The parameters are Upper threshold and Lower threshold The trigger process first waits for the sampled data to go between the thresholds It then monitors the samp
172. endent of each other When a window is duplicated the cursors are also duplicated This menu command keyboard shortcut Ctr1 duplicates the action of the new cursor button at the bottom left of data views The command is available when a data view is the current window and there are less than ten cursors already active in the view A new vertical cursor 1s added at the centre of the window The cursor is given the lowest available cursor number and is labelled with the cursor label style for the window This menu command opens a dialog from which you can select a cursor and its search mode The command is available when a data view is the current window and there are cursors present in the view Normally Signal cursors are static they stay r where they are put Using the cursor mode ee uo BS dialog a cursor can be made active it will move to the position of a data feature if it can dd Threshold El be found This search and move occurs when Search channel 1 ADC 0 waveform the view changes to a different frame or when Start position for search s Cursor 1 the current frame data is changed End position for search ls fxHihD y 25 The cursor mode dialog allows you to define Mlve 100 the feature searched for the channel to be Mess rejection hysteresis searched the limits to the search and to set parameters defining the feature being searched Pesitionifsearch fails s fo for The Curso
173. ent over time Changing pulses Analysis methods In addition to averaging across all stimulus transitions you can average the analysis results that are displayed over a number of sweeps by using the Average of readings control This can be set to any value from 1 to 1000 The area below these controls provides a display of a selected measurement over time The Display selector selects the measurement to be displayed from those available while the count control sets the number of measurements shown over the width of the display area from 10 to 1000 The values displayed in this graph are updated with data from every sweep the Average of readings control has no effect but the A11 transitions checkbox operates The Clear button deletes all saved measurements and restarts the display this can be particularly useful if an aberrant value causes the automatic Y scaling to show too large a data range When the membrane analysis dialog is closed the sampling state sequencing system and writing to disk are both restored to their previous state at the time that the dialog was displayed The pulses dialog can be used to directly edit the pulses including the pulse used for resistance measurements at any point while the experiment is in progress Any changes made to the pulses while the membrane analysis dialog is in use will not cause the membrane analysis to fail in a destructive fashion but deleting the pulse for resistance measurem
174. ents or changing it to a type that is not usable for resistance measurements will prevent the membrane analysis from being carried out Resistance and other membrane measurements are generated by analysis of a transition a step change in stimulation clamped voltage or current level with the new level being maintained for a reasonable period For a simple pulse there are two transitions one at the start of the pulse and one at the end To measure total resistance the static level attained after the transition 1s measured over the final quarter of the stimulus pulse for transitions at the end of a pulse the previous pulse width is assumed The delay of 34 of the pulse width is intended to allow the Capacitive transient to settle this settling time can be adjusted by altering the stimulus pulse width The baseline pre transition level is measured over the same period just before the transition Measurements are taken from both the stimulus and response channels allowing an easy calculation of resistance using Ohms law Once the static levels before and after the transition have been measured voltage clamp data may be analysed further Firstly the response data is searched to find key positions the position and value of the peak overshoot and the points at which the overshoot drops to 90 and 5 of the peak value The decay phase of the overshoot is analysed by fitting a straight line to the log current values against time with all values be
175. er It is used to copy sets of pulse data from the currently displayed state to other states See below for more details on using this This pair of controls is only visible when multiple states controlling dynamic outputs are in use see the Sampling with multiple states chapter They are used to select the state that the dialog shows you can display and edit pulses from only one state at a time This control is also only visible when multiple states controlling dynamic outputs are in use see the Sampling with multiple states chapter It is used to set a label for the currently displayed state a descriptive name for the state up to 10 characters long that will be used in the state control bar buttons and elsewhere within Signal If you leave this field blank the standard state names Basic 0 State 1 as shown in the state selector above will be used Click on this control to see the effect of pulse variations While you hold down the mouse button on this control the limits to pulse variations plus three intermediate values are displayed Click on this button to animate the display to show the effect of pulse variations The display updates to show the effect of each variation in turn Click again on the button to turn animation off 9 2 Pulses Values Dragging and dropping Adding a new pulse Pulse outputs during sampling This is the area at the bottom left of the dialog It holds pulse icons which can n amp
176. er If an analogue Bessel filter is used it 1s worth noting that the 3 dB frequency is often about half the cut off frequency set on the front panel of the filter which uses a different definition for the cut off Sometimes a transition takes longer than expected for a given filter cut off When this happens Signal can either insert an event at a sub level or insert two full height transitions in opposite directions in order to make the resulting convolution approximate to the raw data With the Avoid sublevels checkbox checked then the latter option will be used whenever appropriate though it does not guarantee that no sub levels will be fitted New will take you to the standard process dialog described above but will not generate a result view The results are drawn instead in the file view on top of the original waveform The algorithm used by the SCAN MPRI technique is highly complex The Advanced button allows various parameters used in the trace formation to be changed The basic 2 14 4 eI et Mini tude change of a transiti al principle of the trace formation is id dd J 5 3 60 3 Number of points past critical level for transition Percentage of full amplitude for critical level basar multiple of rise time that might not be a sub level Sensitivity for multimple transitions ve sublevel that an average 1s kept of the point amplitudes of the data When a number of consecutive points fall outside a critical level
177. er interval or the maximum histogram value for Rate display mode Minimum The minimum inter marker interval or the minimum histogram value for Rate display mode Amplitude The difference between the Maximum and Minimum values Extreme The largest absolute value of Maximum and Minimum this will always be the same as Maximum for marker channels You can select areas of this window by clicking them Hold down the Shift key for extended selections You can select entire rows and columns by clicking in the cursor and channel title fields Use the Ctr1 key to select non contiguous rows and columns To copy selected rows and columns to the clipboard right click in the values window and use the Copy command in the popup menu If you use the Log command the selected text is copied and pasted directly into the log window in one operation You can also print the selected portions of the window by right clicking and using the Print command in the popup menu or use the Font command to change the window font The above popup menu commands are also available via the following keyboard shortcuts Ctrl c Copy Ctrl P Print Ctrl F Font Ctrl L Log In addition to the main menu commands it is also possible to access some commands by right clicking on a cursor This produces a popup menu which has a sub menu specific to the cursor which has been clicked on For vertical cursors the sub menu has the following items Copy The position of the selected cu
178. erforms the same function as RATE except that the change 1s postponed until the next time Theta passes through 0 degrees RATEW cannot start output a sinusoid must already be running to pass phase 0 It can stop output but does not remove the overhead for using cosine output This instruction clears the new cycle flag see WAI TC RATEW n expriVn Vn off DAC n n DAC number in the range 0 7 available DACs depend on the 1401 type expr The angle increment in the range 0 000 to 180 degrees The Hz built in function calculates the increment required for a frequency Vn For a variable the value 11930465 is an increment of 1 degree The VHz function can be used to set a variable value equivalent to an angle in degrees OptLB If this optional label is present it sets the next instruction to run 6 15 Signal for Windows version 3 ANGLE PHASE This example starts cosine output at 10 Hz runs for 1 cycle changes to 11 Hz for one cycle then stops ANGLE 0 0 make sure we are at phase 0 RATE OZ 0 Start GUCpPUC at LU Hz RATEW Os HZ LLO request 11 Hz next time around CYCLE10 WAITC 0 CYCLE1O wait for the cycle gt 10Hz CYCLE Lis WALTE 0 CYCLEI1 wait for the cyocle gt 11H4z RATE O54 STOP OUL PUE This changes the cosine angular position It takes effect on the next instruction when the angle increment is added to the value set by this instruction and the result is output ANGLE n expr Vn Vntoff OptLB DAC
179. es testdat000 to testdat999 while testdat10 generates testdat10 to testdat99 If a file name template is set the generated name is used automatically when the data file is saved without the user being prompted to confirm the name A different name can be specified using the File Save As command If this checkbox is set the new data file will automatically be saved to disk when sampling finishes If automatic filename generation is in use the generated filename is used otherwise the usual prompts for a file name from the user are generated This part of the dialog controls three limits Number of Frames Sampling duration and File size Each option has a checkbox to enable the limit plus a field for entry of the limit value If the checkbox for a particular limit is clear or if the corresponding limit value is set to zero then that limit is disabled Note that all of these limits cause sampling to stop not finish sampling can still be continued after a limit is reached In addition to these user defined limits Signal has a built in sampling limit based on the available free disk space If the available free disk space drops below 0 5 Mbytes sampling stops automatically This limit cannot be disabled as it is important not to allow hard disks to get completely full both because this can slow down file accesses considerably and also because of the trouble a full disk gives to an operating system such as Windows
180. es used by Signal into the integer quantities actually used by the DAC hardware This conversion process occurs when generating pulse outputs when waveform data is pasted into an arbitrary waveform pulse and when compiling pulse sequences Full The value corresponding to the full scale output from the DAC For DACs calibrated in volts set this to 5 for a 5 volt 1401 and to 10 for a 10 volt 1401 If your 1401 has a patch clamp scaling card fitted this scales DAC 3 in a 1401p us and DAC 1 in a Power1401 or Micro1401 To calibrate the DAC in volts set Full to 2 048 or 10 24 depending on the scaling card setting Units The units with which the DAC output scaling is specified This is a string from 1 to 6 characters long On the Power1401 DACs 2 and 3 are available on pins 36 and 37 respectively of the rear panel analogue connector If a Signal top box is fitted DACs 2 to 5 will be available on the front panel with DACs 6 and 7 available on pins 36 and 37 respectively of the rear panel analogue connector If a Spike2 top box is fitted then DACs 2 and 3 will be available on the front panel with DAC s 4 and 5 available on pins 36 and 37 respectively of the rear panel analogue connector This section contains a set of checkboxes to enable and disable the individual digital outputs for use Set the checkbox to use this digital output port leave it clear otherwise The fewer digital outputs are enabled for output the more space is available
181. ess this button to reset any states sequencing in operation so that the state sequence restarts at the beginning and also to reset any varying pulses to their initial state Idle Press this button to force sampling and states sequencing to idle It switches manual control of states on sets state zero and turns off writing to disk Pause Press this button to pause any automatic state sequencing that is in progress This option does not pause the sampling itself which continues to run using the current state use the sampling control panel if you want to pause the sampling While states sequencing is paused this button is shown depressed press it again to resume sequencing Manual Press this button to terminate any automatic sequencing in progress and to switch to manual control of states With manual control the frame state is controlled interactively by the user sampling always begins with manual control selected On write Press this button for automatic states sequencing with the states changing only if the previous data was written The states sequencer will have control of the frame states and will move on to the next stage after a sampling sweep only if the sweep was saved to disk Rejected sweeps will repeat the same state until the data is accepted This allows artefact rejection and interactive sweep accept rejection without missing out states from the sequence Cycle Press this button to start automatic states sequencing with the s
182. et this to 1 for high and 0 for low The Id field in the values area boundary displays the name for the currently selected pulse which can be edited as desired The main reason for giving a pulse a name is to make it easy to access the pulse from the script language This specifies a square pulse without any EI eee A variations This is the simplest type of pulse and size vots 3 Stat s 04 TU Noretun is available for DACs and for the digital Length s 0 03 outputs Many of the parameters used for other types of pulse are also used for this type to save space these common parameters are described in detail once only The Size parameter sets the pulse size in calibrated units as defined in the outputs page If absolute pulse levels are in use the item changes to Level and is the level that the pulse goes to for relative levels the size is added to the level prior to the pulse to get the pulse level Either positive or negative values can be used For digital pulses the Size parameter disappears for outputs with a low initial value a high going pulse is produced and vice versa The use or otherwise of absolute pulse levels does not affect digital pulses and overlapping pulses do not invert each other The Start s parameter sets the start time for the pulse If absolute times are not in use this parameter changes to Delay s and sets the delay from the start of the previous pulse to the start of this pulse The Length s par
183. f manual control ower 20 10 pulses at 10 Hz power 30 10 pulses at 10 Hz There are two controls governing overall behaviour State 2 This disables checks on the coil interlock switch on the handle of the MagStim coil so A T Power 20 that the unit will fire without a button being oe fio Feta fro held down MagStim do not recommend disabling these checks as they are a safety Copy to all Copy above Test feature Cancel Help This turns on single pulse mode which allows higher power levels up to 110 In single pulse mode only a single pulse is produced per trigger and the pulse train parameters are ignored The pulse rate can be up to 100 Hz normally depending upon the power level used see the user manual but if you select power levels above 100 the maximum allowed pulse rate is forced to 0 5 Hz With single pulse mode turned off a train of up to 1000 pulses can be generated at the specified frequency from one trigger and power levels above 100 are not available Again the pulse rate can be up to 100 Hz depending upon the power level used The Rapid settings available for the individual states are Set this checkbox for this state to be controlled by the manual controls and not by Signal clear it to give Signal control As you have to disconnect the manual controls in order to connect the serial line used by Signal I fear that this option will not be particularly useful This sets the stimulator
184. f the windows You can sort the list based on the window title type view number as seen by the script language and window state by clicking the title bar at the top of the list Windows Activate Average example frame 1 Maximize lconized MAME OAA chs frame 1 i Normal Seql pls Sequence Normal leonize TEST cts frame 1 File Normal TrendPlot Average example ay Hidden Hide TRIALOAA cfs frame 1 File Normal Restore Close windows i Cancel Using help About Signal 1401 device driver 1401 Monitor revision Working set size Help menu Signal supports context sensitive help and also duplicates the contents of this manual in the help file You can activate context sensitive help with the F1 key or by pressing the Help button from most dialogs to get a description of the dialog and its fields You can use the Help menu Index command to get a dialog holding the help contents an index to help keywords and a word search system to find topics that are not covered by the contents and index From a script view or the script evaluate dialog you can obtain help by placing the cursor on any keyword in the script and pressing F1 To get help on a script function type the function name followed by a left hand bracket for example MemChan then make sure that the cursor lies to the left of the bracket and in the function name and press F1 Pressing the help button the button with a question mark
185. f this window by clicking on them Hold down the Shift key for extended selections You can select entire rows and columns by clicking in the cursor and channel title fields Use the Ct r1 key to select non contiguous rows and columns To copy selected rows and columns to the clipboard by right click the values window and use the Copy command in the popup menu If you use the Log command the selected text is copied and pasted directly into the log window in one operation You can also print the selected portions of the window by right clicking and using the Print command in the popup menu or use the Font command to change the window font Cursor Regions iel Cursor region measurements Cursor menu This command opens the Cursor regions dialog for the current data view The dialog displays values for data regions between cursor pairs One pair can be designated the Zero region by checking the box and selecting the column with a radio button The value in this column is then subtracted from the values in the other columns The pop up menu in the bottom left corner E Cursor regions for example cfs Al ES O 00804196 0 0 0992839 0 103597 0 254639 0 140625 O 0060585 165 057 00597135 i O 0412598 i 0 019932 i O 0164388 i 000712329 0 0 0505854 00935379 i 0 0411643 i 00157335 indicates and controls how the values are calculated The region set by a pair of cursors 1s the data starting at the first curs
186. field The remaining fields are hidden or displayed depending on the channel type For waveform channels the Units may be changed The Full scale and Zero values may also be edited These are the same as the values set in the sampling configuration when the file was sampled originally and can be changed to re calibrate the data For an idealised trace channel the Units may also be changed as well as the 3 dB frequency used for drawing the convolution and fitting the trace to the raw data The Reset Apply and OK buttons are disabled until you make a change to one of the fields The Close button closes the dialog and does not apply any changes This command and the shortcut key Ctr1 1 display information about the current data document Currently 1t just displays the number of sweeps that have been added into a waveform average the number of data blocks in a power spectrum or the number of events from an idealised trace added For an amplitude histogram it is the number of points included which is displayed This command is for XY windows only and opens the XY options dialog The main purpose of this dialog is to control Show key the XY window key The key is a small region that can be key with transparent background dragged around within the XY window that identifies the F Key with border visible XY data channels For each channel it display the name and draws the line and point style This dialog also has a checkbox that controls t
187. field for the change to take effect The spinner to the right of the edit field changes the value in steps of 5 millivolts for voltage clamp and 1 nanoamp for current clamp changes made with the spinner take effect immediately 4 3 Membrane analysis At the bottom is a text field where the total electrode plus membrane resistance is displayed This field is automatically updated whenever the specified sampling state is used you can select the state manually or it can be used as part of normal state cycling or protocol execution Resistance measurements are carried out regardless of whether sampled data is being written to disk or discarded The Membrane button is used to provide the membrane analysis dialog The membrane analysis dialog uses the state for resistance measurements specified in the sampling configuration to carry out a more complete analysis of membrane properties When it is used sampling is automatically switched to the state specified for resistance measurements and writing to disk is turned off The membrane analysis measures the total resistance and total conductance access and membrane conductance the time constant for the decay of the capacitance transient and the membrane capacitance See below for details of the analysis At the top of the dialog is the waveform display area This shows a slice of waveform from the response channel at the time of the start of the stimulus pulse If the analysis has been
188. fields are interpreted in different ways depending on the leak subtraction method described below If base line correction is on the corrected response will also have a DC offset removed so that at the baseline time the response is unchanged The Leak subtraction method can be set to Basic P N or States These three modes are very similar the only real difference is how the leak trace is assembled In Basic mode the leak data is assembled from a fixed contiguous set of frames these frames are set in the dialog by entering a first and last frame All the frames processed use this set of leak data frames contributing to the leak data are automatically skipped In P N mode the frames processed provide the leak traces the first n frames processed are used to make the leak then the next m frames are processed using the leak this cycle continues until all frames are done The values for n and M are entered in the settings dialog In States mode the leak data is assembled from all frames with a given state the state number is entered in the settings dialog All the frames processed use this set of leak data frames contributing to the leak data are automatically skipped As for Auto Average if for instance we only process un tagged frames and there are some tagged frames in the file the process will continue to search for un tagged frames until the required number have been found to complete the formation or subtraction of the leak If Count exc
189. files directly A C library The CFS library is available from CED The library includes all functions necessary to read or write Signal data files from either DOS or Windows programs This library is available along with complete documentation in PDF format from the CED web site http www ced co uk 1 3 Signal for Windows version 3 Licence information CED software is protected by both United Kingdom Copyright Law and International Additional software licences Treaty provisions Unless you have purchased multiple licences you are licensed to run one copy of the software Each copy of the software is identified by a serial number located in the Help menu About Signal dialog box You may make archival copies of the software for the sole purpose of back up in case of damage to the original You may install the software on more than one computer as long as there is No Possibility of it being used at one location while it is being used at another If multiple simultaneous use 1s possible you must purchase additional software licences The original licensee of a CED software product can purchase additional licences to run multiple copies of the same software CED supplies an additional manual set with each licence CED does not supply additional software media As these additional licences are at a substantially reduced price there are limitations on their use 1 The additional licences cannot be separated from the original sof
190. filter applies with the addition that there must be an odd number of coefficients If you try to use an even number your filter will be very poor indeed The example below shows a script for a high pass filter with the same bands and tolerances as for the low pass filter We have added a little more code to draw the frequency response in a result view Signal for Windows version 3 var pim oli var coef 69 prm 0 0 prm 0 1 FIRMake 1 const bins var fr bins FITRResponse fr SetResult bins ArrCons Optimise 0 WindowVisible 1 High pass filter O rE RR Pe Pe Re Pe AAA AAA AAA Hit 1111114 11111111141 111111114 1 111111110 PE A A A AAA NE MO A A AA AAA PE ee A A AAA Pt A A AA AAA NE Hire diri iv 11111111141 111111114 1 1111111141 DAA A AA AAA AAA AA A A A AAA TT T EELO ee eee ERNE Be IE eee RRR ER SERS AAN NA AAA 10 11111111 NO RE T T EST EENE BER Pt ee AAA BERD BERR PT et Iii 11114 AOUT A AAA NO AA NIE PE ee AAA SERS BERR PE et To 4 oy LET AAA A A AE AAA AOS SOTA lt lt S TT IIA PE A Oo S AAA T ee ek co LETT TIN PUTT eT we 66 ITT TT EA PE A Lh HTT T EST T ee A A AAA AO A A A AAA NE PP A AAA AAA AAA MN AA LESA 0 RRR PR Be Pee Ree O Pt ee AA AAA ee eee eee eed RRR A A A eee AA NE O A A A AAA AP A A A AAA AM Aris 141 111111114 1 111111114 MO A A AA AAA reo o be H LUU A O IA AAA CANA A lt a E _____ e808 DOI SS a BED 2 2 CA A A o 2 DO A A AA CEP AA A O DN
191. first frame used for Start offset 0 000000 s each average is offset from the previous average Erames per average Roo start by a set number of frames The Settings dialog holds fields for the channels the width of the average the data start offset and the frames Error bars and frame start offset per average The Channels Width of average and Offset fields New Cancel are all the same as for the standard waveform average processing described above as are the Average x axis starts at zero and Display mean of data checkboxes Count excluded frames is described below Frames between averages f2 Average x axis starts at zero M Display mean of data Count excluded frames The Frames per average item sets the number of source frames that are used to make up the first frame in the average The Frames between averages item sets the number of frames in the source between the start of one average and the start of the next Thus if we are using 2 frames to make up each average setting Frames between averages to 2 will use frames 1 and 2 for the first average 3 and 4 for the second average and so on Setting Frames between averages to 4 means that frames 1 and 2 go to average 1 frames 5 and 6 are used for average 2 while frames 3 and 4 are unused If you set Frames between averages to less than Frames per average then the data for each will overlap with some frames being used for more than one average If Count excluded frames is not ch
192. folder in the folder where you installed Signal and open the file clock sgs Signal will load and run this script which generates an analogue clock in an XY view You can stop the script running and regain control of Signal by clicking on the OK button at the upper right hand side of the Signal window Now use the analysis menu New XY view command to select Trend Plot analysis This opens a dialog for the trend plot settings and leads to a process dialog which is used to select the frames from which measurements are taken As for memory view processing trend plot generation can be saved as part of a sampling configuration You can manipulate the XY view using the Signal menus Most of the Signal commands for example Show Hide channels act on XY views in the same way as for data views You will find that the view menu contains new items Options and XY Draw mode for XY views and the analysis menu is extended to include Delete channel The Change colours dialog is also different for XY views You can read more about XY views in the Edit menu View menu and Analysis menu chapters and in the script language manual If you have followed this chapter you are familiar with the basic actions required to use Signal The next chapter tells you how to configure the system to sample your own data The remainder of the manual covers the menu commands in the system copying data to other applications and printing utility programs and signal conditio
193. g programs or into reports You can use this command with a text file or a script file The file changes back to the state it was in at the time of the last save to disk Send Mail Data update mode Discard changes w Query the user Always save Load and Save Configuration As Default configuration files default sgc last sgc Page Setup File menu If your system has support for Mail installed for example Outlook Express you can send any document from Signal to another linked computer This command vanishes if you do not have mail support Text based documents XY views and memory views can be sent even if they have not been saved to disk Signal writes them to a temporary file if they have not been saved Signal data files cannot be sent as Signal keeps them open so they cannot be copied This command controls how and if changed file view data is written back to the CFS data file Signal holds the data for the current frame in memory where it can be accessed and modified by script commands or by the channel data manipulation commands When the file is closed or the view switches to another frame the data update mode determines what happens Any changes to the frame data can be written back to the disk or the changed data can be discarded or the user can be asked to choose what is to be done The Edit menu Preferences dialog sets the default data update mode for all files use this command to change the mode for a particu
194. ged frames gt The Selected frame state field only appears if you Frame subset Frame state ae gt select Frame state xxx in either the Frame list or the selected frame state gt Frame subset field The channels should all have the same sampling rate as the one you were previewing in the main dialog Channels of a differing sampling rate will be ignored If the filter is of length n then n 2 points around each input data point are used to produce each output point When there is no input data available before or after a point the filter uses a duplicate of the nearest input point as an estimate of the data value This means that the n 2 output points next to either end of the input data should not be used for any critical purpose As applying a filter can be a lengthy process a progress dialog appears with a Cancel button during the filtering operation 17 1 17 2 Signal for Windows version 3 Show details The Show details button increases the dialog size to display a new area in which you can design and edit filters Click this button again to hide the new dialog area aaa M Show aves M Continuous update W Auto GainindB M Free parameters Length lEndpass sd Edge Frequency 306 point Low jizes 1064 High 2270 249 5 Frequency Hz 55 E Transition gap 6 91 The picture in the new area shows the frequency response of the filter The Gain in dB check box sets the y axis scale to dB if che
195. gh frequency content in your signal Filters are usually specified in terms of a corner frequency which is the frequency at which they attenuate the power in the signal by a factor of two and a slope which is how much they increase the attenuation for each doubling of frequency Sampling theory tells us that you must sample a signal at a rate that is as least twice the highest frequency component present in the data If you do not the result may appear to contain signals at unexpected frequencies due to an effect called aliasing As the highest frequency present will be above the corner frequency you should sample a channel at several times the filter corner frequency probably between 3 and 10 times depending on the signal and the application You can choose a range of filter corner frequencies or you can choose to have the data unfiltered for use when the signal is already filtered due to the source 18 2 High pass filter Notch filter Reset calibration AC couple Trigger 1 Setting the channel gain and offset Programmable Signal Conditioners A high pass filter reduces low frequency components of the input signal The high pass filters area is specified in the same way as low pass filters in terms of a corner frequency and a slope except that the slope is the attenuation increase for each halving of frequency If you set a high pass filter a change in the mean level of the signal will cause a temporary change in the ou
196. ght of 10 and the pass band a weight of 1 We will try 40 coefficients to start with so a possible script is Var prml s5 21 5 Array for parameters var coef 40 Array for the coefficients i band start band end function weight pent01 101 000 gt pent ile 1001205 Lo mm 015105 pers 0 0 ci IN 0 Po A o AMM 0 Ao aa f hop Ces FIRMake 1 prm coef PrintLog Pass Band ripple 3 1fdB Stop band attenuation 1f n prem ba 101 perra 1110 If you run this the log view output is Pass Band ripple 28 8dB Stop band attenuation 48 8 The attenuation in the stop band is only 48 dB which is not enough The ripple in the pass band is around 3 of the signal amplitude We can increase the stop band attenuation in three ways by increasing the number of coefficients by giving the stop band more weight or by making the gap larger between the bands We don t want to give the stop band more weight as this would increase the ripple in the pass band We could probably reduce the width of the pass band a little as the attenuation of the signal tends to start slowly but we will leave that adjustment to the end The best way to improve the filter is to increase the number of coefficients If we increase the size of coef to 80 coefficients and run again the output now is Pass Band ripple 58 7dB Stop band attenuation 78 7 A final finesse Digital filtering This is much closer to the filter we wanted You might wonder
197. guments from the table so TABLD 1s not often required TABLD Vm Vn off OptLB load Vm from the table TABST Vm VA otff OptEB Store Vm into the table Vm The variable to load from the table or store into the table off An optional expression that evaluates to an integer in the range 1000000 to 1000000 If omitted the value 0 is used Vn Any offset in the of f expression is added to the contents of this variable and the result is used as an index into the table If the index lies in the table Vm 1s loaded from the table or stored in the table at the index If the index is outside the table TABLD copies 0 to Vm and TABST does nothing Opt LB If this optional label is present it sets the next instruction to run This instruction adds a constant to a variable and detects 1f the result is a valid table index If it 1s a valid index the instruction branches If it 1s not the result 1s reduced by the table size if it is positive and is increased by the table size if it is negative and the instruction does not branch This gives you an efficient way to work through the table TABINC Vn expr OptLB Vn This variable is assumed to hold a valid table index expr This expression evaluates to a positive or negative number that is added to Vn Opt LB If this optional label is present it sets the next instruction to run For example the following codes plays pulses through DAC 0 based on data in the table The table data holds groups of
198. h Micro1401 DACs 0 and 1 Power1401 DACs 0 to 7 and through 1401p us DACs 0 to 1 If you attempt to set cosine output for an unsupported DAC the instruction is treated as a NOP When enabled the cosine value is computed and output every step The time penalty per step per DAC is around 10 us for the 1401p us 4 us for the micro1401 and about 1 us for the Power1401 and Micro1401 mk II The output is output in volts 5 A Cos Theta Phi offset where A is an amplitude scaling factor in the range 0 to 1 Theta an angle in the range 0 to 360 that changes each step set by ANGLE Phi is a fixed phase angle in the range 360 to 360 set by PHASE offset A voltage offset defined by the OFFSET command Theta changes every step by dTheta A cycle of the cosine takes 360 dTheta steps You can change the angle increment immediately or you can delay the change until the next time Theta passes through 0 You can set dTheta in the range 0 up to 360 to an accuracy of about 0 0000001 With the sequencer running at 1 kHz you can output frequencies up to 500 Hz with a frequency resolution of around 0 00012 Hz Ideally the output would be passed through a low pass filter with a corner frequency at one half of the sequencer step rate to smooth out the steps in the cosine wave By adjusting Phi you control the output cosine phase where Theta passes through zero Unless you set the value PHASE it is zero and the zero crossing occurs at the peak of
199. h programmable gains option and the Axon Instruments CyberAmp When you install the Signal software you can chose to install support for one of these types of signal conditioner or you can chose not to install any conditioner support at all You can not choose to install support for more than one type of conditioner and only one type of conditioner can be used at any time You can open the conditioner control panel from either the sampling configuration dialog port setup page or from the Sample menu A signal conditioner takes an input signal and amplifies shifts and filters 1t so that the data acquisition unit can sample it effectively Many input signals from experimental equipment are too small or are masked by high and or low frequency noise or are not voltages and cannot be connected directly to the 1401 Signal conditioners may also have specialist functions for example converting transducer inputs into a useful signal or providing mains notch filters The CED 1902 has options for isolated inputs and specialised front ends include ECG with lead selection magnetic stimulation artefact clamps and EMG rectification and filtering With the Power1401 with programmable gains option only the gain may be set from within Signal You should consult the documentation supplied with your signal conditioner to determine the full range of capabilities Most signal conditioners m ae A asia line Conditioner ees o connection to low COM port for sig
200. hands out time slices of around 10 milliseconds to the highest priority thread capable of running Tasks at the same priority level share time slices on a round robin basis Lower priority tasks rely on higher priority tasks going to sleep when they have nothing to do or when they are blocked for example waiting for a disk read If this did not happen low priority tasks would not run When Signal gets a chance to run it processes pending messages such as button clicks keyboard commands mouse actions and timer events and then updates invalid screen areas Finally Signal is given idle time until it says it does not need any or new messages occur If Signal needs no more idle time it sleeps until a new message appears in the input queue The Minimum gap timer wakes up Signal if nothing else happens This tab is currently used only for enabling the clamping features in Signal If you are not involved with patch voltage clamping etc then use this page to turn off clamping features to avoid confusion The Edit menu Show event details command is available when the current view contains an idealised trace of patch clamp data It opens a dialog that allows information about individual events to be viewed and changed Click on a horizontal portion of the idealised trace to select an event or on a vertical section to select the following event You can also click and drag the idealised trace to change transition times and levels The event deta
201. have been added e Anextra sampling mode has been added to allow for external convert trigger for arbitrary or varying sample rates e Independent colours for channel drawing allow visual grouping of related channels e Cubic spline drawing joins data with smooth curves for better visual presentation and estimation of data values between sampled points e Channel information dialog added e Anew Windows dialog lists all views their types and their view handles and allows rapid view selection e You can now copy and paste data from one XY view to another e Matrix mathematics support including matrix inversion matrix multiply and principal component extraction e The script language now supports multiple display monitors e The script language now supports text to speech TTS conversion for systems with TTS support installed There are many other improvements and more are planned You can find a full list of new features bug fixes and changes in the Revision History in the on line Help Licensed users of version 3 can download updated versions of Signal version 3 from our Web site www ced co uk as they become available 1 1 Hardware required Installation From CD ROM From floppy disk During installation CD ROM and floppy disk After Installation Updating Signal Removing Signal Signal for Windows version 3 The absolute minimum requirement to run the program is a Pentium IT with 64 MB of memory
202. he es El Sample Bar contents from the Sample menu The Sample prept test runexp Bar is a dockable toolbar with up to 20 user defined buttons Each button is linked to a Signal configuration file When you click a button the associated configuration file is loaded and a new data file is opened ready for sampling You can also show and hide the Sample Bar by clicking the right mouse button on any Signal toolbar or on the Signal background The Sample menu Sample Bar EIIE List command opens the z i reni Sac test C signalMtest SGC Sample Bar List dialog Mm ae C signaltramexp SGC Clear l which you control the Sample Bar contents Delete Add opens a file dialog in which you can choose Signal configuration files SGC to Cancel add to the bar If a file holds a label or comment it is used otherwise the first 8 characters Label prep Comment Do preparation of the file name form the label C Asignaliprep SGC and the comment is blank kb You can select an item in the list and edit the label and comment This does not change the configuration file contents You can re order buttons in the bar by dragging items in the list Delete removes the currently selected item Clear All deletes all items The Sample Bar state is saved in the registry when Signal closes and is loaded when Signal opens Each Windows logon account has a different registry configuration If your system has three user accounts each has i
203. he CED 3304 either Current micro mps 50 via a serial communication port or E Copy to all E b Test through a USB port which is setup as a _Cony to al _Copy above te virtual COM port This field sets the Lancel Help COM port that the CED 3304 is connected to If you use the USB connection follow the instructions at the start of the Operation chapter of the CED 3304 Owners Handbook to obtain a suitable device driver for your operating system This field sets the maximum current that you can enter for each state You can select from 10 and 100 microamps and 1 and 10 milliamps This control matches the current range rotary switch on the front of the CED 3304 but does not override its setting the actual range switch setting must match the setting in this dialog or an error will be generated when sampling is begun Check the box for an active high trigger current is generated while the CED 3304 Trigger input is at a high TTL level Clear the box for active low operation current generated when the Trigger input is at a low TTL level When you are driving the trigger input from the 1401 digital outputs it 1s usual to set an active high trigger This field displays and lets you edit the desired current for the state selected in the list of states You can enter any value in microamps from zero to the maximum available for example in the 10 uA range you could enter 0 or 8 234 or 10 The buttons labelled Copy to All
204. he automatic expansion of the axes when new data is added Y Auto expand aves ranges Cancel This example made by the clock sgs script in the Scripts folder shows the key You can choose to make the key background transparent or opaque and choose to draw a border around the key If you move the mouse pointer over the key the pointer changes to show that you can drag the key around the picture Double click the key to open the Options dialog The Draw mode dialog sets the display mode for channels You can set the mode for a single channel all channels or any subset of the channels The Channel field sets the channels to change The next field sets the draw mode to use Click Draw to change the draw mode for the selected channels without closing the dialog click OK to change the draw mode and close the dialog Waveform draw modes Marker draw modes Dots and Lines mode Rate histogram mode Idealised trace draw modes View menu There are five drawing modes for waveform channels Line Histogram SkyLine Dots and Cubic Spline Line joins the data points with straight lines Skyline joins points with horizontal and vertical lines Cubic spline joins the points with smooth curves based on the assumption that the first and second derivatives of the data are continuous at the data points However you must always remember that the only data values that you can rely on are those at the sample point
205. he basic display and analysis controls Instructions that you must follow to keep in step with the text are in bold type with a pointing finger Explanations are in normal type Basic operations The first task is to become familiar with the basic operations that are always needed to manipulate Signal files We use a sample file example cfs The Help menu Index command Getting started topic duplicates this chapter Once you have started Signal you may prefer to follow the remainder of the chapter from the on screen help From the Start button choose Programs then Signal then example Ee a example cts A ES File Edit View Analysis Cursor Sample Script Window Help alela ejej majak ele o mic F B alala 2 ES e lt ample cfs frame 1 Toolbar 5 Keyboard Channel number Control buttons and scroll bar 2 1 0 1 2 2 1 0 1 2 2 1 0 1 2 2 1 0 1 2 Minimised FOS SE window For Help press F1 1 0 00872 1 88767 Frame 1 max 12 0 00 State 0 Flags Status bar You could also have started Signal by selecting the Signal icon and then opened example cfs with the File menu Open command Signal displays the file as it was last saved using information in the file example sgr if it can be found The picture shows the state as shipped You are looking at the raw data in the file we call this a file view This view displays a single frame of the data the axis at the bottom is in seconds The
206. he frequency of the lower edge of the pass band The stop band starts at this frequency less the value set by the Transition gap field Gain da sig 15 20 25 30 3 40 Pi 5 A band pass filter passes a range of frequencies and removes frequencies above and below this range The Frequency field has two editable numbers Low and High which correspond to the two edges of the pass band The stop band below runs up to Low Transition gap and the stop band above from sail oe ib e oa aa ai aa High Transition gap to one half the sampling rate dic ES Gain dB A band stop filter removes a range of frequencies The Frequency field has two editable numbers High the upper edge of the first pass band and Low the lower edge of the upper pass band The stop band below runs from High Transition gap up to Low Transition gap Gain dB 0 5 10 Frequency Hz 17 3 One and a half low pass One and a half high pass Two band pass Two band stop Low pass differentiator Differentiator Signal for Windows version 3 This filter has two pass bands the first running from zero Hz and the second in the frequency space between the upper edge of the first pass band and one half the sampling rate The Frequency field has three editable numbers Band 1 high Band 2 low and Band 2 high These numbers correspond eee sae to the edges of the pass bands heed This filter has two pass bands The second runs up to one half the sam
207. he scaling in the window one page of paper holds the same x axis range as the current display the output spans as many printer pages as are required to show the data selected Y ou must set the print range in a dialog either by typing the start and end times directly or by selecting them from the pop up menus To print an entire frame move to the frame required set the visible width of the view to the X star position Minime Els axis range per page required in the print then lt lt IET select Print Select Mintime for the start position and Maxtime for the end position Print doesn t miya ee l print multiple frames this will be provided in later Cancel versions of Signal All the print commands open the standard print dialog for your printer You can set the print quality you require the better the quality the longer the print takes and you can also go to the Setup page for the printer Once you have set the desired values click the Print button to continue or the Cancel button to abandon the print operation During the print operation which can take some time particularly if you have selected a lot of data a dialog box Printing appears If your output spans several pages the dialog box Do indicates the number of pages and the current page so you can HP LaserJet 4 4M PostScript gauge progress If you decide that you didn t want to print eae ee click the Cancel button This command will print all
208. he start and end of the range For example to find any alphanumeric character use a zA Z0 9 If the search is not case sensitive you can omit one of a z or a z To include the character in a 9 3 Signal for Windows version 3 Select All Replace File comment Frame comment Preferences search place it first or last in the list To search for any character that is not in a list by placing a as the first character To find a non vowel character use aeiou There are three more special search characters that control how many times to find a particular character These characters follow the character to search for Match 0 or more of the previous character So 51 2 matches 52 512 5112 51112 and so on You can also match 0 or more or a character pattern for example h 1 matches hl hel hail horribl and B aeiou r matches Br Bear Beer Beeeaaaooor and so on TE DE Match 1 or more The same as but there must be at least one matching character 2 Match 0 or 1 The same as but matches one character at most The Find Again command repeats the last search with the same options This command is available in text based views and selects all the text usually in preparation for a copy to the Clipboard command This command is used to replace text with different text when you Find What feos E are working with a text based i th Repl
209. hievable a 1401 sampling error will occur and be reported by Signal If Signal detects a Power1401 during program startup it enables a higher timing resolution You can force Signal to allow this higher timing resolution by checking the Assume Power1401 hardware box in the Edit menu preferences dialog The Frame length and Frame points fields set the length of the sampled frame The frame length is always shown in the appropriate time units Changes made to one of these fields automatically cause an appropriate change in the other The Frame length field also updates whenever the sampling rate changes The maximum frame length possible varies with the model of 1401 and the 1401 memory installed each sampled point requires two bytes of memory For a standard 1401 the maximum number of points points per frame times number of channels 1s about 28 000 for a micro1401 or unexpanded 1401p us the limit is about 480 000 while for an expanded 1401p us or Power1401 the limit depends upon the amount of extra memory installed but is at least 15 million For a Micro1401 mk II the limit is either about 480 000 or 1 million depending on the amount of memory the unit was built with The sampling configuration dialog does not apply any limits to the frame length that you enter when sampling starts the 1401 memory required is checked against the memory available Normally zero appears either at the start of the frame or at the trigger time for peri triggered sam
210. hift 1 point left Shift 1 point right Key Shift Zz Shift N Shift R Shift O Shift D Shift I Shift 3 Shift 5 SUE TES SELLET Frame buffer operations Toggle display of frame buffer Clear buffer data Add frame to buffer average Add frame to buffer Add buffer data to frame Subtract frame from buffer average Subtract buffer data from frame Subtract frame from buffer Copy frame data to buffer Copy buffer data to frame Exchange buffer and frame data Multiple frames dialog Key Clic 8 Ctrl Y Enter de Ctrli Ctrl Enter Ctrl Insert Ctrli Ins CRLECELIAS Ctrl M All of these shortcuts are documented with the appropriate menu commands All analysis shortcuts are listed here for convenience There are more keyboard shortcuts for display manipulation see the View menu chapter New Cursor mu Cursor mode Static Maximum and Minimum Maximum excursion Cursor menu A cursor is a vertical or horizontal dashed line drawn in a data view to mark or obtain a position The Cursor menu creates and destroys cursors defines the behaviour of vertical active cursors changes their labelling mode and obtains the values of channels where they cross the cursors and between the cursors Up to 10 cursors of each type numbered 1 to 10 can be active in each data view Cursors can be dragged over and past each other and horizontal cursors can be dragged from channel to channel Cursors in separate windows are indep
211. his box for burst mode sampling leave it clear for equal interval sampling In equal interval sampling the waveform data points are sampled individually in turn The interval between samples is 1 Sample rate number of ADC ports In burst mode sampling all the ADC ports are sampled in a burst as close together as possible the interval between bursts is 1 Sample rate Equal interval sampling has some advantages with the standard 1401 as it loads the 1401 system more evenly while burst mode ensures that the interval between samples on adjacent ADC ports 1s kept to a minimum With a 1401p us Power1401 micro1401 or Micro1401 mk II there is no performance penalty with burst mode Burst mode is generally recommended because it allows greater accuracy in matching the sampling rate used to that required If the first two ADC ports sampled are ports O and 7 or O and 3 for a microl401 or Micro1401 mk II then the second sample and hold circuit optionally fitted to 1401s is enabled If fitted this option causes the sampling on ports 0 and 7 3 to be exactly simultaneous If the 1401 has the 1401 32 multiple sample and hold card fitted then burst mode sampling will be exactly simultaneous on all channels Second sample and hold is not currently available for the Power1401 In Peri triggered sweeps burst mode is always used for efficiency reasons so the state of this checkbox is ignored This checkbox sets the initial state of the Sweep trigger checkbox
212. hroughout Can trigger sweeps or react to Sweep progress Sequencer starts immediately runs throughout Can trigger sweeps via digital and DAC outputs can monitor sampling and react Sequencer starts immediately responsible for triggering sweeps as and when wanted No external trigger or free running sampling N A simulate with Outputs frame mode with free running sequence Sequencer starts immediately runs throughout Can trigger sweeps itself or react to sweep progress N A simulate using Basic mode with free running script 3 13 Signal for Windows version 3 DAC outputs FL Digital outputs a ee ees ee 2 a oo Digital output connections Digital inputs an O py ae E Digital input connections The 1401 DACs Digital to Analogue Converters produce varying voltage outputs in the range 5 volts these can be optionally scaled to 10 volts if required DACs can be used to generate pulses with arbitrary initial values and amplitudes as long as they lie within the DAC output voltage range they can also generate ramps sine waves and arbitrary waveforms To generate complex DAC outputs and particularly for arbitrary waveform output the DACs must update repeatedly with new output values The faster this is done the more accurate the output pulses or waveforms However very high DAC output rates may interfere with data acquisition Signal pulse output is limited to a time resolution of 100
213. ients and must use an odd number you must set a band that stops short of the 0 5 point Remember that by not specifying the remainder of the band you have no control over the effect of the filter in the unspecified region However for an odd number of points the gain at the 0 5 point will be 0 whatever you specify for the frequency band 17 12 Digital filtering The graph below shows the effect of setting an odd number of coefficients when generating a differentiator that spans the full frequency range The second curve shows the improvement when the maximum frequency is reduced to 0 45 Differentiators with 31 HA coefficients E Ep fMax 0 50 as 084 M0 45 99294 A HEE Sa Sas 0 4 EE REE ES EH EEE EH EEE EHEEEEE REECE HEE BREE EERE 1 0 0 0 1 0 2 0 3 0 4 Fraction of sample rate Ih If you must span the full range use an even number of coefficients The graph below shows the improvement you get with an even number of coefficients The ripple for the 0 45 case 1s about the same with 10 coefficients as for 31 Differentiators with 10 Oe ee ee E E E E E a E EHN O O IO IO IO IO ee eee ee ee O EHRE O IO O IO IO IS IO IO IO IO O IO IO AO IO O IO O A O eee OO AO O O IO O IO IO O IO O O IO O IS O A AO IO I HH O ee ee ee 0 4 EEE EEE eee ee Ma0 50 EEE EEH tMas 0 95 HHA m IE oe a E oe DE A A A E ia eet oo A a ee O O O eee O II OI O IO ee ee ee IS AO ISI O IO O IA O IA ANO IS IO IS a O OI O II IO O IO eee ee
214. if there is a formula that can predict the number of coefficients based on the filter specification There is no exact relationship but the following formula worked out empirically by curve fitting predicts the number of coefficients required to generate a filter with equal weighting in each of the bands and is usually accurate to within a couple of coefficients The formula can be applied when there are more than two bands but becomes less accurate as the number of bands increase dB is the mean ripple attenuation in dB of the bands deltaF is the width of the transition region between the bands return An estimate of the number of coefficients Func NCoefMultiBand dB deltaF return dB 23 9 deltaF 5 585 14 41 deltaF 0 0723 end In our example we wanted at least 70 dB attenuation and we weighted the stop band by a factor of 10 20 dB This causes a 10 dB improvement in the stop band at the expense of a 10 dB degradation of the pass band Thus to achieve 70 dB in the stop band with the welghting we need 60 dB without 1t If we set these values in the formula dB 60 deltaF 0 05 it predicts that 67 13 coefficients are needed If we run our script with 67 coefficients we get 70 9 dB attenuation which is close enough If we look at the frequency response of our filter in the area between the pass band and the stop band we see that the curve is quite gentle to start with If you are used to using analogue filters you will rec
215. ignal and a list of signal values an vere a foo ve corresponding amplifier gains The four telegraph settings are selected with the Telegraph set field Telegraph data 2 signals gain of 1 Remove 1 Y signals gain of 2 OY signals gain of 5 1 signals gain of 10 The Scaled item sets the ADC port 2 Y signals gain of 20 rel whose scaling is controlled by a 3V signals gain of 30 _ Caneel telegraph signal Set this to None to Help disable use of this telegraph set The Telegraph field sets the ADC port to read the telegraph signal from 1401 ADC hardware uses 10 volt range The Volts and Gain controls are used to enter a list of telegraph signal voltages and corresponding amplifier gains Enter a voltage value and a corresponding gain then press Add to add this pair of values to the list To remove a pair of values from the list simply select them by clicking with the mouse and then press Remove The checkbox at the bottom of the dialog is used to set the 1401 ADC range it applies to all sets of telegraph information Only set this checkbox if your 1401 is set up to sample voltages from 10 to 10 volts otherwise leave it clear 19 1 MultiClamp 700 telegraphs MultiClamp 700 telegraph configuration Signal for Windows version 3 The telegraph voltage values are not affected by the Full and Zero calibration values set for the telegraph ADC port they are raw voltages as read from the 1401 inputs and c
216. ill remember the position and state of all windows associated with the document This command makes a window invisible This is often used with script windows and sometimes is used to hide data windows during sampling when only the memory views with analysis results are required This command lists all hidden windows Select a hidden window to make it visible You can arrange all the visible Signal windows so that they are arranged in a horizontally tiled pattern by using this command Horizontal tiling arranges the windows so that they tend to be short and wide the exact arrangement depends upon the number of windows You can arrange all the visible Signal windows so that they are arranged in a vertically tiled pattern by using this command Vertical tiling arranges the windows so that they tend to be tall and thin again the exact arrangement depends upon the number of windows All windows are set to a standard size and are overlaid with their title bars visible You can use this command to tidy up the windows that you have iconised in Signal 15 1 15 2 Close All Windows Signal for Windows version 3 This command closes all windows in the Signal application You are asked if you want to save the contents of any text windows that have changed The positions of data document windows are all saved This dialog lists all the document related windows that are open and lets you apply common window operations to one or more o
217. ils dialog gives Event details for constest cfs information h l ormation about the selected E 202 Chan 1 Idealised 38 event in textual form The te time Ea i i Start time 38 396341 me Closed time Level 2 Start time Duration and Fh ceva f T First latency Amplitude can all be changed Any start time you enter must be between the start times of the neighbours Changing the start time adjusts the event duration so that the end time remains unchanged The Undo duration must be a positive Close value and cannot extend the event beyond the start time of the following event Each event also has a number of flags T Level 4 Truncated Assumed amp Level 5 hr 324295 me Amplitude 5 E96546 p mp 4 664639 p mp Duration Bad data Level 6 Baseline 4 Amplitudes relative to baseline Merge Chop Delete Split Fetch Amp Scroll Back Scroll On Fit Wisible Previous Hest Parameterz 9 9 Signal for Windows version 3 Previous and Next Scroll On and Scroll Back Merge Chop Delete Split Fetch Amp Fit Visible Tips for fitting associated with it and these may be set or unset using this dialog These flags and their uses are explained in the Analysis menu chapter The Previous and Next buttons step the current selection of event through the idealised trace Scroll On and Scroll Back will do the same thing as Previous and Next but will do so using a smo
218. in the range 0 to a power of 2 minus 1 then adds an integer constant to it and stores the result in a variable MOVRND Vn bits expr Vn The variable to hold the result bits The number of random bits to generate in the range 1 to 32 The generated random bits fill the variable starting at the least significant bit Bits above the highest numbered generated bit are set to 0 expr An optional expression that evaluates to a 32 bit integer number that is added to the random bits If this is omitted nothing is added Expressed in terms of the script language the random number is one of the numbers in the range expr to expr Pow 2 bits 1 For example MOVRND V33 8 1 generates a random number in the range 1 to 256 The following code fragment implements a random delay of between 1 and 2 024 seconds assuming a 1 millisecond clock MOVEND V1 107993 road Vi O With 0 to 1023 998 DELAY Vik this uses 999 to 2023 steps In addition to generating voltage pulses ramps and cosine waves through the DACs Signal can play arbitrary waveforms The sequencer can start waveform output and test it and branch on the result The sampling configuration sets the size of the area reserved for waveform storage though not all of that area need be used The waveforms are stored in 1401 memory and can be updated just before and during sampling with the SampleSeqWave script command The WAVE instruction starts arbitrary waveform output from the wavefor
219. ing data ends up with half as many bins as the FFT block size As for waveform averaging if the block of data starting at the offset specified runs past the end of the frame the sweep 1s discarded and no analysis is done The data in the memory view spans a frequency range from 0 to half the sampling rate of the source waveform channel The width of each bin is given by the waveform channel sampling rate divided by the FFT block size Thus the resolution in frequency improves as you increase the block size However the resolution in time decreases as you increase the block size as the larger the block the longer it lasts Another feature of the FFT is that the input waveform is assumed to repeat cyclically This means that the maths treats the block of data as though it was taken from an input consisting only of that block repeated over and over again In most waveforms this is far from the case if the block were spliced end to end there would be sharp discontinuities between the end of one block and the start of the next Unless something is done to prevent it these sharp discontinuities cause high frequency components in the result The standard solution to this problem is to taper each data block to zero at the start and end so that the start and end join smoothly This is known as windowing and the mathematical function used to smooth the data is called the window function The use of a window function causes smearing of the data and also
220. ing threshold crossings on a sampled waveform channel Outputs frame the pulse outputs are triggered rather than the sweep pulses can occur before the sweep starts and after the sweep is over and the sampling sweep is started by the outputs Fixed interval similar to Outputs frame but the sweeps are internally timed so that they occur at the specified interval External sweep triggers are not used in this mode Fast triggers like Basic mode except multiple frame states and incremental pulsing are not available This keeps the inter sweep interval to a minimum Fast fixed int has the same limitations as Fast triggers but uses a fixed interval between sweeps rather than requiring an external trigger More details can be found in the Pulse outputs during sampling chapter 3 4 Multiple frame states Use ADC external convert Sample rate Frame length and points X axis zero offset Sampling data This checkbox enables sampling with multiple frame states With multiple frame states disabled all sampling sweeps are the same the same pulse outputs are generated and the new data frames are set to state zero With multiple frame states enabled each sampling sweep can be different from other sweeps in a number of ways and the data frame states are different to indicate what happened during sampling This can be used to achieve a variety of effects and styles of data acquisition The use of multiple frame states is a complex to
221. ion adds one variable to another The SUB instruction subtracts one variable from another In both cases you can optionally add a 32 bit integer constant and optionally divide the result by a power of two The syntax is ADD Va VDrexprrshiftt Va Va VO t expr gt gt SNIfE SUB Vas VOS ExXpEs SOLEC 7 Va Va Vb expr gt gt Shite Va A variable to hold the result It can be the same as Vb Vb A variable to add or subtract expr An optional expression evaluated as a 32 bit integer If omitted 0 1s used shift An optional argument in the range 0 to 31 set to 0 if omitted that sets the number of times to divide the result by 2 The following examples assume that v1 holds 1000 v3 holds 1000 v holds 100 VAR V6 Result 100 ADD V1 V3 V1 O 1000 1000 0 SUB V1 V3 PNAS SG 0 lt OOO lt 4 ADD Vly VOS V1 8 1000 1000 8 SUB Result Y37072 2 VO 225 100 T000 07 4 ADD Result V3 4 1 V6 389 225 1000 4 2 DIV and RECIP divide variables They take around 1 us in the Powerl401 3 in a micro1401 mk II 10 in a micro1401 mk 1 and 5 ina 1401 plus DIV Va Vb Va RECIP Va expr Va Va Vb expr Vb If the numerator is O the result is 0 If the denominator is 0 the result is 2147483647 if the numerator is greater than 0 and 2147483648 if it is negative The 1401p us truncates all results downwards all other 1401s truncate towards 0 So 7 3 or 7 3 is 2 in all 1401s but 7
222. ircle Horizontal Plus ATnangle 1Wertical Font F Signal for Windows version 3 Gaussian function This is because it is assumed that the raw data was filtered using a close approximation to a Gaussian filter The cut off frequency for this filter can be changed by using the channel information dialog For an idealised trace generated using threshold crossing the cut off frequency will be set to the Nyquist frequency by default There is also an optional baseline indicator which will show the level of the baseline for a selected event Both shows both the Basic and Convolution at the same time It is possible to break the drawing of an idealised trace if you get bored waiting Simply hit CTRL Break and the drawing will be abandoned This command is used in XY ind to set the drawin ue o he Ne data Pas Channel Channel 1 Join style not joined Cancel Click OK to make changes and Lmetype sold Width i App close the dialog click Apply to S PPly Marker Dat Size 3 I Ok make changes without closing the dialog Cancel closes the window and ignores any changes made since the last Apply The Channel field sets the channel to edit If you change the channel the dialog remembers any changes you have made so there is no need to use the Apply button before changing channel unless you want to see the change immediately The other fields are LE There are three channel join styles If a channel is Not joined data poin
223. is command appends a new blank frame to the end of a file or memory view This command can be used on a file view to generate an extra frame that will be used to hold processed data for example a frame containing leak subtraction data that will be subtracted from other frames in the file The extra frame can be used as part of a script or it can be manipulated via the frame buffer If the command is used to append a frame to a memory view created by processing the new frame will have its own processing parameters When the frame is appended the process dialog is provided to define the new frame s processing parameters This allows for result views with different frames holding the results of processing different sets of source frames For example if you were sampling with multiple states you might want to produce a multiple frame average with each frame holding the results of averaging source frames with a different state Multiple frame memory views with attached processing parameters can be saved as part of a Signal sampling configuration This command appends a new frame containing a copy of the data in the current frame to the end of a file or memory view This command is not available for memory views created by processing This command removes the current frame from the file or memory view It is only available if the current frame has been appended and not yet written to disk The last frame in a memory view and file view frames st
224. is in use by the arbitrary waveform output in which case the result is undefined DAC n expr Vn Vn off OptLB ADDAC n expr Vn Vn off OptlLb n The DAC number in the range 0 7 Variable 57 n is set to the new DAC value such that the full DAC range spans the full range of the 32 bit variable expr The value to write to the DAC or the change in the DAC value The units of this value are as set in the outputs page of the sampling configuration dialog It is an error to give a value that exceeds the DAC output range Vn When a variable is used the full range of the 32 bit variable corresponds to the full range of the DAC You can use the vDAC32 function to load a variable using user defined DAC units OptLB If this optional label is present it sets the next instruction to run 6 12 RAMP Sequencer outputs during sampling This example ramps DAC 2 from 0 to 4 99 volts in 1 second in steps of 0 01 volts using values and then using variables You can also use the RAMP instruction to ramp a DAC R DAC 240 Ramp 0 TO a gt Ramping MOVI V1l 499 499 steps gt Ramping RAMP1 ADDAC 2 0 01 0 01V increment gt Ramping DBNZ V1 RAMP1 count increments gt Ramping HALT task finished gt Done VM MOVE V3 VDAC32 0 Use variables gt Ramping MOVI V2 VDAC32 0 01 increment in V2 gt Ramping MOVI V1 499 499 steps gt Ramping DAC LFN G set initial value gt Ramping RAMP2 ADDAC 2 V2 add increment gt Ramping DBNZ V1 RAMP2
225. is repeated in turn the specified number of times within each sequencing cycle whereas with Randomised ordering each state 1s repeated the set number of times but the order of states within each sequencing cycle is randomised Semi random ordering does not work well with individual repeats 7 3 Signal for Windows version 3 State sequencing by protocol A protocol consists of a list of up to 10 steps each step pema setting the state that will be used a repeat count anda CO step to use next A protocol can use up to ten steps protocols can loop and be chained together to produce Add protocol Delete Details longer sequences of states Up to 50 separate protocols can be defined within a single sampling configuration giving great flexibility W Create toolbar button for protocol Cycle protocol state only after write W Turn on writing at protocol start Reset pulse steps at protocol start Protocols are defined using the protocols dialog which is State hepeats Next obtained by pressing the Protocol button on the states 7 p page The dialog contains a selector at the top that is 7 7 used to select a protocol for viewing and editing To step4 p p create a new protocol press the Add Protocol button steps o 7 E while protocols can be deleted using the Delete button Step E i h a The protocol name can be changed by directly editing it ce Im in the protocol selector Blank protoco
226. item specifies a state a set of pulse outputs to be used for measurement of resistance and other membrane analyses This value is not checked within the dialog but it must either be zero 1f multiple states are not being used or from zero to the maximum state number in use Whenever this state is used during sampling Signal will automatically calculate and display the membrane resistance In addition this state will be automatically selected when the membrane analysis dialog is used So with state 4 set as in the picture above whenever state 4 is used during sampling the membrane resistance will be recalculated and displayed Signal states are quite a complex topic for more information on them see the chapter Sampling with multiple states A suitable stimulus pulse for resistance measurements should be defined for the control DAC in the specified state this pulse can be added to the pulse outputs or modified during sampling should this prove necessary If there is more than one pulse for this state the pulse to be used for measurements should be labelled by setting the pulse ID to RM otherwise the first pulse in the outputs will be used Currently a square pulse constant or varying amplitude or a square pulse train should be used Varying duration pulses sine waves or ramps and arbitrary waveforms are not currently supported 4 2 Other sampling configuration considerations Running a clamping experiment As menti
227. ith MOVI to set up loops for example MOVI V2 1000 set times to loop WT DIGOUT 00000000 set all digital outputs low DIGOUT RA Sete them all Argh DBNZ V23 WT loop 1000 times These instructions run a labelled part of a sequence and return CALL and CALLV save the next step number to a return stack and jump to the labelled instruction The RETURN instruction removes the top step number from the return stack and jumps to 1t CALLV also sets a variable to a constant CALL LB use LB as a subroutine CALLV LB Vn expr Vn expr then call LB RETURN return to step after last CALL LB The next instruction to run The CALLed section should end with a RETURN Vn CALLV copies the value of expr to this variable CALL1 sets V33 CALL2 sets V34 CALL3 sets V35 and CALL4 sets V36 expr A 32 bit integer constant that 1s copied to a variable You can use CALL inside a CALLed subroutine This is known as a nested CALL If you call a subroutine from inside itself this is known as a recursive CALL The return stack has room for 64 return addresses If you use more than this the oldest return address is overwritten so your sequence will not behave as you expect 6 19 Signal for Windows version 3 JUMP HALT NOP This example generates different pulse widths from DAC 0 The sequence is written to be independent of the sequencer rate However 1t must be high enough so that the widths are possible In this case a sequencer Step
228. ition Signal can also control a pair of MagStim 200 units using two serial lines Transcranial magnetic stimulators are dangerous devices capable of causing serious harm and should only be used by qualified medical practitioners Before using a MagStim device you should read the user s manual produced by MagStim paying particular attention to the warnings and precautions section It 1s your responsibility to ensure that Signal s control of a MagStim is set up in an appropriate and safe fashion for the intended use and to verify that it is operating correctly CED is not responsible in any way for problems caused by use of this software For MagStim devices the auxiliary states system sets the power output and where appropriate inter pulse timing remote control of the hardware can be disabled for specific states to allow the user to control the stimulation manually Options are provided to cause Signal to assume independent triggering mode and to enable high resolution interval timing for the BiStim and to cause the MagStim to ignore the coil interlock switch and to enable single pulse mode for the Rapid This documentation only describes the controls available from within Signal for a complete understanding of the effect these have upon the behaviour of the stimulator you should consult the MagStim user manuals In addition to using a serial line to set the stimulation parameters standard operation with Signal requires that the s
229. itted to the data using the Event details dialog to achieve maximum accuracy for the results The Channel is the waveform channel to be fitted The time range to be fitted is then defined by the Data start time and Data end time If you are analysing voltage activated channels then you should set these times to be the start and end times of the stimulus The first event generated by the analysis will be flagged as a first latency so if you are interested in first latency times you will need to make sure that the first event starts when the stimulus does 11 13 11 14 Advanced Signal for Windows version 3 The Baseline is used at the start of the analysis to tell the event detection where the closed state is expected to fall Baseline track will keep a running average of points in the closed state in order to correct for baseline drift during the recording The next field will be either Open below for an inward current where an opening is downwards on the display or Open above for an Outward current where an opening is upwards on the display This field should be set to just outside the noise level It is used to determine when an event 1s too short to be distinguished from noise Open Level defines the full open level It is used in conjunction with the Advanced parameters to determine what amplitude change is significant enough to constitute a transition The Filter cut off frequency is the 3 dB frequency of the Gaussian filt
230. l Signal and make sure that you select CED 3304 stimulator support If the button is present but disabled you are in External Digital states mode Change the mode to enable the button Click the button to open the CED 3304 configuration dialog From the configuration dialog you can enable use of the CED 3304 set the serial line port used to communicate with the hardware and other overall parameters define the current settings for each state in use set the active level for the Trigger input and test for successful communications with the CED 3304 20 8 Hardware selection Com port for communications Current range selection Active high trigger Settings for state Copy Test Simple use Auxiliary states devices At the top of the dialog are settings that apply to all states In the centre 1s a list of the currents set for each state Below this Ata SS SENET Use CED L53304 gt is a field to edit the current for the Com port for communications com 1 selected state At the bottom are buttons an 100 microamps F to copy settings get help test the CED e 3304 and accept the dialog settings curent 10 mierodmpe te 2 current 50 micro4mpe Set this to Use CED 3304 to use the State 3 current 70 mieraamps stimulator or to Do not use if you do not State 4 current 30 micro mps want to make use of the 3304 support software in this sampling configuration Settings for state 2 Signal connects to t
231. l names are not Steps m 10 allowed and will be rejected Step10 f0 1 Repeat count for entire protocol 1 At end Finish ka W State zero when protocol finishes The checkboxes at the top of the protocol details set general options and what happens at the start of protocol execution The Create toolbar button for protocol M Tum off writing when protocol finishes item enables a separate button for this protocol in the Cancel Help states control bar see the section Controlling multiple states online below Cycle protocol states only after write controls the sequencing of protocol steps if it is checked then the protocol sequence does not advance unless data file frames are written to disk if the data is not written or a sweep is rejected then the protocol does not advance Turn on writing at protocol start causes writing of sampled sweeps to disk normally controlled by the Write to disk at sweep end checkbox in the sampling control panel to be automatically turned on when the protocol starts Reset pulse steps at protocol start causes all varying pulses defined in the pulses dialog to be reset to their initial state when the protocol starts The table below defines the protocol steps There are ten steps in a protocol each one specifies a State Repeats and a Next step These specify the state to be used the number of times to repeat this state and the step to go to from 1 to 10 when this step is done A step is
232. l will be prompted for and this channel will be applied with the appropriate operand to the other channels on a point by point basis This command keyboard shortcut Ctr1 T is used to tag or untag the current frame in the current view When the current frame is tagged this menu item is shown checked All data frames in files handled by Signal can be tagged or untagged the tagged status of a frame is displayed as part of the application status bar and can be interrogated by scripts Frame tagging can be used for any purpose you require all commands requiring a frame selection are able to operate on all tagged frames or all untagged frames The command toggles the tag state of the current frame changing tagged frames to untagged and vice versa 11 19 11 20 Digital filters Keyboard analysis control Signal for Windows version 3 This opens the Digital filtering dialog which can create FIR filters and apply them to waveform channels Windows software is usually orientated towards control by means of the mouse and menus but it is often convenient to use the keyboard instead For interactive analysis of the data using the keyboard can often be much faster With this in mind Signal includes keyboard shortcuts designed to handle most common data manipulation requirements Channel arithmetic Zero channels Negate data Rectify data Subtract DC level Differentiate data Integrate data 3 point smooth 5 point smooth S
233. lar file If Query the user mode is selected the Save changes dialog will appear as required This Adjust data file update mode to match allows changed frames to be individually eee discarded or saved check the Adjust data file update mode to match checkbox if you don t want to see the dialog again These commands manage Signal configuration files These hold the sampling parameters the window arrangement required during sampling the output setup and the types of on line analysis required Signal always has one configuration loaded this is the configuration used for sampling and the sampling configuration dialogs The Load Configuration and Save Configuration As commands transfer the Signal configuration between disk file and the application They both open an appropriate file dialog to select a file for loading or saving to The Save Default Configuration command saves the current configuration as the default configuration that will be automatically loaded whenever Signal is started If the configuration file default sgc exists in the Signal program directory it is loaded when Signal starts You can save the current configuration in this file by using the Save Default Configuration command There is also the standard file last sgc that holds the last configuration that was successfully used for sampling If default sgc cannot be found and last sgc exists last sgc 18 loaded Signal saves the current configuration in last
234. lay though steps have been taken to minimise this When sampling finishes normally the MagStim is disarmed and remote control disabled If the MagStim coil temperature rises too high Signal will stop sampling Once the coil temperature has dropped sufficiently you can press More on the sampling control panel to resume sampling again While sampling is in progress Signal continuously maintains communications to prevent the MagStim from disabling remote control and disarming itself If Signal ceases to communicate with the MagStim because it has encountered a significant problem crashes the MagStim will disarm itself automatically within 1 second but this safety feature only applies if manual control has not been selected by Signal beforehand If manual control is selected the MagStim will disarm itself spontaneously only after 60 seconds have passed without a stimulus trigger so it is your responsibility to make sure the MagStim is disarmed if manual control is used and Signal encounters a significant problem The Rapid stimulator does not appear to disarm itself after 60 seconds in this manner and must be disarmed manually if Signal fails during sampling Because Signal needs to communicate frequently with a MagStim to stop it disarming scripts that operate while Signal is sampling need to be correctly designed If a script carries out a lengthy operation without yielding or using a toolbar or dialog to allow control to pass ba
235. le The front panel Data and Clock inputs are not used with Signal and should be left unconnected CED 3304 PC serial 9 Way D Type 9 Way D Type socket socket CABLE SCREEN A straightforward arrangement to use a CED 3304 with Signal assuming we use digital output bit 0 to control the current pulse timing would be as follows 1 Connect your computer to the CED 3304 using the either a serial or use a USB connection Connect the CED 3304 front panel Trigger BNC plug to the 1401 digital output 0 BNC socket found on the front of all modern types of 1401 If you are using a 1401p us the digital output pulse is available from the 25 way digital output socket on the front of the 1401 2 Inthe Outputs page of the sampling configuration make sure that digital output bit zero is enabled for use Using the pulses configuration dialog set the initial level of digital output bit zero to O and place a pulse which will be high going in the outputs at the time when you want the CED 3304 to fire This output pulse should be as long as the required stimulation 3 The CED 3304 support is designed to work with Signal multiple states sampling which should be set up in Dynamic outputs mode Set one extra state for each output current setting you require and each state should generate digital output pulses to trigger the CED 3304 plus any other outputs required The CED 3304 support will work with manual control of the states or with any style of automati
236. le buttons are hidden because the individual protocols select between these two modes of operation If the number of states is less than ten so that they can all be represented by the individual state buttons the main state selector is also hidden to save space The extra items provided are the protocol selector and protocol run button and the individual protocol buttons Reset aust Basic 0 State 2 State 4 Main seg Y Main seq Idle Manual State 1 State 3 Run Secondary Protocol To the right of the state buttons or the state selector if it is visible is a protocol selector and spinner that can be used to select a protocol from those available Unlike the state selector selecting a protocol does not execute the protocol During execution of a protocol the current protocol in use is shown so you can see what is going on if you use chained protocols If all of the protocols have individual buttons then the protocol selector is not shown Run This button is shown below the protocol selector Pressing it causes Signal to switch out of manual mode and start executing the selected protocol immediately terminating any other executing protocol Like the protocol selector if all protocols have individual buttons this button is not shown Buttons These are buttons created for the individual protocols which are arranged to the right of or below the protocol selector and are labelled with the protocol name A protocol button will
237. le in each band times the weight for the band the same The ripple is the maximum error between the desired and actual filter response in a band The ripple is usually expressed in dB relative to the unfiltered signal Thus the ripple in a stop band is the minimum attenuation found in that band The ripple in a pass band is the variation of the frequency response from the desired response of unity In some situations for example audio filters quite large ripples in the pass band are tolerable but the same ripple would be unacceptable in a stop band For example a ripple of 40 dB in a pass band 1 is inaudible but the same ripple in a stop band would allow easily audible signals to pass By weighting bands you can increase the attenuation in one band at the expense of another to suit your application Differentiators Hilbert transformers Multiband with 3dB octave 17 8 cut Low pass filter example Signal for Windows version 3 The output of a differentiator increases linearly with frequency and is zero at a frequency of 0 The differentiator is defined in terms of a frequency band and a slope The frequency response at frequency f is f slope The slope is usually set so that the frequency response at the highest frequency is no more than 1 The weight given to each frequency within a band is the weight for that band divided by the frequency This gives a more accurate frequency response at low frequencies where the resultan
238. lect an installation directory that contains a copy of Signal for Windows version 3 You can remove Signal from your system open the system Control Panel select Add Remove Programs select CED Signal for Windows version 3 and click Remove This removes files installed with Signal you will not lose files you created 1 2 File icons ilies Using this manual Forthcoming attractions Direct access to the raw data Signal for Windows The various file types in Signal have icons so that you can easily recognise them when you minimise their windows The icons will automatically be used for the relevant files by programs such as Windows Explorer All the icons have a set of waveforms to remind you of the application to which they belong The icon to the left is the Signal application icon that you double click to launch Signal from the Signal program group These icons are for Signal data files The icon on the left represents Signal CFS data files or file views The icon on the right represents an XY data file a saved XY view If you double click on one of these in Windows Explorer it will launch the Signal application if it is not already running and open the data file These icons are for text type Signal documents and files The icon on the left represents a text file which can hold any textual data The centre icon represents a Signal script file Script files hold script programs that execute within Signal to automate an
239. led 3 Signal for Windows version 3 Pre trig time Ports configuration data and triggers when the sampled data value is above the upper level or below the lower level Digital Trigger on a digital input bit state The parameters set the digital input bit from 8 to 15 and select triggering on a high bit or on a low bit The trigger occurs when the bit is in the correct state There is no requirement for the bit to be in the other state first The digital inputs are found on the 1401 digital inputs connector the pins for the digital bits are Digital input bit bit15 bit14 bit13 bit12 bit11 bit10 bit9 bit8 GND Digital input pin 1 14 2 15 3 16 4 17 13 Event Trigger on a TTL pulse There are no parameters the trigger occurs when a TTL pulse is detected on the standard 1401 or 1401plus Event 0 input or the Trigger input on a micro1401 or Power1401 This parameter sets the number of points in the frame before the point at which the trigger occurred This can have any value from 1 000 000 sample interval to the length of the frame 2 sample interval If the value is negative this means that points sampled after the trigger occurs are discarded before the first point in the frame is kept If the value is positive then the specified time must have elapsed before the search for a trigger begins and the resulting frame contains points sampled before the trigger occurred When a non zero pre trigger time is specified the re
240. led in the General tab Open the States tab and make sure that you are in Dynamic outputs or Static outputs mode If MagStim support is installed there will be a button labelled MagStim If this button is not present reinstall Signal and make sure that you select MagStim auxiliary states device support If the button is present but disabled you are in External Digital states mode change the mode to enable the button Click the button to open the MagStim configuration dialog It is meaningful to use the MagStim support without using multiple states but this is not directly supported because the states system is so closely linked to the MagStim support If you want to make use of MagStim support without using multiple states so that the pulse parameters are recorded and MagStim health checks are made but with complete manual control you should turn on multiple states set only one extra state set state zero to use manual control and only make use of state zero while sampling This will allow you access to the MagStim configuration dialog to set the basic MagStim parameters At the top of the configuration dialog there l l TEE EEE is a selector used to select the MagStim type eE You should set this to the type of stimulator ype of stimulator MagStim 200 y you will be using or to Do not use if you do Com port for communications coma w not want to make use of the MagStim support in this sampling configuration If you set the device
241. ling and the y offset changes to keep pace with the mouse pointer You can do this with most x and y axes in Signal This is particularly useful for y axes as they do not have a vertical scroll bar The window does not update until you release the mouse button If you hold down the Ctr1 key the window will update continuously When the cursor is over the axis numbers a click and drag changes the axis scaling The effect depends on the position of zero on the axis If the zero point is visible the scaling is done around the zero point the zero point is fixed and you drag the point you clicked towards or away from zero If the zero point is not visible the fixed point 1s the middle of the axis and you drag the point you clicked towards and away from the middle of the axis In a file view memory view see page 2 9 or XY view you can drag the y axis so as to invert it You are not allowed to invert the x axis Now double click on the time x axis of the display to bring up the X Axis Range dialog box Experiment with the settings to vary the time axis The Left and Right fields set the window start and end times The Width field shows the window efi on Showa width Set the left and right positions or check the parts 0 03 E Width box and set the left position and the width Tan ona Draw Large tick spacing 1 ES You can type new positions or use the drop down f Tick subdivisions po 3 lists next to each field that give you access
242. llowing tasks Display Data Sampling Conditioner Script Scheduler Clamp Background routine limits Minimum gap between timed update routine end and next start 20 me Maximum duration of script idle before Signal sleeps O n0 limit i 0 me Maximum script idle cycles before Signal sleeps O no limit jo The background routine mings penodically to update displays and to keep script idle time functions active This page limits the time used by Signal so that the sistem feels responsive and other applications get a chance to run You must close thie dialog to see the full effect of any changes e When sampling it gives windows a chance to detect that the maximum time has changed which may cause windows to update and processing to occur Any invalidated windows will update the next time Signal gets idle time e Ifa scriptis running it gives any idle function in the script a chance to run e In automatic file naming mode it starts the next file running There are three fields that limit the time used in the background routine They have no effect on the time used when Signal runs a script that does not idle see the Yield or YieldSystem script commands for this The standard values work for most cases 9 8 Minimum gap between timed update routine end and next Start Maximum duration of script idle before Signal sleeps Maximum script idle cycles before Signal sleeps Threads Clamp Sho
243. luded frames is turned on however then even frames that are excluded from the process will count in the total number of frames used The New button or Change if this is used from the Process Settings command closes the dialog creates the new memory view and opens the Process dialog described below Leak subtraction processing uses the same process dialog but because leak subtraction creates a set of frames in the memory view the behaviour is subtly different the Clear bins checkbox if set clears out the entire memory view and if not set doesn t accumulate data into the existing frames but rather appends more frames to the memory view For similar reasons the Analysis menu Append Frame command does not create a second set of process parameters all frames use the same process parameters This command is available when a memory view created using the New Memory View command or Frames Tagged frames a similar XY view is the current view When you 6 Frame subset Frame slate xxx use it a dialog prompts you to select the frames of 1 the source data document to process The Process ea frame state fi dialog 1s also provided automatically when you use Clear memory view data before processing the New or Change button from the Process 4 Re process if source data changes IY Optimise Y axes after processing Settings dialog to create or rebuild a memory view The simplest way to use this dialog is to type in a Cancel
244. lues to produce the floating point data values that Signal uses and displays Signal can also use and create waveform channels where the underlying data are floating point values these are indistinguishable from channels using integer data in nearly all circumstances Floating point data that is the result of analysis can be stored more accurately as floating point values though it requires twice as much disk space to do so The process of converting a waveform into a number at a particular time is called sampling The time between two samples is the sample interval and the reciprocal of this is the sample rate which is the number of samples per second A set of samples taken at regular intervals 1s referred to as a sweep The sample rate for a waveform must be high A Original waveform O O Resulting waveform enough to represent the data correctly You must sample at a rate at least double and PARANA 2 3 4 5 6 7 8 preferably 2 5 to 5 times the 0 1 highest frequency contained in the data If you do not sample fast enough high frequency signals are aliased to lower frequencies as illustrated above The dots in the diagram represent samples the lines show the original waveform On the other hand you want to sample at the lowest frequency possible otherwise your disk will soon be full 9 samples Many users pass waveform data through amplifiers or signal conditioners with filter options to limit the frequency range Some tran
245. m area WAVE Start arbitrary waveform output The waveform output area used by the sequencer can only be loaded up with data by using the SampleSeqWave script language function it can be reloaded during sampling as required The WAVEBR instruction tests the state of the waveform output and branches on the result No branch occurs if there is no output running or requested WAVEBR LB flag LB Label to branch to if the condition set by the flag is met flag An optional single character flag to specify the branch condition S branch if arbitrary waveform output is stopped G branch if arbitrary waveform output is going The following fairly trivial sequence plays the output waveform 5 times MOVI vV1 5 load variable 1 with 5 WL WAVE start output WT WAVEBR WT G wait here while output is going DBNZ V1 WL do this 5 times gt Waiting for cycle The WAVE command starts output it then waits for the output to stop at the wT label Next the sequence repeats this process 5 times When you use the Format or Compile buttons in the sequence editor Signal displays the result of the compilation or format operation in the message bar at the top of the window The messages report either successful operation or the cause of the problem 6 28 Introduction What can multiple states do Enabling multiple states Defining multiple states Sampling with multiple states In the Getting started chapter you will have encounte
246. marked as being the end of the protocol by setting it s next step to zero When protocol execution begins it starts with step 1 The state set by the State field in step 1 is set and is used the number of times set in the Repeats field Following this the protocol switches to the step number set by the Next field This process continues until it is ended by encountering a Next item of zero In the example shown above step 1 repeats state 1 four times and then goes to step 2 This repeats state 3 eight times after which the protocol ends If the Next item for step 2 were set to 1 the protocol would run forever or it could be set to 3 for a more complex sequence The Repeat count for entire protocol item below the step table controls the number of times that the protocol repeats before it ends Set this to 1 for a protocol that goes through the steps only once as in the example above or set it to a larger number for a protocol that repeats a set number of times before ending For example if you set the repeat count to three in the example above the protocol would run for 36 frames before ending If you set this item to 0 the protocol will repeat forever until stopped The items below the repeat count control what happens when protocol execution ends The At end selector selects between Finish and a protocol that will be chained to chaining to a protocol allows more complex sequences than is possible with just ten steps If the At end selec
247. me 11 17 Delete selected text 9 1 DIBEQ output instruction 6 11 DIBNE output instruction 6 11 Differentiate data 11 19 Differentiator 17 8 Differentiator filter 17 4 Digital filter 17 1 Digital filter types 17 7 Digital input test bits 6 11 test saved bits 6 11 Digital markers 3 2 Digital output 6 10 micro1401 and Power1401 6 10 Digital output voltages 3 14 Digital outputs 3 14 Bit numbers 3 14 Connections 3 14 Socket pins 3 14 Digital outputs enable 3 11 Digital outputs for state 7 6 Digital state outputs Bit numbers 3 14 Socket pins 3 14 DIGLOW output instruction 6 10 DIGOUT output instruction 6 10 Directory for file name generation 3 15 Directory for file saving 3 15 Directory for new data files 9 4 DISBEQ output instruction 6 11 DISBNE output instruction 6 11 Display Customise 2 6 DIV output instruction 6 22 DOFF output instruction 6 17 Dots draw mode for markers 10 5 Dots draw mode for waveforms 2 6 DPHASE output instruction 6 16 DRATE output instruction 6 15 DRATEW output instruction 6 15 Draw mode 10 4 Drawing modes Data view 10 4 10 5 DRINC output instruction 6 18 DRINCW output instruction 6 18 DSZ output instruction 6 14 DSZINC output instruction 6 15 Duplicate data view 15 1 Index 2 Index DWAIT output instruction 6 17 DWCLR output instruction 6 18 Dynamic output states 7 2 Hi Edit bar 10 1 Edit menu 9 1 Clear 9 2 Copy 9 1 Copy as
248. memory You turn overdraw mode on or off eestedirame state jo with the View menu Overdraw frame list M Hse colour cycling command In colour cycling mode each Cancel overdrawn frame draws in a different colour otherwise all the display list frames draw in the colour set by the Frame list traces item in the colour setup dialog The current frame draws in 1ts usual colour 1f 1t 1s not in the frame list All the standard mechanisms for selecting frames are available see page 2 3 for details More information on the buffer can be found in the Analysis menu chapter Y ou can use the cursors to take measurements at or between cursor positions Make sure you have some cursors in the window then open the Cursor menu Display Y Values window Experiment with changing cursor positions and channel display types MA Cursors for example cfs The columns show the cursor o Cursor 1 Cursor 2 Cursor 3 Cursor 4 positions and the value at the cursor 0 00844749 00144894 00205479 00276712 positions for waveform channels ong ool 2 Marker channels displayed as Rate 00317383 00927734 00366211 0 012207 0 737505 00146484 0078125 0141602 also show the value at the cursor Tiga Dogo position marker channels in other 0915527 00146484 DO0145484 draw modes show the time of the next marker after the cursor To measure the difference between cursor values use the X zero and Y zero check boxes The radio buttons below each c
249. microseconds which is not fast enough to cause any interference with data acquisition The 1401plus and Power1401 have four DACs numbered 0 to 3 while the micro1401 has two 0 and 1 Both the 1401p us and micro1401 have the DAC outputs available on BNC connectors on the left hand side of the 1401 front panel The Power1401 has DACs 0 and 1 on the front panel and has DACs 2 and 3 on pins 36 and 37 respectively of the rear panel analogue connector The 1401 digital outputs are TTL compatible and can be set high or low When high they generate a voltage in the range 2 6 to 5 volts the usual lightly loaded level is about 4 5 volts When low they generate a voltage between 0 and 0 6 volts The Signal pulse output system and the output sequencer DIGOUT instruction control 1401 digital output bits 8 to 15 of the 16 bit digital output port If you use the output sequencer you can also set bits 0 to 7 with the DIGLOW instruction Signal refers to both sets of outputs as digital outputs bits 0 to 7 The 1401 digital outputs use a 25 way D type socket This is on the right hand side of the 1401p us front panel and on the rear of the Power1401 and micro1401 The Power1401 and micro1401 digital output bits 8 and 9 are also on front panel BNC sockets labelled 0 and 1 the labels match Signal usage If the Spike2 digital 1 O expansion top box is fitted 1t has front panel BNC sockets for digital output bits 10 to 15 labelled 2 to 7 also matching Signal
250. mode the pulse output frame starts at the same time as the sweep triggered or un triggered and is the same length In Peri triggered mode the pulse output frame starts at the time of the trigger which can be before or after the start of the sampled data depending upon pre trigger points and again runs to the end of the sampling sweep In Outputs frame and Fixed interval modes the pulse output frame can be set to any length greater than or equal to the sampling sweep and the sampling sweep starts at a defined point after the start of the pulse output frame P u Ises d i a Og If you press the Pulses configuration Configure Pulses _ naMMMMMMMMMMMMMMMMlMMMMMMMMl i i iiIIssMO button in the outputs page in the sampling configuration Signal displays the Pulses configuration dialog to allow you to view and edit the pulses 000 1 00 EE Del Copy E gt gt The dialog can also be Sine wave DAC 0 at010to0 20s Id4 used to control and LUA TIL size wots 2 Stat s 0 1 cee adjust the pulses while MLL A fa Length is 0 1 _ Cancel Signal is sampling in AL Cycle 2 oa Start phaze so Help which case it is accessed using the sampling menu the toolbar or by using the right mouse button popup menu The pulse configuration dialog can be used to define square pulses pulse trains ramps sine waves and arbitrary waveforms that will be output during the sampling Many of these pulses can automatically v
251. mpling The Event 1 start checkbox allows you to trigger the entire sampling process externally If this checkbox is set clicking on Start will not start sampling directly it enables the start of sampling on an event pulse This allows precise control of the time of the start of sampling the time from which the frame absolute start times are measured While Signal is waiting for an El start pulse the Start button will flash Waiting for E1 The El input is on the 1401p us front panel and is pin 2 of the rear panel Events connector for Power1401 and Micro1401 with a suitable ground on pin 9 Connect the El input to ground or pull it below 0 6 V to start sampling The Sweep trigger checkbox is initially set from the Sweep trigger item in the sampling configuration It enables or disables triggered sweeps while Signal is waiting for a sweep trigger the checkbox text will show Waiting TR The checkboxes in the Sweep end section of the dialog control what happens at the end of a sweep of sampling and how data is written to disk If the Write to disk at sweep end checkbox is set then each frame zero is written to the disk file when the sweep finishes If the checkbox is clear then the frame is not written You can override this behaviour for individual sweeps using the Accept Reject button in the sampling control panel 3 17 Signal for Windows version 3 Sweep end checkbox combinations During sampling The Pau
252. mpling of the next sweep when sampling is paused after collecting a sweep It is equivalent to the sampling control panel Continue button This command toggles the state of the Sweep trigger checkbox in the sampling control panel The menu item displays a checkbox when this option is selected This command toggles the state of the Write to disk at sweep end checkbox in the sampling control panel The menu item displays a checkbox when this option is selected This command toggles the state of the Pause at Sweep end checkbox in the sampling control panel The menu item displays a checkbox when this option is selected This command aborts sampling and discards any sampled data It is equivalent to the Abort button in the sampling control panel This command discards all data returns sampling to the state it was in before sampling started and restarts sampling It is the same as the sampling control panel Restart button This command and its toolbar equivalent hides and shows the pulse w Toolbar definition dialog that is available during sampling if pulse output is in Status bar use This dialog can be used to change the output pulses while data ie acquisition is in progress changes made will be saved in the current Serptbar sampling configuration The menu item is checked when the control Sample bar panel is visible 7 Sampling controls Pulse controls The sampling control panel pulse controls and states control bar can A
253. n drag the pulse out of the display area completely The mouse pointer changes to a hand with a minus sign to indicate a remove operation once you are outside the display area Drop the pulse to complete the removal Alternatively once the pulse has been selected you can remove it by clicking the Del button in the controls area If you cannot find or click on the pulse to start dragging it see Finding a pulse below for how to select it It may be difficult to see a pulse or to click on it because the pulse is too short to see or because it is hidden by another pulse Click on the appropriate output trace then use the lt gt buttons to toggle through all of the pulses on that output At the appropriate point your hidden pulse will be selected and can then be edited or deleted directly Experiment with adding moving and removing pulses to get a feel for how the dialog behaves Note the different behaviour of overlapped pulses with absolute or relative pulse levels When using multiple states with multiple sets of pulses ze 1t may be useful to copy a set of pulses or settings to create duplicated data for other states The Copy button opens a dialog where you can specify items within the currently selected state to be copied and a range of destination states to copy them to The left hand side of the dialog sets the items to copy the right hand side sets the destination If you clear the Outputs frame checkbox the
254. n optionally be put into CEDCOND INT First 4 Last 12 The search for conditioners will start with the channel specified as First or zero if this isn t set and continue until the channel specified as Last has been checked and a missing channel has been encountered If Last is not specified a value of zero is used so the search stops at the first missing channel The arrangement is somewhat different for CyberAmps as they have multiple channels The channel numbers are specified by the rotary address switch at the back of the amplifier note that the first amplifier must have the address set to zero not 1 as normally shipped The channels are numbered off in the order of the address switches assuming 380s and 320s are not mixed the Last value can be used to extend the search Standard 1401 telegraphs 1401 telegraph configuration Amplifier Telegraphs If you are using an external amplifier to scale your signals before they are sampled by the 1401 you can encounter problems if you change the signal gain while sampling the amplitude of the sampled data changes which can make the signal calibrations incorrect Signal can avoid this problem by using amplifier telegraphs to determine the current amplifier settings and adjusting the signal calibration to match Amplifier telegraphs are signals usually analogue voltages from an amplifier that signal the current amplifier settings such as gain and offsets By collecting and inter
255. n the integer data corresponding to zero in the scaled values which can be convenient Signal saves the current preferences to the registry so they are preserved next time the program is run The Save changed data selector sets the default action for file data that has been changed in memory For example the script language could have been used to differentiate the data in a channel You can choose to write changes back always to only write changes after querying the user or to always discard changed data This option sets the default initial state for all data files opened use the File menu Data update mode command to control data write back for a single file Signal saves file memory or XY views as pictures in Same as screen image either bitmap screen image or metafile format A Ea metafile describes a picture in terms of lines and text Screen image x 2 based on a grid of points Metafiles have the advantage Screen image x 4 Screen Image 8 Screen Image x 16 that they can be scaled without losing resolution You can choose the density of the grid The higher the Mica ase a F a density the more detail in the picture The problem for width scaled to 1000 units time and result views with a lot of data points is that the Width scaled to 2000 units higher the grid density the more lines need to be drawn Width scaled to 4000 units and many drawing programs have limits on the number of P Mdth scaled to S000 units lines the
256. nal conditioner control com H software control and T Dump errors to CEDCOND LOG in the Signal application folder communication The serial port used to communicate with any installed signal conditioner is set in the Edit menu Preferences dialog in the Conditioner page In addition to using serial line ports built into your computer you can also make use of serial lines provided by plugging an adaptor into a USB socket USB serial lines generally work fine but you do need one which always appears as the same COM port or where you can set the COM port as which it appears Check the Dump errors to CEDCOND LOG box if you are having problems using your conditioners this will cause a log file detailing communications attempts to be created which will allow the cause of problems to be determined Leave this checkbox clear if you are not having any problems connecting to signal conditioners it s intended for troubleshooting only 18 1 Control panel Port Input Gain Offset Low pass filter Signal for Windows version 3 The control panel is in two halves The left hand side holds the controls that change the conditioner settings the right hand side displays data from the conditioner Signal omits the right hand side while sampling data or if the 1401 interface is not available for any reason If the right hand side is present the Volts check box causes the data to be displayed in Volts at the conditioner input in pla
257. name generation and automatic filing and Sampling limits which can be used to restrict the Save file to disk on sampling finish C Sampling limits amount of data sampled or filed Number of frames There is also a button used to Sampling duration 3 access the artefact rejection File size KByte dialog Artefact reject This sets the directory where the E Sr tat Help automatic file naming looks to produce a unique file name and where new files are saved when sampling has finished This is different from the directory for new files set in the Preferences dialog which sets the location for the temporary files used while sampling If this field 1s blank automatic file name generation and file saving use the current directory You can enter a directory path directly or use the Browse button to select a directory This sets the template for automatic file name generation If this is blank automatic file name generation 1s disabled and normal document names Datal Data2 are used for sampled data If a template is provided it generates a sequence of unique file names based on a numeric code If the template ends with one or more digits these set the length and initial value of the numeric code If the template does not end with digits Signal adds 000 before using it Signal increments the code until it finds an unused name in the directory set by the Path field Thus a template of testdat generat
258. ncertainty in the fit coefficients given that the model fits the data and that the errors in the original data are normally distributed If a coefficient is held the Sigma value will be 0 The Testing the fit section gives more information on the derivation of these values and how to interpret them You can select rows columns or individual cells in this area the use Ctr1 C to copy them to the clipboard This also copies a bitmap image of the page to the clipboard This section of the page displays the differences between the data points and the fitted curve in the large rectangle and a histogram of the error distribution on the right The error plot is self scaling based on the distribution of errors the plot extends from 3 at the top to 3 at the bottom times the RMS root mean square error The grey line across the middle of the plot indicates an error of zero In the case that the data can be o co coo a a modelled by the fitting function plus 3 sE E E A A t normally distributed noise you would expect to see residuals distributed randomly around the 0 error line and the histogram on the right should resemble a normal curve If the data cannot be modelled in this ee ae gt ae way you would expect to see te eer evidence of this in the residuals In i this example generated by fitting a cubic to data that was actually a double exponential you can see that there are clear trends in the errors In extreme cases the error
259. nds for Transistor Transistor Logic and is a method of passing logical High Low information between devices using voltage levels Levels above 3 0 volts are in the High state levels below 0 8 volts are in the Low state Levels in between 0 8 and 3 0 volts are undefined The TTL inputs and outputs on the 1401 are the digital inputs and outputs the event inputs the clock F external frequency inputs the ADC external convert input the clock output the DAC Bri output and the micro1401 Micro1401 mk I or Power1401 trigger input On the microl401 Microl401 mk II and Power1401 the event inputs are not actually TTL but can be treated as such Do not subject 1401 TTL inputs to voltages above 5 0 volts or less than 0 0 volts CED hardware has special circuits on TTL compatible inputs to provide some protection however determined abuse will damage them The 1401 TTL compatible inputs are pulled up by a resistor to 5 volts They require a current of some 0 8 mA to pull them into the Low TTL state Alternatively you can connect them to ground to pull them low useful for the Event inputs See the Owners handbook of your 1401 interface for full details of each input port Signal can sample two types of Marker data keyboard markers and digital markers Signal treats both marker types identically once the data has been captured they differ only in their source A Marker is a 32 bit time value This is the number of sample intervals on each wavefo
260. nel and has zero width the y vertical axis changes to display the selected range and the time axis remains unchanged The mouse pointer changes to a magnifying glass when you hold the mouse button down in the data channel area to show that you are about to drag a rectangle or line to magnify the data If you hold down the Ctrl key before you hold the mouse button down the mouse pointer changes to the un magnify symbol If you drag a rectangle the data in the view shrinks to cover an area the same size as the rectangle you have dragged making this the inverse of the effect without the Ctr1 key Whichever method used to scale the data you can return to the previous display using the Edit menu Undo command or the keyboard short cut Ctr1 Z If you decide not to expand the display after starting to drag return the mouse pointer to the original click position making the rectangle have zero width and height The rectangle will vanish and you can release the button without changing the display If you hold down the Alt key before you click and drag Signal displays the size of the dragged rectangle next to the mouse pointer and does not zoom the display 2 4 Specifying times Getting started Move the mouse pointer over the x and y axes and experiment with clicking and dragging the axes Try it with the Ctrl key held down When the cursor is over the tick marks of an axis you can drag the axis This maintains the current axis sca
261. ner support The Script menu chapter describes the menu commands that control the script system The script language itself is not covered in this manual see the companion text The Signal script language for a full description 2 12 Introduction Types of channel Waveform channels Minimum sample rate Use of filters Input connections Sampling data If you have worked through the previous section you already have most of the skills needed to work with a new data document created by sampling a sampling document A sampling document is much the same as an old document except that the sampling document grows by adding frames to the end The sampling document also has an extra frame frame zero that contains transitory data retrieved as it is sampled We have modified the Process dialog used during sampling to provide mechanisms for automatic processing and updates otherwise you would have to use it constantly in order to follow the course of your experiment Before we discuss the sampling configuration dialog we need to provide some background on the types of data channel that Signal can sample Signal handles two types of channel waveform and marker The waveforms that Signal records and displays are continuously changing voltages Signal stores waveforms as a list of numbers that represent the waveform amplitude at equally spaced time intervals These numbers are 16 bit integers They are scaled using calibration va
262. ng the number of repeats to 1 but a sequencing cycle still consists of each state being used the number of times specified by the repeats Protocol In this mode a separate protocol dialog is used to define arbitrary sequences of states repeated an arbitrary number of times This of course is not a numeric sequencing mode and is documented separately see the section on Protocols below When Protocol mode is selected all controls for state repeats and cycles are hidden and replaced by a Protocols button which provides the protocol dialog When sampling using dynamic outputs Signal provides controls for the state and states sequencing see the section Controlling multiple states online below Parameters C Work Signal3 winD ebug LAS 1 _ SGC When the Individual repeats checkbox is set the eel Patel domo opa States acens Repeats item that sets the repeat count for all awe a cetna Naren I states is hidden Instead the dialog shows a state f indvidualrepeats oie selector and repeat so you can set repeats for Rowe R each state The state selector is derived from the digital output control used in other modes to set the repeat count for a state click on the desired state and edit the Repeats value shown in the state data box Cancel Run now Help Individual repeat counts are used in much the same way as an overall repeat value only each state has a separate count In Numeric ordering each state
263. nificant compared to the errors in the input values and that are not included in the model For example 1f you are fitting exponents to a sampled waveform that includes perceptible mains interference you can get a good fit by eye to the exponential data but with a probability of 0 0000 as far as the mathematics 1s concerned because the model does not include the mains hum and cannot explain why the chi squared value is so high If we assume that the model fits the data we can make an estimate of the standard deviation of the fitted coefficients This means that if we re ran the experiment many times and fitted the data to each set of results what would be the likely variation in the fitted coefficients This is presented as the Sigma value in the results tab If there 1s no error information for each point we assume that all the points have the same normal error distribution and the fit minimises the sum of squares of errors between the model and the data Because there is no independent estimate of the likely spread of the errors in the original data strictly speaking there 1s no way to give a probability of getting an error of at least this size Open Closed times New idealised trace SCAN Analysis menu However we can say though statisticians may shudder Given that the model does fit the data and that the errors all have the same normal distribution then the differences between consecutive errors should
264. nn The channel number is 1 to 80 for sampled channels or 0 to 7 for the last value on DACs 0 to 7 The result is the most recent data available Waveform and DAC data are treated as 16 bit signed values from 32768 to 32767 You also have access to DAC values as 32 bit data in variables V67 to V64 VDACO to VDAC7 without the need to use the CHAN instruction This instruction takes rather longer to execute than other sequencer instructions and may cause sequencer timing problems if used in circumstances when the 1401 is heavily loaded This example waits for a sampled signal to cross 0 05 volts and produces a pulse VAR V1 level VDAC16 0 05 level to cross VAR V2 data FEO hold the last data VAR V3 Low 0 some sort of hysteresis level DIGOUT 00000000 set all dig outs Low BELOW CHAN data 1 read latest data gt wait below BGT data low below wait for below gt wait below ABOVE CHAN data 1 read latest data gt wait above BLE data level above wait for above gt wait above DISOUT ec ane ds 1 pulse output DIGOUT Faggian 0 below wait for below This instruction sets a variable to the current number of points sampled in the current sweep The variable value is set to O if the sampling sweep has not started This instruction is not available for the 1401plus POUNTS Vn Opt LB Vn Sweep points Vn This variable is updated with the sweep point count OptLB If present the instruction branches to this label This
265. nnnannnos 3 20 Signal for Windows version 3 Samping With clamp SUPPONE usina iaa aa 4 1 Introd HoN a na 4 1 Onlime clamp support CAMI Senna a A 4 1 Running a CAMPAL E ista 4 3 Pulse outputs wile SAM DINING asia 5 1 TOC O 5 1 ER T T E A E AAE AA A E E A TA O AAR 5 1 Dae Cin and droppi ile EE 5 3 Addn AMC We PUS o a a a A EE EE 5 3 Movie A PU Ca A A 5 4 Remove a PUll est a E oe Pee Ree 5 4 A 5 4 COP PU ties 5 4 Editine the pulse PALMS ARSS 5 5 SPE QUITE PS E iii 5 5 Vat yin amplitudes PUE vsco da dao 5 5 Varying duration PUISE dd dd 5 6 Sare PUSE A Dana R a a 5 6 Ramp with varying amplitudes ccccccccccccccceeceeceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeees 5 6 MS W AW oi 5 6 ATDA W AVELO naa E E E R 5 6 DCE LS a Wa IO A a ened anne EN 5 7 PulSes with vVanatOOsS nidad necia des 5 7 Outputs frame and Fixed interval SWE pS oooooccnccccooooooononononononnnonnnonononnnonononnnonnnnos 5 8 Controlling pulse outputs during SaMplid8 oooooonnonononononanononononononononnnnnnnnnnnonononononononoos 5 9 Sequencer Outputs during sampling ooccoccocnconnocccnnnonccnnconcnnncnnnonnnnnonnnnnrnancnnrnanenannnnos 6 1 A ORO 6 1 SEQUE COMO io 6 1 sequencer technical AAA 6 1 The seguence EMI O O 6 2 Loading sequencer files for sampling oooccccncnccccaccnnaanonaaaanananonononnnnno nono non n nn nro nro nnnnnnnnnnos 6 3 Ei A a ce oo o ee Ree ee 6 3 DAS GUL CU 10S dad uation 6 6 DIS CU CUIOM Torma 35 2 dances es 6 7
266. nsigned number 0 means a probability of zero and 4294967295 the largest 32 bit unsigned number means a probability of 0 9999999998 BRAND LB 0 3 branch with 50 probability To produce a multiple way random branch you use more BRAND instructions A three way equal probability branch to LA LB and Lc can be coded BRAND LA 0 33333 Split the first route with p 1 3 BRAND Le AUS 50 6067 to here 0 5 35 0 3334 173 iO Sah If neither of the above comes here The following shows the sequence for a five way branch with equal probabilities BRAND PA O22 5 Way UA probability 15 022 1 5 BRAND FX 0 5 Probability to here 0 8 so to FX 0 4 BRAND LB O 5 Probability to here 0 4 so to LB 0 2 Probability to here 0 7 4 z2 LC atini EX BRAND LD 0 5 Probability to here 0 So o LDS UZ GES ES Probability to here 0 The best technique 1s to reduce the branches to a power of two as soon as possible Case 1 of the five way branch is split off probability of 0 2 leaving 4 ways The 4 ways are split with a probability of 0 5 0 4 for each division then the last two routes are split again with a probability of 0 5 0 2 for each division In a Poisson process the probability of something happening per time interval is constant You can generate a delay with a Poisson statistic by POISSON BRAND POISSON prob y poisson delay The probability is given by prob 1 0 1 0 mDelay S 1 where mDelay is the mean delay required in s
267. ntrols frame tagging The last region is the digital filtering command described in the Digital filtering chapter This command is enabled when a file view is selected It opens Waveform Average a pop up menu in which you can select an analysis type from Amplitude Histogram Waveform Average Amplitude histogram Auto Average Auto Awerage Power Spectrum and Leak Subtraction Selecting an analysis PAE AI un Leak Subtraction opens a Settings dialog where you set the analysis parameters and other information needed to construct a memory view to hold the analysis results This dialog can be recalled from the memory view to change the analysis parameters Click New in the Settings dialog box to create a memory view with all data values set to zero and to open the Process dialog in which you select the source data frames to analyse The results of analysing different sets of frames can be summed by repeatedly using the Process dialog to select different frames This analysis averages a waveform across A TEE E ECES multiple frames The Channels field sets the waveform channels to average The Width of Manel SBE El average field sets the width of the new Width of average 04000 e memory view The Offset field sets the start time for the data as an offset from the start of Startoffset fo 000000 s the frame An offset of zero selects data from M Average x axis starts at zero the start of the frame regardless of the frame M Dis
268. o The NEG instruction is identical to MOV except that the source variable is negated first The syntax 1s MOV Va VO expr shift Va Vb expr gt gt shift NEG Vay Vb expr Spa EE A Va AVID ae uexpr gt gt Sine Va A variable to hold the result It can be the same as Vb Vb A variable used to calculate the result It is not changed unless it is the same variable as Va 6 21 6 22 ADDI ADD SUB DIV RECIP Signal for Windows version 3 expr An optional expression that is evaluated as a 32 bit integer If this argument is omitted it is treated as 0 shift An optional argument in the range 0 to 31 set to 0 if omitted that sets the number of times to divide the result by 2 The following examples assume that v3 holds 1000 VAR V6 Result MOV Vi Vv3 set V1 to 1000 NEG V1 V3 set V1 to 1000 MOV V1 Vo pes set V1 to 992 NEG Result V3 0 4 set V6 to 63 MOV Result V3 4 1 set V6 to 502 This instruction adds a 32 bit integer constant to a variable There is no SUBI as you can add a negative number The syntax is ADDI Vn expr OptLB Vn Vn expr Vn A variable to hold the result of Vn expr expr An expression that is evaluated as a 32 bit integer OptLB If this optional label is present it sets the next instruction to run The following examples assume that v1 holds 1000 VAR V1 Result 1000 ADDI Result 1000 set V1 to 0 ADDI V1 4000 set V1 to gt 000 The ADD instruct
269. o ING raw alas 1 3 Licenco tormes 1 4 ins 2 1 o A reer er mene Treen Tener ner Men tnt rere rR ern nae open GeT Troe htt 2 1 BAC O iaa 2 1 SA i 2 1 Toolbar and Status DAT iria 2 2 AS CISCUSSION about AMES aaa 2 2 Contolline the displayen a a ee 2 4 Cursor measure MENIS uti ciand 2 7 Memory VIEWS AS ed odian 2 9 Saving and reloads Memory VIEWS id 2 12 A sacl acuta et ast E OE AE E E ET E EE 2 12 is A IN 2 12 cado 3 1 is she ttoume here E E T E A A 3 1 NANA is E A E II E E E 3 1 Wavelorm channe Doniene ee E E 3 1 TAL eCOm patible AMM E ee RS 3 2 IY TAI fe ICV O A e o E cena EA 3 2 Makor COCE APRIETA oat 3 3 Sala ple COM MOU A OM te atin stas 3 4 General COMM OME AU is 3 4 Peri trio ger CONSULTA LON ii iio 3 7 POR CON UA A AA A AAA AA 3 8 CIAMPCOAH ULA A A E A 3 9 DUCES COM TOUT ALLO MN A AAA A A eases een 3 9 Pulses OL SCOU CI CCR aise AAA AAA AAA AS 3 13 DAC OUMU S Paneer met ee cera ET Or ner eee ee i ea eae 3 14 DISTA CUISINE 3 14 Dista a a a 3 14 Automa nor coni eurai iisas doi ee 3 15 a a a a a a a a a a 3 16 CTE AUCs d mew documen ieri at e a a a a a a a ais 3 16 Ona nas dolls 3 17 Sal Mie control pal tiles 3 17 Dime Sample ss 3 18 SOPPIDE SAMP N eai aa ONE en Oe eRe eee soe ee eee ee ne ee re ern er 3 19 Pino san ne oso 3 19 SR A re oy ee ae nein rene nen em E 3 20 Say io COMI OUT ALOIS ss nas al aaa clin crcl erga ence a E ES 3 20 Sequence of operations to set and save the configuration oooocnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
270. o know if Power1401 hardware is in use Signal tries to detect the 1401 type automatically when it starts up this checkbox sets whether Signal will assume the presence of a Power1401 if it cannot detect the 1401 type directly If a channel display is optimised in the y direction and the Defer on line y axis optimise to sweep end box is checked Signal will wait until the sweep finishes before scanning the data collected in order to perform the optimise If this box is unchecked then the optimisation will be done only with the data collected so far in that sweep This may mean no data has been collected and axes will be set to default limits which may well not be ideal 9 7 Signal for Windows version 3 Maintain display of ADC range online Conditioner Script Save modified scripts Enter the debugger Scheduler When a programmable signal conditioner settings are changed or a telegraph voltage changes during sampling the display y range for the relevant channel can do one of three things Never keep the y axis limits will keep the y axis unchanged In this case there may be no visible indication that the amplifier gain has changed although the resolution of the data may change Maintain it if Showing All range will alter the y axis when the gain changes to show the new full range of the ADC provided the full range was previously displayed Maintain ADC range percentage will maintain the ADC range previously displayed but show th
271. o re create the average if the source data changes and optimise the display after each analysis so that the full data range is visible You can also click Settings to go back to the Settings dialog EN Signal Average1 example Al ES File Edit View Analysis Cursor Sample Script Window Help alsa ejej ea ase ele o de F O alala 2 RS example cfs frame 2 Ml Ed Process Average1 example Frames fan frames y Frame subset an frames y IV Re process if source data changes Y Optimise Y axes after processing i 3 3 Be 5 3 I Clear memory view data before processing a 3 3 e i SOI OLA LION BI CH 0 000 0 005 0 010 0 015 0 020 0 0 z 5 Process Cancel Settings H 1 0 01437 4 11765 Frame 1 max 1 0 00 State 0 Flags 00000000 eae aa AA The dialog closes and the memory window shows the analysed data You can recall the dialog by selecting the Process command from the Analysis menu Do this now and click the Process button again The data in the memory window will not change because this is an average The count of sweeps displayed by using the View menu Info command will 0 010 0 015 0 020 0 025 0 030 0 035 double as long as you have not checked the Clear memory view before process checkbox 2 11 Signal for Windows version 3 Saving and reloading memory views S XY views Summary Experiment with this new window You will find that the new memory view beha
272. o restart the sequencer press a key associated with a sequence step or click a key in the sequencer control panel If you associate a display string with this instruction it appears in the sequencer control panel HALT gt Press X when ready The NOP instruction No OPeration does nothing except use up one sequencer clock tick It can be thought of as the equivalent of DELAY 0 6 20 Variable arithmetic Compare variable MOVI MOV NEG Sequencer outputs during sampling These instructions perform basic mathematical functions while a sequence runs You can also compare variables and branch on the result These instructions compare two variables or a variable and a 32 bit expression and branch on the result of the comparison All comparisons are as signed 32 bit integers BXX Vn Vm LB compare with a variable BXX Vn expor LB compare with a constant BXX Vn Vm toff LB compare with a table entry XX This is the branch condition The xx stands for GT Greater Than GE Greater or Equal EQ Equal LE Less than or Equal LT Less Than NE Not Equal Vn The variable to compare with the next argument Vm A variable to compare vn with or table index variable expr A 32 bit integer constant to compare Vn with This example collects the latest data value from channel assumed to be a waveform waits for it to be in a set range for 1 second then outputs a pulse to a digital output bit START CHAN V1 1 get channel 1 d
273. oe1ases 0098539 and RMS You can select the mode with PP O ae ee cee cS a the popup menu at the bottom left comer of MEE ne eS the window Click on the rectangle showing Mean to see the menu If you want to make measurements relative to one of the regions check the Zero region box and choose a reference region with the radio buttons For a waveform channel or markers as rate Area is the area between the waveform trace and the line joining the intersection points between the cursors and the trace Mean is the mean level of the signal Slope is the gradient of the least squares best fit line to the data Area O read this as area over zero is the area between the waveform trace and the y zero level Sum is the sum of all data points and Modulus is the area over zero but with all amplitudes considered positive the rectified area For a marker channel in other not Rate draw modes Sum is the number of markers in the region Mean is the count of markers divided by the width of the region Slope has no meaning for a marker channel neither does Area Area O Modulus or the others Use the View menu Standard Display command Click on the Keyboard channel CS number and drag it down over the other channel numbers 1 Click 7 HA 2 Drag w o o E Sa Y i ae Y a rit Y EE a 3 Drop DADC1 ADC ADC 3 a es Of 0 0 Do 0 0 eyboard H 7 3 7 3 ms ms As the mouse pointer passes over
274. of the data file as appropriate This button acts on a frame until the next frame is triggered or pulse outputs for the next frame starts Abort This button abandons sampling and discards the file You can use this button before sampling starts or while sampling You are warned if this will lose saved data Restart This button is available once sampling starts It stops sampling discards any saved data then waits for you to start sampling again with the same document You are warned if this will lose saved data 3 18 Other interaction with sampling Stopping sampling Finishing sampling Sampling data If a keyboard marker channel is being sampled you can insert markers into the sampled data by pressing keys on the keyboard You can do this if the new data view is the current view or if the sampling control panel 1s current The current view or window has a highlighted title bar you can make a view current by clicking on it If the sampling control panel is current you should be careful about pressing the space bar which is equivalent to pressing whichever button is currently highlighted or pressing Enter which is equivalent to pressing the Start Continue button If you are using Peri triggered sampling in any of the analogue trigger modes frame zero of the sampling document displays a pair of horizontal cursors that indicate the positions of the two trigger thresholds These cursors are displayed on the highest numbered
275. offsets are set to zero the points to overwrite is set to the points on the clipboard and these two checkboxes are checked the paste operation copies all the clipboard data and changes the waveform output parameters to match Click OK to paste the data or Cancel to do nothing Pulses with variations Some types of pulse can be set up so that they Square pulse varying height at 0 50 to 0 70 1d J vary automatically The pulse types that support S etmv 30 Starts ua M Noretum this are initial level square pulse with varying poe amplitude square ae a varying ia A A M and the ramp pulse All of these use the same three parameters to control the variation only the varying aspect depends upon the type of pulse The pattern of variation used 1s the pulse is generated a number of times without variation then a number of times with one step change added then two steps and so on This repeats until the maximum 5 7 Signal for Windows version 3 Outputs frame and Fixed interval sweeps number of changes has been reached at which point the cycle restarts with the pulse with no step changes The Step change parameter sets the amount by which the varied aspect the pulse amplitude for example changes at a time The Repeats parameter sets the number of times each step is repeated before moving on to the next step and the Steps parameter sets the maximum number of changes to be added This arrangement gives a final value of Ini
276. ollowing Data view background Marker text Axis markings amp text labels Waveform as line Rate histogram Tagged frames background Waveform as dots Rate histogram fill Frame list traces Waveform as skyline Text labels XY view background Waveform as histogram Cursors and cursor labels Standard deviation SEM Markers as dots Controls Fitted data Markers as lines Data display grid See the Edit menu Preferences for text view colours These two pages assign colours to channels EE in the current file or memory view and Channel primary colour Z override the application colours set for drawing modes The primary colour sets the ira drawing colour for lines waveforms markers and histogram outlines The secondary colour is for filling histograms and for drawing the SEM and SD in data views The channel background colour overrides the view background for the area occupied by the channel data An X in a box marks a channel with no colour override Reset All Heset cos Changes made on this page are applied immediately so there is no Draw button You can Reset the selected channels back to the standard colours set in the Application colours page The equivalent script command is ChanColour This page changes the XY channel colours Changes made on this page are applied immediately there is no Draw button The equivalent script command is XYColour This page allows you to override the application colours set for the
277. olumn choose the cursor to make the reference The values for the reference cursor are shown unchanged the values for the other cursors have the value at the reference cursor subtracted You can use this feature to show how data values have changed from a reference point If you move the cursors change frame or change the channel display mode the values in the window update to reflect the change of position Likewise if you show or hide data channels in the display the cursor window display changes to match You can select fields in this window and copy them to the clipboard Click on a field to select 1t or drag across the data area to make a rectangular selection of fields Click at the top or left hand edge to select an entire column or row Click in the top left hand box to select all the fields Hold down the Ctr1 key and click at the top or left hand edge for non contiguous selection of rows or columns You can also print copy to the clipboard change the font or copy to the log window using the right mouse button menu 2 1 Signal for Windows version 3 Now open the Cursor menu Cursor Regions window Experiment with changing CS cursor positions and measurement modes The regions window looks at the data values between cursors There are many A oa ee measurement modes including rea i l eo a eee cea Mean Slope Peak Sum Modulus ree eee jiga SS TEO snes ae Maximum Minimum Amplitude SD banca ox03s97 1 o
278. on BRAND This instruction waits until a given time in sequencer ticks within the sampling sweep WSWP expr Vn Vatotft OPptTLB Wait till expr in sweep expr This is time within the sweep in sequencer ticks Values of expr from 1 to the sweep duration specify a time within a sweep if you specify a time greater than or equal to the sweep duration the sequencer will wait forever The following values of expr have special meanings O Wait until a sampling sweep is in progress 1 Wait until a sampling sweep is not in progress 2 Wait until the sampling sweep is armed the 1401 is ready to accept a trigger but has not been triggered yet Vn When a variable or table entry is used the value is the time within the sweep in sequencer ticks or one of the special values above OPELO If present the instruction branches to this label This can be used to pause sequencer execution until a required sweep time 1s reached it 1s the easiest way of synchronising the sequencer with sampling This instruction causes a trigger to start a sampling sweep If the Free run without restarts box is checked you can use this in Basic Outputs frame and Fast triggers If the box is clear you can use it in Outputs frame mode only Using this in other circumstances may cause sampling problems TRIG Trigger a sweep In Outputs frame sampling mode you must use TRIG to trigger sweeps as external triggers are disabled This instruction sets a variabl
279. oned above because Signal is a very general purpose program it is possible to produce a sampling configuration that does not allow the clamping features to operate correctly though we have tried to make the software as flexible as possible The aspects of the sampling configuration that you need to ensure are correct are Any sweep mode can be used but as the two fast sweep modes do not allow multiple states or changes to the pulse outputs they are very limiting and will not work well External convert sampling cannot be used as this prevents Signal from determining the time course of events The stimulus and response channels specified must exist be generated by the configuration and the ports used to log the channels must use appropriate units as described above The channel calibration scaling factors must be correctly set up for the amplifier gains and any telegraph settings correct or data analysis will give incorrect results e Pulses outputs must be used with absolute levels turned off if you want the holding potential controls to work as expected All DACs specified for stimulus control must be enabled in the outputs and calibrated using suitable units The DAC scaling factors must be correctly set for the amplifier control inputs or the stimuli will be incorrect If multiple states are used which is almost always necessary these should use dynamic outputs variation e As described above the state specified
280. ontrols the appearance of the data and XY views The first region holds commands to show and hide the application toolbar edit bar and status bar The second holds commands that move the display from one frame to another within the data These commands are not available for XY views which have a single frame The third region holds controls for frame overdraw mode and the frame list used for overdrawing The fourth region holds commands for zooming in and out in both the x and y directions The fifth region controls the channels and other items that are displayed on the screen and the waveform and marker drawing mode The final region controls the fonts used and the use of colour in the data and XY windows These three commands enable and disable the display of the application toolbar the edit toolbar and the status bar The toolbar is the array of buttons normally displayed below the application menu bar The edit bar contains buttons for functions available in text views The status bar is always at the bottom of the application window and displays information about the current highlighted view These items display a checkmark if the corresponding bar is displayed click on the item to toggle the display state These bars can also be shown and hidden by right clicking on an unused w Toolbar area of the application window to open a pop up menu This menu can Y Status bar show and hide other types of toolbar The script and sample bars are e
281. onverted to actual volts using the ADC voltage range set in the dialog The scaled port should be calibrated in the main ports configuration page as though the external amplifier gain is set to unity the telegraphed gains with then be applied to this basic calibration It is very important that you enter the value for unity amplifier gain not whatever gain you have in use at the moment During sampling the telegraph signals are read at the start and end of each sampling sweep and used to adjust the signal scaling The MultiClamp 700 amplifiers supplied by Axon Instruments have two sections referred to as channels within them each of which has two outputs giving four separate signals that can be sampled The software controlling the amplifier can generate telegraph information telling interested applications about changes to the amplifier gain settings and other relevant information such as filter settings Using the MultiClamp 700 telegraph configuration dialog you can configure Signal software so that it can receive and make use of this telegraph information and control how you want the information to be used In addition to signalling changes to the amplifier gain the MC700 control software provides information about other aspects of the amplifier setup For each channel in the amplifier we can retrieve The primary output s low pass filter settings The secondary output s low pass filter settings The membrane capacitance The serie
282. oomed well in or well out This option affects only the axis the units used by the cursors etc will still be the same Checking the Logarithmic option will switch the axis from linear to logarithmic modifying the displayed range only if it included negative values In 10 2 View menu logarithmic mode another check box Show powers will appear This allows the big ticks to be labeled with powers of the big tick spacing As with the tick spacing options these changes take place imediately with no need to press Draw Y Axis Range This command double clicking a y axis or the Hi Ctr1 Y shortcut open the Y Range dialog chama 1 ADC 0 Waveform 7 _ Optimise y E The dialog sets the y axis range and style for t nza O data or XY view channels The Channel field potom o1asiezs Show al chooses one all or selected channels see the Getting started chapter for information about fF Large tick spacing a eee selecting channels If more than one channel is Trick subdivisions 5 selected the displayed settings are from the fF Logarithmic I Auto adjust units Close first channel Optimise draws the visible data scaled to fill the display Show All sets the y axis to display the maximum possible range for waveform channels and from 0 to the estimated event rate for marker channels drawn with a frequency axis Both buttons close the dialog To optimise without opening the dialog right click a channel and select the optimis
283. or any text or numeric field in a dialog with the Ctr1 V key combination Clipboard data using the private CED binary format can be pasted into compatible data views an XY view or into the arbitrary waveforms output by the pulse dialog Data is pasted into views within the displayed limits if the clipboard holds more points or channels that the visible data only the visible data is modified This is not true for XY views If the binary data holds fewer points or channels than the displayed data only part of the visible data is modified Pasting from the clipboard does not affect the data stored on the clipboard This command is used to delete the current text selection or if there is no selection it deletes the character to the right of the text caret Do not confuse this with Clear which in a text field is the equivalent of Select All followed by Delete 9 1 Signal for Windows version 3 Clear Copy As Text Data view Text output configuration Output field types Separator Decimal places Field width for all items String delimiter Add header to text block Data output When you are working with editable text this command will delete it all If you are in a memory data view this command will set all the bins in all channels to zero and redraw the window contents If you are looking at frame zero in sampled data Clear erases any overdrawn traces Clear removes everything Delete removes the current selection
284. or up to but not including the data at the second cursor For a waveform channel including one drawn as histogram etc the measurements are Curve area Each data point makes a contribution to the Mean Slope Area Sum Modulus Maximum Minimum Amplitude SD RMS Abs Max area of its amplitude above a line joining the endpoints multiplied by the x axis distance between the data points The picture makes this clearer Curve Area The mean value of all the waveform points in the region If there are no samples between the cursors the field is blank The slope of the least squares best fit line to waveform points in the region The area between the data points and the y axis zero Area is positive for sections above zero and negative for sections below zero Use Modulus if you want areas below the y axis to be treated as positive The sum of all the waveform points in the region If there are no samples between the cursors the field is blank Each waveform point makes a contribution to the area of its absolute amplitude value multiplied by the time between samples on the channel This is equivalent to rectifying the data then measuring the area over zero If a zero region is specified the amount subtracted from the other regions 1s scaled by the relative width of the regions Modulus The value shown is the maximum value found between the cursors o Minimum The value sho
285. ored on disk may not be deleted This command may be used to remove a channel permanently from an XY data document or an idealised trace from a file view Once the channel has been deleted 1t cannot be retrieved The frame buffer is an extra frame of data that is automatically provided by Signal Every open data file and memory document has a separate frame buffer that 1s used in conjunction with the document data this buffer is shared by all of the views of that document The frame buffer can be used to carry out arithmetic on frames for example subtract the average of frame 1 to 4 from all frames either interactively via the commands described below or by using the multiple frames dialog also described below The way to think of the frame buffer is as an extra frame of data that is behind the current frame in the view When you change to a different current frame the buffer moves too so that it is always associated with the current frame Understanding this association is important because all of the frame buffer arithmetic commands described below work with the current frame If you are displaying the frame buffer the buffer moves so that 1t is in front of the current frame but it is still closely associated with the current frame When the buffer is shown the view title changes to show that the buffer is visible the current frame number is still shown in brackets because the user still needs to be aware of which frame 1s current in order
286. orizontal Label Mode Renumber Horizontal 12 4 Display Y values Selecting and copying data Signal for Windows version 3 This command activates a pop up menu from which you can select the horizontal cursor to move to The cursors are listed with their number positions and channel as an aid to identification The Y axis of the relevant channel will be scrolled to display the nominated cursor in the centre of the axis or as close to the centre as possible This command does not change the y axis scaling This command is the equivalent of using the Fetch Horizontal command for all cursors There are four labelling styles for horizontal cursors as for vertical cursors use this command to select the most appropriate mode for your purposes To avoid confusion between the cursor number and the position the number is displayed in bold type when it appears alone and bracketed with the position The style applies to all the cursors in the window You can drag the cursor labels along the cursor to suit the data see the Getting started chapter When created cursors take the lowest available cursor number rather than being ordered by position You can also drag cursors over each other confusing the ordering further This command renumbers the cursors by position with cursor 1 at the top This command opens a new window Tra Toy x containing the values at the position Gusor2 Cusor3 Cursor4 of any cursors in the current da
287. othly scrolling display Repeatedly hitting the scroll buttons will double the scroll speed each time To stop the scrolling either click on the Next Previous or the other scroll button or simply click on the data window Merge will combine the current event with the one to the right to produce a single event with the combined duration of the two events but the attributes of the first Chop will break the current event in two If the current event has an amplitude between those before and after then the first and second new events created by the break will be given amplitudes and attributes from the following and preceding events respectively The Delete button will delete both the current and following event and extend the preceding event to cover the time gap created This event will have an amplitude adjusted to the weighted average of all the events previously covering the time period Spilt will divide the current event into three separate events all having the same amplitude and duration Fetch Amp will scan backwards through the trace to find an event of the same type and adopt its amplitude from there The Fit Visible button will adjust the time and amplitude of the transitions using the filter cut off frequency defined during the SCAN process to build a step response function to fit Idealised traces created using threshold crossings will be fitted using the Nyquist frequency as the cut off frequency unless this frequency is changed in
288. ow contents less the scroll bar are copied to the clipboard as text a bitmap and also as a metafile It is also copied as binary data that can be pasted into an XY view or the pulses dialog See also Copy As Text The visible data is copied to the clipboard as text The text format used is the same as was last used in Export As with text or the Copy as Text command See the Copy As Text command for details of the text output format Make sure that the window is completely on the screen and that it is not covered by any other window before copying Use this option when the required image is an exact copy of the screen If you need to scale or edit the image Metafile format is usually better The screen display is copied to the clipboard as a Windows Metafile You can read an exported image into a drawing program as either a bitmap or as a metafile Metafiles are often the preferred choice as you can treat the image as lines and text for editing Use Paste Special and select Picture in the target application to select the metafile image The visible waveform data is copied to the clipboard as binary numbers using a private CED format This binary data can be pasted into the arbitrary waveform buffer used by the pulses dialog see the Pulse outputs during sampling chapter into another data view or into an XY view This private clipboard format is not usable by other applications You can paste text on the clipboard into any text based document
289. own Script Bar settings Alternatively you can have different experimental configurations by logging on as a different user name Duplicate window Hide Show Tile Horizontally Tile Vertically Cascade Arrange Icons Window menu The Window menu has seven permanently available commands in two sections The first section holds three commands one to duplicate a data document window and two to hide and show windows The second section holds five commands four to arrange windows and the final one to close all windows The remaining space in the menu holds a list of all the windows that belong to the Signal application If you select one of the windows in the list the window is brought to the front and made the current window The list shows the current window checked The last item in the list activates a dialog which lists the windows together with their view handles used by scripts to access them and the window state maximised etc This command creates a duplicate window with all the attributes list of displayed channels display modes colours cursors and size of the original window Once you have created the new window it is independent of the original Duplicating a window allows you to have different views of the same data with different scales and different channels visible You can close all windows associated with a data document using the File menu Close All command see the File menu chapter This w
290. ows version 3 A P A AA E A E A E E E E E E E E T 19 1 Standard TAO lt Oi ap MS 25 di 19 Telegraph configuration 1401 coccccccnnnnnnnnnnonnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnss 19 1 MultiClamp700 Teletipo 19 2 MultiClamp 700 telegraph CONfIgUrAtiON ooononnncnnanooooanaanananannnnnnor nooo nnonnnnnnononnnnnnos 19 2 Auxillary States GOVICCS ii AA AAA AAA AA Ad 20 1 MASSA SUDO la tee 20 1 CEES OFU a Rmmeree eee emer tener ene eae oN ny eer or ea ee ORC aN ttre ee ee ne ene ne 20 8 CED DOAS TOCE a EE 20 8 aint oTa i E TO PPE losa 20 8 CED 3304 support configuration ccccccccccccccccccccccccccccccccccccccccccccccccccccccccceeceeeeees 20 8 Connections and Cab e do 20 10 INOS SON MISS Aia 20 10 Introduction New features in version 3 Signal for Windows The Signal software running under Windows together with one of the CED 1401 family of interfaces gives a PC the power to capture and analyse multi channel waveform and keyboard time marker data Signal is designed to let you manipulate your data using the familiar Windows idioms You can arrange the windows to display the data within them to best advantage and cut and paste the results to other applications Alternatively you can obtain printer hard copy directly from the application When you close a data file Signal saves the screen format and analysis window setup associated with it When you open a file Signal restores the configuration so i
291. pdate the flash firmware without opening the 1401 case New monitor ROMs are available from CED for the 1401p us and the micro1401 You will need to open the 1401 case to replace them we ship detailed instructions with the ROM If you are running Windows NT NT 2000 or Windows XP there is information about the Working Set Size at the bottom of the About box The two numbers describe the minimum and maximum physical memory that the operating system allows Spike2 to use If you use Windows NT and suffer from error 544 when you sample data these numbers are important 16 1 Free system resources Tip of the Day View Web site Other sources of help 16 2 Signal for Windows version 3 If you are running Windows 95 98 or Me the working set information is replaced by the Free 16 bit system resources for the GDI graphic objects and User all other objects The figures given are the percentage of free resources compared to the state when the system started up If either of these figures gets less than 10 you will find that system performance is severely impacted windows may not open and images may be missing from buttons If free resources reach 0 Windows tries to warn you unfortunately as resources have reached 0 the message box may not be legible The usual cause of running out of resources is to open many windows at the same time usually from a script On my Windows 98 system with 256 MB of memory I can generate about
292. period of 1 millisecond set in the Outputs tab of the sampling configuration would be fine The example sets DAC 0 to zero then pulses for 20 milliseconds twice once using CALL and once using CALLV Then after a delay there is a 50 millisecond pulse make sure DACO is zero these two instructions Lave THe same SCILE asec this one 20 ms pulse wait 1 second then a 50 ms pulse DAC 070 MOVI V3 ms 20 2 CALL PUL CALLV PUL V3 ms 20 2 DELAY s 1 1 CALLV PUL V3 ms 50 2 Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Neo HALT So we don t fall into PUL routine PUL DAC 0 1 set DAC value DELAY V3 wait for time set DAC 0 0 set DAC back to zero RETURN back to the caller CALL CALLV and RETURN let you reuse a block of instructions This can make sequences much easier to understand and maintain The disadvantage is the additional steps for the CALL and RETURN If you need to set a variable use CALLV and there is only the overhead of the RETURN instruction The JUMP instruction transfers control unconditionally to the instruction at the label Many instructions allow the use of an optional label to set the next instruction so you can often avoid the need for this instruction JUMP LB Jump to label The HALT instruction stops the output sequence and removes all overhead associated with it It does not stop the sequencer clock which continues to run Any cosine output will stop but will restart when the sequence restarts T
293. pic which is covered in the Sampling with multiple states chapter of this manual ADC sampling is normally done on a clocked interval basis This means that each sample point in the sweep is separated by the same time period With this box checked each point is triggered by a pulse supplied by external hardware On the 1401p us this trigger is connected to the Ext BNC connector on the front panel On more modern 1401s you should use pin 6 of the Events socket on the back of the 1401 The pins are numbered from right to left with pin 6 being the 6 hole along on the top row The Sample rate field sets the sampling rate for all waveform channels in Hz The rate displayed will not always be the preferred rate that was entered it shows the closest rate achievable given the 1401 clocks and the number of ADC ports to be sampled The overall sampling rate in the 1401 is the Sample rate times the number of ADC ports With a Power1401 625 the maximum sampling rate is 625 kHz A Micro1401 mk II will sample at up to 500kHz With a micro1401 or a 1401plus with modern 12 bit ADC hardware the maximum overall sampling rate is 333 kHz A 1401plus with a 16 bit ADC can sample at 400 kHz With a standard 1401 or 1401plus with older ADC hardware the maximum rate 1s 82 5 kHz The sampling configuration dialog does not apply hardware specific limits to the sample rates that you enter If you use a sampling configuration with an overall sampling rate beyond that ac
294. play mean of data T Error bars start time The data from each frame to be a o analysed starts at Frame start Offset and runs up to Frame start Offset Width If this data range extends beyond the end of the frame data for any sweep the sweep 1s not added into the average and an error message 1s generated If you check Average x axis starts at zero the memory view x axis starts at zero If this is clear the x axis starts at the start time of the first section of data added into the average This is Frame start Offset for the first frame analysed The Display mean of data checkbox selects between displaying the mean data value or the sum of all sweeps added into the average If you check the Error bars box extra information is saved with the result so that you can display the standard deviation and standard error of the mean of the resulting data The Waveform Draw Mode dialog controls the display of the error information The New button or Change if the memory view already exists closes the dialog creates the new memory view and opens the Process dialog described below Auto Average Amplitude Histogram Signal for Windows version 3 This analysis averages waveforms as for the ERiiRERT Sx STcens x standard waveform average processing and Channels TA automatically produces multiple averages Each i Width of l average frame is generated from a fixed number ee ee 0 040000 s of frames of source data The
295. pling see below This position can be moved by entering a non zero value in this field This will not alter when sampling takes place but just the way in which the time is displayed on the x axis 3 5 Signal for Windows version 3 ADC ports Keyboard marker Digital marker Burst mode Sweep trigger This field sets the ADC ports to sample You can enter individual ADC ports separated by commas or spaces or a range of ports such as 0 7 or both for example 0 7 1 6 Port numbers between 0 and 31 are accepted Each sampled ADC port creates a separate waveform channel in the resulting data document The ADC ports are sampled in the order specified and the data document will have all waveform channels first so the first ADC port provides data for channel 1 the second for channel 2 and so forth Duplicate port numbers are allowed and will be sampled see page 1 for a discussion of waveform channels The Keyboard marker checkbox enables the keyboard marker channel and logging of keyboard markers If the keyboard marker channel is enabled then it is the first channel in the data document after the waveform channels The Digital marker checkbox enables the digital marker channel and logging of digital markers If this channel is enabled it is the first channel after the keyboard marker channel or the waveform channels if the keyboard marker channel is not present See page 3 2 for a discussion of marker channels Check t
296. pling rate The first band lies in the 2 frequency space between 0 Hz and the lower edge of gg the second band The Frequency field has three 8 editable numbers Band 1 low Band 1 high and 8 Band 2 low These numbers correspond to the edges DS nD Ue E of the pass bands This filter passes two frequency ranges and rejects the remainder Both 0 Hz and one half the sampling frequency are rejected The Frequency field has 4 numeric fields Band 1 low Band 1 high Band 2 low and Band 2 high These fields correspond to the four edges of the two bands 15 20 2 300 35 40 45 5 This filter passes three frequency ranges and rejects the remainder Both 0 Hz and one half the sampling rate are passed The Frequency field has 4 numeric fields Band 1 high Band 2 low Band 2 high and Band 3 low These fields correspond to the four edges of the three bands Se ne EA Frequency Hz Gain dB This filter is a combination of a differentiator that is the output is proportional to the rate of change of the input and a low pass filter The y axis scale is linear rather than in dB although you can display it in dB if you wish There is one editable number in the Frequency field Low pass the end of the Gain gt 0 5 10 15 20 25 30 35 40 45 5 differential section of the filter Frequency Hz The output of the filter is proportional to the rate of change of the input The y axis scale is linear rather than in dB although you
297. points of data on 2 attempt to paste data into the waveform DAC channels at 5000 Hz duration 0 4 seconds buffer It is difficult to ensure that you 3 i 2 have copied exactly the right data onto Modi O the clipboard so Signal provides a Orvermrite 2040 points starting at offset o gt dialog to control the paste W Output length set to end of pasted data W Output frequency set from pasted data The upper part of the dialog describes Cancel the data on the clipboard and provides control of the first data point taken from the clipboard An offset of zero takes data starting with the first point on the clipboard larger offsets cause points at the start of the clipboard data to be skipped The same offset is used for all DACs The lower part of the dialog describes the current waveform output parameters and provides control over what waveform data is changed and how the waveform output is modified to match the clipboard data The upper pair of controls making up the line Modify x outputs start on DAC n set which channels of the waveform are modified If changing more than one channel channels in use starting with the DAC specified are changed The lower pair making up the line Overwrite x points starting at offset y sets the amount of data pasted and where in the waveform buffer it is put The Output length set to end of pasted data and Output frequency set from pasted data checkboxes act as their titles imply If all data
298. power output as a percentage of the maximum available you can set any value from 0 to 100 This control is disabled for manually controlled states This item is only available if single pulse mode is not in use it sets the number of pulses you can set any value from 1 to 1000 This control is disabled for manually controlled states or if single pulse mode is selected 20 4 Freg Dual 2007 devices Manual Power Power 2 Auxiliary states devices This item is only available single pulse mode is not in use it sets the pulse frequency in Hertz you can set any value from O to 100 This control is disabled for manually controlled states or if single pulse mode is selected Two MagStim 200 s can be controlled by Signal to achieve much the same effect as using a BiStim when using this configuration two separate serial lines are used to control the MagStims and two separate trigger pulses are required as for a BiStim in independent trigger mode When the dual 2007 option is selected a separate selector for the second serial line port is shown but as for the 200 there are no controls governing overall behaviour The settings available for the individual states are Set this checkbox for this state to be controlled by the manual controls and not by Signal clear it to give Signal control This control sets the main master stimulator power output as a percentage of Magstim configuration Dual MagStim 200s
299. preting the amplifier telegraphs Signal can automatically adjust for changes in the amplifier gain settings to maintain signal calibration Signal currently supports two types of telegraph signal a standard telegraph method using one or more analogue outputs that are sampled using the 1401 interface and an auxiliary mechanism provided by a customisable DLL that can be installed with Signal Currently only one type of auxiliary telegraph DLL is available with Signal this supports the Axon Instruments MultiClamp 700 amplifiers If you do not install support for an alternative amplifier telegraph then standard telegraphs using 1401 inputs will be available The standard telegraph support uses 1401 ADC inputs to sample telegraph signals in this case analogue voltages generated by the amplifier A lookup table is then used to convert these voltages to the amplifier gain setting which is used to adjust the signal calibration Up to four separate telegraph systems can be defined each controlling a separate ADC input sampled signal Setting up standard 1401 telegraphs consists of defining the ADC ports involved and setting up the table used to convert telegraph voltages to gains ee eee eo een dialog can configure up to four ports whose scaling depends upon telegraph Telegraph set Telegraph 1 outputs For each of these ports you set a port from which to read the telegraph Sa o El saca a E ignal and a list of signal values and the s
300. quence with the TABDAT directive which must come after the TABSZ directive TABDAT expr TABDAT expr expr expr Where expr is an expression that evaluates to a 32 bit integer Each TABDAT directive adds data to the table starting at the beginning The sequencer compiler will flag an error if you define more data that will fit in the table Table data declared in this way is stored separately from the sequence and is transferred to the 1401 when you create a new data file to sample If you do not set the table data with the TABDAT directive or from a script the values in the table are undefined Although you can move data between one of the 64 variables and the table with the TABLD and TABST instructions many instructions access the table directly It takes slightly more time to use a table than to use a variable All references to the table use the contents of one of the 64 variables as an index into the table plus an optional offset as Vn or Vn off or Vn off The offset off is an expression that evaluates to a number in the range 100000 to 1000000 For example if V1 holds 10 V1 refers to the contents of index 10 V1 10 refers to index 0 and V1 10 refers to index 20 Out of range table indices read 0 and are non destructive The TABINC instruction makes it easy to increment a variable used as a table index and branch until the increment generates a value outside the table size The following example generates five DAC outpu
301. query the user Changes made to non channel frame data such as the frame state or flags are always saved Memory views hold all frames in memory and do not save changes until the entire document is saved 8 3 Signal for Windows version 3 Other documents Export As Data file Text file Bitmap file Metafile Revert To Saved Text script and XY documents are held in memory Changes made to them are not permanent until the document is saved to disk This menu item available only RRS Tx when the current window is A Savei 3Signal El lle ws data view 1s used to save some Ml scrdoc a Average1 DataD4 cfs ccomm A sigemd 4 Averagel example cfs or all of the view data to a new ctslayer 1 sigutils 2 Average lotsframes cfs file in one of a number of compiler C WinDebug 4 Average3 Datal4 cfs cond CO winrel E Coldresp cts formats The dialog prompts doc 2 011602ab cts 3 cond8 cts you to choose a file name for the output and lets you select recorder 2 3677c02 cfs 2 cond9 cfs J runtime Averagel Coldresp cfts 8 DataDD cfs the output format You can ficnme choose between Data file saeasype osarierds i cancel cfs to export as a new Signal data file Text file txt Metafile wmf for a scaleable image and Bitmap file omp for a copy of the screen rectangle containing the window Select one of the formats and set a file name then use the Save button
302. r and number alone You can select the most appropriate for your application using the Cursor menu Label mode command To avoid confusion between the cursor number and the cursor position Signal draws the number in bold type when it appears alone and in brackets when shown with the position 27 435 28 2891 4 The mouse pointer changes when it is over a cursor into one of two possible shapes to indicate the two actions you can take with a cursor This shape indicates that you can drag the cursor from side to side If you drag the cursor beyond the window edge the window contents scroll to keep the cursor visible The position vanishes when dragging unless the Ct r1 key is down or you click on the label If you position the mouse pointer over the cursor label the pointer changes to this shape to indicate that you can drag the label up and down the cursor This can be very useful when you are preparing an image for publication and you need the cursor label to be clear of the data There are many ways to use Signal to adjust or customise the display or to control the data that is displayed Move the mouse pointer to a waveform channel clear of any cursor Click and drag a rectangle round a waveform feature then release the button This action zooms the display so that the area within the rectangle expands to fill the entire view If your rectangle covers more than one channel only the time axis expands If your rectangle fits in one chan
303. r so that Signal can tell which channels will be affected by the amplifier set up and gain changes Each MC700 amplifier has two channels each of which has two outputs called Primary and Secondary for the 700B Scaled and Raw for the 700A You should set which ADC ports on the 1401 interface are connected to the various amplifier outputs leave any amplifier outputs that are not being used set to None The checkbox labelled Get signal names units Multiclamp 700 configuration Ea and calibration from 4 Amplifier is Multiclamp 7008 amplifier setup selects Multiclamp 7008 serial number i 06650 how the data channels generated from ports connected to the amplifier ADC ports are set up at the start of Channel Primary fo Secondary E aa E Channel2 Primary fi Secondary None sampling the channel name and units are set using information supplied by the MultiClamp 700 depending upon what Cancel Test Help signal you have set up to be routed to the relevant output In addition the channel calibration is overwritten with values supplied from the amplifier This is the easiest way to use the MultiClamp 700 telegraph data it is generally recommended as it will automatically force the settings to be correct Get signal names units and calibration from amplifier setup 1401 ADC hardware uses 10 volt range If this checkbox is left clear then the channel name units and calibration entered in the
304. r mode item at the top selects Cancel the feature to be searched for or defines the cursor as static The Active channel field sets the channel and the start and end times set the time range to search The start or end time for a search can be a cursor position which itself could be active and changing If the start time 1s later than the end time then Signal will search backwards for the feature At the bottom of the dialog are up to three parameters Noise rejection hysteresis Threshold level and Width for slope used to define the feature more precisely one or more of these may be present depending upon the cursor mode selected The possible cursor modes are When you add a new cursor it starts in Static mode In this state the cursor stays where you put it it is not changed by a change in the position of a lower numbered cursor The result is the position of the maximum or minimum channel value in the search range There is an extra field in this mode for the Reference level The cursor is positioned at the Reference point that is the maximum distance in the y direction away from the reference level 12 1 Signal for Windows version 3 Peak Trough Rising threshold Falling threshold Threshold Steepest rising Steepest falling Steepest slope Slope threshold ve slope threshold ve slope threshold Slope peak Slope trough Turning point Slope The Noise rejection hysteresis field set
305. re are five data channels displayed in the window Channels 1 to 4 hold waveform information and channel 5 holds markers logged from the keyboard Signal arranges channels so that the lowest numbered ones are at the top by default this can be changed using the preferences dialog available through the Edit menu Below the file view window there is a minimised window with the title Log Text This is the log view a text document that is always open in Signal If you try to close the log view it is only hidden and can be re displayed using the Window menu Show command Selecting channels Click on the channel number to select a particular channel Signal highlights the channel number when it is selected Hold down the Shift key and click on a channel to select all visible channels between it and the last selection Hold down Ctrl to select discontinuous channels Click on the blank rectangular area below the y axes and to the left of the x axis to de select all channels Many commands and operations can operate on the selected channels for example y axis optimisation 2 1 Signal for Windows version 3 Toolbar and Status bar A discussion about frames The Toolbar is a shortcut to commonly used menu items Each toolbar button matches an action from the menu The buttons appear in this manual with the menu items To find out what a toolbar button does leave the mouse pointer over the button for a few seconds The Status bar provides info
306. re used to draw user defined graphs with a wide variety of line and point styles Although these views can be created interactively their major use is from the script language They are also created when generating a trend plot The file name extension 1s sxy 8 1 Other file types used by Signal Resource files Configuration files Application preferences Filterbank files 8 2 Signal for Windows version 3 In addition to the document types listed above Signal also uses a number of other types of file Signal creates resource files with the extension sgr Each resource file is associated with a data file of the same name but with the extension cfs The resource files hold configuration information about the data file display so that Signal can restore the display on loading These files are not essential to Signal and if you delete them the associated data file is not damaged in any way a new default display configuration will be used Signal stores sampling configuration information in files with the extension sgc These store all of the information needed to carry out sampling the sampling parameters the position of the sampling control panel and any other windows the position and display configuration of the data document window including any duplicate windows plus any online processing required the online processing parameters and the position and display configuration of the memory views showing the result
307. red the frame state a value attached to each frame in a Signal data file that can indicate a condition or classification of the frame You will also have encountered the Enable multiple states checkbox in the General section of the sampling configuration dialog This chapter describes the uses and control of multiple states in data acquisition Simply put Signal multiple states sampling is a way of generating data files where each frame s state value is set according to external conditions The external condition can either be signalled by external equipment or it can be controlled by Signal generating different outputs for each state in use There are a three different forms of multiple states Imagine an experiment involving measuring responses from a test subject to two forms of stimulation For example you are recording the EEG evoked response to a tone in either the left ear stimulus A or the right ear stimulus B You have external equipment to generate the different stimuli and are using Signal to record the result You want to generate two averaged responses one for stimulus A and the other for stimulus B One way to do this would be to look at signals generated by the external equipment to indicate whether stimulus A or B was used You could then generate two averages one for each stimulus This is what one form of Signal multiple states can do by reading the 1401 digital inputs and using the value read to work out a state code for ea
308. rejection mode item controls what form of artefact rejection to apply It can be set to None for no artefact rejection Tag frames to label frames with artefacts or Reject frames to discard frames with artefacts The Artefact level is the percentage of the ADC range outside which data values will be regarded as an artefact The next two items set the time range for the search for artefacts these are specified as offsets from the start of the data rather than as absolute frame times The final item Allowed bad points percent sets the limit before the frame is rejected or tagged allowing you to avoid rejecting frames with a trivial amount of bad data Unless the mode is None each sampled frame of data is scanned for artefacts All waveform channels are scanned over the time range specified and the number of points that exceed the Artefact level are counted If the number of bad points expressed as a percentage of the total points scanned exceeds the limit for allowed bad points then that frame 1s automatically rejected or tagged as appropriate Once you have set up the sampling configuration you can create a new sampling data document Either click the Run Now button in the sampling configuration dialog or the Run Now button on the Toolbar or Select New from the File menu then Data Document for the file type The exact appearance of the new document view varies depending on the configuration The name for the new document will either be D
309. rements from another for the type of measurement items for the one or two time values needed for the measurements and a measurement Width item A non zero Width will cause a measurement to be taken by averaging the readings from Width 2 to Width 2 of the specified time The types of measurement available are Value at point the channel value at the time specified using the specified width Value difference the difference between the channel value at the time specified and the value at the reference time both using the specified width Time at point the time specified This can be a cursor position 1f that cursor s mode 1s set to move to a feature 1t measures the feature position Time difference the difference between the time specified and the reference time Either or both of these can be a cursor position that can vary Frame number the frame number This is often used as the X measurement to give a plot of measurement against frame Absolute frame time the absolute start time of the frame Often used as an alternative to the frame number to give a measurement against time Fit data Signal for Windows version 3 Frame state value the frame state value Fit coefficients the resulting coefficients from a fit The coefficient index should be specified User entered value a value entered by the user A dialog opens to read the value Cursor regions any measurement available in the cursor regions window can
310. replaced by the settings dialog Processing operations can take quite a time especially in large data documents You can stop a processing operation early with the Esc key The Process command behaves slightly differently when the current window is a j Last n frames filed memory view derived from a sampling 3 Last n frames flea document The command activates a modified version of the Process dialog This dialog is Mumber of frames fid also activated automatically when you create a Frames between updates booo new memory view from a sampling document or when you press the Change button in the Process settings dialog for a similar memory view Clear memory view data before processing e Optimise Y axis after processing Cancel Settings This form of the Process dialog gives you control over when and how the memory view is updated during sampling The Frames field contains extra items that are suitable for processing sampled data Sampled frames and Last n frames filed The contents of the dialog change depending upon which frame option is selected In addition there is a new field Frames between updates Sampled frames all frames that are sampled will be processed This option is not available in Fast triggers or Fast fixed int sampling modes All filed frames all frames saved to disk are processed Last n frames filed process the most recent frames saved to disk the dialog displays a field in which you
311. resent it sets the next instruction to run This example plays a 1 Hz sinusoidal output assuming that the output is not running PHASE Ape DU set the DAC 2 phase angle directly ANGLE 20 prepare to start as a sine wave RATE 2 Hz 1 start the sinusoid 6 16 OFFSET WAITC Sequencer outputs during sampling This changes the cosine output voltage offset for the next cosine output OFFSET n DAC number in the range 0 7 available DACs depend on the 1401 type n expriVn Vn off OptlLB DAC n The offset value for sinusoidal output The units of this value are as set in the outputs page of the sampling configuration dialog It is an error to give a value that exceeds the DAC output range expr Vn When a variable is used the full range of the 32 bit variable corresponds to the full range of the DAC You can use the VDAC32 function to load a variable using user defined DAC units OptLB If this optional label is present it sets the next instruction to run This example ramps DAC 0 from 0 to 1 volt the runs 5 cycles of a sine wave at 1 Hz and finally ramps the data back to 0 volts This example does not work with a 1401plus DAC O 0 use DAC 0 for all output OFFSET 0 1 0 set DAC O offset SZ 0 0 2 1 V sinusoid PHASE O 90 Prepare sinusoid ANGLE 0 0 FSE Start polne RAMP Orto Orde 07 SL Soran to L vort in L Sec RAMPUP WAITC 0 RAMPUP walt for ramp gt Ramp up RATE 0 HZ 1 start sinusoid DELAY
312. rgument In this case the alternatives are separated by a vertical bar for example DELAY expr Vn Vnt off OptLB This means that the first argument can be an expression a variable or a table entry There is no explicit documentation for the use of the table except in TABLD and TABST Where table use is allowed it is written as Vn off If you use a table value in an instruction the effect is exactly the same as using a variable with the same value as the table entry OptLab in instructions is an optional label that sets the next instruction to run If it is omitted the next sequential instruction runs These instructions give you control over the digital output bits and allow you to read and test the state of digital input bits 7 0 The DIGOUT instruction changes the state of digital output bits 15 8 see the Sampling data chapter for the connections The output changes occur at the next tick of the output sequencer clock so you need to use this instruction one tick early DIGOUT pattern Vn Vntoff OptLab pattern This determines the new output state You can set reset or invert each output bit or leave a bit in the previous state The pattern is 8 characters long one for each bit with bit 15 at the left and bit 8 at the right The characters can be O i le GG 99 ce 39 i or standing for clear set invert or leave alone You may omit the square brackets however the Format command will insert them DIGO
313. rids Y 4 ADC 3 Waveform l l J 3 ADC 2 Wavef v Title Title and the horizontal scroll bar ee 4 2 ADC 1 Waveform Y Units Y Units 4 1 ADC 0 Waveform M Small ticks Y Small ticks Check the Chan numbers box M On right F Scroll bar for channel numbers in file and F Scale only F Scale only memory views All On and All Off select all or none of the ELA EA ES channels Draw updates the data window You also have control over the x and y axes You can hide or display the grid numbers on the axes the big and small ticks and the axis title and axis units You can also choose to show the y axis on the right of the data rather than on the left 10 3 Channel Information 10 4 File Information Options Draw mode The Signal for Windows version 3 For publication purposes 1t 1s sometimes preferable to display axes as a scale bar If you check the Scale only box a scale bar replaces the selected axis You can remove the end caps from the scale bar leaving a line by clearing the Small ticks check box The size of the tick bar can set by the Large tick spacing option in the Y Axis Range or X Axis Range dialogs or you can let Signal choose a suitable size for you Use this dialog to view and edit data view channel information You can open it by double clicking a channel title from the View menu or by right clicking the channel to open the context menu You can edit the Title of the channel set by the Channel
314. ritten using Signal State a 32 bit number that can have any value from 0 to over 4 billion it is intended for use in conditional analysis where each state value corresponds to a separate condition in the experiment but may be used for any purpose Signal can sample using multiple frame states logged from external equipment or generated internally to control 1401 outputs see the chapter Sampling with multiple states 2 2 Specifying frames Getting started Start a floating point number holding the absolute start time for the frame This value is the time for the frame x axis zero relative to the start of sampling For files collected by Signal version 1 00 this will always be zero Tag each frame is tagged or not tagged Tagging is intended for any purpose in which specific frames need to be marked for analysis or attention Flags a set of 32 flags available for any user defined purpose They are accessible from the Signal script language Variables 16 floating point numbers that can be read or written using Signal scripts Their meaning is user defined The entire data in the current frame for the view is held in memory This frame is discarded when the view switches to a different frame You may find that Signal s performance when handling large frames 1s improved by installing more memory in your PC Changes made to the current frame must be saved before a new frame is loaded or the changes will be lost You can write the
315. rm channel for keyboard markers and multimples of the outputs resolution for digital markers In addition the marker has 4 bytes of marker data The first of these 4 data bytes is the ASCII code of the keyboard character pressed by the user for a keyboard marker or an 8 bit digital code read by the 1401 for a digital marker The remaining 3 bytes are normally zero Keyboard markers time events to an accuracy of at best around 0 1 second you should use digital markers if you require precise timing The upper and lower case characters a z and the numbers 0 9 are logged but only when the new document window or the sampling control panel is the current window The keyboard marker channel if created is the first channel after the waveform channels These are not available for the Standard 1401 Digital markers are timed as accurately as the outputs and record 8 bits of TTL data These can be used as 8 separate channels of on off information or one channel of 8 bit numbers or any combination in between Digital marker data is sampled when a low going TTL compatible pulse is detected as described below The data is read from bits 0 to 7 of the 1401 digital input 3 2 Digital marker connections Digital input bits O to 7 Digital input other signals Marker codes Sampling data The digital marker data is read from the 1401 digital inputs bits 0 to 7 These inputs are found on the 1401 Digital inputs connector a 25 way D type
316. rm item only one of these 1s allowed 9 3 Signal for Windows version 3 Moving a pulse Removing a pulse Finding a pulse Se Copying pulses To move a pulse select the pulse in the display area Then drag the pulse into the new position required you cannot drag the pulse to a different output The drop position is shown as for adding a new pulse the mouse pointer changes to 7 a hand to indicate taking hold of something without addition or removal To avoid problems with precise positioning of the mouse Signal does not recognise a drag and drop operation until the mouse has moved a certain amount away from the initial position The mouse pointer shows this by only including the pulse icon and showing the drop position when the amount of movement is sufficient You can give up on a move by returning the pointer close to the initial position If you want to move a pulse a small amount but find that Signal will not recognise a drag operation that short move the mouse pointer a larger amount vertically to convince Signal that this is a drag and smaller horizontal movements will be accepted If you cannot find or click on the pulse to start dragging it usually because it is too short or completely hidden by another pulse see Finding a pulse below for how to select it Once the pulse is selected you can change the start time directly by editing it in the values area To remove a pulse select it in the display area the
317. rmation about the current view The status bar holds a number of panes plus an open area on the left This leftmost portion shows prompts from menu items as you move the mouse pointer over the menus the panes each show a particular item of information If the space available is too small for all of the panes panes disappear starting at the right hand side From the left the status bar panes are Cur Pos If the mouse pointer is over part of a view that has axes this displays the pointer X and Y positions The first figure is the channel number from which the Y value is taken Frame This pane shows the current frame number for the current view and the maximum frame number in the view If the current view is not a data document then this pane is blank See below for a discussion on frames Start This pane adjacent to the frame number shows the absolute start time for the frame For files collected by Signal version 1 00 this is always zero State This pane shows the state code for the current frame in the current view as a decimal number It is blank if the current view is not a data document Tag If the current view is a data document and the current frame is tagged described below this pane holds TAG otherwise it is blank Flags This pane shows the flags for the current frame in the current view as an 8 digit hexadecimal number hexadecimal format makes the individual flag states visible each digit shows the state of four flags with
318. rs from the current frame trace They can either all have the same colour set using the View menu Change Colours command or each frame in the list can have a different colour This adds the currently displayed frame to the frame display list If overdraw mode 1s enabled all frames in the frame display list are displayed along with the current frame If the current frame is in the display list this command becomes Remove frame from list Adding or removing a frame from the display list in this way will destroy any dynamic behaviour of the list For eample if you have requested to overdraw all tagged frames and then add another frame using this menu item the display list will then remain fixed even if you subsequently tag or un tagg a frame 10 1 Frame display list Enlarge view Reduce view X Axis range H Signal for Windows version 3 This command opens a dialog that manipulates the frame display list and also controls the Display frame list Frames Tagged frames manner in which the frames are displayed The upper part of the dialog allows you to define Frame subset Frame state wx E which frames are in the display list You can dear pp either enter a list of frame numbers directly or you can use the drop down list to select a category of frames from All frames Current Comcel frame Buffer Tagged frames Un tagged frames and Frame state xxx A second field allows a subset of these frames to be
319. rsor is copied to the clipboard Active mode Puts up the active cursor mode dialog for the selected cursor Delete Deletes the cursor For horizontal cursors the list of commands are as follows Copy Position X Copies the position X of the selected cursor to the clipboard Copy Position HCursor n Y Subtract the position of horizontal cursor n from the selected cursor position and copy the result Y to the clipboard Delete Deletes the selected cursor 12 6 Sampling configuration El Sample Bar Sample Bar List Signal conditioner Sample menu The sampling menu divides into three regions The first configures the channels to sample and provides support for users with a serial line controlled signal conditioner for example the CED 1902 The second region shows or hides the sampling and output control panels during sampling The third region matches the sampling control panel and holds commands to start continue and end sampling enable and disable sweep triggers and to enable and disable data storage to disk This command opens the Sampling Configuration dialog which sets the data capture parameters used when you select the File menu New command see the File menu chapter for details You can load and save the sampling configuration with the File menu Save Configuration and Load Configuration commands You can also access this command from the Signal toolbar You can show and hide the Sample Bar and manage t
320. rt transformer phase shifts a band of frequencies from Fjoy to Fhigh by 7 2 The target magnitude response in the band is to leave the magnitude unchanged Fiow must be greater than 0 and for the minimum magnitude overshoot in the undefined regions Fhigh should be 0 5 Fioy The magnitude response at O is 0 and if an odd number of coefficients is set then the response at 0 5 is also 0 This means that if you want Fhigh to be 0 5 or near to it you must use an even number of coefficients There is a special case of the transformer where there is an odd number of coefficients and Fhigh 0 5 Fjow In this case every other coefficient is 0 This is no help to the MSF and MSF programs but users who write their own software can use this fact to minimise the number of operations required to make a filter It is extremely unlikely that a Hilbert transformer will be of any practical use in the context of Signal so we do not consider them further You can find more information about this type of filter in Theory and Application of Digital Signal Processing by Rabiner and Gold What a signal conditioner does Serial ports Programmable Signal Conditioners Signal can control programmable signal conditioners using the computer serial line ports and can use the signal conditioners to alter input signal gains offsets or filtering before or during sampling Three types of signal conditioner are supported the CED 1902 Mk HI the Powerl401 wit
321. running Windows 95 The more memory you have and the faster the processor the better Signal will run A graphics accelerator will greatly improve drawing and scrolling speeds To sample data you need a CED Power1401 micro1401 Mk II microl401 1401p us or standard 1401 Pulse output during sampling is not available with a standard 1401 USB operation requires Windows 98SE Me 2000 or XP Your installation disk s are serialised to personalise them to you Please do not allow others to install unlicensed copies of Signal Just put the CD ROM in the drive and it will start the installation You can also run the installation manually by opening the folder Signal on the CD ROM then open the disk1 folder and run setup exe If you are installing from floppy disk you may wish to make backup copies of the installation disks before you begin and keep them in a safe place We supply the disks write protected to avoid accidental over writing but they can be physically damaged If we shipped you a CD ROM and you must install from floppy disk you can create the disks by copying the contents of the folders disk1 disk2 etc in the Signal folder to separate floppy disks The resulting disks should hold only the files NOT the folders Place the Signal installation disk 1 in a suitable drive assumed to be a below Click Start then Run then type a setup You must select a suitable drive and folder for Signal and personalise your copy with your name
322. ry and are user specific If you have several different logons set for your computer each logon will have 1ts own preferences 9 4 Display Default font for data views Editor settings Show time as Frame Start as Line widths Channel order Edit menu The Display tab contains preferences relating to the way data is shown on the screen Signal data views record their font so that they appear the same when re opened but they always have a default font when first created This font 1s shown in the rectangular area within the default font box and can be set using the Set Font button which provides a standard font selector Once data views are open their fonts can be individually set using the View menu Font command These buttons lead to dialogs for the standard Rn x settings for script sequencer and text files and to control syntax colouring The Font button Boei fat Tabs Tab size la Insert spaces T Show tabs i Keep tabs opens a font dialog that sets the font used when you open a file You are restricted to fixed pitch fonts in the editor The Tab size field sets how many spaces to move for each press of the Tab key You can also choose to keep the tabs as tab characters or have tabs converted to spaces Syntax Coloring Current Colors Keywords Change The Syntax colouring button opens a dialog where you can enable syntax colouring for script windows Enabling it
323. s Pop py rap py r 2p Pp KV Setup This disables outputs during sampling When selected only a single control is shown This item sets the period of the internal timer used to measure the absolute frame start time and to time digital markers A value of 1 to 10 ms is usually appropriate for these purposes Values from 0 1 microseconds to 250 ms can be entered and they are rounded to the nearest 0 1 microseconds This selects pulse outputs during sampling The pulses can be controlled by the script language or interactively using a dialog The details of configuring and using pulse outputs are covered in a separate chapter of this manual Pulse outputs during sampling When pulse outputs are in use a number of controls to configure the pulses are shown 3 9 Signal for Windows version 3 DAC enables scaling and Resolution ms Absolute levels Absolute times Synchronise sampling Configure pulses units This sets the timing resolution of the output pulses in milliseconds or microseconds and also sets the period of the internal timer used to measure the absolute frame start time and to time digital markers Values are rounded to the nearest 0 1 microseconds The practical limit to the resolution depends upon the type of 1401 in use for a 1401p us the recommended limit is 3 ms for the micro1401 values down to 0 1 ms can be used for the Micro1401 mk II 25 microseconds while for the Power1401 you can go down a
324. s Cubic spline mode becomes Line mode in Windows metafile output If your view has associated error information for example a waveform average with error bars enabled you can set the error display mode as None 1 SEM 2 SEM or SD Error bars only have meaning if the data points that contribute to the average have a normal distribution about the mean Given this 1 SEM shows 1 standard error of the mean 2 SEM is 2 standard errors of the mean and SD is 1 standard deviation If each point of your data can be modelled as a constant real value to which is added normally distributed noise with zero mean then you would expect the measured mean value to lie within 1 standard error of the mean SEM 68 of the time or within 2 SEM 95 of the time The standard deviation represents the width of the normal distribution of the underlying data at each data point This menu item is used to set the display mode for marker data channels You can set the mode for a single channel all suitable channels or any subset of the suitable channels When this command is used the marker draw mode dialog is provided This dialog changes depending on the display mode selected there are three modes available Dots Lines and Rate The simplest method is to draw the Draw mode for example cfs marker channel as dots You can o os 5 choose large or small dots small dots Eme 2 Keyboard Mis E can be very difficult to see on some T Lane Date Cancel di
325. s allows the big ticks to be labeled with powers of the big tick spacing Often in Signal you will need to specify time points within the frame For example specifying the X axis limits the start and end times for file export or the search limits for an active cursor Signal provides a standard control that allows you to enter a time directly or to select the frame limits Mintime and Maxtime the current display limits XLow and XHigh or the position of any cursors You can also use an offset value along with the built in values for example Mintime 0 75 or Cursor 1 0 1 Note that any numerical values entered always use the currently selected time units 2 5 2 6 Signal for Windows version 3 Now double click on the y axis of a waveform channel to open the Y Range dialog Experiment with adjusting the y axis ranges on different channels This dialog changes the y axis range of one or more channels The Channel field is a Channel 112060 f Ootimi drop down list from which you can select any Jessel T Optimise channel with a y axis all channels with y 1 2412791 Show ll axes selected channels or all visible Bottom 01451823 channels You can also type a list of channel numbers and ranges such as 1 4 6 10 l n You can either Optimise the display which mee ee ie makes sure that all the data in the window on 1 Tek subdivisions fs the specified channel s fits in the y a
326. s by how much the data must rise before a peak and fall after it or fall before a trough and rise after it to be accepted as a peak In the diagram of a peak search the first peak is not detected because the data did not rise by Hysteresis Hysteresis within the time range Start These modes have two new fields Threshold and Crossing Noise rejection hysteresis The data must cross Threshold from a level more than Noise rejection Threshold y away from it For a Rising threshold the data must Jy Noise rise through the threshold for a Falling threshold it must fall In Threshold mode the crossing can be in either direction The picture shows a rising threshold The Width for slope measurement field sets the length of data used to evaluate the slope at each data point The result is the position of the maximum minimum or maximum absolute value of the slope Maximum slope Data The result is the position at which the slope Slope crosses a particular threshold The units for the threshold are y axis units per second These modes calculate the slope of the data in the search range and then find the first peak or trough in the result that meets the Hysteresis specification The Width for slope measurement field sets the length of data used to evaluate the slope at each data point The Hysteresis field sets how much the slope must rise before a peak and fall after it or fall before a trough and rise after it to be a
327. s can be 0 1 and meaning match 0 TTL low match 1 TTL high or match anything The bit order in the pattern is 76543210 You may omit the square brackets however the Format command inserts them Vn With a variable the result is input BAND Vn 7 0 BXOR Vn 15 8 A result of 0 is a match not zero is not a match LB The destination of the branch if the input was a match DIBEQ or not a match DIBNE This label must exist in the sequence This example waits for a pulse sequence in which the falling edges of two consecutive pulses are less than 2 v1 2 sequencer clock ticks apart It waits for a falling edge waits for a rising edge with a timeout and then waits for the next falling edge with a timeout If timed out we start again If the input signal has high states less than three ticks wide or low states less than 2 ticks wide this example may miss them WHI DIBNE ode deta ss LIZ WED walt until Aga gt Wait high SETTO MOVI v1 24 set 50 step timeout gt Wait low WLO DIBNE A Ol4WBO ware tor falling gt Wait low TOES DIBEQ A Ll TOLO walt tor high gt Wait high DBNZ Vi TOHL loop if not timed out gt Wait high JUMP WHI timed out restart gt Restart TOLO DIBEQ EEEE O GOTIT jump if found events gt Wait low DBNZ V1 TOLO loop if not timed out gt Wait low JUMP SETTO timed out restart gt Restart GOTIT PR here for 2 close pulses These instructions test digital input bits 7 0 read by the last
328. s far as 10 microseconds This selects between absolute and relative pulse levels With absolute pulse levels the pulse amplitude sets the level directly with relative levels the pulse amplitude is added to the level before the pulse to get the actual pulse level This selects between absolute and relative pulse times With absolute times the pulse dialog allows you to enter the pulse start time directly with relative times you use the delay since the start of the previous pulse This control only affects the way in which the pulse dialog handles pulse start times not the other times shown in the dialog the underlying pulse data or the generation of pulses This selects between mechanisms for synchronising external triggers pulse outputs and the actual sampling sweep Normally the sweep trigger starts the sampling sweep and the pulse output mechanism which is free running throughout sampling is synchronised to the sweep This arrangement gives the maximum accuracy of the sampling sweep start time relative to the sweep trigger about 2 to 5 microseconds but will give a 10 to 20 microsecond delay in pulse outputs relative to the sampling If you select synchronised sampling the sampling sweep start is delayed relative to the sweep trigger by 10 to 20 microseconds but the pulse outputs are more precisely synchronised with the sweep to an accuracy of 2 to 5 microseconds If Outputs frame Fixed interval or Fast fixed int sweep mode is
329. s in the amplifier gain settings Signal supports a standard telegraph mechanism using analogue voltages sampled using the 1401 or 1t can use a custom DLL to support alternative mechanisms if support for an alternative telegraph system has been installed then the label on this button will change For information on setting up and using amplifier telegraphs see the chapter Amplifier Telegraphs Only a few types of amplifier support telegraph signals if you do not have such an amplifier you can ignore this chapter This option only appears when clamping support is enabled in the preferences and is described in the Sampling with clamp support chapter Use this section of the sampling configuration dialog to set the outputs required during sampling to set which DACs and digital outputs are available for use and Type Fuses y to set the DAC units and scaling Resolution ms 0 2 The leftmost area is used to configure the outputs It contains a B cone ae selector for the type of outputs mo mg fo Iv required plus items specific to the DACs 4 to type of outputs The right hand AS areas enable and set up the output a 0 7 onfigure Pulses TI E ports This control selects the type of O carcel Runnow Heb outputs to use either None Pulses or Sequence The controls in the area below the selector vary according to the selection General Port setup Outputs Automate DAC enables scaling and units Full eno Unit
330. s of online processing Signal stores some of its preferences in a file called cfsview sgp This file holds the position of the application window on the screen the colour palette and use colour switch The main preferences information from the preferences dialog is now largely stored in the Windows registry Signal initialises itself with data from this file whenever it is started and saves the current state of the software into this file when it exits These files hold descriptions of digital filters and have the extension cfb They are used by the Analysis menu Digital filters command This command opens a file into a Signal document of any type You Lookin 3 Signal oases can activate this command with the E C sigutils 8 Average3 Datal4 cfs E A cfslayer O WinDebug 8 Coldresp cts E Ctr1 0 shortcut key or from the A compiler A winrel crapres cts la cond e 011602ab cfs 8 DataD1 cfs E menu or toolbar It shows the mue E 3677c02 cts fa Data02 cts E a recorder Averagel Coldrespj cfs 2 Data03 cfs E standard file open dialog for you to runtime la Average1 Datad4 cfs E Data04 cfs k a Jsi i Averagel lotsframes cts Datal cfs a select a file You can open five pa types of file with this command a Signal data file with the standard e extension cfs a text file with the o Me Jo _ omo extension txt a sequencer file with the extension pls a script file with the extension sg
331. s or an XY data file with the extension sxy The type of the file is selected with the Files of type field if this is set to All files Signal will try to open the file selected as a CFS file whatever its extension When you select a CFS data file Signal also looks for a file of the same name but with the file extension sgr If this is found the new window displays the file in the same state and screen position as it was put away Several windows may open if the file was closed with the Close All command See the Close and Close All commands below for more details If a read only CFS data file is opened you will be warned that the file cannot be modified If a text file is opened a simple text edit window is opened This facility can be used as a notepad or as a repository for data copied from other parts of Signal If a script sequencer or XY file is opened a window of the appropriate type is created for it Import data Import Op Cl Close Close All Save Save As Data documents File menu Signal can translate data files from other formats into Signal data files The import data command leads to a standard file open dialog in which you select the file to convert You then set the file name for the result Signal will suggest the same name with the extension changed to cfs The details of the conversion depend on the file type Supported formats include the SON files used by CED programs such as Spike2
332. s pressed during sampling in the keyboard marker channel so you can have a record of where in the sampling you switched to a new portion of the sequence Here is a simple sequence that pulses DAC 0 high for 10 milliseconds ten times starting at 0 25 seconds into the sweep It for use with sequencer execution restarting each sweep 6 3 Signal for Windows version 3 You can start and stop the pulses with keys or by clicking buttons You will find this in Signal3 Sequence MyFirst pls to save typing it in 0000 MOVI Vl 10 Initialise loop counter 0001 DELAY 248 Just less than 0 25 s gt Waiting 0002 LP DAC Ura Set output high 0003 DELAY 8 Wait 9 steps 0004 DAC Q0 Set output low 0005 DELAY 7 Wait 8 steps 0006 DBNZ Vl LP Loop required times 0007 HALT Stop outputs gt Done Open MyFirst pls and click the Check and make current sequence button at the upper right of the window leave the mouse over each button for a second or so to see the descriptive text Open the Sampling configuration dialog and set a configuration with one sampled channel on ADC port 0 a sample rate of at least 10 KHz and a sweep length of a second On the Outputs page make sure the sequencer clock rate 1s set to the default of 1 millisecond and that the Free run without restarts box is not checked To record the output pulses you must connect the DAC 0 output to the ADC 0 input This 1s easily done with a BNC cable on the 1401 front panel You
333. s resistance The external command sensitivity All of these values are retrieved for both channels and the current values are placed in the first ten user frame variables in the sampled data file These values can be retrieved using the script language later versions of Signal may make use of them directly The MultiClamp 700 amplifiers supplied by Axon Instruments can generate telegraph information telling interested applications about changes to the amplifier gain settings and other relevant information The MultiClamp 700 telegraph configuration dialog is used to configure the Signal software so that it can communicate with the MultiClamp Commander software controlling the amplifier and to control how you want the link to the amplifier to be used At the top of the dialog there is a checkbox used to select between a MultiClamp 700A or 700B You should set this according to the type of amplifier you will be using For the 700A you will also have to specify the COM port used to control the amplifier and the Axon bus ID for the amplifier For the 700B you have to specify the serial number of the amplifier If you do not want to make use of the MC700 telegraph in this sampling configuration set the bus ID to Do not use for the 700A or set the serial number to 1 for the 700B 19 2 Programmable Signal Conditioners Below the amplifier identification section there is a section defining the ADC ports connected to the amplifie
334. s settings 11 6 Prompt to save result and XY views 9 4 Pulse output frame 5 1 Pulse outputs start 5 1 Pulses Adding a pulse 5 3 Configuration 5 1 Controlling online 5 9 Current pulse 5 2 Dialog 5 1 Digital bits 5 5 Display 5 1 Drag and drop 5 3 Finding a pulse 5 4 Fixed interval period 5 8 Fixed interval variation 5 8 Initial level 5 5 Moving a pulse 5 4 Paste waveform 5 7 Pulse selection 5 2 Pulse train 5 6 Ramp 5 6 Removing a pulse 5 4 Sine wave 5 6 Square pulse 5 5 Square with varying amplitude 5 5 Square with varying duration 5 6 Step change 5 8 Total variation 5 8 Trigger sampling 5 8 Values 5 3 Varying 5 8 Pulses or sequencer 3 13 R RAMP output instruction 6 13 Randomisation in output sequencer 6 26 Rate marker display mode 10 5 RATE output instruction 6 15 RATEW output instruction 6 15 RECIP output instruction 6 22 Rectify data 11 19 Reduce view 10 2 Regions dialog 2 8 Regular expression search 9 3 Relative measurements 12 4 Remove Signal 1 2 Removing a pulse 5 4 Renumber cursors 12 3 Renumber horizontal cursors 12 4 Replace text 9 4 Repolarisation cursor mode 12 3 REPORT output instruction 6 25 Resource files 8 2 Resources free 16 2 Restart sampling 3 18 Result view Prompt to save unsaved view 9 4 RETURN output instruction 6 19 Revert text document to last saved 8 4 RINC output instruction 6 18 RINCW output instruction 6 18 Rotate data 11
335. same for each band As you can see halving the gap has made a considerable increase in the ripple in both the pass band and the stop band In the first case the ripple is 1 76 in the second it is 8 7 Halving the transition region width increased the ripple by a factor of 5 In case you were worrying about the negative amplitudes in the graphs a negative amplitude means that a sine wave input at that frequency would be inverted by the filter The graphs with dB axes consider only the magnitude of the signals not the sign FIRMake can generate coefficients for four types of filter Multiband Differentiators Hilbert transformer and a variation on multiband with 3 dB per octave frequency cut The other routines can generate only Multiband filters and Differentiators The filter required is defined in terms of frequency bands and the desired frequency response in each band usually 1 0 or 0 0 Bands with a response of 1 0 are called pass bands bands with a response of 0 0 are called stop bands You can also set bands with intermediate responses but this is unusual The bands may not overlap and there are gaps between the defined bands where the frequency response is undefined You give a weighting to each band to specify how important it is that the band meets the specification As a rule of thumb you should make the weight in stop bands about ten times the weight in pass bands FIRMake optimises the filter by making the ripp
336. sampling control dialog Vn When a variable is used for the target the full range of the 32 bit variable corresponds to the full range of the DAC You can use the VDAC32 function to load a variable using user defined DAC units slope This expression sets the DAC increment per sequencer step The sign of the value you set here is ignored as the sequencer works out if it must ramp upwards or downwards to achieve the desired target value If your DAC is calibrated in volts for a slope of 1 volt per second use 1 0 s 1 0 Vs You can also set the slope from a variable or by reading it from the table In this case the full range of the 32 bit value represents the full range of the DAC The DAC changes by the absolute value of this 32 bit value on each step For a slope of 1 user unit per second use VDAC32 1 0 s 1 0 This example ramps DAC 1 from its current level to 1 volt in 3 seconds waits 1 second then ramps it to 0 volts in 5 seconds See the OFFSET command for another example RAMP LO 10753 jStart with zero Size wA Other instructions during ramp WT1 WATTC 1 WT1 wait for ramp to end gt Ramp to 1 DELAY SCL va walt for a second gt Wait 1 sec RAMP AI SSA ramp down WT2 WAITC 1 WT2 walt for ramp gt Ramp to 0 6 13 Signal for Windows version 3 Cosine output control instructions Output as a function of Theta Phi SZ The sequencer can output cosine waveforms of variable amplitude and frequency throug
337. sducers have a limited frequency response and require no filtering Connect your waveform signals to the 1401 ADC input ports Ports 0 7 0 3 for an unexpanded microl401 or Micro1401 mk II are the labelled BNC connectors on the front of the 1401 For the 1401 and 1401p us ports 8 15 are on the 15 way Cannon connector The input voltage range is normally 5 volts some 1401s are supplied set to 10 volts see the 1401 Owners manual Connections are Port 8 9 10 11 12 13 14 15 Ground Pin number 1 2 3 4 5 6 7 8 9 15 For the Power1401 without an ADC expansion box the extra ADC ports are on the back panel labelled Analogue Expansion The connections are Signal for Windows version 3 TTL compatible signals Marker channels 0 Keyboard markers Digital markers Port 15 14 13 12 11 10 9 8 Ground Pin number 35 34 33 32 31 30 29 28 1 19 Users of the microl401 and Micro1401 mk II will find all the ADC ports on BNC connectors If you have a microl401 or Micro1401 mk II ADC expansion box installed ports 4 to 15 are BNC connectors on the expansion box Expansion boxes are also available for the Power1401 For port numbers above 15 you will require a 32 channel expansion card for the standard 1401 or 1401p us or extra expansion boxes for the micro1401 Micro1401 mk II or Power1401 If you sample a port above the number available the result is undefined In several places in this manual we refer to TTL compatible signals TTL sta
338. se at sweep end checkbox controls whether a new sweep is started automatically once the previous sweep has ended If the checkbox is set then sampling will pause until the Continue button is pressed or until the checkbox is cleared again allowing the user to pause for a while or to inspect the data and accept or reject each frame If the checkbox is clear then the next sweep starts immediately Write to disk Pause at Effect No Continuous sweeps for signal monitoring No Continuous sweeps written to disk Interactive sweeps written to disk 1f accepted Interactive sweeps removed from disk 1f rejected The sampling control panel can be hidden or shown using the Sampling menu the Signal toolbar or the pop up menu provided by right clicking the mouse on unused parts of the Signal window It can also be hidden by clicking on the x button at the top right of the control panel or minimised using the button Once sampling has started the sampling control panel changes to show buttons suitable for the sampling process M Sweep trigger If Event 1 start was checked before you click Start the Stop word Waiting flashes until a suitable signal is applied to the Reject Abot Restat E1 input to enable sampling This is distinct from the Sweep end Sweep trigger Use this method to synchronise the start of a a f ee Pause at sweep end sampling with an external event Sampling starts within 1 or 2 us of the external event signal The b
339. sgc each time sampling finishes successfully This opens the printer page setup dialog The dialog varies between operating systems and printers See your operating system documentation for more information For file memory and XY views the important options that are always present include the paper orientation portrait or landscape the paper source if your printer has a choice the printer margins and the choice of printer In text based views in addition to page margins you can set header and footer text to appear on each page If you include f in the text it is replaced by the file name p is replaced by the page number and c is replaced by the file or system time 8 5 Signal for Windows version 3 Print preview Print visible Print and Print selection Sl Print screen Exit This command allows you to view the current document as 1t would be printed You can preview file and memory views and text based windows You can zoom in and out step through pages of multi page documents and print using a toolbar at the top of the screen Use the Close button to leave this mode These commands print data views X Y views cursor windows and text based windows The scroll bar at the bottom of the window is not printed Print selection prints the selected area of a cursor or text window Print visible prints the visible data in the current window Print or its Ctr1 P shortcut prints a specified region of a data view at t
340. side the search Position range The search start position defines the Hera 80 0 repolarisation level The 100 position PO o rre 100 and Width fields identify the 100 level this can lie outside the search range Repolarise drawn at 80 in the picture sets a threshold relative to the 0 and 100 levels The position is the first point in the search range to cross 1t The cursor position is obtained by evaluating the Position field This field will normally hold an expression based on cursor positions for example Cursor 1 2 5 This command opens a pop up menu in which you select a cursor to remove or you can delete all cursors The cursors are listed with their number and position as an aid to identification Deleting a cursor removes 1t from the view other cursors are not affected This opens a pop up menu where you select a cursor to place in the centre of the x axis This command activates a pop up menu from which you can select the cursor to move to The cursors are listed with their number and position as an aid to identification The window scrolls to show the cursor in the screen centre or as close to it as possible This command has no effect if there are no cursors If there is a single cursor the command behaves as though you had used the Move To command and selected it When there are multiple cursors the window is scrolled and scaled such that the earliest cursor is at the left hand edge of the window and the la
341. splays You can also select Lines ET in place of Dots The picture above left shows the result of both types of display If you select lines mode the display of marker values 1s suppressed The rate display mode counts how Pr many markers fall in each time period and ane the result as a histogram annel 5 Keyboard Rate 3 The result is not divided by the width Binsize 1 s sane of the bin so strictly speaking it is a count histogram not a rate This form of display is especially useful when the marker rate before an operation 1s to be compared with the rate afterwards ok e Elo Dravy There are three draw modes for idealised traces Basic Convolution and Both The Basic mode draws the 9 3 201 Chan 1 idealised Both trace as horizontal lines representing Trace offset 8 78997 pAmp M Show baseline _ Cancel event amplitudes separated by ae vertical lines for the transitions EA Closed states are drawn in a different colour There is an optional offset to allow the trace to be drawn below or above the raw data when displayed with a shared locked axis The Convolution is a continuous curve formed by the convolution of the Basic idealised trace and the step response function The step response function is the error function erf defined as the integral of the 10 5 XY Draw Mode Join style Line type and Width Solid Diotted Dashed Marker and Size Dot Cross Diamond OBox C
342. st time you calibrate the channel you must tell the system how to scale the signal into y axis units For example to set up the y axis scales in microvolts you do the following Open the Sampling Configuration dialog Select the ADC port in the Ports setup page Press the Conditioner button to open the conditioner control panel Adjust the gain to give a reasonable signal Make a note of the gain G you have set Close the signal conditioner control panel Set the Units field of the Channel parameters to uV Set the Full field to 5000000 G You only need do steps 6 and 7 once Any subsequent change to the conditioner gain will adjust the channel Full value to leave the units in microvolts ae oe a For the more general case imagine you have a transducer that measures some physical quantity Newtons for example and it has an output of 152 5 Newtons per V If you wanted the y axis scaled in Newtons you would replace steps 6 and 7 above with 6 Set the Units field of the Channel parameters dialog to N 7 Set the Full field to 5 152 5 G To work this out you must express the transducer calibration in terms of Units per Volt in this case Newtons per Volt multiply this by 5 to get the Full value at five volts then divide it by the gain of the conditioner 18 3 18 4 Conditioner connections Signal for Windows version 3 If you have set an offset in the conditioner and you want to preserve the mean signal level
343. stored in variables V57 V64 which you can also refer to as VDACO VDAC7 The values are stored as 32 bit numbers with the full 32 bit range corresponding to the full range of the DAC This high resolution allows us to ramp the DACs smoothly Values written to the DACs are expressed in units of your choice The DAC scaling set in the sampling configuration Outputs page determines the conversion between the numbers you supply and the DAC outputs All DACs are scaled identically The standard settings for a system with 5 volts DACs 1s to set the DAC outputs in volts If you write to a DAC that does not exist the variable associated with the DAC is set as if the DAC were present Output to 1401p us DACs 4 7 is mapped back to DACs 0 3 For the Micro1401 and Power1401 output to DACs that do not exist has no effect The Power1401 DAC 2 and 3 outputs are on pins 36 and 37 of the rear panel 37 way Cannon D type Analogue Expansion connector Suitable grounds are on the adjacent pins 18 and 19 With a Power1401 top box with additional front panel DACs the rear panel DAC outputs are mapped to the two highest numbered DACs For example with a 2709 Spike2 top box DACs 2 and 3 are available as BNC connections on the front panel and the rear panel DACs become DAC 4 on pin 36 and DAC 5 on pin 37 The DAC instruction writes a value to any of the 8 possible DAC outputs ADDAC adds a value to the DAC output The output value changes immediately unless the DAC
344. successful the fitted curve corresponding to the capacitance decay will be drawn on top of the waveform data The two buttons labelled gt lt and lt gt below the waveform display area can be used to increase and decrease the time range shown Immediately below the waveform display area is an area used to show the results of the most recent analysis These results are updated after every frame the results can optionally be the average of all stimulus transitions or just one and can also be averaged over a number of sweeps if desired Membrane readings for clamp set 1 0 2 0 0 0 2 0 4 0 6 0 8 gt lt The holding P otential controls Total resistance 48 3 MOhm 20 7 n5 adjust the membrane holding Access conductance 46 n5 membrane conductance 37 7 nS potential or current in exactly Transient time constant 153 us capacitance 3 16 pF the same way as the equivalent HP Im o M diliransiiong Average ofreadinas f controls in the clamping control 200 aa Sree toolbar LE 0 The All transitions Clear Display Time constant count i 00 checkbox when checked causes 0K the response to all of the edges of the selected stimulus pulse to be analysed rather than just the first edge the one shown in the waveform display If more than one edge is analysed the results shown in the text area are based upon the averaged results for all edges This control also affects the results shown in the graphical display of a measurem
345. sulting data x axis adjusts to start at pre trig time Thus a negative value gives an x axis starting at some positive value because the first point in the frame was sampled some time after the trigger Similarly a positive value gives an x axis starting at a negative value as some points sampled before the trigger are shown The port setup section defines the individual ADC ports You can set the scaling and units for data sampled from a port the name of a General Port setup Outputs Automate Fot ero Full Units Mame Options 5 Wavel data channel taken from a port daria f 5 W Waveform Hone and specify online processing O 5 y ADC 2 None options for data from a port 5 7 ADC 3 None i 5 y ADC 4 None The main dialog displays the 5 y ADCS None current settings for all of the 5 Y ADC E None available ADC ports Double click z i ADC Mone 5 Y ADC 8 None on the entry for a particular port to open the parameters dialog for that Telegraph Conditianer port The entries for each port both in the main dialog and in the parameters dialog are Cancel Run now Help Zero The value in the specified units corresponding to a zero volt reading from the ADC This value along with Full is used to convert ADC data into the floating point values used by Signal Full The value corresponding to the full scale reading from the ADC To scale the data in volts this will be 5 for a 5 volt 1401 and
346. t amplitude will be the smallest Although you can define multiple bands for a differentiator it is unusual to do so Almost all differentiators define a single band that starts at 0 Occasionally a differentiator followed by a stop band is needed A Hilbert transformer is a very specialised form of filter that causes a phase shift of 7 2 in a band often used to separate a signal from a carrier The theory and use of this form of filter is way beyond the scope of this document Unless you know that you need this filter type you can ignore it This is a variation on the multiband filter that can be used to filter white noise to produce band limited pink noise The filter is identical to the band pass filter except that the attenuation increases by 3 dB per octave in the band each doubling of frequency reduces the amplitude of the signal by a factor of the square root of 2 It is used in exactly the same way as the multiband filter A waveform is sampled at 1 kHz and we are interested only in frequencies below 100 Hz We would like all frequencies above 150 Hz attenuated by at least 70 dB A low pass filter has two bands The first band starts at 0 and ends at 100 Hz the second band starts at 150 Hz and ends at half the sampling rate Translated into fractions of the sampling rate the two bands are 0 0 1 and 0 15 to 0 5 The first band has a gain of 1 the second band has a gain of 0 We will follow our own advice and give the stop band a wei
347. t and range to display Coefficients Set the starting point for your fit and optionally fix coefficients Results Display the fitting results and residual errors The three buttons at the bottom of the dialog are common to all pages The Help and Close buttons do what they say Do Fit attempts to fit with the current fit settings This page of the Fit Data dialog controls the type of fit the data to fit and what to display The area at the bottom of the window gives a synopsis of the current fit state Fields are You can select a single channel from the current view If this is a file or memory view the channel must have a y axis If you change the display mode of a marker based channel any fit associated with the channel will most likely become invalid The fit to use 1s defined by its name and the order of the fit a number For example an exponential fit allows single exponents or double exponents The window at the top of the dialog displays the mathematical formula for the fitting function The following fits are currently supported N is the maximum order allowed Name N Comments Exponential 2 This fit includes an offset You can force a zero offset in the coefficients page Set a local reference point for the fit otherwise the even numbered coefficients may become too large to be useful Polynomial 5 These fits do not require starting values for the coefficients Gaussian 2 If you attempt to fit two overlapping peaks you may need
348. t is easy to resume work where you stopped in a previous session You can analyse sections of data by reading off values at and between cursors or by applying the built in frame by frame automated analyses such as waveform averaging and power spectrum More ambitious users can further automate both data capture and analysis with scripts The script language is described in The Signal script language manual If you or your colleagues use the DOS SIGAVG or Patch and VClamp software you can transfer data files to Signal without effort All these programs use CFS data files and Signal will directly read and use them The DOS programs can also read data files generated by Signal but it may not be able to access all the information stored in them Version 3 of Signal is completely compatible with earlier versions it will read data files and sampling configurations created by Version 2 xx without problems Scripts that ran in version 2 should work without modification The main new features in Version 3 are e Curve fitting has been substantially extended to include polynomial and sine fits with enhanced fit quality information including a display of residuals e Curve fitting is now available in XY views e A text sequencer has been added allowing more complex control of the digital and analogue outputs e Single channel patch clamp data analysis including idealised trace formation and dwell time and amplitude histograms e Logarithmic axes
349. t users identify the filter by an index number in the range 0 to 11 The type of the filter is set by the drop down list to the right of the display If you need a filter that 1s not in this list you can generate it from the script language Users of the FIRMake script language command should note that the bands referred to here are pass bands In the script language there are additional stop bands between the pass bands There are currently 12 different filter types This has no effect on your signal This filter type covers the case where you apply a low pass filter designed for a higher sampling rate to a waveform with a much lower sampling rate so that the pass band extends beyond half the sampling frequency of the new file This removes any signal the output is always zero This filter type is provided to cover the case where you apply a high pass filter designed for a higher sampling rate to a waveform with a much lower sampling rate so that the stop band extends beyond half the sampling frequency of the new file This filter attempts to remove the high frequencies from the input signal The Frequency field holds one editable number Low pass the frequency of the upper edge of the pass band The stop band starts at this frequency plus the value set by the Transition gap field eee 16 20 25 30 3h 40 e 5 A high pass filter removes low frequencies from the input signal The Frequency field holds one editable number High pass t
350. ta a alae 0205473 0 0276712 view Columns for cursors that are 00317383 7 00927734 00366211 0012007 absent or for which there is no data 0 737305 00146484 0 078125 0 141602 1 17188 9 0 0830078 are blank 0 915527 00145484 0 0l4 484 The values displayed depend upon the channel type and display mode There is an entry in the table for each channel displayed The displayed values are as follows Waveform The y axis value of the nearest data point that is within one sample period of the cursor or nothing if there is no data point close enough Waveform measurements are not affected by the drawing mode Marker as Rate The height of the rate bin that the cursor crosses If the cursor lies on a bin boundary the cursor is considered to lie in the bin to the right Marker The time of the next marker at or to the right of the cursor The Time zero check box enables relative cursor time measurements If checked the cursor marked with the radio button is taken as the reference time and the remaining cursor times are given relative to it The reference cursor displays an absolute time not 0 The Y zero check box enables relative cursor value measurements The radio buttons to the right of the check box select the reference cursor The remaining channels display the difference between the values at the cursor and the values at the reference The values for the reference cursor are not changed You can select areas o
351. ta included so an offset of zero will use data from the beginning of each frame Again this value is in source x axis units Below these fields are three checkboxes used to enable various options The first checkbox is used to force the start time of the memory view data to zero if this is clear then the memory view data x axis start will be copied from the first data added to the average The second checkbox selects display of the data as a mean value if this is clear then the sum of the data is displayed The third causes error values to be calculated and displayed in the memory view The first thing to do is to select the channels to analyse For this example we use All waveform channels so select this option using the pop up menu Set the other fields as they are in the picture above 0 04 seconds width and a start offset of O 2 10 Getting started Once you have set these values click the New button to generate the new memory view Now set the data frames to analyse When you click the New button several things happen Signal creates a new memory window ready to display the result of the analysis the Settings dialog vanishes and the Process dialog appears You must now set frames from the data document to analyse The three check boxes determine how to treat the result of the analysis You can choose to clear the memory window before you analyse the data otherwise each new average 1s added to the previous one Signal can als
352. tance during a larger pulse and subtracted from the recorded traces to leave only the ion channel effects Normally an average leak trace is Settings for LeakSub1 IVDATA x Leak subtract method P N v Response chan 1 chan 0 1000 Hz v Stimulus chan 2 ADC chan 1000 Hz y Baseline time Cursor y ms Pulse time Cursor y ms Measurement width hooo ms Form from first E Subtract from next fe V Base line correction FT Count excluded frames assembled from a number of small pulses to minimise the effects of noise but this makes no substantial difference to the technique Leak subtraction makes special use of Process Analysis menu two channels the stimulus channel which is used to measure the amplitude of the pulse so it can be scaled but is not modified by the analysis and the response channel which is the only channel modified by the leak subtraction process All other channels are ignored and copied unchanged The Settings dialog holds fields for the leak subtraction mode and the channels for the stimulus and the response to be corrected The Baseline time is a time within the sweep where there is no stimulus and the Pulse time is a time where there is a stimulus These are used to measure the stimulus amplitude the level measurements are averaged over the requested width The last two edit fields specify the frames to use to calculate the theoretical scaled trace and which frames to subtract the leak from These
353. tangles check this box to draw axes as lines We use rectangles to make sure that axes drawn with pens of other than hairline thickness join up correctly The Sampling tab contains preferences for the sampling of data When Signal samples data the new data is stored in a temporary file while it is being collected and only stored in a final CFS file when the file is saved The Directory for new data field sets the directory in which Signal puts this temporary file If this field is blank Signal will use the current directory which may not be where you expect so it is a good idea to set one If you want to save over a network for example or store new data files on media that is slow to write you can use this field to ensure that the temporary file directory which is where all the real time writing occurs 1s on a fast hard disk When you save a new data file Signal prompts you for the file name What happens next depends on where you choose to save the file If the file is on the same volume disk drive as the temporary file Signal just renames the file If the volume is not the same Signal copies the file then deletes the original The Prompt for file comment after sampling item is used to encourage entry of a file comment when a new data document has been created by automatically popping up the file comment entry dialog In order to correctly show the sampling rates attainable and limits to the pulse output resolution Signal needs t
354. tates always changing The state sequencer will have control of the frame states and the sequencer will always move on to the next stage after a sampling sweep regardless of whether the data is written to disk or not Basic 0 Press these buttons when in manual mode to use a state immediately Buttons for unused states are hidden as are buttons for states numbers greater than 9 If you have set labels for the states in the pulses configuration dialog these labels will be used in the buttons instead of the standard names During automatic states sequencing these buttons are disabled but they are pressed automatically to show the state in use State n To the right of the individual state buttons is a state selector and spinner that can be used when in manual mode to choose any state from those available The selector is most useful when there are more than 9 states the limit for individual state buttons As for the buttons if you have set a label for a state this is used instead of the simple state name Selecting a state uses it immediately During automatic states sequencing the current state is shown Signal for Windows version 3 Protocol ordering Starting a protocol Stopping a protocol Auxiliary state hardware When protocols are used some of the states bar controls are hidden and other controls are now displayed Those controls that are retained behave in the same way as described above while the On write and Cyc
355. te levels 3 10 DAC 3 14 DAC enable 3 10 DAC scaling 3 10 DAC units 3 10 Delay relative to sampling 3 10 Digital 3 14 Digital outputs enable 3 11 Pulse output frame 5 1 Relative levels 3 10 Start of pulses 5 1 Synchronisation with sampling 3 10 Time resolution 3 10 Type 3 9 Outputs dialog Absolute times 3 10 Outputs frame length 5 8 Outputs frame sweeps 3 4 Overdraw mode 10 1 p Pass band 17 6 17 7 Paste data into view 9 1 Paste text 9 1 Patch clamping 4 1 Pausing sampling 3 18 Peak between cursors 12 6 Peak search cursor mode 12 2 Peri trigger Configuration 3 7 Level adjust online 3 19 Peri trigger sweeps 3 4 PHASE output instruction 6 16 pls file extension 8 1 Point style in XY views 10 6 POINTS output instruction 6 24 Port Configuration 3 9 Full value 3 8 Options 3 9 Zero value 3 8 Power spectrum 11 3 Number of sweeps 10 4 Power1401 configuration adaption 9 7 Preferences 9 4 Assume Power1401 hardware 9 7 File comment at sampling end 9 7 Font for text views 9 5 Line widths 9 5 Time units 9 5 Preferences file 8 2 Preferences folder 3 20 Previous frame 10 1 Button 2 3 Print Line thickness 9 4 Line widths 9 5 Preview printed output 8 6 Print screen 8 6 Printing data 8 6 Selected cursor values 12 4 12 6 Print control Line thickness 9 4 Printing 8 5 Process dialog 2 11 11 5 for new file 3 17 11 6 Process frames dialog 11 5 Proces
356. ted It functions identically to the Pulses outputs Resolution ms control it has the same hard limits of 0 1 microseconds to 250 milliseconds and the recommended limits for the various types of 1401 are the same as documented for Pulses outputs Note however that some instructions will have a speed penalty such that they will start to limit just how fast you can run the sequencer on modern 1401 s DIV and RECIP in particular may take twice as long to execute If this item is left clear sequencer execution will be restarted at the first instruction at the start of each sampling sweep specifically at the time that the data point at time zero 1s sampled This allows you to easily produce sequencer outputs at a particular time in the sampled data but the sequencer is halted between sampling sweeps If this item is checked the sequencer starts running at the time that sampling starts before the first sampling sweep is started and continues to run until sampling is stopped Sometimes you may want to stop users activating sequence sections with the keyboard or from the sequencer control panel for example when an inadvertent change in a DAC output controlling a force feedback device might hurt the subject This item allows you to do this The script language SampleKey command can always activate sequencer sections This control defines the file holding the instruction sequence to be used You can either enter a file name directly or you can us
357. terrupt is a request to the 1401 processor to stop what it 1s doing at the earliest opportunity and do something else then continue the original task The time delay between the interrupt request and the instruction running depends on what the 1401 is doing when the clock ticks and the speed of the 1401 This delay is typically a few microseconds so instructions do not occur precisely at the clock ticks but changes to digital output bits 15 8 do Changes made by the sequencer to the 1401 DACs and digital output bits 7 0 occur a few microseconds after the clock tick The table shows the minimum clock interval the approximate time per step and the extra time used for cosine and ramp output for each 1401 Notice that the first two are in units of milliseconds and remainder are in microseconds Power Micro mk Il micro1401 1401 plus Minimum tick ms 010 025 10 3 0 Time used per tick us lt l lt 8 lt 10 Cosine penalty tick us 0 55 5 10 Ramp penalty tick us 0 5 0 7 3 10 The Minimum tick is the shortest interval we recommend that you set The Time used per tick is how long it takes to process a typical instruction The Cosine penalty tick is the extra time taken per cosine output The Ramp penalty tick is the extra time taken per ramped DAC Time used by the sequencer is time that is not available for sampling transferring data back to the host or arbitrary waveform output To make best use of the capabilities of your 1401 you shoul
358. test is at the right hand edge The four cursor labelling styles are None Position Position and Number and Number Select the most appropriate mode for your purposes from the pop up menu To avoid confusion between the cursor number and the position the number is displayed in bold type when it appears alone and bracketed with the position The style applies to all the cursors in the window You can drag the cursor labels up and down the cursor to suit the data see the Getting started chapter 27 435 29 2891 1 This command renumbers vertical cursors by position with cursor 1 on the left This menu command is available when a data view is the current window and there are less than four horizontal cursors in the view A new horizontal cursor is added at the centre of the data for the lowest numbered visible channel The cursor is given the lowest available number and is labelled using the horizontal cursor label style for the window The delete command activates a pop up menu from which you can select a horizontal cursor to remove or you can delete all of them The available cursors are listed with their number position and channel number as an aid to identification Deleting a cursor removes it from the window other cursors are not affected This opens a pop up menu where you select a horizontal cursor that is placed in the centre of the visible y axis for the relevant channel 12 3 Move To Level Display All Horizontal H
359. th Fixed interval mode in use two extra SL eae ACE parameters Interval s and Vary appear Frame s fi Trigger s o Interval s E With Fast fixed int mode the Interval s field tewlM f Vaz 50 is available but not the Vary These set the IO timed interval between pulse output frames which cannot be less than the pulse output frame length and the percentage variation in the interval from 0 to 100 If the variation 1s non zero the frame interval used while sampling will vary randomly between Interval Vary and Interval Vary 5 8 Overview Sequencer control panel Sequencer technical information Sequencer outputs during sampling An output sequence 1s a list of up to 2047 instructions that are executed by the 1401 ata constant user defined rate to produce the outputs required Sequences are defined by a sequence file which is a text file with a pls file extension The Signal sequencer has the following features e It controls digital output bits 15 8 to produce precisely timed digital pulse sequences In the Power1401 and Micro1401 it can also control digital output bits 7 0 e It controls the 1401 DACs Digital to Analogue Converters to produce voltage pulses and ramps e Itcan play cosine waves at variable speed and amplitude through the DACs e Itcan test digital input bits 7 0 and branch on the result e It supports loops and branches and can randomise delays and stimuli e It has 64
360. that need to be recorded so that the analysis can distinguish data recorded under the various conditions The General page in the sampling configuration dialog contains a checkbox labelled Multiple frame states which is used to enable multiple states in data acquisition With this checkbox clear sampling does not use multiple states and all sampled data frames are set to state zero With this checked sampling will use multiple states and set the data frames to the appropriate state value When multiple frame states are enabled the sampling configuration dialog gains another page labelled States This page holds controls used to set up multiple states The State variation selector at the top left of the States page select the type of multiple states to use from External digital Static outputs and Dynamic outputs The controls shown in the dialog change dramatically as each type of multiple states is selected so the three forms are described separately below 7 1 Signal for Windows version 3 What type of states do I need Dynamic outputs states State sequencing Numeric non protocol sequencing modes For almost all purposes Dynamic outputs states are the most useful as this allows you to set different output pulses for each state External digital states can be used only with external hardware that generates digital codes to indicate what is happening while Static outputs states are limited to unchanging digital and DAC o
361. the Windows clipboard 9 6 Pulse outputs during sampling The DAC select parameter specifies which DACs are used Its value is made up by summing codes for each DAC wanted the code values are shown in the table Thus for DACs 0 and 1 enter the value 3 The Rate Hz parameter sets the output rate for the data points for each DAC and the Points parameter sets the number of data points for each DAC The maximum output rates possible vary according to many factors such as the ADC rate and the pulse output timing resolution Tests carried out at CED with two channels of ADC data sampled at 10 KHz show that waveform rates of 70KHz are possible using a 1401 plus and rates of up to 275KHz can be achieved with a micro1401 Setting a waveform The Signal data view copy operation places data onto the clipboard in a private format which can be used by the pulse dialog To place suitable data on the clipboard open a data file and adjust the display so that the required time range and channels are visible then use the Edit menu Copy command to place the data onto the clipboard If you then open the pulse outputs configuration dialog select or add a Clipboard holds 2040 points on 2 channels sampled at a waveform output item and then press frequency of 10000 Hz duration 0 204 seconds Ctrl V the Paste command Use clipoard data starting at data offset E shortcut Signal recognises this as an aveform pulse item outputs 12 000
362. the Channel information dialog Amplitudes having the assumed amplitude flag set are held fixed then a second pass of fit is made freeing up assumed amplitudes above a certain duration This duration is defined in the Advanced Parameters dialog which can be accessed by clicking the Parameters button Often the fitting routines may produce unexpected results and there can be a number of reasons for this 1 There is too much level data at the start and end of the fit Restrict the visible area to only view the transitions you are interested in fitting 2 The initial guess is too good When the initial guess for the fit is so close to the correct solution the fitting routines can be unable to detect which direction to move the parameters to improve the result This can be overcome by worsening the initial guess manually first Dragging the amplitude of a principle event usually works best 3 The initial guess is too poor Use the other edit functions to produce a more plausible guess 4 An amplitude needs fixing If you flag an event as having an assumed amplitude it will be held fixed for the fit If 1t is a very short event the assumed amplitude flag will remain set otherwise the fit routine will clear the flag 9 10 Toolbar Edit bar Status bar Next frame Previous frame lt lt gt gt Goto frame Show buffer Overdraw frame list Add frame to list View menu This menu divided into six regions c
363. the digital bit controls set or modify the state bit patterns to various useful ways These are intended to allow simple patterns to be set up quickly and to help to produce more complex ones Invert This inverts all of the bits for all states This is useful for converting all zeroes to all ones and a walking one into a walking zero 1 5 Signal for Windows version 3 Static outputs states 0000 This sets all of the bits for all states to zero This is useful for clearing the control before entering new patterns 0100 This sets the bits to zero with a walking 1 This leaves state zero all clear sets bit O for state 1 bit 1 for state 2 and so forth The pattern 1s not adjusted to skip disabled digital input bits 1234 This sets the bit values to count the states using binary code Thus state 1 has just bit 0 set state 2 has bit 1 only and state 3 has both bits 0 and 1 Again the pattern is not adjusted to skip disabled digital input bits When sampling using External digital states Signal behaves much as it does with multiple states disabled The only difference is that the state value for sampled data frames varies according to the digital inputs Static outputs states are more complicated than aad General Port setup Outputs States Automate External digital states as Signal controls the state of State variation Static outputs Ordering Numeric each sampled data frame Humber of extra states E At the st
364. the file memory XY and text based views on the _ Fase A Heading Screen printed on e screen to one printer page The screen is o Portrait Landscape Include c for date or ZHc for long date scaled to fill the printer page and the views are scaled to occupy the same proportional positions on the printed page as they do on the screen The command Ye ema leads to a dialog in which you can set a page title If you use c in the page title then the date and time will be included in the title in compact form e g 23 11 2004 15 47 31 Using c will include the date and time in long form e g 23 November 2004 15 47 31 The exact format will depend on the regional settings and preferences set in the operating system Landscape print orientation often looks better than portrait as the output aspect ratio is closer to that of the screen You can also choose to print view titles and choose to draw a box round each view The fonts used by the printer for text views may appear stretched in order for the view s proportional size to be preserved If you prefer the font to have an aspect ratio closer to that used on the screen then check the box to have it do this gt Views V Print Titles Print Borders Preserve text view character Aspect ratio This command will close all open files and exit from Signal If there are any data documents or text based files open containing changes that have not been saved you will
365. the highest on the left This pane is blank if the current view is not a data document view Caps If the keyboard Caps lock is on this pane displays the text CAPS Num If the keyboard Num lock is on this pane displays the text NUM Record This pane displays REC if Signal is recording user actions into a script You can hide and show the toolbar and status bar by using the View menu You can also drag the toolbar and stick it to any of the 4 sides of the application window This is known as docking the toolbar Frames are a central concept within the Signal software A CFS file or Signal data document consists of a number of sections or frames each frame corresponds to a sweep of data A data document view normally displays data from a single frame at a time this is the current frame for that view You can have duplicate views of the same data document each view can have a different current frame so you can examine separate parts of the file simultaneously Each frame in a Signal data document has the same number and type of channels but may have varying frame start and end times Each frame holds channel data from the various channels in the source data file Usually all of the waveform channels will have the same number of data points while the number of markers can vary In addition to this channel data there are a number of other data items attached to each frame Comment a line of text of up to 72 characters that can be read or w
366. the histogram The two regions labelled Include and Exclude represent the flags associated with each event A burst duration begins with the Settings for OpClAmpl example 201 ADC 0 Idealized Maximum amplitude High y Low y 0 022 y 200 Exclude Open time Closed time Y Firstlatency Level 3 Truncated T Level 4 Assumed amp Level 5 e Bad data Level 6 Channel Minimum amplitude Bin size Mumber of bing Include e Opentime Level Closed time Level 2 PT Firstlatency Level 3 l Tmncated Level 4 Assumed amp Level 5 Bad data T Level 6 Level 1 FT Level 2 lt 1 lt 1 41 Settings for Burstifconstest x Channel 201 Chan 1 Idealised Bin width jo sms Maximum duration hoo me Critical interval oa me Include I Open time I Level Open time T Closedtime Level Closed time TT Firstlatency Level 3 Firstlatency Level 3 TN Trnested Leveld Truncated Leveld Assumedamp Level5 Assumedamp Level5 Bad data T Level E Y Bad data T Level 6 Level T Level 2 start time of an event included using the rules described for the Open Closed time histogram and ends at the start time of an excluded event having a duration longer than the critical interval 11 16 Append frame Append frame copy Delete frame Delete channel The frame buffer Clear buffer Analysis menu Th
367. then a a OK as transition is deemed to have taken ES E Ge place The critical level is defined as a percentage of the difference between the baseline and the full open level To begin scanning for the next transition imediately would probably result in another transition being detected straight away when in fact it was just part of the same transition already found For this reason the scanning jumps ahead by an amount to get past the transition just found This amount is specified as a percentage of the filter length the filter length being defined as the time taken for the step response to get from 1 to 99 of the step amplitude For a Gaussian filter this turns out to be 0 6165062 f where f is the 3 dB filter cut off frequency Rise time is defined as 0 3321412 f Percentage of filter length to jump after a transition Free amplitudes longer than this multiple of rise time An average of the data point values found while looking for a transition is kept and used as the default amplitude for an event If this amplitude is less than the critical level from the baseline then the event is flagged as closed The data skipped over by the transition detection is checked for turning points that might indicate that a transition in the oposite direction to the previously detected one had been missed If a turning point is found a transition is inserted at a time calculated from the amplitude of the turning point and an assumption that the original
368. tial Steps Step change The total number of pulse forms generated is one more than Steps as the variation includes the basic pulse without any step changes In the example shown above seven pulses will be generated with amplitudes of 30 20 10 0 10 20 and 30 mV Each pulse will be generated twice in the order shown the entire sequence will repeat after 14 pulses If you wanted a sequence that ran 30 20 10 0 10 20 and 30 you would set the basic pulse amplitude to 30 and the step change to 10 With Basic Peri triggered Fast triggers and ER a Fast fixed int sweep modes the length of the Framets Tmigger s 1 ulse output frame is set by the data frame In Hevi Mots jo eae Fast triggers a Fast fixed int Seemee Reese sio modes the pulse frame length is the same as the data frame length in Peri triggered mode the pulse frame length is the data frame length less any pre trigger points With Outputs frame sweep mode in use the pulse frame length can be set independently of the data frame and the data acquisition sweep starts at a fixed time within the pulse output frame These times are set by extra parameters shown with the Initial level data for all outputs The Frame s parameter sets the length of the pulse frame this value can not be less than the sampling sweep length The Trigger s parameter sets the time relative to the start of the pulse outputs frame of the start of the sampling sweep Wi
369. ticks because step 6 adds an extra step into this part of the sequence so we get 10 ticks overall for the low phase too Step 6 is DBNZ V1 LP which causes the sequencer to decrement the value of variable number 1 and if it has not reached zero branch to the label specified so that the next instruction executed will be step 2 As we have put the value 10 into variable 1 at the start of the sequence this will cause the loop that creates a pulse to execute ten times so we get ten pulses Finally step 7 is HALT which causes sequencer execution to stop until the start of the next sweep where it will be automatically restarted at step 0 This step displays the text Done in the sequencer control panel while it is executing The simple way to use the sequencer is with restarts so that the sequence restarts at the first step at the start of each sweep However this mode of operation may prove insufficiently flexible for some situations If you check the Free run without restarts box in the Outputs page you get a different style of sequencer operation You will find a sequence that operates in this manner in Signal3 Sequence MySecond pl1s this also demonstrates some other aspects of sequencer use 0000 FL F DAC Diz Start fast pulse gt 29 LOL SLOW 0001 DELAY ms 9 1 Wait 9 ms gt S for slow 0002 DAC 0 0 Set sOUEDUG Nigh gt S for slow 0003 DELAY ms 8 1 Walt 8 ms gt S for slow 0004 JUMP FL Continue gt S for slow 0005 Sh S
370. timulation be triggered using one or more TTL pulses generated by the 1401 digital outputs This allows the pulse timing to be precisely controlled relative to the Signal sampling A Rapid can be configured to either generate a train of pulses in response to a single trigger or to generate one pulse per trigger allowing complete control of the patterns of stimulation Manual triggering of the 20 1 Signal for Windows version 3 MagStim configuration stimulator is also possible as is triggering from other external hardware in which case you would use the MagStim trigger to trigger the sampling sweep as well The MagStim support monitors the stimulator health both waiting until the MagStim is ready for use before allowing a sampling sweep to proceed and terminating sampling if a hardware problem develops To aid data analysis Signal saves the MagStim stimulation parameters used in the data section variables of the sampled file The parameters saved are the power level the pulse interval BiStim only the secondary power level BiStim and dual 2007 the pulse frequency Rapid only and the pulse count Rapid only These values are placed in the first user frame variables in the sampled data file The saved values can be retrieved using the script language later versions of Signal may make use of them directly To configure a MagStim in Signal open the sampling configuration dialog and make sure that Multiple states is enab
371. tion to wait at this step for the specified number of sequencer ticks into the sampling sweep plus one more for the delay instruction itself this could be changed but we have kept it this way to match the behaviour of the Spike2 sequencer The sequencer control panel displays the text to the right of the gt for the current step so while this step is executing the text Waiting is displayed Note that we wait for 249 ticks 0 249 seconds This means that the next step will execute at precisely 250 steps into the sweep or 0 25 seconds Note also that we could have written DELAY ms 249 1 to achieve the same effect and that this would have the advantage of being immune to changes in the sequencer tick interval Step 2 1s DAC 0 2 which sets the output of DAC zero to two volts This step has a label at the start of the instruction text LB so we can branch back here from elsewhere in the sequence 6 4 Free running sequence Sequencer outputs during sampling Step 3 is DELAY 8 which causes sequencer execution to wait at this step for nine sequencer clock ticks thus setting the width of the pulse The width will include the time taken by the DAC instruction as well giving ten ticks 0 01 seconds overall Note that as for step 1 we could have written DELAY ms 9 1 to achieve the same effect Steps 4 and 5 are the same as steps 2 and 3 they set the DAC output low again and wait for 8 sequencer ticks We only wait for 8
372. to manually adjust the guesses for the peak centres to get convergence Sine i You can fit a single sinusoid with an offset If the frequency guess in radians is not reasonably close the fit may not converge Sigmoid l A single sigmoid may be fitted You fit data over a defined x axis range set by the between and and fields You can choose values from the drop down list or type in simple expressions for example Cursor 1 1 There must be at least as many data points to fit as there are coefficients For example to fit a double exponential which has 5 coefficients you need at least 5 points Most fits will use many more points than coefficients This 1s the x axis position to use as the zero value of x in the fitting function The most common value for this would be the start point of the fit However in some cases you may want this to be elsewhere For example in exponential fitting you may want to calculate the likely amplitude of a trace at some position Making this position the reference point makes it easy to calculate the amplitude it is the sum of the even numbered coefficients All fits except the polynomial are done by an iterative process Each iteration attempts to improve the coefficient values The iterating stops when improvements in the fit become insignificant the iteration count is exceeded the mathematics of the fitting process suggests that the fit is not going to improve or there is a mathematical problem This
373. to use the buffer This command keyboard shortcut Ctr1 0 clears the data in all channels of the frame buffer to zero This is the initial state of the buffer after a CFS data file has been loaded 11 17 Signal for Windows version 3 Copy to buffer Copy from buffer Exchange buffer Add to buffer Subtract buffer Average into buffer Multiple frames This command keyboard shortcut Ins copies the data in the current frame into the frame buffer This command keyboard shortcut Ctrl Ins copies the data in the frame buffer and any count of sweeps averaged to the current data frame This command keyboard shortcut Shift Ins exchanges the data in the frame buffer and any count of sweeps averaged with the data in the current frame This command keyboard shortcut adds the data in the current frame to the data in the frame buffer There is an alternative form of this command available only through the Ctr1 shortcut and the Multiple frames dialog which adds the buffer data to the data in the current frame This command keyboard shortcut subtracts the data in the frame buffer from the current frame There is an alternative form of this command available only through the Ctr1 shortcut and the Multiple frames dialog which subtracts the current frame data from the buffer This command keyboard shortcut Enter adds the data in the current frame into an average accumulating in the frame buffer The addition
374. to variable Vn and sets the initial value when the sequence is loaded Anywhere in the remainder of the sequence where Vn 1s acceptable name can be used name can be up to 8 characters must start with an alphabetic character and can contain alphabetic characters and the digits O to 9 Variable names are not case sensitive A variable name must not be the same as an instruction code or a label There 1s no need to specify a name or an initial value If no initial value 1s set a variable is initialised to 0 even if not included in a VAR statement Signal automatically assigns v56 the name VDigIn and variables v57 through v64 the names VDACO through VDAC7 The following are all acceptable examples VAR V1 Waitl ms 100 Set name and initial value VAR V2 UseMe Set name only so value Is 0 VAR V3 200 No name initialise to a value VAR V4 No name initialised to O When a variable is used in place of a bit pattern in a digital input or output instruction bits 15 to 8 and bits 7 to O have different uses In the expressions that describe these operations we write vn 7 0 and vn 15 8 to describe which bits are used BAND means bitwise binary AND if both bits are 1 the output is 1 otherwise 0 BXOR means bitwise exclusive OR if both bits are different the output is 1 otherwise 0 When used in one of the cosine output angle instructions the 32 bit variable range from 2147483648 to 2147483647 represents 180 up to 180 degrees The Var
375. tor is set to Finish protocol execution finishes otherwise execution switches to the protocol selected and carries on When a protocol is chained to it starts off completely normally so that the Turn on writing and Reset pulse steps checkboxes take effect These checkboxes do not take effect if the last step in the 7 4 External digital states Sampling with multiple states protocol has a Next item of step 1 so it is meaningful to allow a protocol to chain to itself The checkboxes at the bottom control what happens when a protocol actually finishes and are disabled if the protocol chains to another protocol State zero when protocol finishes enables automatic switching to state zero at the end if it is clear then the last state set by the protocol will continue to be used Turn off writing when protocol finishes controls disabling of writing sampled data to disc External digital states are the simplest but most a General Port setup Outputs States Automate limited form of multiple states External equipment State variation E sternal digital gt generates a digital code of Number of extra states E up to 8 bits corresponding to the current experiment state Signal reads this code from the 1401 digital inputs 8 15 at the end of a sampling sweep and uses it to set the state for each sampled data frame There is no software control over ran Cae oe A the state values so all of 0 Digital inputs enable the st
376. tput but the output will return to zero again after a time that depends on the corner frequency of the filter The lower the corner frequency the longer it takes for mean level change to decay to zero A notch filter is designed to remove a single frequency usually set to the local mains power supply 50 Hz or 60 Hz depending on country The Reset Calib button resets the calibration information to show raw volts taking into account the current gain and offset The units for the calibration will be set to V and the port Full and Zero values adjusted as appropriate On the CyberAmp this option will use the native calibration information and units specified by a SmartProbe if present The remaining options are for the 1902 only This is present for the 1902 only and can be thought of as a high pass filter with a corner frequency of 0 16 Hz However it differs from the high pass filters as it is applied to the signal at the input the high pass filters in the 1902 are applied at the output The 1902 provides two conditioned trigger inputs and one output This control selects which of the inputs is connected to the output If you change the gain or offset in the control panel Signal will adjust the port Full and Zero values in the sampling configuration to compensate so as to keep the y axis showing the correct values This means that if you change the gain the signals will still be correctly calibrated in the file However the fir
377. tputs page of the configuration Static output states can only be used with the outputs type set to None in the Outputs page as otherwise the state values would conflict with the other outputs When sampling using Static output states controls for the state and states sequencing are now provided see the section Controlling multiple states online below 7 6 Controlling multiple states online Non protocol ordering Sampling with multiple states When sampling using dynamic outputs or static multiple states Signal provides a states control bar to allow you to control the sampling state and sequencing The control bar can be docked at the edges of the Signal application or be left floating When sampling starts the states control bar will be shown initially and can be hidden or shown during sampling by using the sample menu or a popup menu that is displayed if you right click on an unused part of the toolbar area The states control bar 1s set up differently depending upon whether numeric random or protocol sequencing are being used without protocols extra buttons for controlling states sequencing are provided whilst when using protocols extra items to select and run a protocol are provided along with optional individual protocol buttons The states control bar contains a number of buttons and controls Reset Fause On write Basic 0 State 2 State 4 Basic O Y Idle Manual Cycle State 1 State 3 Reset Pr
378. ts are not connected If the channel is Joined each point is linked to the next by a straight line If the channel is Looped the points are joined and the final point is linked back to the first point The Line type and Width fields define the lines joining the points These two fields set the type of line used to join data points The Width field determines how wide the line is in units of half the data line width set in the Signal Preferences dialog If you set a Width of other than 1 the Line type field is ignored and a solid line is drawn The Size field sets how far in points the markers extend around their screen position A size of 0 makes the markers invisible There is a wide range of marker styles to choose from including boxes circles triangles and vertical and horizontal bars This command sets the font B that is used for each window Font Font style Size The font selection dialog is Times New Roman Regular E de me 8 de Cancel Small Fonts generated by the operating system and varies with the version of Windows Terminal Times New Roman F Wide Latin F Wingdings x The font size changes the space allocated to data channels in a data view Smaller fonts give more space to the channels however fonts Sample AaBbYyZz need to be large enough to Script read easily You can set Western y different fonts in each data or text window 10 6 Use Colour
379. ts at 5 different intervals TABSZ 10 table of 10 items TABDAT ms 1000 3 VDac32 1 1000 ms 1 volt TABDAT ms 100 3 VDac32 2 100 ms 2 volts TABDAT ms 50 3 VDac32 3 E 50 ms 3 volts TABDAT ms 500 3 VDac32 1 500 ms 1 volt TABDAT ms 200 3 VDac32 0 200 ms 0O volts MOVI vV1 0 use Vl as table index set O TLOOP DELAY V1 programmed delay DAC 0 V1 1 set DAC O to the value TABINC V1 2 TLOOP add 2 to V1 branch if in table If you have a very long data sequence you should consider using the table as a buffer The basic idea is to divide the table into two halves and use a script to transfer new data into the half of the table that the sequence is not using To find out where the sequence has reached look at the value of the variable used as an index with SampleSeqVar Set a large enough table size so that the time taken to use half the table is several seconds 6 9 Signal for Windows version 3 Sequencer instruction 6 10 reference Digital I O DIGOUT DIGLOW Each instruction below is followed by an example The examples show the preferred instruction format however the system is flexible For example a comma should separate arguments but a space is also accepted The patterns used for digital ports should be enclosed by square brackets however you may omit the brackets if you wish Many of these instructions allow you to use a variable or a table entry in place of an a
380. ts own Sample Bar settings Signal supports serial line controlled programmable signal conditioners These devices amplify and filter waveform signals and can provide other specialist functions If a suitable conditioner is installed in your system this command opens the conditioner dialog see the Programmable signal conditioners chapter for a full description 13 1 Signal for Windows version 3 Show Sampling controls Ml Start sampling Stop sampling Continue sampling Triggered sweeps Write to disk at sweep end Pause at sweep end Abort sampling Restart sampling Show Pulse controls pl Sample now E H Show Sequencer controls This command or its toolbar equivalent hides and shows the sampling control panel the menu item is checked when the control panel is visible The main controls within the control panel are duplicated in this menu as the Start sampling Continue sampling Triggered sweeps Write to disk at sweep end Pause at sweep end Abort sampling and Restart sampling commands see the Sampling data chapter for full details of the control panel commands In summary the commands are This command starts sampling It is the same as the sampling control panel Start button When sampling has started the Start sampling command changes to Stop sampling This is equivalent to the sampling control panel Finish button There is no warning before this command takes effect This command enables sa
381. tures If you check the Free parameters box dragged features are not limited by the next band and will push bands along horizontally If you clear the box the horizontal motion of a dragged feature is limited by the next moveable object To the right of the frequency response display is a slider that controls the number of filter coefficients In general the more coefficients the better the filter However the more coefficients the longer it takes to compute them and the longer to filter the data If you check the Auto box the program will adjust the number of coefficients for you to produce a useful filter The traffic light display above the slider shows green if the filter is good amber if the result is usable but not ideal and red if the result is hopeless If you change a filter or create a new filter you will be prompted to save the filter bank when you close the digital filter dialog Filter bank Filter types All pass All stop Low pass High pass Band pass Band stop Digital filtering A digital filter definition is complex and it would be tedious to specify all the properties of a filter each time you wanted to apply one to data To avoid this Signal contains a filter bank of 12 filter definitions This filter bank is saved to the file Filtbank cfb when you close Signal and reloaded when you open it When you use the digital filter dialog you specify which filter you want by the filter name Scrip
382. tware and are recorded at CED in the name of the original licensee 2 All support for the software is expected to be through one nominated person usually the original licensee 3 The additional licensed copies are expected to be used on the same site and in the same building laboratory and by people working within the same group 4 When upgrades to the software become available that require payment both the original licence and the additional licences must be upgraded together If the upgrade price is date dependent the date used is the date of purchase of the original licence If some or all of the additional licences are no longer required you can cancel the unwanted additional licences before the upgrade 5 If you are the user of an additional licence and circumstances change such that you no longer meet the conditions for use of an additional licence you may no longer use the software In this case with the agreement of the original licensee it may be possible for you to purchase a full licence at a price that takes into account any monies paid for the additional licence Contact CED to discuss your circumstances 6 If you hold the original licence and you move all licences are presumed to move with you unless you notify us that the software should be registered to someone else 1 4 Getting started Introduction h this section you will open a Signal data file manipulate the contents and familiarise yourself with t
383. tween cursors 12 5 Maximum waveform output rate 5 7 Mean value between cursors 2 8 12 5 Membrane analysis 4 4 Memory Minimum required in Windows 1 2 Memory view 2 9 Behaviour 2 12 Clear 9 2 Creating a new view 11 1 Loading as file view 2 12 Number of sweeps 10 4 Saving to disk 2 12 Menus Analysis 11 1 Cursor 12 1 Edit 9 1 File 8 1 Help 16 1 Sample 13 1 View 10 1 Windows 15 1 Metafile image export 8 4 Metafile output scaling 9 4 Microseconds time 9 5 Milliseconds time 9 5 Minimum between cursors 12 5 Modify channel data 11 19 Modulus of data 12 5 MOV output instruction 6 21 MOVI output instruction 6 21 Moving a pulse 5 4 MOVRND output instruction 6 28 ms sequencer expression function 6 7 MUL and MULI output instructions 6 23 Multiband filter 17 7 Multiband with 3 dB octave cut 17 8 MultiClamp 700 telegraph configuration 19 2 MultiClamp 700 telegraphs 19 2 Multiple frame operations 11 18 Multiple frame states 3 5 7 1 3304 20 8 Auxiliary devices 20 1 DAC outputs for state 7 6 Digital output bits 7 6 MagStim 20 1 Number of states 7 2 Online control 7 7 Randomised 7 3 Sequencing 7 3 Set number 7 2 Static outputs 7 6 Uses 7 1 Multiple states enable 3 5 7 1 N NEG output instruction 6 21 Negate data 11 19 New document 3 16 8 1 Temporary directory for data files 9 4 New file from existing file 8 4 New memory view 11 1 New XY view 11 7 Next fram
384. tween the 90 and 5 points being fitted This gives us the time constant for the decay The area underneath the overshoot is also measured to give the total amount of charge 4 5 4 6 Having the total charge and the amplitude of the voltage step the membrane capacitance can be calculated as charge delta volts We are then able to estimate the access resistance by Rs time constant capacitance which allows us to separate the membrane and access components of the total resistance For current clamp experiments only the overall resistance is measured When all transitions are being analysed the results of analysis of each transition are averaged to give a single set of measurements for the sweep Pulse outputs during sampling Introduction In the Sampling data chapter you will have encountered pulses as an option within the outputs section of a Signal sampling configuration and details of the connections for these outputs Signal can generate outputs from the 40 plus Power1401 and micro1401 using the DACs and the digital outputs This chapter describes the pulses configuration dialog and its use during sampling Signal pulse output like Signal data acquisition is arranged as fixed length frames Depending upon the sampling sweep mode the pulse output frame may be the same length as the sampling sweep longer or shorter In all circumstances the pulse outputs are fixed in time relative to the sampling sweep In Basic
385. type incorrectly this will Ste 0 maru contal not always be detected though errors during ALE sampling will almost certainly result Below the device selection is a selector for the COM port used to control the stimulator you can select any port from COMI to COM9 As many PCs only have one COM Spe al Power 20 port you may have to use a USB serial port expander to provide a spare COM port we have found that most of these USB serial Test ports work satisfactorily Cancel Help ower Ls power 0 Below the basic stimulator configuration there are controls that vary according to the type of stimulator selected These are in three sections first there are controls which govern the overall behaviour for all states next is a display of settings for all states which is used to select a state for editing by clicking on the information for that state and finally an area where the settings for the selected state may be edited 20 2 200 device Manual Power Copy Test BiStim device BiStim in independent trigger mode Use Hi res timing mode Manual Auxiliary states devices This 1s the simplest type of MagStim and does not require any controls for overall behaviour The settings available for the individual states are Set this checkbox for this state to be controlled by the manual controls and not by Signal clear 1t to give Signal control This sets the stimulator power output as
386. uration make sure that digital output bit zero is enabled for use Using the pulses configuration dialog set the initial level of digital output bit zero to 0 and place a pulse which will be high going in the outputs at the time when you want the MagStim to fire This output pulse should be at least 10 microseconds long 1 millisecond works well 3 The MagStim support uses Signal multiple states sampling which should be set up in dynamic outputs mode You can use any number of extra states each extra state providing separate MagStim settings Each set of pulse outputs should include digital output pulses to trigger the MagStim along with any other outputs required The MagStim support will work correctly with manual control of the states or with any style of automatic states sequencing including protocols When Signal begins sampling with the MagStim support enabled it checks for a correctly functioning MagStim device as part of the process of initialising for sampling If a MagStim is found then Signal will carry out the initial configuration of the MagStim and arm the device While sampling is in progress Signal will set up the MagStim using the current state data before each sweep it will then delay the start of each sweep until the MagStim reports that 1t is armed and ready At the end of each sweep the MagStim health is checked to make sure it is OK Note that the checks on MagStim readiness can impose a significant extra inter sweep de
387. usage Signal output bit number bit7 bit6 bit5 bit4 bit3 bit2 bit1 bitO GND Pulses and DIGOUT pins 1 14 2 15 3 16 4 17 13 DIGLOW pins 5 18 6 19 T 20 8 21 13 Digital input bits 8 to 15 are used in External digital states mode and in Peri trigger sampling Digital input bits 0 to 7 are read by the output sequencer and will be used in the future for digital markers Signal refers to both sets of inputs as bits 0 to 7 The 1401 digital inputs are on a 25 way D type plug this is on the right hand side of the 1401 and 1401plus front panel and on the rear of the Powerl401 and microl401 The Power1401 and micro1401 digital inputs 8 and 9 are also on front panel BNC sockets labelled as event inputs 0 and 1 matching Signal usage If the Spike2 digital I O expansion top box is fitted it has digital inputs 10 to 15 on front panel BNC sockets labelled as events 2 to 7 also matching Signal usage Signal input bit number bit7 bit6 bitS bit4 bit3 bit2 bit bitO GND External digital peri trig pin 1 14 2 15 3 16 4 17 13 Sequencer Marker pins 5 18 6 19 7 20 8 21 13 Automation configuration Path File name template Save file to disk Sampling limits Sampling data This section of the sampling configuration dialog controls the General Fort setup Outputs Automate Signal automation features There E Filing are two areas of the dialog Filing m O m which controls automatic file File name template
388. used this item is ignored as the sweep is triggered directly by the pulses system giving the same effect as if synchronised sampling was selected When using Basic sweep mode it is possible to start sampling at exactly the time of the sweep trigger by providing the trigger pulse to both the 1401 EO and E4 inputs This mode of operation is only available when synchronised sampling is disabled On the micro1401 Microl401 mk II and Power1401 in this circumstance the trigger input is automatically routed to both EO and E4 internally thus guaranteeing a precise sampling start relative to the trigger Press this button to configure the output pulses using the pulse configuration dialog Details of this are given in the chapter Pulse outputs during sampling This section contains four sets of controls one for each DAC users of microl401s and Micro1401 mk IIs should ignore DACs 2 and 3 These control if a DAC is available for use and set the scaling and units with which DAC values are defined Enable These checkboxes enable the DACs for use Set a checkbox to use a DAC leave it clear otherwise The fewer DACs are enabled for output the more space is available for the display of each DAC in the pulse dialog 3 10 Digital outputs enable Outputs type Sequence Sampling data Zero The value in calibrated units corresponding to a zero value output from the DAC This value along with Full is used to convert the floating point valu
389. uted noise is unlikely to model the data or that the errors in the original data have probably been over estimated 11 11 Signal for Windows version 3 11 12 Context menu Testing the fit Chi square fits Least square fits If you right click on this page you are offered a context menu that contains Copy Log and Log Titles commands The Copy command copies selected sections of the results or all the results if there is no selection to the clipboard as text It also copies the page as a bitmap The Log command prints a one line synopsis of the current fit to the log window The Log Titles command copies a suitable set of titles for the logged data When you fit a model to measured data to obtain the best fit coefficients there are two questions you would like answered 1 How well does this model fit the data Put another way how likely is it that this model plus some degree of random variation can explain my data set 2 Given that the model does fit the data how much confidence can I place in each of the fitted coefficient values When we talk about fitting curves to data we are making the implicit assumption that you took measurements from some process that follows a model and that this model can be expressed as a mathematical function with adjustable parameters which are our fitting coefficients Further we assume that the measurements you make are not perfect they have random variations with a known probability
390. utputs so both of these are more limited The main part of this chapter will cover Dynamic outputs states with the other forms of multiple states described afterwards Dynamic outputs states are the most useful and general form of multiple states each state generates a separate set of output pulses The actual digital and DAC outputs for each state along with some other information including the state label are set by using the Configure Pulses dialog available from the Outputs tab of the sampling configuration the states page is purely concerned with defining how many states there are and how they are sequenced how Signal will switch from state to state during sampling The Number of extra states item controls the number of extra states from 1 to 256 Note that this item sets the number of extra states in addition to state zero Thus in the example shown below there are 5 states possible with codes running from 0 to 4 In many circumstances Signal will use only states 1 to 4 state zero is treated as an background or idle state This also controls the states available in the pulses configuration dialog you should set the extra states you want here before you set up the pulse outputs themselves Because in Dynamic i G t outputs mode Signal is controlling the states as it General Fort setup Outputs States Automate is in Static outputs mode R P State variation Dynamic outputs Ordering Numeric we need ways of
391. uttons available are Continue This is labelled Start before data capture starts It is only enabled when sampling is paused at the end of a sweep When you click on the button sampling of the next sweep is enabled If the Sweep trigger box is checked the 1401 system waits for the sweep trigger before starting to collect data Stop This is displayed after data capture starts If you click on this button the data capture stops in the same manner as if one of the Automate limits was reached Once the data capture has been stopped no more sweeps will be collected but it is possible to resume sampling again Accept Clicking on this button writes unwritten frame O data to disk If frame zero is still being collected it overrides the Write to disk at sweep end checkbox so that the frame is written to disk at sweep end it does not affect subsequent sweeps If the frame has been collected and sampling is paused this writes the frame to disk immediately The button acts upon a sweep up until the point that the next sweep is triggered or pulse outputs for the next sweep begins Reject If Write to disk at sweep end is checked or sampling is paused at the end of the sweep and frame 0 has been written then the label on the Accept button changes to Reject Clicking on Reject either overrides the Write to disk at sweep end checkbox to cause the currently sampling frame not to be written automatically or removes the current frame 0 data from the end
392. ves just like the original file view but has only one frame This is a good time to experiment with manipulating the data in the memory view without worries about overwriting file data Use the Analysis menu Modify channel command and try out the options note that most of the options have keyboard commands assigned You can save memory view data to disk as a CFS file When the CFS file is reloaded into Signal it appears in a file view Now select the File menu Save As command This displays a standard Save As dialog to allow you to select a name for the file to hold the memory data The memory view data will be saved as a CFS data file Once you have entered a suitable name and saved the memory view close it using the File menu Close command You can open the file holding the memory view data by using the File menu Open command it is now opened as a file view with a single frame In addition to file and memory views Signal also uses XY views These hold multiple data channels up to 256 that share the same x and y axes Each channel is a list of x y co ordinates and has its own point marking style line style and colour XY views have a wide range of uses ranging from user defined graphing to drawing pictures XY views can be used from the script language Signal can also create XY views holding data taken from measurements from data files Use the Script menu Run Script command and select the Load and run command Locate the Scripts
393. ving configurations 3 20 Scale data 11 19 SCAN method 11 13 Screen dump 8 6 Script Bar 14 2 Script menu Evaluate 14 1 Script Bar 14 2 Scroll bar show and hide 10 3 Search for text 9 3 Selecting a channel 2 1 Selecting channels 2 1 Send Mail 8 5 Sequencer Compile sequence 6 2 Example 6 3 Expressions 6 7 Format text 6 3 Format with step numbers 6 3 Instruction format 6 7 Instructions 6 6 Set file to use 6 3 Variables 6 8 Sequencer control panel 6 1 Sequencer or pulses 3 13 Sequencer outputs 6 1 Sequencer technical information 6 1 Serial number 16 1 SETS output instruction 6 25 SGC Configuration file extension 8 2 SGP Preferences file extension 8 2 SGR Resource file extension 8 2 sgs file extension 8 1 SGS standard file extension 8 1 Shift data 11 19 Show hidden window 15 1 Show Hide Axes 10 3 Channels 10 3 Edit bar 10 1 Grid 10 3 Scroll bar 10 3 Status bar 10 1 Toolbar 10 1 Signal conditioner 18 1 Connections 18 4 Sample menu 13 1 Sine wave output 6 14 Skyline draw mode for waveforms 2 6 Slope of line 2 8 12 5 Slope peak and trough cursor modes 12 2 Slope search cursor modes 12 2 Slope cursor mode 72 2 Smooth data 11 19 Software help desk 16 2 Standard 1401 telegraphs 19 1 Standard deviation curve fitting 11 12 Standard display settings 10 3 STATE output instruction 6 25 States sequencing 7 3 Static output states 7 6 Static outputs states 7 6
394. w event details Edit menu If the time interval set by this field passes without the background routine running it is scheduled to run as soon as possible You can use values in the range 1 to 200 milliseconds The standard value is 20 milliseconds The lower the value the more time Signal will spend on background processing relative to other applications This field also limits the time that Signal will sleep for see the discussion of threads below When Signal gets idle time see the discussion of threads below you can limit the time it uses before Signal goes to sleep Signal uses idle time to run script idle routines such as those created by ToolbarSet 0 You can set from 0 to 200 milliseconds 0 means no time limit The standard value is 10 milliseconds The larger the value the more the script idle routine runs at the expense of other applications As an alternative to limiting the script idle routine by elapsed time you can limit it by the number of times it is called You can set from 0 to 65535 times 0 means no limit The standard value is 0 Setting both this and the Maximum duration field to 0 is unkind to other applications Setting this to 1 is the most generous to other applications A thread is the basic unit of program execution a thread performs a list of actions one at a time in order To give you the impression that a system with one processor can run multiple tasks simultaneously the system scheduler
395. with a frequency a 4 4 d 17 10 General multiband filter Band pass filter with 111 coefficients Digital filtering You can define up to 10 bands However it is unusual to need more than three The most common cases with three bands are called band pass and band stop filters In a band pass filter you set a range of frequencies in which you want the signal passed unchanged and set the frequency region below and above the band to pass zero In a band stop filter you define a range to pass zero and set the frequency ranges above and below to pass 1 You must still allow transition bands between the defined bands exactly as for the low and high pass filters the only difference is that now you need two transition bands not one Also if you want a non zero response at the Nyquist frequency you must have an odd number of coefficients For our example we will take the case of a signal sampled at 250 Hz We want a filter that passes from 20 to 40 Hz 0 08 to 0 16 with transition regions of 7 5 Hz 0 03 If we say it is 10 times more important to have no signal in the stop band than ripple in the pass band and we want 70 dB cut in the stop band we will get 50 dB ripple in the pass band because a factor of 10 is 20 dB To use the formula for the number of coefficients we need the mean attenuation ripple in dB and the width of the transition region The two stop bands have an attenuation of 70 dB and the pass band has a ripple
396. without being aborted or failing the sampling configuration is held in memory for use the next time you sample You can use the File menu Save Configuration As command to save the configuration used to disk You re use a saved configurations by loading it with the File menu Load Configuration command before you use the File menu New command to create a sampling document 3 20 Introduction Online clamp support features Sampling configuration Sampling with clamp support Signal incorporates specialised features directly supporting whole cell and single channel clamping experiments In order to avoid confusing users who are not using this type of experiment these features can be hidden or shown by using a checkbox in the Signal preferences dialog the initial state of this option is set up by Signal if it detects that it has not already been set The features controlled by this option are the online clamp support as documented below leak subtraction analysis and the generation and analysis of idealised single channel traces Unlike many applications used for clamping experiments Signal is a very general purpose program that can be used for many other types of experiment This can make it harder for new users setting up clamping experiments to find the aspects of Signal that they need In addition to this chapter and general familiarity with Signal clamping researchers should look at the amplifier telegraph system Sampling d
397. wn is the minimum value found between the cursors The value shown is the difference between maximum and minimum values found between the cursors The value shown is the standard deviation from the mean of the values between the cursors If there is no data the field is blank The value shown is the RMS level of the values found between the cursors If there are no values between the cursors the field is blank The value shown is the maximum absolute value between the cursors If the maximum was 1 and the minimum was 1 5 this mode would display 1 5 12 5 Selecting and copying data Context menu commands Signal for Windows version 3 Peak The value shown is the maximum found between the cursors measured relative to the baseline formed by joining the two points where the cursors cross the data Trough Trough The value shown is the minimum value found between the cursors measured relative to the baseline formed by joining the two points where the cursors cross the data The measurements available for marker type channels are Mean Sum Maximum Minimum Amplitude and Extreme If you select other measurements the result is a blank field The values calculated for the measurements are Mean The count of markers between the cursors divided by the time difference between the cursors This could be thought of as the mean marker rate Sum The total number of markers between the cursors Maximum The maximum inter mark
398. work Once an idealised trace exists it is stored in the sgr file associated with the data file and re loaded each time the data file is opened An idealised trace is one in which the raw data trace is converted into a set of events Each event has a start Channel 1 Chan 1 time amplitude duration and a set of flags to identify Coe M the type of event a closed time first latency etc The settings dialog holds fields to determine how this trace 55 574tme Marimon BE should be generated The SCAN method of generating 55s HCusot ie such a trace makes use of an assumption that a Peelmetrack 10000 points Gaussian filter was used to remove noise from the Peenbelow HCursode Amp signal to produce a high time resolution guess of what Orenlewel HCursor2 pAmp the original unfiltered noise free waveform was This Fiter cutoff frequency 4 kH2 is a form of reverse convolution of the idealised trace 3 avoid sublevels 7 Outward curent with the step response function for the amplifier The default draw mode for the resulting trace 1s to show the ie al SEN convolution over the top of the raw data with the idealised trace drawn below It is worth noting that the usual multiple poles Bessel filter used by most patch clamp amplifiers is a good approximation to a Gaussian filter so produces quite satisfactory results with this technique Once this process has been used to produce a rough idealised trace the trace will need to be f
399. xis Logarithmic Auto adjust units MEM range use Show All which adjusts the display to the data limits if possible or you can type in the y axis limits You can also control the other features as for the x axis Close Open the View menu Waveform Draw Mode dialog Experiment with different drawing modes for the channels Signal data files hold two basic 5555 teh example data channel types waveform T and marker Waveform data Erans 2ADC1 5 khz Lie El channels hold values that are balas the amplitude of the waveform at equal time intervals Marker data channels hold the times at which something happened If a waveform has any error information associated with 1t then the display of this information may also be manipulated here There are several different ways to display waveform data click the Draw button to cause an update without closing the dialog Draw Cancel dew The most common draw mode for waveform data is Line mode In Line mode successive data points are joined with straight lines You can also select Histogram Skyline Dots or Cubic Spline modes In Dots mode you can choose large or small dots small dots can be very difficult to see on some displays See the View menu chapter for a complete description of waveform and marker draw modes Open the View menu Customise display dialog Experiment with the channels axes and grid This dialog sets the Display for example cfs
400. y can cope with You can set a grid based on the screen resolution or a grid such that the width of the image is a fixed number of points If you are not sure what setting to use start with Same as screen image and adjust it as seems appropriate for your use 9 6 Use enhanced Metafile format Prompting to save views Use lines in place of rectangles Sampling Directory for new data Prompt for file comment Assume Power1401 hardware Defer on line optimise Edit menu Signal supports two metafile formats Windows Metafile MF and Enhanced Metafile EMF WMF is a relic of 16 bit Windows and has limitations but is widely supported EMF is the standard for 32 bit programs and has many more features However some graphics packages do not support this fully this was written in late 1998 but is still true in 2004 Several users pointed out that it was very irritating to be asked if you want to save data that is derived from other data and that can easily be derived again This is especially true when you are developing a new script application If you check the Do not prompt me to save unsaved result and XY views box Signal will close and throw away result and XY views without requiring a confirmation As this is potentially destructive we suggest that you don t use this option when you care about the result or XY data This only affects metafile output Some graphics programs cannot cope with axes drawn as rec
401. y than and add and subtract This means that 1 2 3 is interpreted as 1 2 3 and not as 1 2 3 Apart from this evaluation is from left to right unless modified by brackets The sequence compiler evaluates expressions as real numbers so 3 2 has the value 1 5 If expr is used as an integer for example DELAY expr it is rounded to the nearest integer Values in the range 3 5 to 4 49999 are treated as 4 These functions can be used as part of expressions to give you access to Signal sequencer step timing and to convert between user units and DAC and ADC values S expr The number of sequencer steps in expr seconds milliseconds and MS VRB E microseconds For example with a step size of 200 milliseconds us expr Y 5 s 1 1 returns 5 5 This is often used with the DELAY instruction Each instruction uses 1 step so use DELAY s 1 1 for a delay of 1 second Hz expr The angle change in degrees per step for a cosine output of expr Hz For a 2 Hz cosine use RATE Hz 2 To slow the current rate down by 0 1 degrees per step use RINC Hz 0 1 Use in RATE RATEW RINC and RINCW instructions VHz expr The same as Hz but the result is scaled into the 32 bit integer units used when a variable sets the rate MOVI V1 VHz 2 followed by RATE V1 will set a 2 Hz rate Vangle expr Converts an angle in degrees into the internal angle format The 32 bit integer range is 360 degrees The result is expr 11930464 71 6

main manual

Contents

Download Pdf Manuals

Related Search

Related Contents