Home

Touch Screen Based Measuring Equipment

1. end CTube get_m_DAF_gun get inline void CTube set_m_DAF_gun const boolean value begin CTube set_m_DAF_gun set preserve no m_DAF_gun value 14 end CTube set_m_DAF_gun set 4 4 Get and Set Operations for Associations inline Touch screen based equipment Page 89 inline const CMeasuring Equipment CTube get_measuring_equipment const fit begin CTube get measuring equipment get preserve no return measuring equipment f end CTube get measuring equipment get inline void CTube set_measuring_equipment const CMeasuring Equipment value begin CTube set measuring equipment o set preserve no measuring equipment value end CTube set measuring equipment set inline const CField CTube get the CField const int index const H begin CTube get the CField96 get preserve no retum the CField index H end CTube get the CField9o get inline void CTube set the CField const int index const CField value ftit begin CTube set_the_CField set preserve no the CField index value end CTube set_the_CField set endif Touch screen based equipment K C code file of tube class begin module cm preserve no LXE EQS SZ EWS end module cm g begin module cp preserve no H end module cp Module CTube Pseudo Package body Subsystem Top Level Source file E CTube c
2. 7 5 2 15 CRemote The currently operational COPER T is able to execute most of its functions by remote control The demonstrator uses this remote control to obtain the measurements All possible functions can be found in the user manual for the COPER T1 Mullekom 1996 For the remote control class all functions possible with the remote control are implemented as member functions Each of these member functions have the appropriate attributes to receive the parameters to measure with and to return the measured values The member functions call another member function to send the command to the serial port which is used to communicate with the COPER T1 The serial port is also implemented as an object of type PORT This PORT class is developed by another section at EED 7 5 2 16 General To accomplish the functionality of the classes extra functions are developed Those functions are not essential for the design of the demonstrator so they are not explained any further Because of the automatic generation of the blueprint for the demonstrator and because the demonstrator is only a part of this study the C code is not revealed in this document It is however possible to obtain the CD touch screen feasibility to look at the complete C code The CD can be obtained by Philips EED section MMM The dialog classes are generated by making the view with Microsoft s application studio and then convert them to a class with the classwizard The code fo
3. For normal measurements always select Mains locked The Crystal locked mode l is used only for special investigations To change the II setting go to submenu Clock pulse type and press lt enter gt Make the appropriate choice Mains l locked Crystal locked and store according procedure Change edit data const enum get m ClockPulseType const void set m ClockPulseType const enum value H Attribute m DAF gun const boolean get m DAF gun const void set m DAF gun const boolean value Additional Private Declarations 4 begin CTube private preserve yes end CTube private private implementation Data Members for Class Attributes Page 86 begin CTube m_name attr preserve no private string U string m_name end CTube m_name attr f begin CTube m_MeasDelayFor2A attr preserve no private int U int m_MeasDelayFor2A If end CTube m_MeasDelayFor2A attr fti begin CTube m_Type attr preserve no private enum U enum m_Type end CTube m_Type attr l begin CTube m_DeltaThreshold attr preserve no private int U int m_DeltaThreshold end CTube m_DeltaThreshold attr 4 begin CTube m_DeltaGain attr preserve no private int U int m_DeltaGain f end CTube m_DeltaGain attr begin CTube m_VideoSensFactorX attr preserve no private int U int m_VideoSensFactorX end CTube m_VideoSensFactorX attr 4 begin CTube m
4. Frame Frequency Clock Pulse Type System has new tube parameters which do not belong to a specific Edit position parameters User is logged in and is authorised to edit parameters The user is able to edit the mask pitch of a measure field If the measured field is not defined an error message is displayed The mask pitch is known by the system Edit colour recognition settings User is logged in and is authorised to edit parameters The user is able to fill in the starting gain value for automatic gain adjustment The user is also able to change the threshold for noise reduction Start gain for auto gain and threshold value are known by the system Edit convergence settings User is logged in and is authorised to edit parameters The user is able to edit the initial gain and initial threshold values used The initial gain value and the Initial threshold value are known by the Edit Eccentricity settings User is logged in and is authorised to edit parameters The user is able to edit calibration parameters gain sense and offset and Camera parameters gain and threshold used for eccentricity measurements Sensitivity offset gain and threshold for eccentricity measurements Edit linewidth settings User is logged in and is authorised to edit parameters The user is able to set the process parameters and the gain threshold parameters for the line width process Touch screen based equipment a E Exceptions Result Name
5. User selects one of the above mentioned SBIP functions System activates the selected functions User selects field to be used in relation with the SBIP functions System performs function System shows measurements if any Un Jd C2 bh2 Touch screen based equipment Page 32 10 SBIP Functions with field selection in multi selection mode Scenario may be executed for boot SBIP Initialise SBIP Live video Fixed video 1 User selects field to be used in relation with the SBIP functions 2 System makes selected field active if it currently is not and the system makes the field inactive if it currently is active 1 and 2 are repeated until user satisfied with selection 3 User selects one of the above mentioned SBIP functions 4 System performs selected function 5 System shows measurements if any Before starting the selected functions the system asks the following function dependent data Set Gain Gain Threshold Auto Gain Gain Delta gain Threshold Auto Threshold Delta threshold Mask in plane Threshold white For the function Mask in plane the currently selected colour is used It may be changed in the Video control window 11 Measurement utilities in single selection mode Scenario may be executed for Convergence Gravity XY Gravity X Gravity Y Brightness Brightness colour Linewidth User selects measurement utility System activates the selected measurement utility User selects field System performs s
6. is fixed this parameter is in fixed proportion to 1 2 Horz field of view By Auto cali bra tion the Vert lI field of view is set to Il 1 2 Horz field of view 3 4 default Il For TVT 4500 mm ForCMT 2250 mm const int get m 1X2VertFieldOfView const void set m IX2VertFieldOfV iew const int value Attribute m LineFrequency Under the tube parameter Line frequency the l appropriate line deflection frequency must be stored Therefore check the actual line deflection frequency setting of the time base unit and go to Line frequency and press enter Make your choice in the submenu and store according procedure Change edit data it NOTE In case the Line frequency is changed from 16 I KHz into 32 62 kHz or vice verse the jumper setting on I the video gener ator card must be changed also const int get_m_LineFrequency const void set_m_LineFrequency const int value II Attribute m_FrameFrequency Under the tube parameter Frame frequency the appropriate frame deflection frequency must be stored Therefore check the actual frame deflection setting of the time base unit const int get m FrameFrequency const void set m FrameFrequency const int value H4 Attribute m ClockPulseType With the tube parameter Clock pulse type a selection can be made between Mains locked or Crystal locked video synchronising
7. known As long as users keep their fingers in contact with the screen no selection will be made After dragging the cursor when the users are satisfied with its placement they confirm the selection by removing their finger from the touch screen 4 1 Target Location It is recommended that for applications having established key input areas positions along the lower and right hand borders of the touch screen should be used to minimise activation time and error Bering 1989 Use of the lower border exclusively can accommodate users with either a right or left hand preference Another study only investigating accuracy and target locations revealed the highest accuracy in the top middle section the screen was divided into 9 sections of the screen This was not surprisingly because using a slight 15 degrees backward tilt of the screen the line of sight normal to the screen intersected just below the centre of the top row Parallax shifted the lower rows downward increasing the error distance A side result from a study concerning screen angle versus accuracy showed larger Mean Y errors at the top of the screen Beringer 1985 4 2 Target Size Only a few studies handle about target size but information about optimal target size is reported as side results in several other studies concerning touch screens The main problem is that no study used more than 5 target sizes to retrieve their information so there is not an l explicit function co
8. the menu style and the direct manipulation are a good choice for dialog style For the process maintenance engineer all dialog styles are appropriate Because of the huge rate of input and output variables used by middle to large sized MMM equipment fill in forms and command Touch screen based equipment Page 10 language would be appropriate A mix of Menu Fill in forms Command language and Direct manipulation seems to be optimum for MMM equipment When using a mix it is very important to apply the right style for the right dialog Menu style and direct manipulation are most efficient for navigation while Fill in forms and Command language are most efficient when entering big amounts of data Looking at the operator tables in appendix D the Fill in forms and command language has positions with no marks Both dialog styles should not appear in the MMI when an operator is at the controls For the process maintenance engineer no limitations hold Operator Operator Maint Manual Auto Process User o QD ub wee emu Ea se Psychology se eRSGR CU wm Attitude Positive Motivation Moderate Low moderate Moderate Knowledge and experience Typing skill Moderate Moderate Hig gh High System experience High Task experience Moderate Moderate Moderate Mee uer n Application experience Moderate Use of other systems High Computer literac Ha Hii High Job and task characteristics Frequency of use M
9. Setmode Prerequisites Description Exceptions Result Name SelectRaster Prerequisites Description Exceptions Result Name SetDosPosition Prerequisites Description Exceptions Result Name SearchDot Prerequisites Description Exceptions Result Name Normal VideoPos Prerequisites Description Exceptions Result Name InitPattern Prerequisites Description Exceptions Result Measure functions Name Convergence Prerequisites Description Exceptions Result Name Eccentricity Prerequisites Description Exceptions Result Name LineWidth Prerequisites Description Exceptions Result Name ColourRecognition Prerequisites Description Exceptions Result XY controller Name sent XY coordinates Prerequisites Description Exceptions Result Page 80 Touch screen based equipment G List of candidate classes AQUA camera colour simultane configuration convergence delay disk eccentricity Fields file gain Histogram ID number recognition image line linewidth Mask pitch Measure normalise video parameters password Plane remote system SBIP screen image screen picture Page 81 spot systeem threshold tube tube supplies user 2A mon video control identification classes brainstorm convergence date eccentricity linewidth maintanence engeneer measurement system operator proces engeneer power power suppl
10. Video search Video move Video colour Video picture Colour recognition User selects video function System shows input window for desired control function User enters new video properties User confirms or refuses newly entered data System makes video changes according to the confirmation 7 SBIP Functions with field selection in single selection mode Scenario may be executed for set gain auto gain auto threshold mask in plane pixel calibration show planes histogram User selects one of the above mentioned SBIP functions System activates the selected function User selects field to be used in relation with the SBIP functions System performs function System shows measurements if any pneus pu re 8 SBIP Functions with field selection in multi selection mode Scenario may be executed for set gain auto gain auto threshold mask in plane pixel calibration show planes histogram 1 User selects field to be used in relation with the SBIP functions 2 System makes selected field active if it currently is not The system makes the field inactive if it currently is active 1 and 2 are repeated until user is satisfied with the selection 3 User selects one of the above mentioned SBIP functions 4 System performs selected function 5 System shows measurements if any 9 SBIP Functions with field selection in single selection mode Scenario may be executed for boot SBIP Initialise SBIP Live video Fixed video
11. Xue mieQ w i l tooma p nae u M AIDS IUS OSDIA w ihe DH peg w Jou9 501 00g ua ave y i L AAO OTS w lcd ARA a Winata u Keepin on u p uegony w du A eee aed i 5 i egies Tw M onere rp Fianondo e Dewey eg l e rwgeg w edis d po An mre e6suje g w M4 vt04faieQveoy w T enexug RAD amp i OS Juu iu Ginas lt ouav ui i 3 i t K agioog u CUN Hc p 1 S mmm 215 E a d H o cC OPE 3 l oO E raved aware L un Sos PAD E eye fi ue sso ee i o E s c Undiovcu Luieithguon ad s S t uegtrowang w SE pA C i Ke Veeoog wnbyojmeQ u ynewooh seousuajad ui Oevones7u l e vewoog xwQormQ w Bous piwsssdou OengvoN ur S Wy sOumaguisoyw unue i voganu o om 808079 l mm rekopunieow 2 f i t e e i A ERA t Choses w O ye e deiseuTw 3 1 Uvns ur o be z VI j00q apowdess w fe ez O eiu Sew i a a US NY asl veemoq Wueusnseew w i b A Mi knoomeg w E T4UPWJOPQUANTM i nT ee we westgogepew 4 DA OWM Uu ir 1 Touch screen based equipment Page 94 O Internal scenarios User interface r m Measuring m Measure m Tube NS m Field i Equipment acer i t select i selec gt I Cd measure i t set measure l l i o l i l l i l l l H dg i l i i i 1 1 i p what i Fi i i 1 i um
12. appropriate action In the first release 5 pointers are implemented The first four are the parameters for the measurements colour recognition convergence eccentricity and linewidth The fifth is added to store the field related parameters and measurements of the meas utility dialog In figure 7 the currently implemented version is shown In figure 8 the desired implementation is shown As with the tube class two functions are available to load and save the tube dependent parameters The argument to these two functions is the filename After executing a measurement it is visualised in the tube dialog The involved measuring object initialises a measurement and it is not desired that those objects write directly in the tube object The field object is the place to store a flag to the last updated parameter object Using this construct the tube class itself is able to visualise the last measurements Some measurements have multiple values which can not been displayed at the same time in the tube dialog The indication of which value to show and how to get this value is handled in the involved object See appendix P for the user interface layout of the previous mentioned dialog Page 36 Field m_ColourRecogpara CColourRecogpara m_ConvPara CLineWidthPara Figure 6 Actual implementation Base class CMeasureParameters A Field abject CMearureParameters m_Porameters CColourRecogpara cc CEccePa
13. be as usable as categorical ordering Yet another choice of ordering is to group the most frequently used items McDonald amp Stone 1983 A combination is probably the best way to obtain high performance Note that in the McDonald amp Molander study a touch screen was used to select the items Searching for the desired item is one thing but selecting it is something else It depends on the selecting device how long it takes to select the item Menu choice selection A variety of selection mechanisms are available The most common are single character letter or number selection codes Alternatives include moving a cursor through the list of choices with cursor control keys or pointing directly to the choice with an alternative input device such as a mouse joystick or touch screen Selection codes provide faster selection times as compared to number codes or cursor selection except for first time users in combination with a small number of options Shinar amp Stern 1987 When using selection codes clear label choice is very important Cursor selection took more time because it depends on the item to select and the place of the cursor Touch screen based equipment Page 56 how many keystrokes must be done This drawback can be overcome with touch screen Albert Albert 1982 compared a large set of pointing devices including touch screen Menu navigation Navigating in a menu system is facilitated by a number of design techniques T
14. e The user is able to execute basic measurements to detect possible errors in the system 8 e The user is able to access the basic functions of the frame grabber to detect errors or to remedy possible errors 9 e The user must be able to change the image type 10 e The user must be able to select the data collection system 11 e The maintenance user is in addition to the previous tasks able to add new users to the system 12 Touch screen based equipment Page 21 e The maintenance user is able to edit the following parameters and variables program table program parameters tube parameters position parameters colour recognition parameters convergence parameters eccentricity parameters and linewidth parameters 13 e Both users must be able to see the measurements 14 The remote control is used to access COPER T1 The use case for that actor can be substituted by a functional diagram with the functions available by the COPER T1 remote control access The table containing those functions is available in the users guide of the COPER T1 Mullekom 1996 The tube actor interacts with the system in two directions e The image on the tube is captured by the camera 15 e The image is composed on the screen by tube supplies 16 In Table 3 an overview of the identified use cases is given The complete use cases are given in appendix F Use case Comment l Login 2 Select tube file 3 Choose measurement 4 Choose
15. field 5 Choose colour 6 Execute video search 7 Execute colour recognition 8 Execute basic measurements 9 Execute basic framegrabber functions 10 Change image type 11 Select data collection system 12 Add new user 13 Edit parameters and variables 16 Compose ima e on measured object tube Table 3 Use case of actors Touch screen based equipment Page 22 7 1 10 Context diagram Login Sdect tube file Choose measurement Choose fidd Compose imag oh measuring object tube Choose colour Execute video search Execute colour recognition Ez cute basic measurements Excoute basic framegrabber fimetions Change image type Sdect data collection system Add new user Edit parameters and vanables User nen View measurements commands pos Y pos casures XY Frame Figure 2 Context diagram With the definition of the major objects and actors and the determination of the fundamental communications from and to the actors the system scope has been outlined The system scope captured in a context diagram is seen in figure 2 7 1 11 Problem statement A textual description of the problem to be solved can be given as follows There are two types of users The first one is the operator The second one is the maintenance engineer The system must know which user is using the system Before the user starts
16. infrared touch system can be sealed to NEMA 4 requirements preventing water and dirt from penetrating the display electronics Infrared Disadvantages Resolution Scanning infrared touch systems typically provide a resolution of eight touch points per inch Parallax Parallax occurs when a touch is detected while the stylus is still some small distance from the surface of the display For scanning IR touch systems parallax accurs because the invisible grid of IR beams can be interrupted before actual contact is made with the display The amount of parallax is dependent on the type of integration and display type Typically the flatter the display face the lower the amount of parallax Resistive Advantages Touch screen based equipment Page 76 e Resolution Resistive overlay touch systems can deliver resolution of up to 4069 x 4069 touch points e Stylus Requirements Resistive overlay touch systems have no stylus limitations e Sealability Resistive overlay touch systems can be sealed to NEMA 4 requirements preventing water and dirt from penetrating the display s internal electronics Resistive Disadvantages e Transparency Resistive overlay touch systems have a transparency of 55 to 78 due to multiple layers of different materials found in the resistive overlay sensor They are also more susceptible to glare and reflection than any other touch technology e Calibration The 4 wire resistive overlay touch system are subjec
17. int CTube get_m_VideoSensFactorY const f begin CTube get_m_VideoSensFactor Y get preserve no return m VideoSensFactorY i end CTube get_m_VideoSensFactorY get inline void CTube set m VideoSensFactor Y const int value IBR begin CTube set_m_VideoSensFactorY set preserve no m VideoSensFactorY value end CTube set_m_VideoSensFactorY set inline const int CTube get_m_DeltaCentreX const begin CTube get_m_DeltaCentreX get preserve no retum m_DeltaCentreX 4 end CTube get_m_DeltaCentrex get inline void CTube set_m_DeltaCentreX const int value begin CTube set_m_DeltaCentreX set preserve no m_DeltaCentreX value end CTube set_m_DeltaCentreX set Touch screen based equipment Page 88 inline const int CTube get_m_DeltaCentreY const Ifl begin CTube get m DeltaCentreY 6 get preserve no return m DeltaCentreY H end CTube get m DeltaCentreY 5 get inline void CTube set m DeltaCentre Y const int value H4 begin CTube set m DeltaCentreY set preserve no m DeltaCentreY value ii end CTube set_m_DeltaCentre Y set inline const int CTube get_m_VideoShiftStepSizeX const f4 begin CTube get_m_VideoShiftStepSizex get preserve no retum m_VideoShiftStepSizeX Ifi end CTube get m VideoShiftStepSizeX o get inline void CTube set_m_VideoShiftStepSizeX const int value Ifi begin CTube set_m_VideoShiftStepSizex s
18. is to place the mark on their responsibility The manufacturer supplier draw a standard certificate in which they state that the product satisfies all the European standards The manufacturer supplier is advised to argument the certificate in a product file The second way is for manufacturers suppliers who are forced to engage with a competent body This may happen because they want to differ from certain aspects in the standard The third route is to test single products separately and approve them with the CE mark Finally there is a route for manufacturers with the EN29001 Certificate As a consequence of the ISO9001 standard they are obliged to handle according to the standards consequently they are only checked on the application of the Quality Assurance System There are a few measures that can be taken to reduce the EMI Franey 1995 Laan 1993 The first one is good PCB design This is a measure we have to believe from the manufacturer The second is to avoid loops or large wiring Another important measure is the path of the wiring If possible the wiring must follow the grounded enclosure to avoid coupling with the interfering field It is not sufficient to use coax cables because noise currents are induced on the cable shield by the magnetic fields Schelkunoff 1934 Probably the most important measurement against ESD is the grounding system There must be no loose ends and the grounding wiring must be of excellent quality Finally
19. made for a specific apparatus When using touch screen technology the only hardware necessary is a touch screen All output goes via the touch screen The main problem using a touch screen is the amount of real estate on the screen In practice this does not seem to be a problem because of the flexibility the software panel offers Most of the controls are made in software During normal process operation the touch screen only has to display relevant parts during that process so real estate is shared during the life cycle of a process In this way apparatus space is gained because large switching panels are superfluous Small MMM equipment could also profit from integrating touch screen technology In those apparatus a small LCD display with touch screen may be applied This also holds for distributed controls in a big apparatus In this way the maintenance engineer can control local parts of the apparatus This is a space and costs saving solution Touch screen based equipment Page 13 4 Touch screen style guide Accuracy of input using touch screen devices is affected by a number of variables that include device type target size target location touch screen position and selection strategy Several studies have investigated these variables and produced sometimes complementary and sometimes conflicting results The conflicting results however may in many cases be a problem of interpretation or a difference in hardware configuration The p
20. that direct manipulation interfaces are easier to use for experts and easier to learn for novices than non direct manipulation interfaces Other studies showed a slight advantage of command language interface over the direct manipulation interface Karat Karat 1987 found a clear advantage for the direct manipulation system when tested on novice users Two other studies Card amp Moran 1983 Shneiderman 1982 showed clear efficiency advantages when using direct manipulation interfaces All the studies more or less indicate whether or not direct manipulation will provide a performance advantage will probably depend on how well the direct manipulation interface maps to users goals intentions and tasks and what alternative interface it is being compared to Icons must be carefully designed to allow natural and meaningful associations to be drawn between icon attributes and important object attributes Touch screen based equipment Page 65 Natural Language Natural language allows users to express requests to a software application in there native language For this propose a keyboard as an input device is assumed but it is also possible to provide voice input Natural language interfaces are not yet widespread so only the advantages and disadvantages are given Advantages and Disadvantages of natural language Advantages Disadvantages Easy to learn Assumes knowledge of the problem domain Easy to remember Lengthy confirmation and c
21. the Pulse width depends on the line deflection frequency the type of tube and the Test array on which the tube is measured The proper value is determined empirically by increasing or decreasing the value until the Video dot on the camera image comprises 4 to 5 phosphor dots lines of one colour in the horizontal direction const nt get m PulseWidth const void set m PulseWidth const int valuc Attribute m 1 2HorzFieldOfView The tube parameter 1 2 Horz field of view stands for half the Field of view of the camera in mm in the X direction For Coper T this parameter is not fixed depending to the choice of the enlargement by the camera s default ForTVT 6000 mm Touch screen based equipment ForCMT 3000 mm The above value is fixed as tube parameter for the Tube type TVT CMT in question i This value can be determined with l select blank green video focus the camera select live video l count the number of horizontal trios calculate 1 2 horizontal field of view using the formula if V2 num of trios maskpitch const int get_m_1X2HorzFieldOfView const void set m IX2HorzFieldOfV iew const int value Attribute m 1 2VertFieldOfView The tube parameter 1 2 Vert field of view stands I for half the field of the camera in mm in the Y direction Because the CCD pixel length height ratio
22. the Eight and Sixty four choices Another study Wallace 1987 also investigated the depth versus breadth trade off and included some kind of time pressure This study makes it clear that time pressure will affect performance Interfaces to be used under such conditions should be tested under comparable conditions Touch screen based equipment Page 55 It is also important to organise within menu screens Snowberry Snowberry 1983 evaluated the same depth breadth conditions as Miller but they organised the 54 items into particular groups within one screen The performance was slightly better than the two level eight items per level structure Landauer Landauer 1985 studied depth versus breadth using a touch screen in combination with structured menu items numbers and letter ranges This study indicated that maximum performance is reached if the number of selection items is between the 8 items on a screen with more levels and 64 items on a screen providing lower levels Paap Paap 1986 found that optimum breadth values ranged from 16 to 78 items on one screen That is when menu choice items on a given screen can be ordered into meaningful groups Collecting more information about depth breadth trade off Lee amp Chao 1989 a good guideline for making the depth breadth trade off in a menu hierarchy might be stated as follows Choose a maximum breadth of 6 to 12 items per screen if no useful grouping can be imposed Or if the menu ite
23. the wave motion by absorbing part of the wave The Page 68 X Transmitter Touch absorbs energy causing a dip in amplitud Touch o ouch M vr region 3 M KH iid Y Receiver Y Transmitte X Receiver anem Figure 10 Surface Acoustic Wave received signal is compared to a stored signal The change recognised and a coordinate calculated This process happens independently for both X and Y axes By measuring the amount of the signal that is absorbed a Z axis may also be determined Capacitive Capacitive overlay technology uses a glass overlay with a thin metallic coating over the surface of the display screen Voltage is applied to the four corners of the screen creating a uniform voltage field The user must touch the overlay with a conductive stylus such as a finger to activate the system Touching the overlay surface causes a capacitive coupling with the voltage field drawing a minute amount of current to the point of contact The current flow from each corner is proportional to the distance to the finger and the ratios of these flows are measured by a controller and used to locate the touch Clear Glass Overlay y u Protective Cover Thin Transparent Metallic Coating CRT Face Figure 11 Capacitive Touch screen based equipment Infrared Scanning infrared IR technology relies on the interruption of an IR light grid in front or the display screen The to
24. touch screen systems the choice is narrowed to three technologies surface acoustic wave capacitive and five wire resistive In appendix L there is a list of the touch screen manufacturers from which information is received for this feasibility investigation The surface acoustic wave is exclusively manufactured by ELO The capacitive system is manufactured by Micro Touch and the five wire resistive is manufactured by ELO and Micro Touch These two manufacturers are the world leaders in touch systems Both developed their own technologies to improve accuracy and lifetime It is therefore that surface acoustic wave and capactive overlay are well suited in industrial environments According to ELO and Micro Touch all three of the touch screen technologies are unaffected by EMI from other nearby touch screens CRTs other displays and environmental EMI Electrostatic Protection is covered by the IEC 801 2 standard For the five wire resistive system proper transient protection must be applied to withstand high discharges Even for the capacitive overlay an additional metallic layer is placed between the CRT surface and the active metallic layer This additional layer protects the touch screen technology against the high ESD coming from the CRT but also contributes to the low transparency Looking at the technologies in combination with the EMC ESD constraints the surface acoustic is in favour Both 5 wire resistive and the capacitive touch systems use an over
25. uo siujejisuo e 9 e E Uotle 20Sss y paapag incid A suoje12dO jo uopebedosg SSE1D p aapaqg SPINQUHY peapiag Suojeiadg 29 pue 31nqpily SSL PP Seej2qQns eee SASSEIIINS UAaMAQ Sainaueiewe esas ir Z 88e QedNsS rennes einquue ue si 0IeUNUHISIG ieousipayul etdpiny WSR oe SASSr IANS ALOK diusiequou fiwolsipuou Buiddeueao E AALY sesseipqno s eizedns 2 isajuadoig uopeziyejouan a UO ndo jo 5 Suoneivewerhur eo 9101200 apiaoid hi cnet ee Isnw sessejoqng vouesedo yu Uu a x o0 S S8 ssri2iedns eui ur gt as ese S vore edo ol E 2 n SSe Q S uOTe ose 5 O ner v suojlesedg pensqy 3 E side2uo paoueapy E gt UONEION epo joalqo owen UOJPr20 Sy suopepossy Aseusay ejnqpiv ur p Iperepio Bupepig 9710W 10 BUD peyrseds Aje2uewnw e bet pug euo 20 0102 evondo mop 0101 30 0 02 Avery m b auo ji2ex3 OWEN VONE DOSSY suojlepossy enjea swe einqu e menor owen ste2 s92uejeut 129 q0 stel Aiquessy w30 ajeusayje UopeBabby 8 5 1 Ui ed suonebas6by E 3d e ueqp19 qu uopez e3u00 BdA vines 194 Gse vonesodo vonezedo enea ny ed L eiep einquue ed vjep einquie einqube 186819 side2uo o1seg UONEION aPOW 19a qO RNC blest tcc aceasta ty P Mc prousoni ww wy YED w ptt erygsueS w eau BOBO SUES w pto peysueg w RA d iG OIQNU DC A w wr dexguipc6 _w yy U
26. value end CTube set_m_MeasDelayFor2A set inline const enum CTube get m Type const Iff begin CTube get_m_Type get preserve no retum m Type fitt end CTube get_m_Type get inline void CTube set m Type const enum value i ft begin CTube set_m_Type set preserve no m Type value end CTube set_m_Type set j inline const int CTube get m DeltaThreshold const begin CTube get_m_DeltaThreshold get preserve no return m_DeltaThreshold end CTube get m DeltaThreshold 6 get inline void CTube set_m_DeltaThreshold const int value begin CTube set_m_DeltaThreshold set preserve no m_DeltaThreshold value end CTube set_m_DeltaThreshold set inline const int CTube get_m_DeltaGainQ const begin CTube get_m_DeltaGain get preserve no return m DeltaGain end CTube get_m_DeltaGain get inline void CTube set_m_DeltaGain const int value ff begin CTube set_m_DeltaGain set preserve no m_DeltaGain value end CTube set_m_DeltaGain set inline const int CTube get_m_VideoSensFactorX const begin CTube get_m_VideoSensFactorX get preserve no retum m_VideoSensFactorX end CTube get_m_VideoSensFactorX get inline void CTube set_m_VideoSensFactorX const int value begin CTube set_m_VideoSensFactorX set preserve no m_VideoSensFactorX value 1 end CTube set_m_VideoSensFactorX set inline const
27. visible Touch screen based equipment Page 54 Menus are most appropriate for USER PSYCHOLOGY Negative attitude Low motivation KNOWLEDGE AND EXPERIENCE Low typing skill Little system experience Low task experience Low application experience Frequent use of other systems Low computer literacy JOB AND TASK CHARACTERISTICS Low frequency of use Little or no training Discretionary use High turnover rate Low task importance High task structure Design issues in menu systems can be divided into five areas e Menu structure e Menu choice ordering e Menu choice selection e Menu invocation e Menu navigation Menu structure One important issue regarding menu design is the number of choices and the number of levels to select from There is a trade off between the two Miller 1981 One extreme is to have only a few choices on any individual menu screen and therefore many or at least more levels in the menu hierarchy An other extreme is to have many choices on a single menu screen The problem is to find the optimum of user performance when navigating through a menu system to a target item at the lowest level According to Miller 8 items per menu screen was the optimum He only uses 4 menu structures to retrieve this optimum Two choices per screen six levels Four choices per screen three levels Eight choices per screen two levels Sixty four choices all on one screen one level The real optimum might lie between
28. void CTube m Load begin CTube m Load 884685594 body preserve yes end CTube m Load 884685594 body void CTube m GetSelectedField begin CTube m_GetSelectedField 884685595 body preserve yes end CTube m_GetSelectedField 884685595 body void CTube m GetNextSelectedField begin CTube m GetNextSelectedField 884685596 bod y preserve yes end CTube m GetNextSelectedField 884685596 bod Y Additional Declarations begin CTube declarations preserve yes end CTube declarations Touch screen based equipment Page 91 L Touch screen producers PostalCode Carroll Touch P O Box 1309 Round nd 78680 jU DEC Data Concepts roce ex i E US Edmark ro oum n Elo Touch Systems 41752 Christy St 94538 3114 Micro Touch Unit 163 Milton Park Didcot OX14 4SD UK e e et Pixel Labs West Oceanside 92056 Touch control CONTEC Microelectronics San Jose 95131 USA Troll Touch Ste 106 CyberTouch 853 Lawrence Drive Nawbury Park 91320 2232 US Touch Technology Astin TX US 1120 nm Avenue Suite 2a Building 101 Avenida de la Plata 2188 Bering Drive 25510 Avenue Stanford a zu T 02 1 quue e iesus Z que tsse3 1 t qune 2 iv 1 s8e 9 amp sUOfjej20S S Y UIEMIAQ ujeisuo 122 qO
29. 8 the following guidelines may be drawn e Provide consistency in syntax Use an action object syntax Avoid arbitrary use of punctuation Avoid positioned grammars The syntax should be natural and mnemonic Defaulting of optional parameters Avoid use of control keys Command language lexicon The language lexicon in a command language is mainly command names and or abbreviations The studies from Bernard amp Grudin have investigated the naming behaviour of both users and designers There is a very low probability that any two users or designers will suggest the same name for a given function The main problem here is to discover what properties of a command name set will provide optimal performance In one study Caroll 1982 users performed better on congruent as opposed to non congruent commands The best performance was reached if the commands are hierarchical and congruent Several more studies Furnas 1985 Wixon 1983 have been conducted and the following rules could be learned from all of them e Command names must be hierarchical congruent specific familiar consistent e Use user jargon and avoid computer jargon e Use consistent rule for abbreviations e Allow full command names Touch screen based equipment Page 62 Command language interaction The structure of the dialog or interaction can be designed to enhance the usability of a command language system One study Granda amp Teitelbaum 1982 investiga
30. 85 pp 73 78 LedGard H F 1980 The natural language of interactive systems Communications of the ACM 23 pp 556 63 Lee E amp Chao G 1989 The increase utility of incorporation keywords in menu systems and users increase experimence Behavior amp information technology 8 no 4 July August 1989 pp 301 308 Lockheed M 1996 Succeeding with the booch and omt methods Addison Wesley ISBN 0 8053 2279 5 MacGregor J amp Lee E 1986 Optimizing the structure of database menu indexes A decision model of menu search Human Factors 28 no 4 August 1086 387 400 Mayhew J D 1987 Principles and Guidelines in Software User Interface Design Prentice Hall inc ISBN 0 13 721929 6 1987 Touch screen based equipment Page 46 McDonald J E amp Molander M E 1988 Color coding categories in menus CHI 88 Proceedings May 1988 00 101 106 ACM McDonald J E amp Stone J D 1983 Searching for items in menus Proceedings of the human factors society 27th annual meeting 1983 pp 834 37 Moran P and Newell A 1983 The Psychology of Human Computer Interaction Hillsdale N J Lawrence Erlbaum Associates 1983 Miller D P 1981 The Depth Breadth Tradeoff in Hierarchical Computer Menus Proceedings of the Human Factors Society 25 annual Meeting 1981 pp 296 300 Mori T 1996 Analysis of ESD Immunity of Electronic Equipment Based on Ground Potentioal Variations JEICE Trans Comm
31. CO Pm 7 2 1 Tit ODM ATES TA A T T TT E T 7 2 2 The User Profile scscsccceissccisccisciscerateavsviescesccusrutseibiGecvveisvectesasssucdcsdeceldseaascenlsnscvgatsieccvotdsedesvocsezse 7 PER E Psychological Characteristics Of user seien enne 8 2 2 2 Knowledge and experience of user csssssssesctceececesceceeceesetsceseaseetecsseseseseceaecs 8 2 2 3 Job dnd PASS uoc rdi ere eid cuit atre eot E liuc ra edu cs ud 8 2 2 4 Tools Tor the USOp o eorr Se ctediudelispe mes athtek sva d eta radial eee uod a E ep iaiia ed a 9 2 3 Selecting dialog style for MMM equipment users ccccssssscerssecssesscersenereorscetsscersnsseresees 9 Jd Inputanmd Output Deylcas esae an aeeainas vencusasasecobancu std veteucnucieneavsesascaastaveuesiontvotacs erase ll 3 1 Input qe e a a oa saet doloso iaai sostes M 11 3 2 OEP UE EEEIEE EREA AE A EE A E E E 12 4 Touch screen style guide eese T c E Misses 13 GLE CGI Strategy fe Pp IRE ror 13 4 1 Target Location eeeeeeee eere etn E TEA n Wevestes ties cons qe edecad 13 4 2 Target Size ie ce rien onis Kassierer oni Ev re itv iu Seve UIN LIE VA E EE T 13 4 3 Target Layout P 14 4 4 Touch screen position esses eere eese e nete tenen enean a etna tn sensa aset e aem a ttes nae seas sm seen eta 14 Dev EMC ESD Constraints sssi sae inizia qiie ener Dici eaae ceto Reels ve
32. ICS EB 681 1S Technische Universiteit 17 Eindhoven Faculty of Electrical Engineering Section of Information and Communication Systems Master s Thesis Touch Screen Based Measuring Equipment Design and Implementation B G F A W van Oversteegen Coach Dr ir A C Verschueren H van Broekhuyzen Philips EED sectie MMM Ir R C H M Overkamp Philips EED sectie MMM Supervisor Prof ir M P J Stevens Examiner ir G L J M Janssen Date 10 April 1998 The Faculty of Electrical Engineering of Eindhoven University of Technology does not accept any responsibility regarding the contents of Master s Theses Preface Most people only read the first pages of a thesis report Therefor this is the right place to thank all people who made my thesis a success During the conduction of my thesis I was fortunate to have the most able guidance of two individuals Henk van Broekhuyzen and Ad Verschueren in carrying out their task as coach Henk van Broekhuyzen was my coach at the Philips Equipment Engineering Division section Measuring Magnetizing and Matching MMM Ad verschueren was my coach at the Eindhoven University of Technology faculty of Electrical Engineeing Section of Digital Information Systems ICS EB Both people not only coached me on my thesis but also handed me the knowledge to make proper decisions for the future Furthere more I like to thank Ru d van Mullekom and Jos van Cranenbroek for their support and information
33. Moderate Moderate One similar system Novice Expert in field Some similar system Native language Use of other systems Computer literacy English Little or none H Non German Frequent Low Frence Moderate Spanish High Taiwanish Chinish Touch screen based equipment J Header file of Tube class begin module cm preserve no if AXA Vo VZLA AW end module cm begin module cp preserve no end module cp t Module CTube Pseudo Package specification ft Subsystem Top Level Source file EXCTube h ifndef CTube h define CTube h 1 begin module additionalIncludes preserve no 48 end module additionalIncludes begin module includes preserve yes 4 end module includes CMeasuring Equipment include CMsrngEq h CField include CField h begin module additionalDeclarations preserve yes end module additionalDeclarations fli Class CTube H Category Top Level IHil Subsystem Top Level Persistence Transient 4 Cardinality Multiplicity n class CTube begin CTube initialDeclarations preserve yes end CTube initialDeclarations public 4 Constructors generated CTube CTube const CTube amp right Destructor generated CTube Equality Operations generated int operator const CTube amp right const int operator const CTube amp right const I Other Operations specified H Operation m
34. Moderate application experience Low to high use of other systems Moderate to high computer literacy JOB AND TASK CHARACTERISTICS Low to High frequency of use Little or no training Discretionary use Moderate turnover rate Moderate task importance Low task structure Limited task domain Little or no research has been conducted that directly studies user performance with function key interfaces The basic guidelines are drawn from other dialog styles e Provide enough function keys to support functionality but not so many that scanning and finding them is difficult Use function keys for generic high frequency important functions Arrange in distinguished groups Base groupings on flow of use Recognizability takes precedence over consistency Place high use keys within easy reach Consistency in function keys Provide feedback when function keys are pressed Grey out soft function keys if not active and use status indicator on mode keys Direct manipulation A direct manipulation interface is one in which users perform actions directly on visible objects The term direct manipulation was first coined by Schniederman Schneiderman 1982 to describe new interfaces having the following characteristics e Continuous representation of objects e Physical actions or labelled button presses in place of command language e Rapid incremental reversible operations with immediately visible results Direct manipulation interfaces ofte
35. ONU DC A Uu RA d iSIOXUCOA w wy dng A w Wa WONCOA w 34 ADAST Tw BA XyOAeT w P SONAS us a AezISdeiSdeisoroIN Cur e PETEERE Sete tur poutauj w x ug w jo i9tuO fveS w W uipeequy w UtduiMeu123 RETE tai c _ rt d 905D Page 93 i i men Ousosdseo Numa Donoru wy ef Scenes sw un onde a www i 22 ya eed cate pe A E ipie i M Mune E buus weu w mame Wu SEAS concur ue i bees oe Moun wus woned w i mimg w haguna nuUo ur H i I M4 h eee as H Ww bemydgs ur 1 _ ed amg ewt Ex ee ose uc a io 288 vonai6008 300102 w 4 T ui Sur w b up eur 8299 Sayy ui rici i r t I t Jeueloesuo 0s eon wu Quent rui ur i Pornon reme t t0g09DARDON w lo bee pe ee eee i t hoquovees w I t woo gtogies w t I 1 beop2 fug u Les euID91057 w Un dva i d heeunu6ug w tapones w ien ee Bo ve y tin fte j oeprary w 10 A2ugnb9 38843 w Ox Amae w i i ipM Sug w uw 3uenbe jeu u Uax Aetio i t poor sseuufug w wA MAOK ue AZIL ur Oe Bois i sno JONI tau bug w qu ADAJON HD w E tuti vous ur H e2u86 0 002 ur 14 Pv eti us vont ang w pepsog w Wt SAAGUYSOIMAON w i i Cheung u utt w OaxAmeg w Ju xezGdegyUgoeD A w poutan ony w i retis 19 aanus meg w uie ony w w AA
36. Prerequisites Description Exceptions Result Name Prerequisites Description Exceptions Result Name Prerequisites Description Exceptions Result Name Prerequisites Description Exceptions Result Name Prerequisites Description Exceptions Result Name Prerequisites The parameters for the linewidth process are known by the system Edit password User is fogged in and is authorised to edit parameters The user can change his hers password User data contains new password Save to disk The user can enable or disable the data savings to disk Data to disk is enabled or disabled Save to Aqua The user can enable or disable data transfer to AQUA On line data collection is enabled or disabled Use colour simultaneous utilities The user is able to perform a complete colour recognition Before this function is performed the use must fill in the field of measurement and the expected white level And the user or XY table must position the camera to the selected field The system will then perform the following sequence Select lt Red gt lt Blanked gt raster Cail SBIP function lt Auto gain gt Call SBIP funktion lt Mask to plane gt for red Select lt Green gt lt Blanked gt raster Call SBIP function lt Auto gain gt Call SBIP funktion lt Mask to plane gt for green Select lt Blue gt lt Blanked gt raster Call SBIP function lt Auto gain gt Call SBIP function lt Mask to p
37. RTs with high accuracy 7 1 1 OMT The OMT methodology has three fundamental models each representing a different view of the system under consideration The three views the object dynamic and functional models allow developers to emphasise distinct aspects of the system as the circumstances direct Associated with each model is a micro process that defines the steps necessary to develop the model 7 1 2 The OMT Object Model The OMT object model is the fundamental model on which the remaining models are based The object model captures the entities that appear in the application and solution domain their structure and the relationships among them Class diagrams showing the existing classes and their relationships along with an occasional object diagram showing the individual instances of classes compose the object model 7 1 3 The OMT Dynamic Model The OMT dynamic model indicates the dynamics of the objects and their changes in state By exploring the behaviour of the objects over time and the flow of control and events among the objects the dynamic model captures the essential behaviour of the system Scenarios are Touch screen based equipment Page 19 captured in a message trace diagram These diagrams which show the life history of objects compose the OMT dynamic model 7 1 4 The OMT Functional Model The OMT functional model captures algorithm requirements design and data flow The functional model is best cap
38. Result Image is collected in memory Use case remote system Sbip Functions Touch screen based equipment Name BootSbip Prerequisites Description Exceptions Result Name InitTMS Prerequisites Description Exceptions Result Name GetImage Prerequisites Description Exceptions Result Name GrabImage Prerequisites Description Exceptions Result Name SetGain Prerequisites Description Exceptions Result Name AutoGain Prerequisites Description Exceptions Result Name AutoTreshold Prerequisites Description Exceptions Result Name Take Image toPlane Prerequisites Description Exceptions Result Name PixelCalibration Prerequisites Description Exceptions Result Name GravityX Prerequisites Description Exceptions Result Name GravityY Prerequisites Description Exceptions Result Name GravityXY Prerequisites Description Exceptions Result Name Border Prerequisites Description Exceptions Result Name Convergence Prerequisites Description Exceptions Result Name BrightnessNoColour Prerequisites Description Exceptions Result Name BrightnessColour Prerequisites Description Exceptions Result Name LineWidth Prerequisites Touch screen based equipment Description Exceptions Result Video Sync functions Name Init Video Prerequisites Description Exceptions Result Name
39. _Save 884685593 void m Save f Operation m_Load 884685594 void m_Load 44 Operation m_GetSelectedField 8 84685595 void m_GetSelectedField Operation m_GetNextSelectedField 8 84685596 void m GetNextSelectedField 4 Get and Set Operations for Associations generated Page 84 Association Has Tube 34846EB700FA Ift Role CTube measuring equipment const CMeasuring Equipment get measuring equipment const void set measuring equipment const CMeasuring Equipment value Association lt unnamed gt 34846F380000 Role CTube lt the_CField gt const CField get_the_CField const int index const void set_the_CField const int index const CField value Additional Public Declarations begin CTube public preserve yes end CTube public protected Additional Protected Declarations 4 begin CTube protected preserve yes end CTube protected private amp 8 Get and Sct Operations for Class Attributes generated Attribute m name Under tube parameter Tube name This text is also added to the Data collection on disk if enabled To edit this name go to submenu Tube name and key in the name e g 66 FS 32 kHz 50 Hz Confirm your change via Change edit data const string get m name const void set m name const string value Attribute m MeasDelayFor2A With the program parameter Meas delay for ILA mon t
40. _VideoSensFactorY attr preserve no private int U int m_VideoSensFactorY 4 end CTube m_VideoSensFactorY attr begin CTube m_DeltaCentreX attr preserve no private int U int m DeltaCentreX 4 end CTube m DeltaCentreX attr begin CTube m DeltaCentreY attr preserve no private int U int m DeltaCentreY 48 end CTube m_DeltaCentreY attr Iii begin CTube m_VideoShiftStepSizeX attr preserve no private int U int m VideoShiftStepSizeX Iff end CTube m_VideoShiftStepSizeX attr begin CTube m_NrOfVideoShiftSteps attr preserve no private int U int m NrOfVideoShiftSteps end CTube m_NrOfVideoShiftSteps attr f4 begin CTube m_PulseWidth attr preserve no private int U int m_PulseWidth end CTube m PulseWidth attr Ifi begin CTube m_I 2HorzFieldOfView attr preserve no private int U int m_1X2HorzFieldOfView Ift end CTube m_1 2HorzFieldOfView attr iiH begin CTube m_1 2VertFieldOfView attr preserve no private int U int m 1X2VertFieldOfView HH end CTube m L 2VertFieldOfView attr HA begin CTube m_LineFrequency attr preserve no private int U int m_LineFrequency 4 end CTube m_LineFrequency attr Touch screen based equipment Page 87 begin CTube m_FrameFrequency attr preserve no private int U int m_FrameFrequency f end CTube m_FrameFrequency attr begin CTube m_ClockPulseType attr preserve no private en
41. about MMM equipment Nico de Boer for killing my test PC during the first flash over tests and Anton Vervoort for its knowledge support about ESD and EMC Special thanks to Ir G Janssen working at the University of Eindhoven I was very honoured that he was willing to read my paper for approval I acknowledge with gratefulness the help of Theo Veltman with whom I shared a room at Philips He showed to be a great source of information Finally I am grateful to Ir Richard Overkamp and Prof Stevens They offered me the opportunity to conduct this thesis containing both hard and software components There is one person who I do not thank This man is right BILLL GaTeSSSS whose software has a mind on its own Bertrand Oversteegen Eindhoven The Nederlands April 1998 Contents MEC DD EP dae r iosia i 3 1 1 The cathode ray tubes iseer eaea nadana cocus aad Sp concra eroe pene sae e ira ace PUR oae e d epa eode Nee SR doE 3 1 2 MMM quipment i co cosi sonvnsasnarevsasevscccossunssesss as eseekexevenseudeue cotensnstayntstnsbentoveanstantuasteaassaysbive 4 1 3 Problem statement sseicccssccsiss cccscoaesesceshtesisesasesdosaassssassccsosavesdatsecsvaassvaseacscesesadestecbevasuevete seceded 5 1 4 DIgIngdeee e P 5 1 5 Organisation of this report a eceece sees se ene eese en ntn stesso tame ase ten M epos sa se sena sso Pa sebo essen reseau 6 Ze M n MACHINE LACE A
42. aces Selecting items is done by specially designed dialogs with push buttons to select the item For the first version of the demonstrator the associations to the methods of the measurements are implemented statically But for reusability and flexibility these associations must be implemented dynamically The available methods should be detected during program start as mentioned in previous sections All classes are developed to be able to use the dynamic linking Because of this dynamic nature on the amount and kind of measurements the user dialog for the available measurements is controlled in the CMeasuringEquipment class The current COPER T1 has grouped several elementary functions for analysing development purposes This group contains functions provided by several different classes To make a proper design the initiations of those functions are handled by a user dialog controlled in the CMeasuringEquipment class and visualised as Meas utilities For the same reason as mentioned above the dialog to handle video operations is controlled by this class The dialog that gives access to the various parameters distributed over several other classes is also controlled in this class This dialog is dynamic because the measurements also have parameters that can be changed The actual dialog that provides the parameter manipulations for the dynamic classes is controlled by the classes itself To avoid distraction the operator only sees the d
43. ain subject of this thesis was to investigate the possibility to redesign an existing user interface for industrial production equipment This comprises three major parts The first part deals with the selection of appropriate dialogs to use when operating with industrial equipment A comparison is made between all possibilities offered by several dialog styles The results show that a menu system with direct manipulation is highly recommended to satisfy the needs for a big group of different users The literature also shows that de desired dialog is best performed with the utilisation of a touch screen input output device The second part handles about the constraints of user interfaces if a touch screen is used as a communication media A comprehensive inquiry has been made to design guidelines for a touch screen biased user interface The electrical characteristics of a touch screen device are important Industrial environment is very harsh for delicate electronics For present study the electric static discharges during measuring were of great importance The surface acoustic wave technology proved to be reliable touch screen technology in those harsh environments Finally the last part handles about the design of a touch screen based industrial equipment An Object Oriented method is used to design the measuring system The implementation is done with Visual C The design is made flexible and reusable to satisfy the demands of future equipmen
44. ange of displays without expensive custom fees Capacitive Disadvantages Transparency Capacitive touch systems typically have a transparency of 85 This can affect image quality particularly for high resolution video mode Stylus Requirements Capacitive touch systems require a conductive stylus to operate Materials that will not conduct a current such as a pencil fingernail or insulated glove will not activate the system Although a capacitive system s sensitivity can be adjusted to activate with a thin cotton or surgically gloved hand it cannot recognise a gloved and an ungloved hand using the same sensitivity setting Sensor Drift Calibration Capacitive touch systems are subject to drift where the touch active zones move from the graphic targets representing them Periodic calibration is required The drift is denied by the manufacturer of capacitive touch systems Infrared Advantages Transparency Because there is no overlay covering the display the Transparency of infrared touch systems is 100 Stylus Requirements Scanning infrared touch systems require that the stylus have a minimum diameter of 5 16 However there are no limitations on the type of material the stylus is made from Calibration Since the optoelectronics determining the touch location are fixed in one place infrared touch systems are not subject to sensor drift They need only to be aligned with the corresponding display Sealability Scanning
45. anise and simplify the model Though INDTE incase thee tequecy is changed am 18 i 1 1 kHz into 32 62 kHz or vice verse the jumper setting on the it can be performed at any time this process EE M Ee Cho d ed is usually delayed until enough details are vi ve found to detect commonalties Figure 5 Rational Rose attributes input dialog Touch screen based equipment Page 29 In object oriented systems besides association and aggregation there is a special relationship that may exist among classes This relationship called generalisation is used as an organised sharing or reusing of features among the classes A class that defines the common structure attributes behaviour operations relationships associations and meaning for a set of classes is the generalisation of this set of classes This process of finding the common abstraction and making a superclass is done by examining the found classes Super classes or better base classes are also used to describe the interface to sub classes When using this kind of interface design it is possible to add classes at a later time The system knows the base class so it knows how to communicate with it Newly designed classes must be derived from the bass class The latter is important for the design of the demonstrator The demonstrator is designed to perform the tasks of a COPER T1 But the design must also be able to perform tasks of other MMM equipment if possible even without r
46. are susceptible to direct exposure to water dirt and other corrosives e Parallax Strain gage touch systems have some degree of parallax caused by the flat glass overlays combined with curved displays Guided Acoustic Wave Advantages e Resolution The resolution of guided acoustic wave technology is more than 100 000 touchpoints per square inch e Calibration Since the reflector arrays determining the touch location are fixed in one place guided acoustic wave systems are not subject to sensor drift The touch system needs only to be aligned with its corresponding display e Z axis In addition to the typical x and y coordinates guided acoustic wave systems can provide a z axis component which is determined by the amount of pressure applied to the sensor e Sealability Guided acoustic wave touch systems sensors can be sealed to meet NEMA 4 requirements preventing water and dirt from penetrating the display s internal electronics Guided Acoustic Wave Disadvantage e Transparency Guided acoustic wave touch systems typically have a transparency between 90 and 95 This can affect image quality particularly for high resolution video mode e Stylus Requirements A hard stylus such as a pen will not absorb the acoustic energy and will not be recognised as a touch Surface Acoustic Wave e Resolution The resolution of surface acoustic wave technology is determined by the physical placement of the reflector arrays Typical resolu
47. at flash overs are normal although not desired during the measuring of the CRTs Touch screen based equipment Page 5 Because of the fact that there has not been any drastic technology change in CRTs the equipment to calibrate them hasn t changed very much either Most of the newly designed equipment were upgrades on older ones Not only the lack of technology change is responsible but also the existing equipment is very stable and accurate 1 3 Problem statement The variety in MMM measuring equipment is very high There is fully automated test and calibration equipment but also equipment that is operated manually Due to the upgrade policy a great variety in input output devices are used to operate the equipment Varying from VT220 terminals to push button panels and from VGA monitors to Led displays A touch screen could replace this mixture The architecture for the most complex equipment is based on one or two VME systems together with 68K processors Sometimes the mechanical movements are controlled by a PLC The intention is to migrate the measuring systems to PC based platforms Electrostatic charge of operators and electromagnetic radiation from the equipment may damage the new equipment This is one of the greatest concerns of the section MMM Currently if a flash over is detected the system is rebooted to avoid errors in the proceedings of the calibration Booting may only consume a little time because the amount of CRTs is hi
48. ata Y 1 gravity for data Y 1 2 const int get m VideoSensFactorX const void set m VideoSensFactorX const int value f4 Attribute m VideoSensFactorY Under the tube parameter Video sens factor Y is meant the dis tance in mm the horizontal video line or video dot is shifted per video step y on the screen This parameter enables the program to position the horizontal video line or video dot in front of the camera centre The value of Video sens factor Y depends of the tube size and the vertical deflection scan amplitude and frequency This parameter is determined by Auto calibra tion as indicated by c see Tube parame ters For the determi nation method see topic Video sens factor X const int get m VideoSensFactorY const void set m VideoSensFactor Y const int value Attribute m DeltaCentreX The tube parameter Delta centre X is the permissible searching tolerance for the vertical video line or video dot position to the centre of the camera If this parameter is adjusted too critical the video line or video dot cannot be positioned An excessive Delta centre X value may cause part of the Video dot to extend outside the camera Field of view The Delta centre X is determined by Auto calibration as indicated by c see Tube parameters using the formula I Delta centre X Video sens factor X 3 4 cons
49. ation the reader should read the book Mathew wrote 2 2 3 Job and Tasks Job and task characteristics drive MMI design in many ways Menus system controlled forced choice question and answer dialogs system free choice are highly structured interface styles and should only be used in tasks that are themselves highly structured input and output modes Command languages user controlled free choice are highly flexible unstructured dialog styles and so are appropriate for unstructured tasks processing mode Ease of learning should be compatible with the turnover rate and primary training In general the ease of use ease of learning trade off should be guided by frequency of use task importance and system usage as in Table 1 Touch screen based equipment Page 9 Frequency of use High Ease of use Low Ease of learning and remembering Task importance High Ease of use Low Ease of learning and remembering System use Mandatory Ease of learning and remembering Discretionary Ease of use Table 1 Compatibility 2 2 4 Tools for the user The performance is affected by the kinds of tools available to the user e g a user with a calculator is more effective than one who must do calculations by hand More importantly operator performance may differ accordingly to the effectiveness of the user interfaces of the different systems Now the simple checklist provided in appendix A and B can be used to develop a questionnaire wh
50. ay be executed for linewidth convergence eccentricity and all other newly to add measurements 1 User selects measurement 2 System makes the selected measurement active 3 User selects field 4 System makes field active and starts selected measurement on selected field 5 System shows measurements 4 Execute measurement in multi mode Scenario may be executed for linewidth convergence eccentricity and all other newly to add measurements 1 User selects fields for next measurement 2 System makes selected field active if it currently is not And the system makes the field inactive if it currently is active 1 and 2 are repeated until user satisfied with selection 3 User selects measurement 4 System measures the selected fields 5 System shows measurements Touch screen based equipment Page 31 For the measurements convergence and eccentricity the system asks the user to remeasure complete colour calibration before it starts the measurements 5 Edit data Scenario may be executed for all possible data available 1 User selects edit data input System shows edit data input window User selects the group of data to edit System shows the selected group input window User selects the data to edit System makes it possible to edit data User confirms or refuses newly entered data System makes data changes according to the confirmation pu TION DA du vc pa 6 Video control Scenario may be executed for Normalise video
51. ble This means that in countries where labour is cheap most of the activities have to be done by people On the other hand if labour is expensive a fully automated production line is the most profitable one MMM offers a scale of equipment to perform the desired measurements From hand operated equipment to fully automated production lines Not only equipment for the three disciplines is made but also equipment to perform quality measurements on finished CRTs Most of the equipment projects images on the screen for evaluation The more automated equipment has vision capabilities to interpret the projected picture The low automation equipment leaves the interpretation to the operator This certainly does not imply that a CRT calibrated with a fully automated equipment performs better than a CRT calibrated manually The calibration always is a mean over the entire screen The middle of the screen and the corners are each opposite The electron beam has to travel a log way to hit the phosphor in the corner Making the perfect image in the corners generally gives a miserable image in the middle of the screen The equipment that projects images on the screen has high tension power supplies to activate the electron beams Not only the high tension voltages put constraints on equipment design but also the fast switching modes of those power supplies Most of the CRTs are tested under stress higher voltage than during normal operation The consequence is th
52. cal Evaluation of Entry and Selection Methodes for Specifying Dates Proceedings of the Human Factors Society 32 Annual Meeting 1988 pp 279 83 Granda R E amp Teitelbaum R C 1982 The effect of VDT command line location on data entry behavior Proceedings of the Human Factors Society 26 Annual Meeting 1982 pp 621 24 W D Greason 1994 Analysis of Human Body Model for Electrostatic Discharge ESD wih Multiple Gharged Sources JEEE Transactions On industry Applications vol 30 NO 3 May June 1994 Greene S L 1988 Entry based verses selection based interaction methodes Proceedings of the Human Factors Society 32th Annual Meeting 1988 pp 284 87 Hollands J G amp Merikle P M 1987 menu Organization and user expertise in information search tasks Human Factors 29 no 5 Oktober 1987 pp 577 89 Karat J 1987 Evaluating user interface complexity Proceedings of the Human Factors Society 31th Annual Meeting 1987 pp 566 70 Karat J amp McDonald E J 1986 A comparison of menu selection techniques touch panel mouse and keyboard Man Machine Studies 25 pp 73 88 Kraut R E amp Stephen J H 1983 Command use and interface design CHI 83 Proceedings pp 120 24 ACM P C T van der Laan Electromagnetic Compatibility college dictaat 5785 Landauer T K Nachbar D W 1985 Selection from Alphabetic and Numeric Menu Trees Using a Touch Screen Breadth Depth and Width CHI 85 Proceedings April 19
53. classes obtained in the list from appendix G is shown in table 4 Tube Picture g pre Camera ERES XY Frame Table 4 Candidate classes Video search Colour recog Convergence Eccentricity Line width Measuring equi nition Touch screen based equipment Page 24 7 2 2 Model dictionary Now a semi formalised textual description of the candidate classes can be given That list is placed in appendix H At this stage the tool rational rose is used to develop the demonstrator application The classes found in the previous section are entered into the class diagram of rational rose Because of the big amount of classes only a few classes are shown in figure 3 which shows the input screen of rational rose During the proceedings of this document only a part of the classes from the demonstrator will be shown As an example of how rational rose works the complete input screens are shown The final release with all classes is available in appendix N CMaasuring Equipment p i e M M CVideo Seatch om o CColou Recognition MUNDI RDUM MEN MT tS MEAN CEDE p I L J _ op Microsoft Vi EJ Micrasol Figure 3 Rational Rose class input Touch screen based equipment Page 25 7 2 3 Associations In Rumbaugh s book Rumbaugh 1991 the definiti
54. ction Auto gain can be changed via the tube parameter Delta gain The default value for Delta gain is 5 const int get m DeltaGain const void set m DeltaGain const int value itt Attribute m VideoSensFactorX By the tube parameter Video sens factor X is meant the distance in mm the vertical video line is shifted per video step X on the screen This parameter enables the program to position the vertical video line or video dot in front of the camera centre The value of Video sens factor X depends on the tube size and the horizontal deflection scan amplitude and fre quency This parameter is determined by Auto calibration as indicated by c see Tube parameters As autocalibration value it will be calculated for the centre of the tube The method is as follows Il 1 First the Video dot is positioned in front of the camera and the nominal video data X and data Y addresses are stored 2 The Video dot position is changed to the nominal data X and data Y 1 The centre of Gravity XY is measured 3 The video dot position is changed to the nominal data X and data Y 1 Again the centre of gravity XY is measured 4 The Video sens factor X and Video sens factor Y are calculated as follows Il Video sens factor X gravity for data X 1 gravity for data X 1 y2 Video sens factor Y eravity for d
55. ds in menu driven computer programs Human Factors 29 no 4 August 1987 pp 453 60 Touch screen based equipment Page 47 Somberg B L 1987 A Comparison of rule based and positionally constant arrangements of computer menu items CHI 87 proceedings May 1997 pp 255 260 Stroustrup B 1997 The C Programming language third edition ISBN 0 201 88954 4 Addison Wesley Teitelbaum R C amp Granada R E 1983 The effects of positional constancy on searching menus for instructions CHI 83 Proceedings pp 150 53 Valk M A 1985 An experiment to study touchscreen burron design Proceedings of the Human Factors Society 29 annual Meeting 1985 pp 127 131 Wallace D F Anderson N S and Shneiderman B 1987 Time stress Effects on Two Menu Selection Systems Proceedings of the Human Factors Society 31 annual Meeting 1987 pp 727 31 Wallace D F Anderson N S and Shneiderman B 1987 Time stress Effects on Two Menu Selection Systems Proceedings of the Human Factors Society 31 annual Meeting 1987 pp 727 31 Weisner S J 1988 A touch only user interface for a medical monitor Proceedings of the Human Factors Society 32 Annual Meeting 1988 pp 435 39 Whiteside W amp Jones S 1985 User Performance with Command Menu and iconic Interfaces CHI 85 Proceedings April 1985 pp 185 91 Whitfield D and R G Ball 1983 Some comparisons of on display and off display touch inpu
56. e cameras is situated here but also the conversion from an image on the measured object to digital data is handled here The basic calculations from data to measurement values are also situated here The vision class itself is only a collection place The real actions take place in the aggregation objects CVideoProcessor and CCameraCabinet 7 5 2 8 CVideoProcessor Currently MMM uses a very powerful processor based on a single board image processor card built around the TMS 34010 34022 Graphics system processor It contains instructions for pixel manipulations on video memory The video memory can store an image supplied by a CCD camera On the card a system memory contains the program for the processor In the future other image cards may be used It might be possible that the image grabbing and the image processing is split Doing so creates a diversity of implementation choices Due to the fast Pentium processors used in PC based industrial computer systems the choice could be to use a frame grabber and process the images on the main processors itself multiple processor systems are available in industrial computers If the processing is so complex or time constraints are important special DSP cards can be used to process the images The base class CVideoProcessor provides a flexible basis in the future The necessary operations are defined in the base class In the future other classes can be derived from this class to provide the needed func
57. echnology is banned because of the disadvantages in combination with CRT s Although this technology is frequently used in industrial environment but only in combination with LCD and no constrains on ESD Touch screen based equipment F circumstances use case Use case user Name Prerequisites Description Exceptions given Result Name Prerequisites adjusted Tube file is open Description Exceptions Result Name Prerequisites Description Exceptions Result Name Prerequisites Description Exceptions Result Name Prerequisites one to be used as a new one Description Exceptions Result Name Prerequisites Description Exceptions Login System is running No user is logged in One of the users logged in with his name and password The system verifies user data The system displays startup screen some users may not like to be logged in If user data is inconsistent with system user data an error message i5 One user is logged in Measure Convergence User is logged in Colour recognition is done Tube supplies are Measure Eccentricity Measure Linewidth Select tube file name User is logged in One can select the right tube file or one can alter The system displays the directory with the available tube types and asks the user to select one or alter one 1 The user selects a tube The system connects the tube file to the tube 2 The user selects a file to a
58. ecompiling and linking The measurements used in the analysis for the demonstrator are also basic measurements for MMM equipment The class Method is added as a base class to the classes Video search Colour recognition Convergence Eccentricity and Linewidth The class Measurement parameters is added as a base class for the classes Colour Recognition parameters Convergence parameters Eccentricity parameters and LineWidth parameters If in the future a new measurement is necessary or another MMM equipment must be designed two new classes may be derived from the base classes and added to the system In the future it might also be necessary that the measuring equipment is used for other measured objects therefore the measured object Tube is added as a derived of a new base class Measured Object As mentioned before the currently used image processing card is the Sbip working in a VMS environment But in the future it might be possible that a new processing card is available or the entire system is to be PC based Maybe new algorithms are designed to process the image In all those cases a super class representing the interface to the card object makes it possible to adapt the system in a short time to make the system even more flexible for newly or redesignable objects The measuring dependent subsystems are added as inheritors of the super class used systems In case of the demonstrator a Vision system is needed to obtain the images from the CRT A V
59. eedback Human Factors Society 1985 pp 445 58 Callahan J amp Hopkins D 1987 An empirical comparison of pie vs linear menus CHI 88 pp 95 100 Card S K amp Moran T P 1983 The Psychology of human computer interactions Hillsdale N J Lawrence Erlbaum Associats 1983 Card S K 1981 The Model Human Processor A model for making Engineering Calculations of human Performance Proceedings of the Human Factors Society 25 annual Meeting 1981 pp 301 305 Ezzell B 1992 Grafisch programmeren in windows 3 1 ISBN 90 229 3808 5 Foltz P W 1988 Transfer between menu systems CHI 88 Proceedings May 1988 pp 107 122 ACM Franey J P 1995 Field Induced ESD from CRT s Its Cause and Cure JEEE Transactions On Components Packaging And Manufacturing Technology PART A vol 18 NO 2 June 1995 Franik E P amp Kane R M 1987 Optimizing visual search and cursor movement in pull down menus Proceedings of the Human Factors Society 31 Annual Meeting 1987 pp 722 26 Furnas G W 1985 Experience with an adaptive indexing scheme CHI 85 Proceedings April 1985 pp 131 36 ACM Galiz W O 1989 Handbook of screen format design 3rd ed Wellesley Mass QED information Science Inc 1989 pp 173 77 Touch screen based equipment Page 45 Good M 1985 The use of logging data in the design of a new text editor CHI 85 Proceedings April 1986 pp 93 113 ACM Gould J D 1988 Empiri
60. elected measurement System shows measurements if any we ee 12 Measurement utilities in multi selection mode Scenario may be executed for Convergence Gravity XY Gravity X Gravity Y Brightness Brigtness colour Linewidth 1 User selects field to be used in relation with the utility functions 2 System makes selected field active if it currently is not and the system makes the field inactive if it currently is active 1 and 2 are repeated until user satisfied with selection 3 User selects measurement utility 4 System performs selected measurement 5 System shows measurements if any Before starting the utility linewidth the system asks the user to enter the focus direction and the level For the functions linewidth and brightness colour the currently selected colour is used It may be changed in the Video control window Scenarios can also be depicted graphically using a message trace diagram to capture the pattern of interactions for each one In the message trace diagram columns are used to indicate the participating objects in the scenario usually with the actor s to the left A message trace diagram shows the events and messages passing among the objects by the use of ordered labeled arrows The internal scenarios are given in appendix O Touch screen based equipment Page 33 7 5 System design After determining the system s requirements the overall approach to the solution its system architecture and style m
61. electronic or mechanical protection can be applied to the sensitive parts of the electronics This is not always the solution because if the protection is starting to work the properties of the interfering signal are changed and the destructive properties of the signal may be even bigger To make a good decision on the touch screen to use it is important not only to look at the CE marker but also on the construction and technology of the touch screen Touch screen based equipment Page 16 6 Touch Screen technology In order to select the touch technology that best fits the MMM equipment it is important to take a brief look at how each technology functions There are seven basic types of touch technology capacitive overlay force vector guided acoustic wave resistive overlay scanning infrared strain gage and surface acoustic wave Each type of touch technology has attributes that are desirable for specific applications All types of touch systems are attached to a display unit whether a terminal CRT flat panel display static graphic or combination of flat panel display and static graphic The difference between the technologies lies largely in the way the touch is detected and the method used to process the touch data Appendix E gives an introduction to the types of touch technologies available today It is advised to read this appendix because it contains valuable information 6 1 Choice of Touch Screen By investigation of the
62. elta threshold Delta gain Video sensitivity factor x Video sensitivity factor y Delta centre x Delta centre y Video shift step size x number of video shift steps pulse width 1 2 Horz field of view 1 2 Vert field of view Line frequency Frame Frequency Clock pulse type ID one of the 25 fields on the display face representation of the picture on the tube representation in the system of a real person function password preferences representation in the system of a remote system two directions representation of the hardware tube supplies to generate an image on a tube Touch screen based equipment Page 83 I Questionnaire User Profile Checklist Job and Task Characteristics Job catagories Turnover rate Frequency of use Operator High Low Production chief Moderate Medium Process specialist Low High Maintenance man Sytem use Primary training Mandatory H None Discretionary Manual only Elective formal Mandatory formal Task importance Task structure High High Low Moderate Low Psychological Characteristics Cognitive styles Attitude Motivation Verbal Spatial Positive High Analiytic Intuitive Neutral Moderate Negative Low Knowledge and Experience Reading level Typing skill Education Non Non Non Low Low Low Medium Medium Medium High High High System experience Task experience Application experience Expert Novice in field No similar systems
63. eps The lifecycle of the software starts with the initial formulation of the customer s wish and continues via analysis design and implementation to testing Followed by the installation and maintenance of the application There are a lot of tools to guide a developer through the bumpy road of software development At the section MMM a pilot project was developed as a windows application In the future MMM wants to continue developing in windows so for the demonstrator the window environment is used The compiler to use is C which allows object oriented design There are several software engineering methods for developing OO structured designs The pilot project used OMT developed by Jim Rumbaugh Rumbaugh 1991 Another famous OO tool developer is Grady Booch Lockheed 1997 Currently they are working at the Rational Company developing a method using all advantages of several development tools The demonstrator is developed with OMT from Rumbaugh and the tool used is Rational Rose The equipment selected to be used for the demonstrator is the COPER T1 COPER T1 stands for COmbined PERformance Tester with 1 camera This name indicates that different measurements can be executed In the basic software version only the convergence measure algorithm is implemented Options like landing sharpness and raster measuring algorithms can be delivered as add on measurements on the basic version The measurements can be done on CMT and TVT colour picture C
64. equipment Page 98 Use Interface m Measuring m VideoSystem m Remote Equipment t 1 1 Iselect specific video do l 1 l make l l execut j e g color si selection pp i color dialog l do color l l select color l i i colo Li Video control l a H Touch screen based equipment Page 99 P Dialogs Measure Fy ei an ct 1 Li 1 t 1 I 4 L L t Q K 7 i 1y DID 0 L4 9 d T s Wied 00m zo Mim min zinc t Am omm d mom nny oem t Li i 1 L L 4 t t 1 L t 1 t Ld 1 1 i t 1 L t L 1 a P ing op fend L 1 ond Measurement Utilities Mava atarra ARPS iajyaweied u nuaui Jaweed P3 001 eSeq yuatudinb a p seq usa15s yono Wi H ee ee WMH seamn mn ng ep ipa eweuayaqny amseay e 5 oe Soe LA dGueudmb3 WAAL Jequunu p 10 en eA Meu 139 43 10I e8eq jueuidmbo poseq usoa15s qono a GTSIJ 0 PTOusezui 0 UT 5 0 Uuoa3raxseg id GIG Oo 101d 0 pToyusezuL 0 2J FAASSEN 0 ure5 Ll GT3IJ 5 qT3I4 0 uoaTaXsekH 0 pToysezyL M Q1914 0 ure5 T QT131I4 Q uoardxsexd S qT3I4 Q0 prouseaur 0 uTeD 0 uoiraxXseW a qT4I4 ido qI4I4 0 pToys
65. eruL 0 uoiTdxXseW 0 ure5 D qTAI4 X qT3I4 Q uo3rdXSek 0 prousezul T dqT31I4 0 ure5 4 QG I3I43 Q uoardxXsew a qT314 s1e3eurPzequor3tufo2eu1noTo2 0 qoaraxsew 0 YORTAXSeW X3 QTaIg D qT4I4 0 YOITIYSeN 0 uoaraxsen 4 qT4I4 N qT41I4 seTqetzeapTeta 0 YD3TAXSeW a qT4I4 pexooT SUTEN d L STNAXDOTI Q Aduanbserzqowerg Q YOITaAXSeW 0 AduaenbezJSuTT W QT3I4 O MeTAJOPT9T432J9AZXT O MBTAJOPTOETAZIOHZXT 0 uoiraxsew 0 uaprMesTnhd H QTI3IJ 0 sdeiS33TUuSO9PTAJOZN 0 Xezrsde1S331USOSDTA 0 uoaraxsew Q0 Aeaauepeired if ATIA 0 Xeaquejgeired 0 AX1O32P4SUSSOSDTA 0 u2a3rdxsen 0 X1TOQIeTSUSsSOSPTA A Q I3I4J Q Uure5eired 0 PTlouseduLeiTed 0 uoaraxXseq LW29 ed L z qT4I4 WEN 0 YoITdysen seTqetTzeaeqny fl aTaI4 Q deistbA Q uoaradxsew 0 woneba 1H GISIG 0 deasgo1queBA 0 deqsudpeba Q YOITaXSen Q wonudpeba 29 Q I3I3 0 deaS3OAQNU DEDA 000000 0 peusues 0 uqoirdxXsew 000000 0 ueeassuess u GTS3I43 000000 0 enTasues 0 deasgoiNdeasoi2TW Q YOITAXSeW 0 Xxezrsdeasdea3soio2IN X Gaird Q ATeAe7T 0 XT 4 T 0 uoardaxseW AX UOT3291IGQ d qT4I4 ON unbayad 0 YOITdxSeW epouaeKua3prMeurT y 193oureaed woy 31ed jo o dure xq Oo TOT 33ed jusuidinba paseq uoo1os qno
66. et preserve no m_VideoShiftStepSizeX value Ifl end CTube set m VideoShiftStepSizeX o set inline const int CTube get m NrOfVideoShiftSteps const begin CTube get_m_NrOfVideoShiftSteps get preserve no return m NrOfV ideoShifiSteps 4 end CTube get m NrOfVideoShiftSteps o get inline void CTube set_m_NrOfVideoShiftSteps const int value begin CTube set_m_NrOfVideoShiftSteps set preserve no m_NrOfVideoShiftSteps value H end CTube set_m_NrOfVideoShiftSteps set inline const int CTube get_m_PulseWidth const Ifl begin CTube get_m_PulseWidth get preserve no return m_PulseWidth Ifl end CTube get_m_PulseWidth get inline void CTube set_m_PulseWidth const int value Ift begin CTube set m PulseWidth 6 set preserve no m PulseWidth value f end CTube set_m_PulseWidth set inline const int CTube get_m_1X2HorzFieldOfViewQ const 4 begin CTube get m 1X2HorzFieldOfView 6 get preserve no return m 1X2HorzFieldOfView amp H end CTube get_m_1X2HorzFieldOfView get inline void CTube set m 1X2HorzFieldOfView const int value begin CTube set m 1X2HorzFieldOfV iew set preserve no m IX2HorzFieldOfView value end CTube set_m_1X2HorzFieldOfView set inline const int CTube get_m_IX2VertFieldOfView const H begin CTube get m 1X2 VertFieldOfView get preserve no return m_1X2VertFieldOfView end CTube get_m_1X2VertFieldO
67. etal on the interior of the glass of the neck The getter material adsorbs much of any remaining gas molecules left over from the evacuation of the CRT e When the CRT is ready it is measured to investigate if there are any loose contacts or if there are conducting paths that should not exist e The next step is to calibrate the three separate electron beams red green blue on the appropriate phosphors dots The CRTs of Philips are equipped with an IMACO ring at the front of the electron gun The IMACO ring is basically a programmable 8 pole magnet The programming is done with the aid of 8 magnetic coils which magnetise the IMACO ring e Finally the CRT is matched with a deflection coil that provides optimum purity It takes some ingenuity to get a good match between the light used for exposure and the future electron optical system in order to get good purity The mentioned technology used in CRTs has not dramatically changed during its existence 1 2 MMM equipment The Measuring Magnetising and Matching MMM is done with equipment developed and build at the MMM section of Philips The section MMM is part of the Equipment Engineering Division EED of Philips The EED develops and builds equipment to produce CRTs CRTs are produced in a lot of countries all over the world Every country has its own characteristics about wages education culture etc Because of the amount of produced CRTs it is important that the costs are as low as possi
68. evices appropriate to optimise performance of the chosen MMI As the title of this report proves a touch screen is a well suited IO device to support the optimal MMI The use of a touch screen in a graphical user interface puts certain constraints on the design These constraints are pointed out in chapter 4 As mentioned in the problem statement ESD and EMC are of great concern when introducing new parts of the equipment In chapter 5 the ESD and EMC constraints are evaluated to make a good decision what touch screen technology is appropriated for MMM equipment Chapter 6 evaluates all possible touch screen technologies Together with the desired constraints a touch screen technology is chosen to be used for MMM equipment Chapter 7 shows the design of a demonstration model of aMMM equipment using a touch screen The demonstrator is designed to serve as a blue print for other MMM equipment Rational Rose OMT is used as case tool to design the demonstrator Finally chapter 8 contains the conclusions of the study Touch screen based equipment Page 7 2 Man Machine Interface 2 1 Introduction Industrial systems are really more than the software alone and the scope and purpose of the system is wider than the functionality provided by the software In fact the software is only one component in a larger system and provides only a subset of the functionality that is desired This larger system includes at least one human user and may in fact include
69. f other systems Moderate to high computer literacy JOB AND TASK CHARACTERISTICS Moderate to high frequency of use Little or no training Discretionary use Low to moderate turnover rate Moderate task importance High task structure The design of fill in form systems can be divided into five separate issues e Fill in form caption and field design Fill in form input formats Fill in form prompts and instructions Fill in form navigation Fill in form error handling Fill in form organization and layout There are several areas to consider in the design of captions and fields One study Savage compared four field indicators in combination with several cursor types A conclusion that can be drawn from this study is that blinking block cursor in combination with broken underline field indicator is an optimal way to design input fields on a fill in form Note however that the performance study did not include actual speed or correctness of input Touch screen based equipment Page 58 Fill in form input formats To gain speed and reduce input errors the following design guidelines are appropriate If possible provide completion of unambiguous partial input Provide pop up or pull down menu s for fill in fields with many entry options Avoid complex entering fields Break up long input fields Provide defaults Make high frequency inputs easy to express Let the user specify the units of measurement Allow abbreviated input A s
70. f the same type The pointer is NULL if the referenced Object Var 12 22 98 1 field is not selected During development a second attribute is added to indicate that a dialog may be visualised without the need of an already opened dialog This feature is added to Object2 Var 12 12 avoid the need to control the handling of the dialog in another object In this way more objects can initiate a request to show data in the dialog if it was open and do nothing if it was closed Now depending on the second attribute three actions may take place a dialog is opened the contents of an existing dialog is changed or nothing happens It depends on the contents of the values in the objects what is going to be displayed E g three pointers in the array are not NULL meaning show the data of the objects where the pointer points to The first step is to compare the data in those three objects If a value is the Figure 8 Update mechanisme same in all objects that value is shown If one differs an unknown sign question mark is shown In figure 9 this example is shown graphically The previous can be done for one no comparison is necessary to many objects The values to display are stored If the dialog is already shown a comparison is made between the new values in that dialog and the old values stored in the same manner as just mentioned If they differ the user has made changes If this is the case the user is asked to save or cance
71. fView get inline void CTube set m 1X2VertFieldOfView const int value begin CTube set_m_ X2VertFieldOfView set preserve no m IX2VertFieldOfView value H end CTube set m 1X2VertFieldOfView set inline const int CTube get m LineFrequency const f begin CTube get_m_LineFrequency get preserve no return m_LineFrequency f end CTube get_m_LineFrequency get inline void CTube set_m_LineFrequency const int value fil begin CTube set_m_LineFrequency set preserve no m_LineFrequency value end CTube set_m_LineFrequency set inline const int CTube get_m_FrameFrequency const begin CTube get_m_FrameFrequency get preserve no return m_FrameFrequency end CTube get m FrameFrequency 6 get inline void CTube set_m_FrameFrequency const int value begin CTube set_m_FrameFrequency set preserve no m_FrameFrequency value fit end CTube set_m_FrameFrequency set inline const enum CTube get m ClockPulseType const Iii begin CTube get m ClockPulseType 6 get preserve no return m ClockPulseType It end CTube get m ClockPulseType o get inline void CTube set_m_ClockPulseType const enum value 4 begin CTube set_m_ClockPulseType set preserve no m ClockPulseType value end CTube set_m_ClockPulseType set inline const boolean CTube get m DAF gun const begin CTube get_m_DAF_gun get preserve no retum m DAF gun
72. for member and the first letter of each additional word is capitalised The demonstrator is designed in visual C In that language there are several ways to define the attributes The most common are Public Protected and Private To preserve data hiding and to make sure the class is the only responsible for the attributes the attributes for the tube class are declared private The tool Rational Rose will automatically produce if desired public access functions Private declared attributes are only accessible by operations of the class itself Public declared operations are accessible by every class 7 2 4 2 Operations The tube class must be able to load and save its parameters As one can see in figure 5 the parent of the field class 25 fields is the tube class In the field class an attribute m_Selected is available to know if the field is selected If an object wants to know what field is selected he asks it to the tube class because it knows how to access the fields Therefore a third operation is added to the tube object It might be possible that more fields are selected at the same time so a fourth operation is added The four operations are written in table 8 m_GetSelectedField m Load filename m GetNextSelectedField Table 8 Operations in the tube class These operations must be public because they are initiated from the parent class Touch screen based equipment Page 28 7 2 5 Rational Rose After enter
73. general purpose vibration glass visible damage cleaning solutions overlay Chemicals that susceptible to affect polyester shock Manufacturer s published data Explanation of used words Drift The gradual movement of the touch active zones away from the graphic targets representing them Touch technologies that are subject to drift require periodic calibration to should not be used restore the touch active zones to the correct coordinates NEMA 4 12 complance National Equipment Manufactuters Association NEMA 4 compliance indicates that a device can withstand hose directed water and still operate NEMA 12 compliance indicates that a device is for industrial use Parallax An optical phenomena in which a touch zone registers slightly differ from the graphical target Parallax is caused by space between the display surface and the plane of the touch sensor and varies according to the type and architecture of the touch technology Parallax tends to be unnoticeable on flat displays and increases relative to the curvature of the display Resolution The physical spacing between the adjacent touch coordinates Touch screen based equipment Page 73 Response time The time required by the touch system to locate the touch and transmit the coordinates to the host system Stylus A finger digital pen or any device used to activate the touch system Transparency The clarity of the image measured in percentage of light that
74. gh so stalling the production costs money A PC based platform requires new development environments The section MMM has chosen to migrate to Windows NT This means a Microsoft environment for future developments 1 4 Objective The objective of this study is to obtain more information about novel approaches to user interfaces and system design This is part of a broader study to investigate the feasibility for the section MMM to design a new generation of MMM equipment Important aspects in this study are Desired functionality for the users Description of man machine interfaces MMI for middle to large sized MMM equipment Information model of MMI Market scan of available touch screens Selection of suited touch screen technology Style guide for MMI with touch screen Verification of ESD and EMC requirements Designing an implementation of a working application in a windows environment Evaluation of the application Touch screen based equipment Page 6 1 5 Organisation of this report The organisation of this report is as follows Chapter 2 handles about the Man Machine Interface MMI applied in MMM equipment A thorough investigation is added in the appendix C about the possibilities in MMI s This data is mapped on the characteristics of the operators using the MMM equipment This strategy points out which MMI is applicable in the user s point of view Based on the results chapter 3 handles about the selection of the IO d
75. gh task structure Touch screen based equipment Page 60 Because questions and answer interfaces are relatively rigid structured dialog style they are best suited to tasks that are themselves highly structured There is not much research on question and answer interfaces The guidelines that follows are based on basic research and aspects of other dialog styles Maintain system titles for navigating proposes Use clear and simple languages Provide brief prompts and instructions Make a visual difference between questions prompts instructions and user input Minimise typing requirements Allow flexible navigation Command languages The original style of computer human interaction is command language The user types in requests through an artificial language Advantages and Disadvantages of command language dialog style Command language dialog styles have particular advantages and disadvantages to other dialog styles These are included in the following table Advantages Disadvantages Powerful Difficult to learn Flexible user controlled Difficult to remember Fast efficient Assumes typing skill Uses minimal screen real estate Error prone Enhancements are invisible Command language is most appropriate for USER PSYCHOLOGY Positive attitude High motivation KNOWLEDGE AND EXPERIENCE Moderate to high typing skill High system experience High task experience High application experience infrequent use of other sys
76. gital four and five wire designs are available for sensing the position of the touch In the digital design one layer is divided in to r horizontal lines and the other layer is divided Spacer Dots in to vertical lines Together the two layers Thin Transparent form a raster When a finger or other stylus Metallic Coating presses the two layers contact is made between a horizontal and a vertical line The controller detects a specific block in the raster Figure 13 Resistive In a four wire resistive touch screen electrode arrays at opposite sides of the substrate can establish a 1 D voltage gradient across the substrate s resistive indium tin oxide ITO coating Similar electrodes can establish an orthogonal gradient across the membrane s ITO coating Both sets of electrodes also allow the ITO coatings to act as high impedance probes When the user touches a four wire system screen the controller establishes a gradient across the substrate The controller then measures the voltage at the point of touch using the membrane as a probe Similarly the controller establishes a gradient across the membrane and uses the substrate as a probe The two voltages provide the x and y coordinates of the touch point In a five wire system the substrate has a resistive ITO coating and electrodes on all four sides The membrane has a single electrode and a conductive coating When a finger or other stylus presses the two layers together the cont
77. he selects a tube file that corresponds with the CRT he is going to measure In that tube file CRT specific parameters are stored The user is able to execute measurements on a measured object In case of the demonstrator the measured object as in all cases at the current time is a cathode ray tube The demonstrator is a copy of the COPER T1 and the main measurements are convergence eccentricity and linewidth These measurements are also base measurements for other MMM equipment CRT calibrations are very complex and the optimal calibration of the CRT is an average measured at several places on the screen For measurements with COPER TI the screen is divided in 5 x 5 measuring fields The user must be able to select one of these fields For several measurements a specific colour picture or location is needed so the user must be able to make a selection If the COPER T1 is not equipped with an XY frame the camera is placed Touch screen based equipment Page 23 by hand This is very inaccurate so a video search must be available In the current MMM equipment CCD cameras and LDR s are used to measure the CRT The CCD cameras black and white and the LDR s can only detect grey levels so the equipment is able to conduct a colour recognition cycle to determine which dot is of what colour The images are composed on the CRT by tube supplies The main measurements all use elementary measurements to calculate the results For error detec
78. he delay time of the first measurement after switching over to Video dot is defined This parameter is only required to usc Coper T on the lLA monitor because of the Vk Ik circuit behaviour in the IIA For non ILA applications use zero seconds const int get m MeasDelayForZA const void set m MeasDelayFor2A const int value 88 Attribute m Type Under the tube parameter Tube type TVT or CMT tube types can be selected This choice is important for the Colour recognition algorithm CMT has a hexagonal dot structure TNT has an Inline structure const enum get m Type const void set m Type const enum value li Attribute m DeltaThreshold This value will be added to the value as found after the function auto threshold Normally this function will be used after the auto gain function and before the convergence measurement to optimise signal noise ratio The default value for Delta threshold is 10 To check ifthis is sufficient visually observe the picture area outside a Video dot If pixels light up outside the Video dot increase the Delta threshold value to e g 15 Changing the setting can be done via submenu Delta threshold by keying in the desired value const int get m DeltaThreshold const void set m DeltaThreshold const int value Touch screen based equipment Attribute m_DeltaGain The Gain as determined under the fun
79. he most important guidelines are drawn from several research studies e Establish conventions for menu design and apply them consistently on all menu screens within a system e Use context labels menu maps and place markers as navigational aids in complex menu systems If possible use direct access to menus and create macros to facilitate navigation for expert users e Think about backward navigation These guidelines are found in several studies Teitelbaum amp Granada 1983 Foltz 1988 Lee amp Chao 1989 Fill in forms A fill in form interface is similar to a paper fill in form The only difference is the presentation which is a computer screen Advantages and Disadvantages of Fill in forms styles Fill in forms have particular advantages and disadvantages to other dialog styles These are included in the following table Advantages Disadvantages Self explanatory Assumes knowledge of valid inputs Efficient use of screen real estate Assumes typing skill Many possible input values Error prone Provides context Knowledge of special keys Enhancements are visible Inflexible Touch screen based equipment Page 57 Fill in forms are most appropriate for USER PSYCHOLOGY Negative or neutral attitude Low or moderate motivation KNOWLEDGE AND EXPERIENCE Moderate to high typing skill Little to moderate system experience Little to high task experience Low to moderate application experience Moderate to frequent use o
80. he object is responsible to know while operations are the behaviours that the object is responsible for performing At this stage it is important not to analyse in great detail because one might loose overview 7 2 4 1 Attributes The attributes for the demonstrator are found in the problem statement and in the user manual for the COPER Mullekom 1996 A list of the attributes with the owner classes is found in the model dictionary which can bee seen in appendix H To follow the tool input and to clarify the demonstrator design the tube class is worked out in detail Touch screen based equipment Page 27 The tube class represents the measured object In the user manual of the COPER T1 one can read that it is possible to edit the general tube parameters These parameters are specific for one kind of CRT The tube object is responsible to know these parameters so they are added to the attributes of the tube class The problem statement does not add more attributes The attributes found for the tube class are listed in table 7 m 1 2VertFieldOfView m_LineFrequenc m DeltaCentreX m DeltaCentreY m DeltaThreshold m VideoShiftStepSizeX m DeltaGain m NumberOfV ideoShiftSteps m ClockPulseTube m VideoSensitivityFactorX m PulseWidth PO m_VideoSensitivityFactorY _ m_1 2HorzFieldOfView eo o S Table 7 Attributes of tube class Following a style guide attributes are started with m
81. his study is part of a larger study to investigate and migrate to that vision Older MMM equipment is normally not replaced by a newer one s but most of the time it is updated The hardware layer in the first steps of the migration forms the connection between PC and VME systems In later times or for new equipment it is necessary to develop PC based hardware Only if a fully PC based platform is desirable The policy for MMM is to avoid risk as much as possible Therefore they use proven development environments and only change if another has proven to be stable In lambert 1996 Windows version 3 1x was used and the newest relevant developing environment was MSVC 1 51 In the future they want to migrate to if proven stable Windows NT Currently they are investigating newer development environments but until a proper decision is made MSVC 1 51 is the choice to make The demonstrator will run on a stand alone PC but if the hardware subclass is implemented and the proper measurement code is added it can run on a VME PC PC The subsystem user interface contains all objects that interact with the users These objects accept the events of the users The user interface contains classes that are added in the design phase Because of the visual capabilities of the development environment used for the demonstrator and because of the lack of knowledge of its capabilities the user interface classes are designed with the aid of the application studio The u
82. ialog with the available measurements But for analysing debugging and system set up more dialogs must be available Displaying available dialogs is handled by an extra dialog The CMeasuring Equipment class controls the extra dialog Touch screen based equipment Page 35 As can be seen in the massage trees printed in appendix the CMeasuringEquipment object is responsible for the execution of the next action It depends on the execution mode if a selected measurement is to be executed immediately or it has to wait until a field is selected To tackle this problem three functions are added The first defines that the next action to be taken is the selected one And the second executes the selected action If the execution mode is Multiple selection the second function is called immediately If this is not the case execution is started if a field is selected Calling the third function that executes the selected action does this See appendix P for the user interface layout of the previous mentioned dialogs 7 5 2 2 CMeasuredObject The Class CMeasuredObject represents the measured object For now the tube is the only object to measure but in the future there may be other objects so the class CTube is derived from the class CMeasuredObject The CRT is divided in 25 fields The only time for a field to be visual on the user interface is for selection To avoid a lot of programming and make it possible that every field displays itself the OO wa
83. ich can be answered by a representative sample of users or people who know the users The questionnaire is printed in appendix I The questionnaire is answered by 4 people from EED who s task it is to keep in touch with the end users The conclusions drawn from these questionnaires are displayed in table 2 To decide what design rules are to be used and what the constraints are first a more detailed explanation about dialog styles is appropriate This explanation can be found in appendix C 2 3 Selecting dialog style for MMM equipment users To make a selection between the dialog styles all the analyses in the previous sections can be gathered in selection tables see appendix D Each cell in the table holds a particular value of the user characteristic Using the results from the questionnaire a first scan marks several dialog styles The next step is to look if an unmarked cell has a serious disadvantage If not the cell is marked Operators for MMM equipment can be separated in two groups The first group comprises of people working at equipment operated by the operator e g COPER T1 In those cases the operator is constantly interacting with the MMI of the equipment The second group has a more controlling part Those operators supervise one or more automatic equipment Only if a failure has occurred or a preventive action has to be taken the operator will interact with the MMI of the equipment In appendix D it can be seen that for both groups
84. ideo system is needed to keep track of the characteristics of the images on the CRT Finally a Supplies system is needed to drive the CRT to produce the image on the screen Some MMM equipment also performs calibration operations like magnetising For this equipment a magnetising system may be added to the measuring equipment together with the needed measurement classes Five new super classes are added to the design The classes Method and Measure parameters describe the interface between measurements and the system The class Measured object describes the interface between the measured object and the system The class Used systems makes it possible to add new subsystems needed for other measurements The class Video processor forms the interface between the system and the imaging processing part One note must be made The design thus far is certainly not the final design It is very well possible that subsystems derived from the super classes are complexes of other classes For the demonstrator all measurements and actions take place via the remote control The classes as they are derived now directly place their requests to the remote control while in a fully operational system some classes must be more refined to fulfil the proper actions As an example the hardware design is not available at the current state of the design process Especially the classes derived from the used System class are built with that design Touch screen based equipmen
85. ing the attributes and operations in the tool Rational Rose the input screen looks like figure 5 G i n m ri 3 n Be CTube m name string tn MeasDelayFor2A int m Type enum m DeltaThreshold int m DeltaGain int VideoSensf actoiX int VideoS ensF actorY int DeltaCentrex Int measured object m DeltaCentreY int Gym VideoShifiStepSizeX int Gym NrOfVideoShiftSteps int Gym PulseWidth int 7 m 1 2HorzFieldOfView int amp ym 1 2VetFieldOfView Int rU amp ym LineFrequency int m FrameFrequency int Gom ClocPulseType enum m DAF gun boolean m_Save CColout Recognition CEccentticity m Load m_GelSelectedField m GetNextSelectedField _CConvergence_ CLineWidth AS UE acm VS measuring equipment FERRE E E Has Tube Aas Video search N a CColcur Rec Pata CConwPara Se L7 The lock in front of the attributes indicates that it is a private attribute At this stage one can fill in all possible knowledge about the attributes and the operations The input screen for the attribute m_LineFrequency is given in figure 6 z nder ha tube parameter Line frequency the appropriate 7 3 Generalisation line deflection frequency must be stored Therefore check the actual line deflection frequency setting of the time base choica in the nd st edur As analysis progresses it becomes necessary eer ee to org
86. is allowed to pass through the touch system In overlay based touch systems clarity is reduced relative to the thickness of the overlay Touch systems that do not use overlay based technology have 100 transparency Z axis The capability to measure pressure against the touch sensor in additiond to x and y coordinates Specific pressure thresholds can be set to emulate mouse button clicks and activate different touch events MTBF Mean Time Between Failure A statistical estimate of how long a component or a system is expected to perform before a failure occurs Comparison of the Touch Technologies After giving an overview of the various touch technologies and the factors that can affect their suitability for use in various application environments let us compare each technology s advantages and disadvantages Force Vector Advantages e Resolution Force vector touch systems provide a resolution of 40 Points per inch e Z axis In addition to the typical x and y coordinates force vector touch systems can provide a z axis component which is determined by the amount of pressure applied to the display e Transparency Since no overlay covers the surface of the display the transparency of the force vector technology is 100 e Stylus Requirements Force vector touch systems have no stylus limitations e Integration Since the display rests on the force vector touch pedestal it does not require an invasive integration Force Vecto
87. isting COPER T1 is placed outside the system boundary The user is also outside the system boundary In section 2 3 one can see that there are two types of users with a different privilege in the system One is the operator and the second is the maintenance engineer 7 1 8 Actors For general MMM equipment the external actors User and Tube are identified The demonstrator needs a remote control That remote control can be seen as an actor For some MMM equipment a XY Frame is needed so that forms a fourth actor 7 1 9 Use case The actor identified as User is the one that initiates all activities in the system There are two types of users possible The first is the operator mostly initiating the main measurements secondly the maintenance user This user is able to fully use the possibilities of the measuring equipment and to freely change process and program parameters To make this possible the user must login 1 this is the first use case First the use case of the operator is given e The user can select a tube file name according to the tube to be measured 2 e The user chooses a measurement to conduct 3 e The user chooses the field to be measured 4 e For some measurements it is necessary to select a colour 5 e The user must have the ability to execute a video search to place the image in front of the camera 6 e The user must be able to execute a colour recognition to enable colour detection by the system 7
88. ith a fine array of holes one for each trio of phosphor dots positioned about 1 2 inch behind the surface of the phosphor screen With some CRTs the phosphors are arranged in triangular formations called triads with each of the colour dots at the apex of the triangle With many television and some monitors they are arranged as vertical slots with the phosphors for the 3 colours next to one another The location at which the electrons hit the phosphor landing and the relative position of the three separate electron bundles convergence is determined by the mechanical parts of the electron gun The following is a greatly simplified description of the general process of colour shadow or slot mask CRT construction e The screen and envelope glass pieces are molded separately and then glued together as one of the last steps of assembly prior the baking and evacuation The shadow mask is manufactured through a photo etching process Since a position error of even a tiny fraction of a mm would result in purity errors each shadow mask is unique for its faceplate They are not interchangeable l e The CRT is evacuated This takes several hours at the vacuum pumps The assembly is then sealed by heating and melting e The getter part of the electron gun assembly is then activated via induction heating from a coil external to the neck of the CRT This vaporises and deposits a highly active Touch screen based equipment Page 4 m
89. ks like figure 4 The multiplicity of an association is the number of instances that participate in the association Likewise the association has a role at each end of the association For clearness only the aggregation between class measuring equipment and class tube is added with roles and aggregation name Aggregation is stronger form of association It shows the relationship of a Touch screen based equipment Page 26 whole to its parts The multiplicity is 1 to 1 in which case nothing has to be added For the aggregation between class tube and field the multiplicity is 1 to 25 Rational Rose vers mdl Class Diagram Logical View 7 Main HEIE x ius as View Browse Report Quer Jools Window Help ae Siue veer c7 va x Dem mle S efe alala ae ae e Has User C Measuring Equipment Dy cx Has Video search measured object CVidao Search C Colour Recognition CEccentiicity CConvergence Te CLineWidth SS CField D f f ie E Ld d dnt For Help press F1 i MOUTH LL d Ql NUME Figure 4 Rational Rose with classes and associaties 7 2 4 Attributes and Operations To distinguish an object from others and further define the roles in which the objects participate essential settings and behaviours must be identified There are two types of features that need to be identified OMT attributes and operations Attributes are the things that t
90. l the new Touch screen based equipment Page 38 data The latter is done by a separate function because it can also be used if the dialog is terminated If a save action is desired the previous selected objects are updated Trivially a storage array of pointers is necessary to keep track of the previous selected parameter objects After these actions the newly found data is given to the dialog If no dialog was open and it is allowed a dialog is opened Checking for a change by the user is done before comparing the new data variables in the objects In this way only one mirror of data variables is necessary Another possibility is to create a new object of the same type and store the data in that temporary object A note is to be made The object that creates the pointer array uses pointers of type CMeasures Params Because that object is part of the system it does not know the type of the parameter class But within the receiving object a C type cast is made to its own type because he must access the measurements and parameters This is no problem because an object knows its own class New classes for other measurements must all contain the same structure This can be guaranteed by deriving these classes and a new to design class from the base class Cmeasureparameter defining the needed functions virtually By making a more sophisticated design it is possible to make a set of utility functions that can be used to perform the desired operatio
91. lane gt for blue The system knows which dots are red green or blue Use normalise video search utilities Video address is set by the system to the normal video position Video address is at the normal video position Use Video control utilities Page 78 Description The user is able to select the colours to set the video position and to choose a video picture more info in manual Exceptions Result A screen colour is selected A screen picture is selected and the line s or spot s are moved to specific positions Name Use SBIP functions Prerequisites Description The user is able to reboot the SBIP Initialise the SBIP Put the SBIP to live video Set Gain and threshold for specific fields manually Process auto gain Process auto threshold Mask in Plane make a one bit plane of a one coloured image Pixel Calibration Show planes Histogram Exceptions Result Name Use measurement utilities Prerequisites Description The user is able to execute some elementary measuring functions for analysing development purposes Convergence Gravity XY Gravity X GravityY Brightness Bright colour Linewidth Exceptions Result Use case tube Name Project image Prerequisites Camera is in front of the field to measure Description The system is able to project an image on the screen The system is able to measure the image or to collect the image in a video memory for further analysis Exceptions
92. larification dialogs Less transfer problems to other languages Assumes typing skill Powerful Error prone Flexible Enhancements are invisible Fast Expensive to implement Uses moderate screen real estate Natural language is most appropriate for USER PSYCHOLOGY Negative attitude Low motivation KNOWLEDGE AND EXPERIENCE High typing skill Low system experience High task experience Low application experience High frequency of use of other systems Low computer literacy JOB AND TASK CHARACTERISTICS Low frequency of use Little or no training Discretionary use High turnover rate Low task importance Low task structure Touch screen based equipment Page 66 D Evaluation form Menu Fill in forms Question and Command Function keys Direct 27 Natural 2 x guage 12 RE 5 i lation languape Attitude Motivation Typing skill System experience Task experience Application perience Use of other systems Computer literacy Primary training Little ohh e II Hoos oS Little offe ee ae eee eco System use pues EI NET UM Discreti Turnover rate High Low EM Dscce me Disa eee High 000 Other s IS Task importance n Low 999 low 000 Task structures Maint Process 4 Touch screen based equipment Page 67 E Touch Screen technologies Force Vector In force vector technology a CRT or flat panel di
93. lay with a metallic coating so there is a big surface and the Zt surface transfer impedance 3 is also rather big The noise voltage induced by the magnetic field is accordingly high In the surface acoustic wave touch screen technology the only place where noise voltage can be induced is in the transmitters receivers and the wiring to the PCB If the Transmitters and the PCB are well grounded the influences of an interference signal is smaller in surface acoustic wave than in both five wire resistive and capacitive touch screen technologies Touch screen based equipment Page 17 Another important selection criteria is the price A touch screen with surface acoustic wave costs 2182 DM a 1 12 fl 2 443 84 The same monitor equipped with a capacitive touch screen system costs fl 3 972 00 So not only in technology the surface acoustic is preferred but also in price competition the surface acoustic is the preferred one 6 2 EMC ESD test The selected touch screen is tested in a simulated environment After adding one extra protection to the communication cable the touch screen did not fail if flash over tests were conducted The complete test report is available at Philips EED section MMM Touch screen based equipment Page 18 7 Demonstrator design Using a set of defined techniques and sign conventions to convert wishes to a software product is called software engineering S E A S E method always uses sequential predefined st
94. le for the selection and positioning of the camera or cameras In this object a member function takes as arguments the desired location on the measuring object Now if the apparatus is automatic that function will take actions to select or move the right camera When this is not available this function will send a message to the user to move the camera to the right spot This object is the only one responsible for the selection so it is aware of the current position of the camera 7 5 2 11 CVideoSystem The Object of type CVideoSystem is responsible for the picture to be shown on the measured object For flexibility in the future the picture properties on the tube are separated for every field By doing so it is possible to adjust all properties of the picture in a single field Currently only the move property is changed separately The others Kind of picture and colour are all updated at once This is due to the fact that the video card that produces the image is unable to show different pictures with different colours on the screen The changing of the picture properties is therefore handled in a single object with type CVideoSystem In that object an array variable is defined of length 25 This array contains pointers to the objects that represent the picture properties in one field The system is able to alter these properties by the object of type CVideoSystem The user must alter these properties too To avoid difficult programming the dialogs t
95. lective formal Mandatory formal Turnover rate Page 52 System use Mandatory Discretionary Other tools High Telephone Moderate Calculator Low Adding machine Other Task structure High Moderate Low Handedness Gender Right Female Left Male Ambidextrous Touch screen based equipment Page 53 C Dialog Styles A dialog style is an overall style of interaction At least seven distinct dialog styles exist and are addressed here e Menus e Fill in forms e Question and answers e Command languages e Function keys e Direct manipulation e Natural language Most MMI s employ more than one dialog style and each of these may not appear in its most pure form In the following sections however a distinction is drawn to represent research findings design principles and guidelines Menus In menu driven user interfaces the primary form of interaction is a sequence in which the user is repetitively presented with sets of choices and asked to select one or more from among them Choices may be presented as words in a list or as collections of icons Advantages and Disadvantages of menu dialog styles Menus have particular advantages and disadvantages to other dialog styles These are included in the following table Advantages Disadvantages Self explanatory Inefficient Easy to learn Inflexible Few keystrokes Impractical for numerous choices Easy error handling Take up screen real estate Enhancements are
96. lter Alters the contents of the file and saves it with a new tube name The system connects the new tube file to the tube If a new tube file has the same name as an old one an error message is displayed and the user is asked to enter a new tube file The parameters in the tube file are connected to the tube Edit program parameters User is logged in and is authorised to edit parameters The user is able to change ID number Fields for system Number of measurements Meas delay for 2A mon SBIP flag Configuration and Measurement For more explanation see topics in manual Result Name Prerequisites Description Exceptions Result field Name Prerequisites Description Exceptions Result Name Prerequisites Description Exceptions Result Name Prerequisites Description in the function search dot Exceptions Result system Name Prerequisites Description Exceptions Result are known by the system Name Prerequisites Description Page 77 System has new program parameters Edit tube parameters User is logged in and is authorised to edit parameters The user is able to change parameters which do not belong to a specific field Tube name Tube type Delta Delta gain Video Sens Factor X Video Sens Factor Y Delta Centre X Delta Centre Y Video Shift Step Size X Nr Of Video Shift Steps Pulse Width 1 2 Horz Field Of View 1 2 Vert Field Of View Line frequency
97. magnetic immunity or susceptibility EMS The EMI depends on the device that is causing the electromagnetic interference If one part of an equipment is measured and meets the EMI requirements it will not violate those requirements if that part is used in a complete set up The EMS depends on the environment Other devices may transmit electromagnetic waves but it is also possible that interference is caused by a process in the working environment The MMM equipment is situated on the production floor and the process involves high voltage switching and demagnetising sequences Sometimes a jump over occurs and an electromagnetic pulse is generated these pulses are called System Generated ElectroMagnetic Pulse SGEMP This is a non repetitive interference and in this category one can also place the Electro Static Discharge ESD ESD events are caused by familiar triboelectric processes Greason 1994 in which personnel accumulate static charge by friction between their clothing and other objects in the workspace When a charged person touches a metal object such as a piece of MMM equipment the charge flows off in a spark There are several other ways by which a person may by charged e g by switching on and off CRT s Franey 1995 To show their conformity to the EMC requirements prescribed by law all electrical devices have to be marked with the CE conformity mark A manufacturer supplier has several routes to obtain the CE marking The first one
98. ms are complex of nature chose a breadth of up to 25 if an organisation can be imposed to effectively group items and the items are not complex Menu orientation is also an aspect of menu structures One study Backs 1987 investigated the horizontal and vertical orientation Vertical menus performed faster but for better layout and compatibility with existing structures no generalisation can be made A pie structured menu performed better for 8 items Callahan 1988 However some important qualifications must be considered Larger amounts of items become impractical It is hard to imagine how pie menus could effectively be presented in deeper menu structures Finally pie menus require a graphic user interface It sometimes happens that some menu items may not be selected To maintain the learning patterns navigating through the menu it is better to grey those items than to delete them Somberg 1987 Franik amp Kane 1987 although for experienced uses deletion performs a little faster Menu choice ordering The optimal ordering in user search and selection time is studied by several people McDonald amp Molander 1988 Hollands amp Merikle 1987 MacGregor amp Lee 1986 Given that menu systems are generally chosen to support novice casual users these studies suggest that ordering by logical semantic groups is a good organisation principle to employ If the users however are high frequency more experienced users alphabetic ordering might
99. n X measur i show li cc n I L t i j i i 1 i i l i Execute measure in multi single mode 2 i A UU TET user interface m Measuring m Measure m Tube i m Field Equipment 1 l measure i gt gt 1 set measure what action measur l l l l l l l l l I l l l i l l Execute measure in multi selection mode Touch screen based equipment Page 95 User interface m Measuring m User CUser CMeasuring Equipment i l i Login Do Dialog i a Enter password l Set preferences zt D I l i i l l Login MM 4 m User interface m Measuring XXXX equipmet l select al LJ I I Edit data V selec User interface m Measuring Object m Tube m Field m Remote Equipment 1 I i t select i l select start selected L 1 i l l execute l i l l l Measurement utils multi selection mode i E i ES l 1 l 1 Touch screen based equipment Page 96 i EET uipment User interface m_SBip m measuring _ m Tube m ANIM m Remote fe oe i 1 select function i ship select selec I i i what acti
100. n include pointing devices such as a mouse trackball or touch screen and often make heavy use of graphics in displaying objects and actions Touch screen based equipment Page 64 Advantages and Disadvantages of direct manipulation Direct manipulation dialog styles have particular advantages and disadvantages to other dialog styles These are included in the following table Advantages Disadvantages Easy to learn and to remember Not self explanatory What you see is what you get Can be inefficient WYSIWYG Flexible Difficult to design recognisable icons Provides context and instant visual Icons take more screen real estate than feedback words Exploits human use of visual spatial cues Less error prone Direct Manipulation is most appropriate for USER PSYCHOLOGY Negative attitude Low motivation KNOWLEDGE AND EXPERIENCE Low typing skill Moderate system experience Moderate to high task experience Moderate application experience High frequency of use of other systems Low computer literacy JOB AND TASK CHARACTERISTICS Low frequency of use Moderate training Discretionary use High turnover rate Low task importance Low task structure Care must be taken in the design of a direct manipulation interface in order to realise the potential advantages Several studies have tried to directly compare direct manipulation interfaces to non direct manipulation interfaces One study Robert amp Moran 1982 seems to suggest
101. n ioo rN em Vp QU E IR RR UE caadiauseonades 15 6 Touch Screen technology aiite eva ginseen ks NUS AS Ia SENSA VA UNES Aka ux opas aR Po URN ed 16 6 1 Choice of Touch Sereen aua eed err tr pe REFIERE anb va Tu eU rH EERXN VERE V RO nd Nea EC ursa ubi cede app MaShGE 16 6 2 EEMIC ESD fest EI 17 7 Demonstrator ESI BN D 18 7 1 1 rg per 18 7 1 2 The OMT Object Mod lsiriounnnuiinnnn onran a 18 7 1 3 The OMT Dynamic Model sseennseenersensosenernsosresnssonssssourersrsrsrrasanssreeronsenreeses 18 7 1 4 The OMT Functional Model uiscera nsctseestacesves dent eed oa erp feas sages 19 7 1 5 Deyelopment SCH VIL S MR TO 19 7 1 6 Conc ptualisaion sse or Cd dU EC iaeiae aE aa lu Ee uu Eae vae 19 TAT System boundary annesini aa AE A T A a oau et aane 19 7 1 8 Va Co oI EA EEEE ES E istos cite E T 20 7 1 9 VISESCdSe Lido a a i almi o anes a eine 20 TETO Context diagranm sissid e PES Geb Qoae teh Sin e iaaa dire estou USE 22 7 1 11 Problem statement oce eec eden paste ded ur Redon Fase HR at nei 22 7 2 Domain amallysis sscssssssessscesessoserssees AE seca Sasvavedasscecees we 23 7 2 1 Candidate class s n ni oss asgo ea reeeo ta E aeai aa el A eaae ioi Doctus xe 23 7 2 2 M odeldict nary sariren aen araea eenaa aiaa aa usd epa Roques 24 7 2 3 PRSSOC TAH ONS s oa ciel e Aea E eflec cS aneni ais aE Ee E a a Ea 25 7 2 4 Attributes and Operations essesssesssesreesessrererstasesnsertssetasrsrorosen
102. ncerning target size verses accuracy trade off Touch screen based equipment Page 14 One study Leahy 1990 developed a graph to show expected accuracy as contour lines with touch sensitive region dimensions as axes Tree visual target sizes were used 7 5 mm2 12 2 mm2 and 20 mm2 As a worst case reading they stated that a target size of 36 mm would result in an accuracy between 50 and 99 Most studies indicate that keys may be closely spaced provided the individual keys are adequately sized Beaton et al 1995 They also state that the corresponding touch sensitive area extending beyond the visible target was found to be optimally 2 7 mm above and below target and 1 7 mm to the right and left of the target Others Leahy 1990 Beringer 1985 noticed a general bias to touch low and to the right of targets was found If clear blunder errors could be eliminated a target size of 18 mm in x and 15mm in y would be appropriate Bering 1989 For small target selection purposes it is even possible to use 5 mm in x and 3 mm in y Sears 1991 but one has to use an advanced selection technique taking longer to select a target Looking at all studies a target size of 20 x 20 mm is a good trade off between real estate and performance 4 3 Target Layout Two studies investigate the design of the targets The Valk 1985 study revealed that users of touch screen buttons consequently aimed at the labels or pictures on the button The labels and pic
103. nology is really factory resistant in not proven yet This test should be the first to execute because if it fails it will be very hard to find a technology that can resist the harsh environment A lot of people indicate that there is a lot of dust in the factory and that might effect the capabilities of the touch screen According to the studies the dust should not be a problem for the chosen touch screen The demonstrator is developed with Visual C 1 51 This is very out dated New development platforms are tested by section MMM The newer development platforms offer more functionality so if the MMI for MMM equipment is further developed a study about the newly offered functionality can be used for touch screen applications Touch screen based equipment Page 44 references Albert A E 1982 The effect of graphic input devices on performance in a cursor positioning task Proceedings of the Human Factors Society 26th Annual Meeting 1982 pp 54 58 Backs R W amp Walrath L C 1987 Comparison of Horizontal and vertical Menu Formats Proceedings of the Human Factors Society 3 1 Annual Meeting 1987 pp 715 17 Beringer D B and Bowman M J 1989 operator behavioral biases using high resolution touch input devices Proceedings of the Human Factors Society 33 Annual Meeting 1989 pp 320 22 Beringer D B and Peterson J G 1985 Underlying behavioral parameters of the operation of touch input devices baises models and f
104. ns on random variables By implementing the classes in the described way a truly OO concept is used for viewing and editing the measurements values and parameters The object is the only one responsible for displaying the contents The objects that request such a view only have to send the appropriate pointer array of the base class for the parameters See appendix P for the user interface layout of the edit view dialogs 7 5 2 6 CVideoSearch CColourrecog CConvergence CExcentricity CLinewidth The objects that contain the measurement methods are very simple for the implementation of the demonstrator The demonstrator does not have to execute the real measurement code Only a call with the appropriate attributes to the remote control object is sufficient to retrieve the desired measurement values For the remote contro line it is not possible to place more than one request So to reduce overhead and excessive programming the calling measurement object waits until the measurements are received The retrieved values thus may be passed by the attributes in the call In the remote control link it is not possible to send the measurements specific parameters stored in the measurement parameters For the time being there is no access necessary to those objects But if the measurement codes are implemented in these objects complete working equipment calls to the parameter classes are necessary A measurement object always knows the type of meas
105. o alter position kind of picture and Colour are controlled by the same object of type CVideoSystem This is the object that communicates with the remote control object to change the properties of the image on the remote controlled COPER T1 The dialogs can be seen in appendix P 7 5 2 12 CPicture The CPicture class is used to enter the 25 field dependent picture properties This class only contains the three properties colour type location of the picture By adding pointers to the Field class and to the Video class the pointers can be referenced during creation to point to the corresponding field and to the object controlling the dialogs This class contains information about preferences during measurement so in this class the load and store member functions are used Touch screen based equipment Page 41 7 5 2 13 CSuppliesControl The class CSuppliesControl is completely empty in the demonstrator design 7 5 2 14 CUser For touch screen based user interfaces left handed persons need a left oriented user interface and right handed need a right oriented user interface To optimise the performance of the user it is important that the user has a personal calibrated touch interpretation For MMM there is a difference in normal operators and maintenance operators As mentioned in chapter 2 this lies in the ability to change process parameters All these properties are handled in the object of type CUser The dialog is shown in appendix P
106. oderate Moderate Primary training high System use High Turnover rate Moderate Low Other systems Task importance EE Task structures High Han o Low Table 2 User characteristics Touch screen based equipment Page 11 3 Input and Output Devices At the current time various input and output devices are available to designate data locations movements and signalling including keyboards mice light pens trackballs joysticks screen buttons lights gages and touch screen devices Voice systems are just beginning to gain popularity as input device but for industrial environments voice systems are not applied 3 1 input device In the previous section four dialog styles are chosen to be appropriate for MMM equipment In table 3 the suitability of input devices is given for the desired dialog style Helander 1988 Debora 1988 keyboard mouse light pen trackball joystick touch screen Table 3 Input devices for MMM equipment appropriate dialog styles for limited alphanumeric data entry a touch screen may prove to be useful Plaisant 1992 There are three possible input devices that might be appropriate for MMM equipment e keyboard e mouse e touch screen A mouse in an industrial environment is by far not an optimum solution Keyboards tend to get stuck in polluted areas Another mayor drawback as can be seen in the section ESD EMC test is the fact that keyboards are a pre
107. on perfor l f i ask function depended data 1 El enter requested gt If no data is S Rese sear ee MENEIEEN execute selected sho gt E 5 SBIP functions single selection mode i i depends on requested util User interface m Measuring Object m Tube m Field m Remote Equipment select i start selected util execute function Measurement utils single selection mode l 1 s 1 Touch screen based equipment f User interface i m_SBip Page 97 selec j q m measuring m_Tube if m Field m Remote Equipment I I be selec select function 4 SBIP functions multi selection mode what action l i 1 i l i i i execute selected l i show measures if i i i i lu i Select tube User interface m Measuring m VisionSystem m Picture m Tube m Measurement CMeasuring Equipment CVisionSystem CPicture CTube CMethode Select tube file 1 open file i l open file l gt open file S E a E i E open file sp 3t l l I l I l l l l I I l l I l t l I i I I i I I l l l Touch screen based
108. ons of association aggregation and generalisation are explained Looking at every pair of classes on the candidate class list all associations in the demonstrator model can be found In table 5 the associations are drawn F A EEENEREERS Sbip Camera Video search X X Colour recognition X X Convergence I qe ree uEE X ccentricit ERG Re ERE X X Linewidth Lie J4 n gn x X Measuring equipment X X X X X X X X X X X aa ae X E JEEP TES uen X TEE HERLAN eT E B JE i aa a es ee ube supplies ee std ee Tube Iam pq i s sq ae ee e ee EE ield ee Ed E 3 20 Jr Ee eee Hawe LM H9 H X User a ee SR TA EET E Loggin LE ye a a ee Remote control E hm hi Table 5 Associations note has an association with X To obtain a better design the measurements are separated in two parts First the measurement itself how it is executed secondly the parameters used to execute the measurements and the measured values The first is associated to the measuring equipment and the second is associated with the fields of the system Table 6 shows the four extra classes added to the list of candidate classes Colour Recognition parameters Eccentricity parameters Convergence parameters LineWidth parameters Table 6 Extra classes After adding the new classes and associations in Rational Rose the input screen loo
109. other users and other software systems Considering the two main subsystems the computer and the human a few initial observations can be made The human is flexible and adaptable Most importantly the human can learn how to operate in new environments By contrast the computer system is not flexible or adaptable at least most commercial systems currently are not Inputs must be made in particular format outputs are predefined Thus for a given computer system the human can learn and adapt while the computer cannot On the other hand people build computers and computers can be redesigned In the past the design of the MMI was heavily biased to accommodate the weaknesses of the computer system More recently due to advances in the technology the bias toward accommodating the deficiencies of computers in the MMI is slowly shifting Initially the idea has been to shift the responsibility for the success of the MMI interaction to the computer designer Or better yet to the MMI specialist That is without actually making computers more adaptable or more flexible it is nevertheless possible to make them more compatible with the way people work and communicate The touch screen feasibility study is part of a research project to use new technology so few restrictions are made on which MMI to use The method used in a book about user interfaces Mayhew 1987 is adopted to guide the choice for a possible MMI 2 2 The User Profile Perhaps the f
110. pp begin module additionallncludes preserve no end module additionalIncludes begin module includes preserve yes end module includes CTube include CTube h begin module additionalDeclarations preserve yes end module additionalDeclarations Class CTube CTube CTube begin CTube CTube hasinit preserve no end CTube CTube hasinit begin CTube CTube initialization preserve yes end CTube CTube initialization begin CTube CTube body preserve yes end CTube CTube body CTube CTube const CTube amp right begin CTube CTube copy hasinit preserve no end CTube CTubetcopy hasinit begin CTube CTubetcopy initialization preserve yes end CTube CTubetcopy initialization begin CTube CTube copy body preserve yes end CTube CTube copy body CTube CTube Page 90 begin CTube CTube body preserve yes end CTube CTube body int CTube operator const CTube amp right const begin CTube operator body preserve yes end CTube operator body int CTube operator const CTube amp right const begin CTube operator body preserve yes end CTube operator body Other Operations implementation void CTube m_Save begin CTube m Save 884685593 body preserve yes end CTube m Save 884685593 body
111. puter must be protected at any time because a flash over can kill the electronics PC side For the design of the demonstrator Rational Rose OMT is used as object oriented design tool and Visual C is used as developing environment Basically this meant for the author learning OO by scratch using an unknown design tool and programming in a new environment despite these problems the results were very good The visual aspects of C contribute to a nice graphical user interface without concerning too much about the underlying code Rational Rose is a tool that closely matches with the literature about OMT design The code generation as used was not optimal But with the right settings a nice code body can be produced The design of the demonstrator in combination with the design of LabMag Lamber 1996 may be used as an indication on how MMM equipment can be designed in the future Some of the classes designed for the demonstrator can easily be used by other designs The latter is especially true for the classes in the user interface subsystem like virtual Keyboards and input dialogs Touch screen based equipment Page 43 8 2 Recommendations Due to time pressure an evaluation with end users is not available With the current demonstrator this evaluation can take place in the future The people which played with the demonstrator are very enthusiastic about its capability A real field test is not executed Whether the touch screen tech
112. r Disadvantages e Response time Due to the complex calculations that are required by the force vector technology the system s response time is relatively slow e Calibration The calibration of a force vector touch system is complex and time consuming Periodic adjustment will be required e Sealability Force vector touch systems cannot be sealed to NEMA 4 requirements and are susceptible to direct exposure to water dirt and corrosives e Environmental resistance Force vector touch systems are sensitive to shock and vibration Strain Gage Advantages e Resolution Strain gage touch systems can deliver a resolution of up to 4096 x 4096 touch points Touch screen based equipment Page 74 e Z axis In addition to the typical x and y coordinates stain gage technology can provide a z axis component which is determined by the amount of pressure applied to the display The harder the user presses the more energy is absorbed by the force sensing transducers e Stylus Requirements Strain gage touch systems have no stylus limitations Strain Gage Disadvantages e Transparency Strain gage touch systems typically have a transparency of 92 This can affect image quality particularly for high resolution video mode e Calibration The calibration of a strain gage touch system is complex and time consuming Periodic adjustment may also be required e Sealability Strain gage touch systems cannot be sealed to NEMA 4 requirements and
113. r those classes can also be found on the CD Touch screen based equipment Page 42 8 Conclusions and Recommendations 8 1 Conclusions The study reveals that users of MMM equipment are very divers There is a great difference in education level and in the turnover rate of the users Because of those properties a menu structured dialog in combination with direct manipulation is a very well suited dialog for users of MMM equipment This dialog performs well with the aid of a touch screen IO The developers of MMM can gain from this result because currently an update in the button panels means a new design With touch screens only a new graphical user interface layout is necessary When designing a touch screen based application it is important to follow the constraints that are put on the design by the touch screen A user friendly system is necessary By frustrating users because they can not touch a specific point easily an antipathy against touch screen technology may be created A new user has to take some time to learn the interaction between user and screen approx 20 min The new user must be positioned in front of the touch screen so only little parallax error occurs After such a session the user is able to adapt the way he touches if the angle of sight is not orthogonal to the screen The technology chosen Surface acoustic wave is shown to be stable in a simulated test environment The communication between touch screen and com
114. ra CLineWidthPara Figure 7 Future implementation Touch screen based equipment Page 37 7 5 2 5 CColourRecogpara CConvPara CEccePara CLineWidthPara CDumpVar These classes contain the measurement parameters and measured values These values differ per class but the total structure of the class shows a lot of comparisons in structure They all have a function that handles the user dialog for changing and viewing the data They have load and save functions As mentioned in the previous chapter the value to be displayed in the tube dialog after a measure can be received by calling one function which automatically returns the desired value The value to return if more possibilities are possible can be set in the user dialog of this class The load and save functions are the same as in the other classes The Microsoft Foundation Classes MSFC also contains a base function to store the data of a class The only problem is the fact that a generated file is unreadable for a human user By designing a new load and save function this problem is solved Although it is a frequently used function the implementation differs per class because of the difference in data The dialog handling of these classes is much more interesting As mentioned in section Array NULLINULL SXXX NULLI XY XINULL SY Y Y 7 5 2 3 the function that handles the request to show the dialog is accompanied by an attribute that contains pointers to classes o
115. roblem of interpretation would merely disappear if a standard of interpretation could be adopted When interpreting results from previous studies it is important to look at the touch screen technology used resolution parallax Selection strategy Basically there are three used strategies Potter amp Weldon 1988 which allow a user to select one of a set of displayed predefined areas The simplest strategy land on uses the initial touching of the touch screen for selection Sears amp Shneiderman 1991 If a selectable item is under the initial touch it is selected otherwise nothing is selected All further contact with the touch screen is ignored until the finger is removed The second strategy first contact was designed to work basically the same as land on but take advantage of the continuous stream of touch data provided by the touch screen controller Users make selection by dragging their fingers to the desired item In this strategy it is not what position the user lands on that becomes selected rather is is whatever selectable item the user first contacts If the user makes first contact with some undesired item before reaching the desired item the undesired item will be selected The third strategy take off was designed to utilise the contours stream of touch data and give more user feedback Whenever users make contact with the touch screen a cursor appears slightly above their finger so that the specific position of selection is
116. roller establishes first an x axis and then a y axis gradient across the substrate The controller uses the membrane as a probe at all the times The two voltages that the probe senses reflect the points x and y coordinates Touch screen based equipment Touch System Comparison Page 71 Specific data regarding the impact of environmental factors on touch systems may be found in the following technology comparison charts Table 1 Touch System Comparison 1 Force Vector Strain Gage Guided Acoustic Wave Surface Acoustic Wave Capacitive Infrared four wire resistive five wire resistive Stylus type No stylus limitation No stylus limitation Requires soft energy absorbing stylus Requires soft energy absorbing stylus Requires conductive stylus No stylus limitation Minimum stylus diameter 5 16 No stylus limitation No stylus limitation Resolution and Z axis 40 points inch plus 256 z axis levels 4096x4096 physical plus z axis 21 904 points square inch plus 256 z axis levels 100 000 points square inch plus 256 z axis levels 1024x1024 physical 64 points square inch No Z axis 100 000 points square inch 100 000 points square inch Denied by the competition Transparency 100 92 90 95 90 95 85 100 55 75 55 75 Sensor Drift and Calibration Subject to drift Requires repetive calibration Subject to drif
117. rs Fill in form error handling Inevitably users will make errors entering data To make those errors as small as possible the following guidelines can be drawn Touch screen based equipment Page 59 Allow character edits in fields Place the cursor in the error field after error detection Provide semantic and syntactic information in error messages Question and answer As in a menu system the user is posed a single question Like in fill in form interface however the user is expected to type in an answer Advantages and Disadvantages of Question and answer dialog style Question and answer dialog styles have particular advantages and disadvantages to other dialog styles These are included in the following table Advantages Disadvantages Self explanatory Inefficient Simple and nonintimidating Assumes typing skill Many possible input values Error prone Accommodates hierarchical task structure Inflexible Complete and clear prompting No forward context Enhancements are visible Questions and answers are most appropriate for USER PSYCHOLOGY Negative attitude Low motivation KNOWLEDGE AND EXPERIENCE Moderate to high typing skill Little to moderate system experience Low task experience Low application experience Moderate to frequent use of other systems Low computer literacy JOB AND TASK CHARACTERISTICS Low frequency of use Little or no training Discretionary use High turnover rate Low task importance Hi
118. s on a CRT TVT or CMT The CRT can be positioned on the inside or outside of the measuring system When generalising the system the question rises is ita CRT that is to be measured By placing the CRT measured object outside the system borders it is easier to adapt the system for future use The measurements are properties of the images on the CRT As mentioned earlier these images are taken at predefined places on the screen A Coper can have several automated forms It may be equipped with 12 cameras for a fully automated system or it may be equipped with one camera This camera may be positioned manually or it is integrated in a semi automatic positioning system In the last case an XY frame is used to position the Touch screen based equipment Page 20 camera in front of the desired place on the CRT For the measuring system it is not important how an XY frame works For this reason the XY frame is placed outside the system boundary The complexity of MMM equipment makes it almost impossible to completely redesign a MMM equipment and give full attention to all parts of the design The main part of this study lies in the user interface and how to adapt the current software to be used with the new user interfaces The system is designed with full functionality but at the place where specific measuring code should to be implemented a call to an existing COPER T1 is executed to retrieve the desired measurements For simplicity the ex
119. ser interface must comply with the design rules found in chapter 4 Touch screen based equipment Page 34 7 5 1 Code generation For the demonstrator this is the last stage in which the development tool is used The design at this point is converted to code by the code generator of RATIONAL ROSE The default property sets are used to generate the code for the demonstrator In appendix J the H file of the generated code for the tube class is shown And in appendix K the C file is shown In those files one can see that there are a lot of inline statements This is one of the switches in the property set Inline code is faster in execution because every declaration of the inline function is replaced by the code itself 7 5 2 Classes 7 5 2 1 CMeasuringEquipment The class CMeasuringEquipment is the main class of the demonstrator It contains the associations to the measurements the used systems the measured object tube the user and especially for the demonstrator the remote control Further it manages several user interactions In the windows environment the common way to select items is by the menu system But for a user interface with touch screen input the screen dimensions needed to design a menu system would be enormous Another argument not to use the standard menu structure is the performance gain when using symbols and labels for selection purposes See Chapter 4 for more information about touch screen constraints on user interf
120. splay is placed upon a platform The platform consists of a top plate that rests upon three dimensional springs moving in all directions relative to the base of the pedestal Internal sensors measure the distance between the top plate and the base Each sensor consists of two parallel plates that form a capacitor one attached to the top plate and one attached to the base When the display is touched the force causes a slight change in distance between the top plate and the base This causes the distance between each set of sensor plates to change slightly changing the capacitance across the plates and determining the amount the top plate moved in the x y and z directions The platform then determines the level of force exerted to cause this movement and the location where the force was applied Strain Gage The strain gage system consists of a CRT mounting shield a clear glass overlay and four strain gage transducers hooked to the clear glass overlay Touching the glass overlay causes the strain gage transducers at the four corners to measure compare and calculate the touch location Guided Acoustic Wave Guided acoustic wave is based on transmitting acoustic waves through a glass overlay placed over the display surface A transducer High frequency signal mounted on the edge of the glass emits an drives transducer acoustic wave The wave packet travels along _ e the reflector array is redirected across the zm eae i l overlay to the reflec
121. t Requires complex repetive calibration Not subject to drift Not subject to drift Not subject to drift Not subject to drift Subject to drift Requires repetive calibration Not subject to drift Reliability 80 000 hours MTBF Not published by manufacture Sensor Unlimited Controller gt 180 000 hours MTBF Sensor 50 000 000 touches per point Controller gt 65 000 hours MTBF Sensor 20 000 000 touches per point Controller gt 572 600 hours MTBF gt 138 000 hours MTBF Sensor 2 000 000 touches per point Controller gt 180 000 Sensor 35 000 000 touches per point Controller gt 65 500 gt 107 500 hours MTBF Activation Parallax and Response Time Tactile activation no parallax 200ms Tactile activation parallax 100ms Tactile activation no parallax 18 50 ms Tactile activation no parallax 18 22 ms Tactile activation 8 15 ms Proximity activation parallax 18 40 ms Tactile activation 7 12 ms Tactile activation Touch screen based equipment Table 2 Touch System Comparison 2 Page 72 Scratch Wear NEMA Ratings Dust and Dirt Chemical Vibration and Resistance Moisture Resistance Resitance Shock Resistance Resitance Force Vector Extremely NEMA 12 Not affected by Not effected by Sensitive to resistant dust and dirt general purpose vibration and cleaning solutions shock Strain Gage Extremely NEMA 12 Not affected by Not effected by Sensitive to resis
122. t The dialogs used are designed using the style guide made in the second part of the thesis Touch screen based equipment Page 3 1 Introduction Main subject of this thesis develop a touch screen user interface for control of colour CRT production equipment In this introduction chapter we will describe the CRT manufacturing process pose a problem statement specify the objectives and explain the organisation of this report 1 1 The cathode ray tube The colour cathode ray tube CRT found in television computer and video monitors utilise a shadow mask or aperture grill a fraction of an inch 1 2 typical behind the phosphor screen to direct the electron beams for the red green and blue video signals to the proper phosphor dots Since the electron beams for the R G and B phosphors originate from slightly different positions Individual electron guns for each and thus arrive at slightly different angles only the proper phosphors are excited when the purity is properly adjusted and the necessary magnetic field free region is maintained inside the CRT Note that purity determines that the correct video signal excites the proper colour while convergence determines the geometric alignment of the 3 colours Both are affected by magnetic fields Bad purity results in incorrect colours Bad convergence results in colour fringing at edges of characters or graphics The shadow mask consists of a thin steel or InVar a ferrous alloy sheet w
123. t Page 30 7 4 Application analysis Application analysis use modelling techniques similar to those used in domain analysis but now it includes those classes that represent the interface between the application system and the user The demonstrator is designed with the use of the user manual of the current COPER T1 This means that classes like the Video Processor are already in the design Normally they are added in the application analysis A way to begin application analysis is to take each use case in turn and define the details of the interaction between the actor and the system The recommended way by OMT is to start with a scenario that is a specific sequence of events being exchanged There are two types of Scenario s in OMT An external scenario is a scenario where the events between the actors and the system are modeled An internal scenario also shows the events between the objects in the system 7 4 1 Scenario In OMT the external scenarios are presented by a numbered list of events The following list represents the external scenarios for the demonstrator 1 Login 1 User enters name 2 User enters password 3 System looks for valid combination and starts the measuring system with the privileges of the current user 2 Select tube file 1 User selects file input screen 2 System shows file input name 3 User selects tube file 4 System loads the selected tube file 3 Execute measurement in single mode Scenario m
124. t devices for interaction with computer generated displays Ergonomics 26 no 11 pp 1033 1053 Wijdeven L 1996 Labmag Hogeschool Eindhoven 1996 Afstudeerverslag Wixon D 1983 building a user defined interface CHI 83 Proceedings December 1983 pp 24 27 ACM Touch screen based equipment A User Profile Checklist Psychological Characteristics Cognitive style Verbal analytic Spatial intuitive Attitude Positive Neutral Negative Knowledge and Experience Reading level No reading capabilities Less than fifth grade Fifth to twelfth grade Above twelfth grade System experience Expert Moderate Novice Native language English German French Spanish Taiwanese Chinese Typing skill Non Low Medium High Task experience Novice in field Moderate Expert in field Use of other systems Little or none Frequent Page 51 Motivation High Moderate Low Education Elementary school High school degree College degree Advanced degree Application experience No similar systems One similar system Some similar system Computer literacy High Moderate Low Non Touch screen based equipment B Job and Task Characteristics Frequency of use Low Medium High Job categories Operator Production chief Process specialist Maintenance man Task importance High Low Physical Characteristics Color blind Yes No Primary training None Manual only E
125. t int get m DeltaCentreX const void set m DeltaCentreX const int value fiti Attribute m DeltaCentreY The tube parameter Delta centre Y is the permissible searching tolerance for the horizontal video line position to the centre of the camera If this parameter is adjusted too critical the video line cannot be positioned An excessive Delta centre y value may Page 85 cause part of the video dot to extend outside the camera Field of view The Delta centre Y is determined by Auto calibration as indicated by c see Tube parameters using the formula Il Delta centre Y Video sens factor Y 3 4 const int get m DeltaCentreY const void set m DeltaCentreY const int value fit Attribute m_VideoShiftStepSizeX The tube parameter Video shift step size X is the step size between two video steps during Micro stepping This parameter is determined by Auto calibration as indicated by c see Tube parameters based on the Number of video shift steps as follows i The Video dot is positioned in front of the camera centre and the video data X address is stored 2 The video position is set to the data X value 1 The centre of gravity is measured 3 The video position is stepwise changed with n 5 video shifts so 5 10 15 until the delta Gravity X is exceeding two times the phosphor pitch 4 The Video shift step si
126. t to drift Periodic calibration is required The 5 wire resistive overlay touch system is not subject to sensor drift e Environmental Resistance The 4 wire resistive overlay touch system will degrade accuracy over time by constant flexing The life time of the 4 wire is much shorter than the 5 wire technology The exposed polyester top layer is susceptible to cuts scratches and abrasions Selection of touch technologies Looking at the advantages and disadvantages of the various touch technologies and considering the application environment not the EMC ESD constrains there are three technologies that suits the MMM equipment best Surface Acoustic Wave Capacitive and Five wire resistive The surface acoustic wave has a long life time gt 50 000 000 touches per point is difficult to destroy has an image clarity that is good for an overlay and drift free operation is guaranteed Like the surface acoustic wave technology the capacitive overlay is difficult to destroy and the life time is long gt 20 000 000 touches per point The Image clarity is less but still good for an overlay and the response time is very fast 8 15 ms Finally the five wire resistive with its ability to react on any kind of stylus and its response time of 7 21 ms The Transparency is low 55 75 There is no sensor drift All three technologies deliver very high resolutions and proved their durability in industrial environments As may be noticed the Infrared t
127. tant dust and dirt general purpose vibration and cleaning solutions shock Guided Acoustic Difficult to NEMA 12 Not affected by Not effected by Wave scratch glass NEMA 4 dust and dirt general purpose overlay is cleaning solutions breakable Surface Acoustic Difficult to NEMA 12 Will operate with Not effected by Tolerant of Wave scratch glass moderate dust and general purpose vibration glass overlay is dirt excessive cleaning solutions overlay breakable accumulation susceptible to may affect shock performance Capacitive Difficult to NEMA 12 Will operate with Tolerant of scratch NEMA 4 moderate dust and vibration thick conductive layer dirt excessive glass overlay is subject to wear accumulation moderately glass overlay is may affect susceptible to breakable performance shock Infrared Extremely NEMA 12 Will operate with Not effected by Sensitive to resistant NEMA 4 moderate dust and general purpose vibration and dirt excessive cleaning solutions shock accumulation Chemicals that may affect affect performance polycarbonates should not be used four wire Sensitive to NEMA 12 Not affected by Not effected by Tolerant of resistive damage of NEMA 4 dust and dirt general purpose vibration glass conductive cleaning solutions overlay coating on Chemicals that susceptible to membrane affect polyester shock should not be used five wire resistive Resistant No NEMA 12 Not affected by Not effected by Tolerant of failure before NEMA 4 dust and dirt
128. ted the location of the command line The head movement seemed to be an important time factor The performance was optimal with the command line at the bottom of the screen Most command languages start at the top and move down to the bottom Several more guidelines may be used e Interactive support through defaults commands editing intelligent interpretation type ahead and feedback On line quick help e Function keys for high frequency commands If possible use tailorable language Function keys When using a function key interface commands and sometimes objects are specified by pressing special keys on the keyboard Advantages and Disadvantages of function keys dialog style Function key language dialog styles have particular advantages and disadvantages to other dialog styles These are included in the following table Advantages Disadvantages Self explanatory Limited number of keys available Easy to use Hardware approaches to expansion are expensive Flexible Software approaches to expansion sacrifice screen space efficiency and ease of use Requires little or no screen real estate Makes keyboard system or application Specific Low typing requirements Touch screen based equipment Page 63 Function keys are most appropriate for USER PSYCHOLOGY Negative to positive attitude Low to high motivation KNOWLEDGE AND EXPERIENCE Low typing skill Low to high system experience Moderate to high task experience
129. tems High computer literacy JOB AND TASK CHARACTERISTICS High frequency of use Formal training Mandatory use Low turnover rate High task importance Low task structure Touch screen based equipment Page 61 A very powerful dialog style is created if care is taken in designing command languages Because they are the oldest and were the most common used command languages have been more extensively studied The design of command languages can be divided into four separate design issues Command language semantics e Command language syntax Command language lexicon Command language interaction Command language semantics Semantics describes the set of functionality the language provides and how that functionality is broken down into language elements Two extremes are rich language and minimal language A study Kraut 1983 looked at the amount of richness to provide into a language They stated that without reducing the richness of functionally the performance could be improved This could be done by changing the design such as more consistency in syntactic rules easy access to status information a more interactive style better error feedback and even alternative interfaces to more and less frequent used commands Another study Good 1985 covered the same area Command language syntax The syntax of a command language involves both the format and punctuation From two studies Ledgard et al 1980 Barnard amp Grudin 198
130. ter human interaction is not such that we can yet apply models such as the Model Human Processor or GOMS Moran 1983 with complete confidence eliminating all need for testing The determinants involving the human performance are summarised in appendix A and B The main part from the tables in the appendix are drawn from Mayhew but they are adapted to the users of MMM equipment Most determinants are trivial but some need extra attention 2 2 1 Psychological Characteristics of user Designers can design interfaces to address differing cognitive styles negative attitudes and low motivation or to exploit and maintain positive attitudes and high motivation for instance consider the following goals User characteristic Design goal Low motivation discretionary use Ease of learning Low motivation mandatory use Control power High motivation due to fear Ease of learning robustness control power High motivation due to interest Power ease of use E g users who are highly motivated out of fear for example of losing their job or of appearing incompetent need the reassurance that the system is not overly complex and will not be overly complex to learn 2 2 2 Knowledge and experience of user User experience in not just a simple binary dimension novice and expert The dimension of knowledge and experience is a continuum A number of characteristics that are relatively independent are listed in appendix A and B For more detailed inform
131. ting edge and returns to region S Im the array where it is reflected back to the oY transducer The first reflector will send a ees signal back first then the second and so on When a stylus such as a finger comes into Reflection contact with the wave it attenuates the wave motion by absorbing part of the wave Control D PADI A LDIYI AVIVA 7 E electronics detect the location of the dip in the ndm Touch absorbs energy wave amplitude thus determining the touch train causing a dip in amplitude position Figure 9 Guided Acoustic Wave Touch screen based equipment Surface Acoustic Wave Surface acoustic wave technology is based on transmitting acoustic waves across the surface of a glass overlay placed over the displays surface A touch screen controller sends a five megahertz electrical signal to the transmitting transducer The transducer mounted on the edge of the glass converts the signal into ultrasonic waves within the glass These waves are directed across the front surface of the touch screen by an array of reflectors Reflectors on the opposite side gather and direct the waves to the receiving transducer which reconverts them into an electrical signal Since the speed of the wave is known and the size of the glass overlay is fixed the first reflector will send the first signal back first then the second and so on When a stylus such as a finger comes into contact with the wave it attenuates
132. tion and engineering purposes these elementary measurements are available to execute Due to the fact that image processing is the basis on which the equipment works a frame grabber or other vision card is used to process the images received from cameras or LDR s For evaluation and engineering purposes the basic functions of these cards must be available Currently in MMM equipment SBIP cards are used The functions of currently implemented SBIP cards can be found in the Users guide Mullekom 1996 For manual measurements it must be possible to select different images If a user has maintenance privileges he is able to edit program parameters tube parameters position parameters colour recognition parameters eccentricity parameters convergence parameters and linewidth parameters With those privileges he is also able to enter new users to the system The user must see the measurements made to judge the quality of the CRT 7 2 Domain analysis 7 2 1 Candidate classes The standard OMT approach to finding the domain classes is to examine the problem statement for candidate classes Typically nouns pronouns noun phrases and implied nouns are underlined for further evaluation To gain a complete list of candidate classes the user manual of the existing COPER T1 is also evaluated In appendix G a complete list of candidate classes is available Trivially unsuitable candidate classes are not included in the list A filtered list of candidate
133. tion is 33 touch points per inch e Calibration Since the reflector arrays determining the touch location are fixed in one place surface acoustic wave systems are not subject to the phenomenon of sensor drift The touch system needs only to be aligned with its corresponding display e Z axis In addition to the typical x and y coordinates surface acoustic wave systems can provide a z axis component which is determined by the amount of pressure applied to the sensor e Stylus requirements Guided acoustic wave touch systems can be operated with gloved hand e Flexibility Surface acoustic wave touch systems are adaptable to a wide range of displays without expensive custom fees Surface Acoustic Wave Disadvantages Touch screen based equipment Page 75 Transparency Guided acoustic wave touch systems typically have a transparency between 90 and 95 This can affect image quality particularly for high resolution video mode Stylus Requirements A hard stylus such as a pen will not absorb the acoustic energy and will not be recognised as a touch Capacitive Advantages Resolution Capacitive overlay systems can deliver a resolution of up to 4096 x 4096 touch points Sealability Capacitive overlay systems sensors can be sealed to the display preventing water and dirt from penetrating the display s inner electronics NEMA 4 requirements can be met Flexibility Surface acoustic wave touch systems are adaptable to a wide r
134. tionality For the demonstrator using the SBIP the SBIP class is immediately derived from the base class The defined methods in the base class are a little coloured because the are retrieved by looking at the current capabilities of the SBIP card Extra methods may be needed but were not implemented because of the limitations of the SBIP card For the demonstrator the available methods are sufficient 7 5 2 9 CSBIP For the demonstrator and currently all MMM equipment the card to use is the Sbip card For video processing a lot of preferences are available They are accessed by the dialog controlled in the derived class from the base class CVideoprocessor A different Touch screen based equipment Page 40 implementation asks for a different dialog in this way the system does not have to know anything from the available preferences The member functions are not implemented for the demonstrator because all measurements are handled by remote control access and in the demonstrator no SBIP is available Actions initiated by the user in the dialog are interpreted by a member function and then redirected to the remote control object 7 5 2 10 CCameraplacing The task of the class CCameraplacing is to place the camera on the correct spot of the measured object If a fully automated system is built more cameras or an XY table are available If none of this gear is available hand placing is required The object of type CCameraplacing is responsib
135. tunesosterennsreseaeses 26 7 2 5 Rational RoSeusia narnia eden dan e R E ead 28 73 GME FaAlIS ALI CREER Ls NEN Situs 28 7 4 Application analysis ia 30 7 4 1 SCENArl O m IT S P 30 7 5 System design 7 5 1 Code generation 7 5 2 E E E A E TE E A ET 34 8 Conclusions and Recommendations oossosoooosesosesesseossooooooososasonessasoseesosososososossesessse S2 8 1 Conclusions PORE E IEEE EE vues EEEE EE EEEE P 8 2 Recommendations ET INE PATO sou esee sees ess EEEE EOR E PE s 43 references PEN me setae AO Appendix w zoo m v y FY n oz s tr m User profile Checklist Job and Task Characteristics Dialog Styles Evaluation form Touch Screen technologies Use Case List of candidate classes Model dictionary Questionnaire Header file of tube class C code file of tube class Touch screen producers OMT notations Design internal scenarios Dialogs Example of part from parameter file 51 52 53 66 67 77 81 82 83 84 90 91 92 93 94 99 102 Touch screen based equipment Page 1 Abstract Today a lot of user interfaces are redesigned to satisfy the needs of industrial operators Man machine interfaces are optimised to offer a great functionality to the users The m
136. tured in textual form as a functional specification In the early phases of system development use cases capture the system s functional requirements During later development phases object message diagrams are used to capture design level detail of the functional model 7 1 5 Development activities ae breue ainknowledge Non functional constrains m ul The OMT process is best described as a set me of stages or activities that need to be Use cases aliz performed to construct a system Each activity involves using multiple models and has several deliverables Figure 1 shows the high level approach to the m dynamic model activities domain Object model domain 7 1 6 Conceptualisation To capture the system demands OMT uses Z use case modelling An use case describes functional model object model dynamic model paw an actor s typical use of the system An application application application igen actor is a role played by a physical person aes ct or object when interfacing with the system implementation An actor may engage in more than one Sio typical use case if the uses are separated in object code time or place Figure 1 OMT structure 7 1 7 System boundary The system boundaries are defined by the responsibilities of the system For the Coper the responsibility is to executing measurement
137. tures should be positioned away from the edge of the touchable portion of the buttons to reduce invalid touches The same study showed that it is important to have a feedback when a touch is detected This fastens the learning curve that is always detected in studies with touch screens The Egido et al 1988 study shows that the performance of the user is higher when he can make a decision based on labels and pictures This means that equipping the button with pictures and labels will optimise the performance 4 4 Touch screen position Most of the studies involving touch screens mention the parallax problem The D Beringer 1989 study investigates the effect of placing the touch screen to different line of sights with the user The results show that even a 17 degree angle altered the touch positions with 5 mm Not only the touch location is affected by poor touch screen placement But also the muscular fatigue is huge if the arm and hand of the user is unable to rest in between touches It is even optimal if the arm remains in contact with an anchoring point Touch screen based equipment Page 15 5 EMC ESD Constraints Electromagnetic compatibility EMC is the capability of an electrical device or system to operate in its electromagnetic environment without disturbing or being disturbed by it EMC is an important criterion of product quality According to the definition EMC is subdivided into electromagnetic interference EMI and electro
138. uch frame or opto matrix frame contains a row of IR light emitting diodes LED s and photo transistors each mounted on two opposite sides to create a grid of invisible infrared light The frame assembly is comprised of printed wiring boards on which the optoelectronics are mounted and is concealed behind an IR transparent bezel The bezel shields the opto electronics from the operating environment while allowing the IR beams to pass through The IR controller sequentially pulses the LED s to create a grid of IR light beams When a stylus such as a finger enters the grid it obstructs the beams One or more phototransistors detect the absence of light and transmit a signal that identifies the x and y coordinates Page 69 Photo sensor A Touch Gridof Activation Infrared Light Figure 12 Infrared Touch screen based equipment Page 70 Resistive Resistive overlay technology consists of a glass overlay substrate with a thin metallic Clear Glass Polyester coating over which a layer of polyester is Overlay Membrane with placed membrane The polyester layer has a Thin Transparent similar metallic coating on the interior Metallic Coating surface Tiny spacer dots of non coated polyester prevent the two surfaces from contacting each other A final hard coating is CRT Face usually applied to the external surface of the 3 polyester to reduce damage from sharp styli B e Di
139. um U enum m_ClockPulseType t end CTube m_ClockPulseType attr begin CTube m_DAF gun attr preserve no private boolean U boolean m DAF gun 4 end CTube m_DAF gun attr Data Members for Associations Iff Association Has Tube 34846EB700FA begin CTube measuring equipment role preserve no public CMeasuring Equipment UHGN CMeasuring Equipment measuring equipment end CTube measuring equipment role Ift Association lt unnamed gt 34846F380000 fit begin CTube lt the_CField gt role preserve no public CField 1 gt 25UHN CField the CField 25 l end CTube lt the_CField gt role Additional Implementation Declarations begin CTube implementation preserve yes end CTube implementation h Class CTube Get and Set Operations for Class Attributes inline inline const string CTube get m name const begin CTube get_m_name get preserve no return m_name end CTube get_m_name get inline void CTube set_m_name const string value begin CTube set_m_name set preserve no m_name value H end CTube set_m_name set inline const int CTube get_m_MeasDelayFor2A const begin CTube get_m_MeasDelayFor2A get preserve no cetum m_MeasDelayFor2A 88 end CTube get_m_MeasDelayFor2A get inline void CTube set m MeasDelayFor2A const int value begin CTube set_m_MeasDelayFor2A set preserve no m_MeasDelayFor2A
140. un Vol E79 B NO 4 April 1996 Mullekom R 1996 User manual software for copert T1 Internal Report Philips 7322 055 2177 Norman K L Chin J P 1988 The effect of Tree Structure on Search in a Hierarchical Menu Selection System Behaviour amp Information technology 7 no 1 January March 1988 pp 51 66 Paap K R amp Hofstrand R 1986 The Optimal Number of Menu Options per Panel Human Factors 28 no 4 Aug 1986 pp 377 86 Potter R L amp Weldon L W 1988 Improving the accuracy of touch screens an experimental evaluation of three strategies CHI 88 pp 27 32 Roberts T L amp Moran T P 1982 Evaluation of text editors proceedings Human Factors in computer systems March 1982 pp 136 41 Rumbaugh J 1991 Object Oriented modeling and design prentice Hall ISBN 0 13 630054 5 Schelkunoff S A The electromagnetic theory of coaxial transmission lines and cylindrical shields BSJT Vol 13 1934 Schneiderman B 1982 The future of interactive systems and emergence of direct manipulation Behavior and information technology 1 pp 237 56 Sears A and b Shneiderman 1991 High precision touchscreens Design strategies and comparison with a mouse International Journal of Man Machine Interaction 34 1991 pp 593 613 Snowberry K amp Parkinson S R 1938 Computer Display Menus Ergonomics 26 1983 pp 699 712 Shinar D amp Stern I H 1997 Alternative option selection metho
141. undamental principle from which all others derive is to know the user It is wrong to assume that all users are alike and that all users are like the developer So to derive an efficient and effective MMI it is important to make an inventory of the users User performance may be described in terms of a number of general determinants including the human information processing system the user s psychological characteristics the user s knowledge and experience the user s job and tasks the user s physical characteristics the user s physical environment and the user s tools The first determinant tends to be relatively constant across individuals while the last six may vary significantly across user groups or individuals From the software designer s point of view the first six are predetermined while the last is under the designer s control Touch screen based equipment Page 8 Knowledge of human information processing strengths and weaknesses leads to a number of design goals for interactive systems It is also true that knowledge of the characteristics of human information processing allows people to make predictions regarding human performance on interactive systems The idea behind a model such as the Model Human Processor Card 1981 is that a good validated model or theory allows us to make predictions about behaviour without actually having to collect empirical data Unfortunately the state of science of compu
142. urement parameter object because they are designed together But it does not know the location of the object During start up of the program a member function in the measurement method object initiates a call to the system to request for the appropriate pointer The system then asks the field object to generate the array of field dependent measurement parameter object pointers As mentioned before there are several ways to implement the dynamic linking And the generation of those pointers depends on the implementation If the Touch screen based equipment Page 39 list of pointers is generated it is returned to the object that asked for it Before performing a measurement the measurement method object asks the system which field is next to be measured Accordingly a call to get the correct parameters is initiated Parameters that are not field dependent are kept in its own object By touching the button of the measurement twice in a short time the dialog to edit these variables is shown The measuring equipment object controls the measurement selection dialog so that object requests the measurement method object to show the appropriate dialog In case of the demonstrator only the linewidth object contains field independent parameters The linewidth property dialog is shown in appendix P 7 5 2 7 CVisionSystem The vision system represents a collection of objects that is responsible for the collection of measurements Not only the placement of th
143. ust be determined For large systems this is normally the stage at which a lot of system organisation is handled For the demonstrator a one man design however this stage only comprises partitioning the system into subsystems and allocating subsystems to components e g hardware software and operations Subsystems are typically chosen in a way that organises the systems into distinct layers and partitions For the demonstrator three layers are chosen On top there is the user interface In the middle the task dependent layer is situated and at the bottom the hardware dependent layer is found One can identify more subsystems but for the demonstrator these three are sufficient For future use and increasing reusability more subsystems are a must The hardware layer is not further explored The demonstrator as implemented performs the measurements by accessing another COPER T1 with a remote link No hardware dependent code is necessary at this stage In lambert 1996 a MMM equipment is designed using OO in combination with visual C In that design the subclass hardware platform can be identified The design is very solid and it forms a great blueprint for a lot of MMM equipment At this time all major MMM equipment is based on VME platforms in combination with PC based platforms For the interaction with the users a great variety of input output devices are available In the future one likes to if possible go to one uniform user interface T
144. vergence Once again this is not hard coded The type is passed as an argument to a function in the field object Two functions are available to load and save the CRT dependent parameters The argument of these two functions is the filename The two functions work independent of the file contents The arrangements of the parameters are solely determined by this function and therefore object independent Another measured object most probably has other parameters So the implementation can not be inherited But the function must be implemented for the system so it is declared virtual in the measurement base class In appendix Q a part of the file is given as an example See appendix P for the user interface layout of the previously mentioned dialogs Touch screen based equipment 7 5 2 4 CField As with most other parameter dependent classes the dialog for the field dependent parameters is controlled in its own class As mentioned earlier this is the object in which the measurement parameter objects are linked For the first development the objects are linked statically This means there is a variable for every measurement parameter To fulfil the needs this must be changed to a dynamic construct or to an association of a linked list of measurement parameters In the last case the name of the measurement is passed as an argument By passing the name of the desired measurement every member of the linked list can determine if it must take the
145. y the dialog of the tube class is also responsible for the visualisation of the fields If a field is selected this is passed to the appropriate field object 7 5 2 3 CTube The parameters for the CRT are situated in this class so the dialog to edit those parameters is also controlled from within this class The measurement parameters are associated with the fields In multi selection mode more than one field can be selected The parameters that are the same in all those fields must be displayed in the measurement parameter dialog Or if a change is made it must be changed in all selected fields The locations of those parameters must be available to the dialogs To overcome this problem the request to do a parameter update of the measurements is situated in the tube object This object knows the available fields and knows the locations of the available measurement parameters It is neither wanted nor necessary that this object has knowledge about the interior of the measurement object dynamic If a update request is done single or multi mode The function that executes the dialog request sends an array with pointers of type CMeasureParameter This is the base class If field X is not selected the array location X is filled with a NULL pointer The array is sent to a arbitrary measure parameter object of the right type For instance if the user wants to change the convergence parameters of the selected field the type to send to is of type con
146. y to H fields present at ESD In general keyboards are if possible avoided in industrial equipment Touch screens do not cover all the appropriate dialog styles but as can be seen in the Plaisant study for limited alphanumeric data entry a touch screen even has some advantages over a keyboard The mayor drawback of touch screens is the target size See chapter 4 for more information For large data entries a keyboard is preferred to get the highest performance from the process or maintenance engineer If a specific MMM apparatus needs large data entries an additional plug in keyboard or a shielded drawer keyboard is most appropriate Process continuation buttons are used to navigate back and forward in semi manual controlled apparatus These buttons are used intensively For those buttons it is very important that they can be located and accessed easily To get a reference for location they might be situated in the corners of a touch screen but this is not an optimal solution To get the highest performance for that control external buttons are most appropriate although the drawback with touch screen might be solved if the sensitive area around the button is relatively large Touch screen based equipment Page 12 3 2 Output The traditional output device for MMM equipment is signalling lights gages analog digital LED displays and screens More recently LCD s are used as an alternative output device Every MMI is specially hardware
147. y tube platform vaccum video memory XY controller Touch screen based equipment Page 82 H Model dictionary Video search Attribute operations Color recognition Attribute operations Convergence Attribute operations Eccentricity Attribute operations Line width Attribute operations measuring equipment Attribute operations SBIP Attribute operations camera Attribute gain operations XY controller Attribute operations Tube Attribute operations Field Attribute operations Picture Attribute operations User Attribute operations remote system Attribute operations fube supplies Attribute operations search the video dot on the screen sequence to detect the color of the spots measuring sequence to measure the convergence measure the data collection system for quality assurance measuring sequence to measure linewidth Measure Convergence Measure Eccentricity Measure Linewidth frame grabber card grabs a frame from the camera live video gain threshold image map Boot initialise Fixed video auto gain auto threshold mask in plane pixel calibration show planes histogram colour simultane an apparatus to grab an image from the tube representation in the system of a XY controller to position a single camera x position y position ready move representatation in the system of a real tube CYT or TYT Tube name Tube type D
148. ystem should be case insensitive when it really does not matter Keep input fields short Do not combine letters and numbers if possible e Do not require leading zeros These guidelines are drawn from several studies Galiz 1989 Gould 1988 Greene 1988 Fill in form prompts and instructions Fill in forms have prompts or brief syntactic or semantic instructions associated with individual fields Those are instructions for cursor movements the use of special function keys or screen acceptance and cancellation The following guidelines apply Provide prompts when use will be relatively infrequent Prompts should be brief and unambiguous Place prompts to the right of fields or in a MicroHelp line at the bottom of the screen Provide instructions for navigation and completion Place instructions in a consistent location across the screen Use consistent terminology and consistent grammatical form Fill in form navigation The user must be able to move from field to field to accept a filled in screen to cancel a screen and to move forward and backward between screens The following guidelines can be applied When a form is entered position the cursor in the most likely default position Arrange field groups consistently with default cursor movement Allow forward and backward movements Make protected areas on the screen inaccessible Do not use auto tab unless fields have fixed lengths e Provide titles and page numbe
149. ze X is calculated as follows Il Video shift step size X total number of video shifts W phosphor pitch Il delta gravity X 2 number of video shift steps Il This results in a Video shift step size X which is required to reach a Video dot position change of one phosphor pitch by the chosen Nr of video shift steps const int get m VideoShifiStepSizeX const void set m VideoShiftStepSizeX const int value Attribute m NrOfVideoShiftSteps The tube parameter Nr of video shift steps is the number of video steps the spot is shifted during Micro stepping in X direction For convergence measurements this is also the number of measure ments for averaging This parameter is not defined by Auto calibration but set by the user The value depends of the required Accu racy the higher the better but be aware that the measuring time is influenced The default value for the Nr of video steps is 5 To change this parameter go to submenu Nr of video shift steps and key in the desired value To store see topic Change edit data To check if the Accu racy is within limits see topic Accuracy const int get m NrOfVideoShiftSteps const void set m NrOfVideoShiftSteps const int value Attribute m PulseWidth The tube parameter Pulse width determines the width of the Video dot or vertical line in horizontal direction The value of

Touch Screen Based Measuring Equipment

Contents

Download Pdf Manuals

Related Search

Related Contents