User`s Manual
Contents
1. 2 DI Ovenlennretn enten ett ocak ENDE DUE UE 2 2 2 1 Power and Output Connections sersssssssssssessssssssssesssnssesassssssessnssesnnseessnssssnsesasssssnseesnsssseon 4 2 2 2 Temperature Reference snotteren 4 2 2 3 Thermocouple Status eet ettet ttt ttti ttt ttt ttt ttt ttt ttc 5 2 2 4 Thermocouple Connections cesssssasssssssssssssssssnsssssssesssssssnssessnsssessesnssssnsssssnssssnssetnssessnssesn 5 2 2 5 Placement of the Model 1 3 essssssssssssssssssssssssseesnssssssssesnssesnassesnssssnsssesnssesnaseesnssssnaseesnssseees 5 2 2 6 Mounting the Model 1 3 sessssssssssssssssssssssssssssssssssnssssesssesnssesnassevnssssnassesnssstnasessnssssnassevneseeees 6 3 THERMOCOUPLE BASICS 1 eie na opua Ewuv xd chin Gs na CUu mi Rad Ea n ERE REY E wa RnB iER GE Ed 6 3 1 What is a Thermocouple sasscsnssosnssesonseosansosonseosansosonsoosansosonssosansosonssosansossnssosoansosonsoosen 6 3 2 Thermocouple LV pes sms o a pet D M se UL M D EDU M C D ME SEE aT 3 2 Making a Thermocouple eessssssssssssssssssssssnsssssnsessnssesnnssssnssssassesnsssssnssesnssssasessnssesasessnsssee he 3 3 Mounting a Thermocouple on a Rocket cssssssssssssssssssessssssnssesesssesnssssnssssnssssnassesnssssnasessnessenes 8s 3 4 Converting Thermocouple Output Voltage to Temperature assssssssssssssssssesssssssensessnssssen 9 3 5 Examples of Rocket borne Temperature Profiles eet 11 APPENDIX A CALIBRATION PROCEDURE n nennen nnn un2. selected The Standard data Figure B4 Select are not selected and will not which data are to be be plotted displayed 17 B1 6 Produce Data Plot Once valid calibration data are loaded and data are selected for display the Plot Reset button becomes active Click to plot the selected data Figure B5 shows a plot of the same data set plotted in Figure 10 Differences in smoothing see section B2 0 cause differences in the appearance of the two plots Plot Temperature Data 36 41 Temperature F Figure B5 Plot of Ascending and descending thermocouple temperature data To add or delete data from the plot select or deselect the data of choice and then click the Plot Reset button again B1 7 Enlarge Plot Area To look at a subset of data click on the Select button A rectangular area of the plot can be selected by holding down the mouse and dragging A box appears showing the selected area Releasing the mouse fixes the box dimensions If a different box is required just click and drag the mouse again 18 To enlarge the area in the box select Zoom Select Plot Reset to redisplay all of the data To replot an enlarged area without resetting the plot ranges useful if you are selecting or deselecting data click successively on Select and then Zoom without using the mouse to create an enlargement box B1 8 Enable Cursor To read temperature values at specific altitudes select the Enable Cursor button Windows w
3. tedious or by using the DisplayData Program from Glitch Figure 4 Model 1 3 14 pin box header plan 3 Laboratories that is provided with every Model 1 3 Use of this software is described in Appendix B This software also has the ability to read and or convert and display data in English and Metric units 2 2 1 Power and Output Connections Capable of being powered by a 9 volt battery the sensor board draws very little current and can connect directly with an RDAS via a 14 pin box header J1 a connecting ribbon cable is not included Power V is supplied to pins 1 and 2 while pins 13 and 14 are ground Output can be routed to any one of the six RDAS analog input channels pins 7 12 by re arranging an onboard jumper which is set to channel 1 at the factory If you are already using this RDAS input channel for other things the jumper may be soldered to another channel Because it is designed to work with an RDAS the output of Model 13 is constrained to never exceed the 5 volt input limit of the RDAS regardless of supply voltage It is mechanically possible to insert some connectors into the box header the wrong way causing among other things a reversal in the polarity of the supply voltage Obviously the Model 1 3 will not work Figure 5 Thermocouple connected this way but on the other hand it and temperature standard isot ted d outputs can be jumpered to WES DUL BE afer any of six channels If not used with
4. an RDAS power leads can be soldered directly to the bottom of the board at pins 1 2 and 13 14 Small and symbols on the bottom side of the board identify these pins The output voltage can be sampled at test point To relative to test point G or ground see Figure A1 in Appendix A On power up the output of the Model 1 3 will read low by about 0 02 volts In the course of 5 minutes it will rise and stabilize to the calibrated values 2 2 2 Temperature Reference The Model 1 3 contains its own LM34 temperature reference which is useful for monitoring board temperature and performing bench top calibration checks Output from this reference is jumpered to analog channel 2 at the factory it can also be rerouted to a different channel if desired This output can also be sampled at test point So see Figure A1 in Appendix A 4 2 2 3 Thermocouple Status A thermocouple status indicator is included This LED lights when a thermocouple is absent or if the temperature sensed by the thermocouple exceeds the upper temperature limit which is about 20 F 105 C in this design 2 2 4 Thermocouple Connections The thermocouple connects to the Model 1 3 via a 5mm terminal block J2 Figure 6 For proper functioning it is important to connect the thermocouple with the proper polarity For a type T thermocouple see section 3 2 on thermocouple types this means connecting the 4 n copper wire to the t
5. radiation and or convection One fairly successful attempt is shown in Figure 8 To minimize contamination of the thermocouple junction with heat from the supports the rule of thumb is that the supports must be at least 20 wire diameters from the junction My experience has been that this may not be enough In the 8 mount shown here the thermocouple is suspended between two posts here made from toothpicks of unequal length about 1 25mm apart corresponding to about 100 wire diameters support unction separation The junction is about 4 6mm away from the rocket body to minimize heat transfer from the body into the thermocouple A variation of this mount surrounds the sensor with a white painted shroud made from 4 6mm brass tubing aligned parallel with the intended direction of travel of the rocket This shroud shields the thermocouple from direct sunlight as well as providing protection from impact during dogm j Nd preparation and at the end of the flight The supports are thermocouple The of unequal length so that the airflow across the sensor is junction is visible near not impeded by the upper support Moreover the upper the center of the ME Hl sensor wire is the Constantin lead because Constantin is stronger than copper and so is better able to withstand the high G s and drag during ascent The thermocouple leads continue into the electronics bay they are insulated with heat shrink tubing wh
6. size of the wire matters very little meaning that thermocouples can be routinely made from wire as small as 0 001 25um which is the diameter of a human hair This small size means that the thermocouple has very little thermal inertia it can change it s temperature very quickly To minimize thermal inertia my initial work was done with 0 001 wire type T thermocouples Though successful I found the fine wire difficult to handle On one occasion I dropped a short length of 0 001 Constantin wire on the bench in front of me After searching for 10 minutes I found it only to discover you can t solder human hair 3 2 Making a Thermocouple Making a thermocouple is a fundamentally simple procedure Take any two chunks of dissimilar wire and connect them together Bingo you have a thermocouple You will get a voltage if there is a difference in temperature along the S9 wires However it may be hard to accurately convert that voltage into a temperature because the metals may not be pure and accordingly will not match the behavior of the standard metal pair Assuming standard thermocouple wire is in hand then a usable thermocouple is readily formed by joining them together Thought should be given to the way the joint is formed Theoretically butting one piece against the next will suffice Practically enced Lu remove Soldered surface impurities while forming a strong mechanical bond requires the formation of some form of weld Mos
7. the metal of the wire Two different metal wires in contact will generate no signal if the same temperature is maintained down the length of the wires However a signal develops if the tem perature changes along the length of the wires away from the junction It is important to understand that the signal that develops is related to the difference in temperature along the wires not the absolute temperature of the thermocouple Moreover the strength of this signal is related to the net temperature difference along the wires If one or both wires pass through in and back out of a local hot or cold spot the net effect of this local temperature change is zero the voltage shift induced going in is exactly cancelled by the voltage shift coming back out To get a signal related to the absolute temperature a second junction must be inserted into the circuit that is held at a known temperature Historically this reference temperature was usually 32 F 0 C because making crushed ice baths using pure water was easily done in the laboratory Outside the laboratory it was less convenient The use of thermocouples in the field was accelerated by the development of semiconductor devices that can mimic the behavior of a 32 F 0 C reference junction The AD595 thermocouple chip incorporates just such a device 3 2 Thermocouple Types Useful thermocouples are made from highly refined metals so that their behavior will match those of standard thermocoup
8. this manual Because the use and application of the Model 1 3 unit is beyond my control the purchaser or user agrees to hold me Robert Brigham harmless from any and all claims demands actions debts liabilities judgments costs and attorney fees arising out of claimed on account of or in any manner predicated upon loss or damage to property of or injuries to or the death of any and all persons arising out of the use of this equipment Due to the nature of electronic devices the application and environments for those devices the possibility of failure can never be totally ruled out Life support applications This product is not designed for use in life support appliances devices or systems where malfunction of this product can reasonable be expected to result in personal injury My customers using or selling this product for use in such applications do so at their own risk and agree to fully indemnify me for any damages resulting from such improper use or sale If you need to contact me your best bet is by e mail I have been slogging along with the same Compuserve account rbrigham1 cs com for over 12 years and don t feel like changing just yet Robert Brigham 26500 Agoura Road PMB 336 Calabasas CA 91302 June 2004 NAR 79579 L2 ii Table of Contents T INTRODUCTION ais seorezenesoresanrenindnnwenitiandennt vendohwtsardehtddnedenitinndonrd deka 1 2 MODEL 1 3 THERMOCOUPLE SENSOR BOARD
9. Glitch Laboratories Thermocouple Sensor Board Model 1 3 User Manual n Glitch Laboratories If it works it s a Glitch v1 31 9 17 2004 Note Glitch Laboratories is a fictiteous company It has no web site buildings or other physical presence though there are some mass spectrometers at NASA that are Were using Glitch Laboratories controllers It merely exists for those moments when I get to putter at the electronics bench in my apartment Consequently it cannot extend any kind of formal warrantee But I will do my best to offer one Product warranty I warrant that the Model 1 3 unit is free of defects and that it will operate at a satisfactory level of performance for a period of one year from the original date of purchase If the unit fails to operate as specified notify me within the warranty period Modifications to the unit void the warranty On the other hand I have tried really hard to produce something robust useful and worthy of the Glitch name er well anyway So if you have a problem with a Model 1 3 sensor board please let me know I will do all I can within reason to make it better especially if I like you and you seem to be intent on having fun This is a hobby after all However here is some more legalese Product disclaimer and limit of liability The Model 1 3 unit is intended for use in model high power rockets only Do not use this device for any other purpose than specified in
10. and test point locations wires touch any of the component leads Step 2 Power up the board wait 5 minutes and read and record the voltage relative to ground of the LM34 at test point So The output of the LM34 by devious design corresponds to the temperature in F For example an output of 0 67 volts corresponds to a temperature of 67 F An output of 1 volt means the air conditioner isn t working Step 3 Being careful not to touch or breath on the thermocouple read the output voltage at test point To Note for example that f the temperature is Calibration may also be performed with a type K thermocouple but the resulting values are best used with the DisplayData program see Appendix B section B1 4 since an equation of the form shown in Equation 3 does not provide a great fit to the type K thermocouple calibration curve Mi Le 67 F then an output voltage around 1 67 volts can be expected Record this value Step 4 You may want to repeat steps 2 and 3 a few times just to be sure temperatures are stable If you are satisfied you have a good temperature voltage pair go on to step 5 Step 5 Prepare a crushed ice slush It is best if pure water is used and only a small amount of water exists between the lumps of ice This ice bath serves as a 32 F 0 C reference While observing the thermocouple output voltage at test point To place the thermocouple into the ice bath Look for and record the lowest o
11. atuitous photo of my of these qualities is quite slow Response Ultimate Endeavor flying on an time of these devices is drastically reduced Aerotech K700 at Lucerne Dry Lake Dec 2002 by the thermal inertia of the encapsulation material and housing For example a LM34 in a plastic TO 92 housing make take in excess of 8 minutes to thermally equilibrate with its surroundings A typical high power rocket flight from ascent to touchdown is usually over in half this time The best type of temperature sensor that is simple rugged and fast enough to work in a rocket borne probe is a thermocouple 2 Model 1 3 Thermocouple Sensor Board 2 1 Overview A thermocouple consists simply of the junction of two pieces of wire made from different metals Change the temperature around the junction and a DC voltage appears at the end of the wires A small DC voltage The Model 1 3 Thermocouple Sensor Board is the result of experimentation over more than two years with various circuits that could stably amplify this small input voltage into a larger more convenient output voltage Many different configurations were tried but I finally settled a a circuit based on the Analog Devices AD595 thermocouple sensor chip Since I already had an RDAS it was a simple Figure 3 Model 1 3 matter to route this output voltage to one of the six Thermocouple Sensor Board E recording analog input channels of the RDAS The RDAS also supplies power t
12. ave and then Save Settings A small file named Settings txt containing your preferences is loaded into the same directory as your data Any time you go to load in data from that directory these preferred settings are read in too Note the DisplayData program will leave a Settings txt file in any directory you access data from However unless you actually choose to save your preferences in that directory the settings file is empty and can be deleted B1 16 Exit When you are done select the big Exit button in the lower left corner of the main window It will take you out of the DisplayData program and back to reality or the Windows version thereof Good Luck Eu APPENDIX C Mounting Templates The templates below may be cut out and used to drill mounting holes for the Model 1 3 sensor board MODEL 1 3 MOUNTING TEMPLATE MODEL 1 3 MOUNTING TEMPLATE 22 MODEL 1 3 MOUNTING TEMPLATE MODEL 1 3 MOUNTING TEMPLATE
13. bits of information in the RDAS file header so it knows how to read the rest of the data If it can t find these bits of information it gives up and pouts Two types of data can be exported by the RDAS software Raw and Interpreted The DisplayData program only reads Interpreted Data To load data select Files then Open A dialog box will appear that allows location and selection of the interpreted data text file to be displayed Files of the wrong type or format cannot be loaded After data are loaded the program determines which channels contain analog data and highlights them as shown in Figure B2 The data set shown here contained records on channels 0 and 1 but the program has no way of knowing the source of the data In this example channel 0 contained the thermocouple data while channel 1 contained standard temperature data from the LM34 Clicking the ADC 1 button in the standard channel window is required to tell the program that the standard temperature data are on this channel B1 3 Select Thermocouple Type Plot Temperature Data Figure B2 DisplayData main window after loading data Channels containing data are highlighted The type of thermocouple that produced the data must be specified Click Select Thermocouple Type in the main menu or just click on the box indicating the type of thermocouple selected A dialog box will appear allowing selection of either a type T or K thermocouple If needed later version
14. en 13 APPENDIX B DISPLAYDATA SOFTWARE eene nene unen 15 B1 DISPLAYDATA PROGRAM ss renenersrsnnenoarnesensnanden ene kenenednndin dend tat bene 15 B1 1 Loading the DisplayData Progrartacesssssssssssssssssssssssssnssesssssessnsessnsessnsssssnssesnsssssnssesnsseen 15 Bio Toaduie DEE On NE E INL D DD NE 16 B1 3 Select Thermocouple Typeessssssssssssssssssssssssssssssssssesassesssssssnsssssnssssnsssssnssssnssesnssesnsssse 16 iii BL 4 Enter Calibration Values c cccsssccsssosssssecssssessscossnsosvsessvasssosoussscscesssvevevsosasacbesconsstcscosesdescosssasssbeesouasese 17 B1 5 Select Data to be Displayed eee sooseneeseenensenensenensenenseneneneenensensenensenensenenseneneneeseneenee 17 B1 6 Produce Data Plot 18 B1 7 Enlarge Plot Area nsesenenonsosenenevensenenenensenenensenenenoenenenenenseonsenenenenoonenoenenenenenseneneneneneneesenensenenenevoesenen 18 B1 8 Enable Cursor 19 B1 9 Burgi c P oco sa resos 20 B1 10 Plot Title 20 B1 11 Plot Colors nssesenonensenenenenonsenenenensenenenensenenenenensenenenensenenenenensenenenenseneneneneosenenenensenenenenensenenenensenenenene 20 B1 12 Smoothing Data nsesenenonsenenenonsenenenenoosenenenensenenenenoesenenenensenenenensenenenenoosenenenensenenen
15. enensenenenensenenenene 20 B1 13 Saving Plots c 21 B1 14 Saving Data 21 IAN EEAUPANIGUDC 21 unnr 21 APPENDIX C MOUNTING TEMPLATES cerner 22 iv 1 Introduction The Glitch Laboratories Model 1 3 temperature sensor board is designed to be used in high power rockets as a means of gathering temperature data It has a very fast response time on the order of a tenth of a second is small and light Without offset the board produces a voltage resembling temperature in F for both type T and K thermocouples i e 45 F produces a nominal signal level of 0 45 volts The board operates from a single sided power supply so it cannot produce a negative voltage for a negative temperature This is circumvented by adding 1 volt to the output offset so in the example above 45 F actually produces a nominal value of 1 45 volts Allowing for the non linearities of thermocouple response this allows measurement of temperatures below 100 F 73 C The Model 1 3 is a product first of my desire to put a high power rocket to good use and second my curiosity about the weather and atmosphere borne of my experience as a pilot My high power rocketry activities have been carried out at Lucerne Dry Lake in Lucerne Figure 1 Lucerne Dry Lake Lucerne Vall
16. ere they connect to the mounting terminals on the sensor board To facilitate connection it is advisable to solder the thermocouple leads to 4 6mm lengths of heavier gauge solid wire which can be reliably gripped by the terminal set screws So long as the leads terminal block and sensor board all remain at the same temperature within the electronics bay the presence of dis similar metal junctions at these connections will not contaminate the signal from the external junction with additional spurious signals Surrounding the board with a spongy piece of foam rubber is an effective way to prevent temperature gradients during flight Again the arrangement shown in Figure 8 represents just one possibility Experience will doubtless give rise to improvements 3 4 Converting Thermocouple Output Voltage to Temperature Uh Oh lookout Equations The output of thermocouples is very small DC voltage All thermocouples have their output referened to a standard junction held at 32 F 0 C Measurements have been made of high purity standard metal pairs at carefully controlled temperatures by organizations like NIST These data have been tabulated modeled with high order polynomials and made available to the general public One source is through the Omega website http Avww omega com Aemperature Z Zsection asp PDF files are included on the CD with this manual that include tables for type T and K thermocouples as well as coefficients for the p
17. erminal and the Constantin lead to the terminal For a type K thermocouple the Chromel wire is connected to the terminal and the Alumel lead to the terminal Inverting the polarity will not hurt Figure 6 Model 1 3 i thermocouple terminal anything It just won t generate a meaningful plock output Three feet about one meter of 0 005 copper and Constantin type T thermocouple wire are provided with each Model 1 3 Sensor Board unless type K wire chromel alumel is requested Additional wire is inexpensive and may be purchased from a vendor such as Omega For example Omega designates their 50 foot spools of 0 005 wire as follows Metal Part Number Copper SPCP 005 50 Constantin SPCC 005 50 Chromega Chromel SPCH 005 50 Alomega Alumel SPAL 005 50 I have not done this yet but finished thermocouples can also be purchased from Omega Note if you like to make your own igniters SPCH 005 is a form of Nichrome that makes excellent bridge wire 2 2 5 Placement of the Model 1 3 Thermocouple signal levels are very small and require stable high gain DC amplification to make them usable This means that other small signals 5 introduced along with the thermocouple signal get amplified too For example the RDAS is a significant source of pulse noise To offset this an RC low pass filter has been incorporated into the input of the Model 1 3 to remove much of this noise However the best remedy
18. ey CA It s big and it s flat Valley California This location Figure 1 at the edge of the Mojave Desert lies at an elevation of 2900 880m is surrounded on three sides by mountains of varying heights note background in Figure 2 the highest being the San Bernardino Mountains to the south which reach over 8000 2400m Except during occasional winter storms weather at this location is marked by calm conditions that extend from the early morning hours to early afternoon at which time cooler air flowing from the coast through Cajon Pass moves in to displace warmer desert air In the winter the temperature difference is small and little or no afternoon breeze arises In the summer months the afternoon zephyr can be ferocious Meteorologists and pilots will tell you that an enclosed basin under still clear starry skies will fill up with cold stable air as the night progresses This gives rise to a temperature inversion ie temperature goes up with altitude not down Noticing the slow and often complex drift of rocket exhaust plumes in the almost dead calm air of many Lucerne Valley mornings I became curious about the air that was moving them Thus was born the idea of flying a sensor that could measure and record air temperature during flight Have you figured out by now I m a geek The sensor had to be light simple stable and fast The average semiconductor temperature sensor while possessing most Figure 2 Gr
19. g M Eq 2 The coefficients M and b are unique but similar to each Model 1 3 sensor board and must be determined through a simple calibration procedure see Appendix A Combining equations 1 and 2 gives T A MeN ee B M Eq 3 Which is the form I routinely used in a spreadsheet to convert my voltage data into temperatures Examples are shown in the next section 10 3 5 Examples of Rocket borne Temperature Profiles In February of 2004 I made two flights that recorded atmospheric temperature profiles The first took place at about 9AM Conditions at the surface were dead calm and chilly but warming quickly under a clear sky The temperature sensing payload was carried aloft in an airframe made from standard 2 5 hardware Figure 9 and was powered by an Aerotech 1366 Redline motor The thermocouple assembly Figure 8 protruded from the side of the electronics bay and was protected by a metal shroud omitted here to show the structure of the thermocouple resembling 4 launch lug The white painted metal shroud is designed to limit heating of the thermocouple by direct sunlight However in this instance I forgot to orient the electronics bay with the thermocouple in the shade the rocket sat on the pad long enough for the sun to heat the shroud causing readings at liftoff to be about 13 F too high Upon liftoff temperature readings dropped abruptly as ambient air moved past the accelerating rocket borne sensor F
20. ighly stratified Plot produced in a spreadsheet program Noon 2 14 04 Lucerne Dry Lake CA 5000 m 4500 4000 3500 d 3000 7 2500 2000 1500 1000 500 O T T T T T T T Ascent T 30 40 50 60 70 80 90 100 Descent T T F Figure 11 Data collected at noon Air stratification is gone 12 APPENDIX A Calibration Procedure Calibrating the Model 1 3 Temperature Sensor board is a fairly simple procedure if the following materials are at hand Ground 1 Crushed ice bath best if in a thermos bottle o 2 Volt meter S a Output ed 3 Isothermal workspace i e not in the sun near a Standard radiator or open window f Temperature Output The following steps assume the Model 1 3 circuit board has reached the same temperature as T it s surroundings and that the i m i surroundings are not changing E uw T Mb CNN gt THERMO on the board unless you are LM34 COUPLE working in a 95 F 37 C or so Temperature BOARD Reference lt V1 3 temperature Tip Don t breath environment Step 1 Attach a type T B thermocouple to the Model 1 3 and place the thermocouple junction in the vicinity of the LM34 reference as shown in Figure Al Be Figure Al Calibration of Model 1 3 careful not to let the s Thermocouple Sensor Board Note position of bare thermocouple thermocouple
21. igure 9 Apparent temperature rose quickly as the rocket accelerated due to frictional heating of the thermocouple by the air The RDAS deployed the main at apogee just shy of 3000 910m above ground The rocket then descended at a rate of about 20 Sec 6m ec Frictional heating of the thermocouple is negligible at this speed so data acquired Figure 9 Rocket used to gather during the descent magenta is the most accurate Several temperature data inversion layers are evident in the cool air that had in Figures 10 and 11 accumulated in the basin overnight By noon the surface had warmed up to 55 F 13 C and a light breeze was blowing in from Cajon Pass stirring the air column A second flight on an Aerotech J420 Redline took the rocket to almost 5000 1500m The descent data shown in Figure 11 show that the cool inversion layers had warmed and or been swept out of the basin The ascent data are dominated by frictional heating of the thermocouple and with calibration could be used as a direct record of the rocket s velocity which in this case was estimated to be in excess of 500mph 800 KM hr 1 5 seconds after liftoff 11 9 15Am 2 14 04 Lucerne Dry Lake CA 3000 4 2500 d 2000 um i 1500 10007 500 OA T T T 1 A 30 0 35 0 40 0 45 0 50 0 55 0 60 0 scent T Descent T T F Figure 10 Data collected at 9am Air column below 900 is cooler and h
22. ill appear showing Cursor Altitude and the Time and Temperature of whatever data curves are plotted An example is shown below The cursor consists of a horizontal line The mouse controls the vertical position of the cursor so long as the mouse points to a location within the plot The cursor freezes if 1 the mouse is dragged out of the plot area or 2 the left mouse button is clicked Clicking the left mouse button again unfreezes the cursor Plot Temperature Data A l t i t u d e o 1 mro 36 41 Temperature F Figure B6 The cursor line can be used to read time temperature and altitude data 9 B1 9 Plot Units Default altitude units are determined by the units specified in the original data file Once loaded however they may be changed at will by selecting the Feet or Meters option at the bottom of the main window The average of the first 100 altitude measurements corresponding to all or part of the data baseline before liftoff is calculated and shown in the Launch Site Elevation box If it is within 20 6m of 0 the DisplayData program assumes that the altitude measurements are referenced to the local surface Above Ground Level or AGL If the launch site elevation is known it can be entered into the Launch Site Elevation box Selecting MSL Mean Sea Level in the Select Elevation Reference box recalculates and displays altitudes relative to sea level On the other hand if the average of the fi
23. is to keep the sensor board and especially the thermocouple leads as far from electronic noise sources as possible It may even be necessary to shield the leads or the board depending on the layout of the electronics bay 2 2 6 Mounting the Model 1 3 The Model 1 3 measures 1 5 x2 0 38mmx51mm and has mounting holes at all 4 corners that will accommodate 4 or 6 2 5mm or 3mm screws These holes are separated by 1 25 32mm and 1 75 44mm respectively For convenience Appendix C of this manual contains several copies of a mounting template Note that though the thermocouple can measure a wide temperature range the electronics themselves must remain within a 32 120 F 0 50 C operating window Outside this window the AD595 will not provide accurate cold junction temperature compensation see below However unless launching in below freezing conditions this is not a problem even if below freezing temperatures are encountered aloft the temperature of the typical electronics bay does not change significantly during the brief interval of a rocket flight This can be verified by monitoring the output of the LM34 during flight 3 Thermocouple Basics 3 1 What is a Thermocouple Thermocouples consist basically of two joined wires and the signal they generate though small is dependent only on the temperature and the composition of the wires since it is these parameters that govern the number of free electrons roaming around in
24. isplayData Program Data gathered using the Model 1 3 Temperature Sensor Board connected to an RDAS unit can be processed and plotted using a standard spreadsheet program such as Excel but the conversion process can take in excess of half an hour and is subject to error The DisplayData program was written to streamline this process as well as provide a way to save the data and plots for use by other programs B1 1 Loading the DisplayData Program The DisplayData program is supplied on the included CD and has been written in Visual Basic I have tested it on several computers running Windows XP and Windows 2000 and encountered no problems It is small and consists of two parts the executable file DisplayData exe and the interface support file COMDLG32 ocx The executable may be placed in any convenient location on your computer The COMDLG32 ocx file should be placed in the system folder which will have a path OEE nn MM CAWINDOWS system32 although in my experience just having it in the vicinity of same folder as the DisplayData exe file was enough on some machines Once the program is on your computer start it by double clicking the icon The main Plot Temperature Data window will appear Plot Temperature Data 15 B1 2 Loading Data The DisplayData program can only read text files exported by the RDAS software It will get confused by other text files and tell you to try again This is because it wants to see specific
25. les that have been very carefully characterized by national and international organizations such as the National Institute of Science and Technology NIST Metals for thermocouples have been selected that can be highly and easily refined easily worked and are chemically and mechanically stable in a variety of environments Thermocouple types correspond to a particular pair of metals and by international agreement are designated by a particular letter For example a particular alloy of chromium 10 and nickel 90 called Chromel when joined with another alloy of nickel 95 Mn 2 Al 2 called Alumel forms a type K thermocouple This metal pair is easily fabricated to high purity is inexpensive resists corrosion and is good to temperatures in excess of 2500 F 1390 C Another alloy of nickel 45 and copper 55 is called Constantin When joined with pure copper it forms a type T thermocouple Because of the ease with which high purity copper and Constantin are fabricated the type T thermocouple gives very good accuracy at low even cryogenic temperatures The fact that copper melts at 1980 F 1083 C prevents the use of type T thermocouples at high temperatures Other common thermocouples are the type E Constantin Chromel which produces the highest signal level of any standard metal pair the type J Iron Constantin as well as the types S R and B using Platinum and Rhodium alloys that find use at very high temperatures The
26. o run the sensor 2 2 Features The specifications for the Model 1 3 are listed in Table 1 At 1 5 the board is slightly wider than an RDAS Compact is 2 long and weighs unconnected 0 5 oz Table 1 Model 1 3 Thermocouple Board Specifications Model Number 1 3 Width 1 5 38 Length 2 0 51 Height nominal 0 625 16 Weight 0 5 15 Supply Single sided 7 16 Supply Load 9v 4 Supply Load 9 16v 11 Input T or K type Thermocouple 32 F 0 C 1 31 Volts Output DC 83 F 28 C 1 85 Volts nominal Min lt 100 F 73 C 0 05 Volts values Max gt 220 F 105 C 3 7 Volts Max Output Load 5 mA Accuracy 1 0 6 F C Operating Temperature Range electronics 32 120 0 50 F C Standard LM34 Output Range 0 3 Volts Standard LM34 Output F 00 in Volts Standard LM34 Output Accuracy 0 8 F 0 5 C 77 F 25 C Standard Max Output Load The Model 1 3 can amplify the output from any thermocouple It is designed to provide an output voltage from a type T or K thermocouple input that is easily deciphered in the field The voltage out is 1 volt plus the temperature More accurate results can be had by calibration and conversion of the output with standard thermocouple formulas This can be done by setting up a spreadsheet in Excel or similar program
27. olynomials One inconvenience of the polynomials is their range Depending on the thermocouple type different coefficients are used on either side of freezing This is not a problem with the 9 DisplayData software which automatically accounts for the change This makes things a little awkward if data are reduced with a spreadsheet since on a cool day the temperatures aloft may drop below freezing producing a data set that must be reduced with two sets of equations The following comments are directed more to folks interested in playing with spreadsheet data reduction For the type T type thermocouple I have gotten around this inconvenience by fitting the tabulated standard data from 100 F 73 C to 200 F 93 C with a fairly simple equation that is accurate to within 0 5 F 0 3 C over this range It is accurate to within 0 1 F 0 06 C from PF 16 C to 174 F 79 C This equation takes the form T A mv C B Eq 1 where mv Thermocouple output in millivolts A 271 B 753 4 C 8 4 and T is in F I have not modeled the type K thermocouple with this equation because one it doesn t fit an equation of this form as well as the type T does and two I haven t been using the type K thermocouples much The Model 1 3 Thermocouple Sensor Board is essentially a linear DC amplifier This means that the output voltage V is related to the input voltage mv by a simple linear equation V M mv b Rearrangin
28. rst 100 values is greater than 20 6m the DisplayData program assumes that the altitude data are referenced to MSL Selecting AGL in the Select Elevation Reference box recalculates and displays altitudes relative to local ground level The units of temperature can be changed at will between F to C by appropriate selection of the Temperature Units buttons It is important that the correct units are selected before entering calibration values B1 10 Plot Title A title can be entered on the plot Under the Plot Properties menu item is the option to Enter Plot Title Titles up to 30 characters long will be displayed at the top of the plot B1 11 Plot Colors Colors used in the plot can be changed from their default values Colors of the Plot Background Plot Frame Plot Ascent Plot Descent and Plot Cursor can all be changed by selecting the desired option under the Plot Properties menu item Sorry your stuck with the colors on the rest of the window B1 12 Smoothing Data There is usually significant random noise in the altitude and especially the thermocouple data Display of the data is improved by smoothing the data using a running average technique The number of data points to be averaged depends on first whether ascending or descending data are being smoothed since the data density per foot meter of altitude is usually much sparser on the way up than on the way down Second data sampled at 200Hz or 100Hz can be averaged over a
29. s of this program can be written to accommodate additional types of thermocouples 16 B1 4 Enter Calibration Values The voltage and temperature values from the calibration procedure Appendix A must be entered before temperature plots can be produced Be sure the type of thermocouple selected matches the type used for the calibration See also section B1 7 Plot Units to set temperature units Select Set Calibration from the menu bar and enter the values Selecting Calculate will show the calculated values of m and b from Equation 2 above Selecting Ok performs the same calculation but promptly closes the dialog box Enter Calibration Values Voltage Temperature F SR SER First Measurement Second Measurement Calculated Parameters Figure B3 Entry of calibration values The procedure for measuring the calibration values is given in Appendix A For reference the calibration values are shown in the main window above the Exit button Note that calibration values cannot be edited in the main window Standard data do not need to be calibrated to be displayed B1 5 Select Data to be Displayed It is often useful to distinguish between the data acquired during ascent from those acquired during descent Checking the appropriate Ascent and Descent boxes in the Thermocouple and Standard channel windows determines which data are plotted In the example in Figure B4 both the Ascent and Descent thermocouple data are
30. t thermocouples can form strong bonds by simply fusing their component wires together in a i flame The size of the weld bead Welded see Figure 7 needs to be Type K minimized the thermal inertia of the bead limits the response time of the thermocouple while at the same time making it large Figure 7 Thermocouple junctions made from 0 005 130um wire enough to mechanically withstand the rigors of the environment to be measured This simple technique works for the type K thermocouple shown above I have found welding type T thermocouples more difficult A bead can be formed but when made with 0 005 wire the copper lead is very weak and breaks at the bead with even a minimum of handling I have resorted to joiningmy type T thermocouples with solder This may make purists shudder because this introduces additional metals Pb and Sn of dubious purity into the junction However so long as the junction is isothermal the presence of these impurities will have little effect 3 3 Mounting a Thermocouple on a Rocket Several things need to be considered when mounting the thermocouple on the rocket and there is still a lot of experimentation that can be done to find the best way to do this Obviously the thermocouple junction needs to be mounted where it can sample the air outside the rocket The trick is to keep it from sampling other heat sources such as the thermocouple supports sunlight and the rocket body itself through
31. utput voltage This will correspond to 32 F 0 C and will be found right at the surface of the ice where it is in contact with the water Water away from the ice surface can be measurably warmer You now have all the data you need to calibrate your Model 1 3 sensor board Recall equation 3 V b T A C B Eq 3 M Given two pairs of temperature and voltage measurements T V and T V this equation can be rearranged and solved for M de AMY T v Eq 4 T K B 7 T B 2 b v 4 2E Eq 5 The Excel spreadsheet I used for data reduction has equations 4 and 5 built into and then b it I just plug in the temperature and voltage pairs and the spreadsheet does the rest The DisplayData program also accepts the same calibration data The Model 1 3 sensor board is designed to be mechanically rugged and electronically stable that is why trimming potentiometers and dip switches were omitted from it s design so the calibration obtained in this way should be valid for a considerable length of time Intrinsic accuracy of this technique is limited by the 0 8 F accuracy of the LM34CAZ reference the purity of the thermocouple wire and the intrinsic error of equation 3 relative to the actual response only about 0 1 F in this temperature range Additional error might creep in if the board was changing temperature during the calibration 4 APPENDIX B DisplayData Software l B1 DisplavD DisplayData D
32. wider interval than 50Hz data Third personal aesthetics or experimental needs may dictate more or less smoothing 20 To change the smoothing intervals from their default values select Smooth Data from the menu A dialog box will appear allowing you to specify smoothing intervals for the ascending and descending altitude and temperature data B1 13 Saving Plots The contents of the Plot Window not the whole window may be saved as a bitmapped graphic file Select Files from the menu then Save then Plot A dialog box will appear allowing the location and name of the plot graphic to be specified B1 14 Saving Data The raw and processed altitude and selected temperature data can be stored as a csv file for further work in a spreadsheet program Temperature data are stored if either the ascending or descending data for the temperature source thermocouple or standard are checked for display I have found that csv comma separated value files are much easier to import into spreadsheets than the space delimited text files exported by the RDAS software B1 15 Saving Settings After you have gone to all the trouble of setting up the DisplayData program with your preferred RDAS data channels altitude and temperature units calibration launch elevation and plot colors it seems a shame to lose it when you close down the program Well you don t have to When you have things just the way you like it under the Files menu option select S
Download Pdf Manuals
Related Search