Okay, now, with that wiring you're not going to get anything on the computer program, because you don't have the wires connecting the 3202 with the BS2, and you don't have any power going to the 3202. You knew that, right?
I'm not familiar with the kind of pot you have there, but I assume you were able to identify the right leads from the packaging, so you have the Vdd and Vss going to the two "outer" leads (as I described them) and the yellow wire is coming from the wiper. It looks to me as though you have four wires going to the pot - I'm not sure why that would be.
I have to pack up and get out to the launch, and probably won't be able to check in again until tomorrow. I hope you have a productive day - I look forward to seeing the progress tomorrow.
Yes, the two 'outer' wires on the pot go to Vss and Vdd on the BOE. Vss is GND and Vdd is +5 volts. That gives power right? The yellow wires goes from the wiper to CH-0
We'll (that is, THEY) will get it figured out at 1 PM if I don't.
Thanks for the early morning tutorial! (yawn) Go launch rockets, and keep the pointy end up as they say! I'm eager to hear all the details of today's launch.
Before I leave for practice, could you tell me the format in which the numbers for the voltages came up. Where they in labeled columns, rows, or just number separated by commas?
This is a response to your question that you asked us a few days ago, we are very sorry we are late. We copied and pasted the program you posted, which was fairly straightforward, and when we ran the program on our homework boards, we got these results:
When we multiplied 100,000 by 1.2207 we were surprised to get an answer instead of an overflow error. However, it multiplied 34464 by 1.2207 instead of 100,000 by 1.2207. We understand that it probably gave us an answer instead of an error because the STAMP used simple logic, and it computed 1*1.2207 and added the zeroes on afterwards. However, we are wondering why it decided to use 34464*1.2207 instead of 100000*1.2207? We realized that the last equation (100,000*1.2207) was different on the STAMP then the calculator. We are very confident that the calculator was correct, but why wasn't the STAMP? We also noticed that the STAMP rounded the numbers. All of our answers on the STAMP were rounded up to the next number except when it would go above 53687, which resulted in an overflow error.
Mike and Sean
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ "Learn to obey before you command."
I tried to explain the overflow error and why the Stamp turns 100000 into 34464 a couple of days ago, in this message: http://forums.parallax.com/showthread.php?p=915797
I know it is hard to keep track of all the posts in this thread! PBasically, it comes down to the fact that the BASIC Stamp can only work with numbers up to 16 bits, which is numbers in the range of 0 to 65535. You know that 216 = 65536?! Memorize that number! If you want to continue to work with computers, it is a number you should learn to recognize instantly, and the whole series, 1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024, 2048, 4096, 8192, 16384, 32768, 65536. You should learn to write down that series quickly from memory. See the number 4096 there? That is the binary resolution of the MCP3202 analog to digital converter. 212=4096
The program had no trouble to identify the numbers between 53688 to 65535 as overflow (when multiplied times 1.2207), but it actually takes a more complicated PBASIC program to detect that someone enters a larger number, such as 100000, and flag it as an error.
Did you get the MCP3202 running?
I'm still busy today, setting up a system with 49 temperature probes to monitor the grassland habitats of wolf spiders.
As to world cup. My wife is Italian by birth, and her brothers expound the minutiae of the play and of the players. One of them is married to a Brazilian, so guess that there might be a divided household there. Sorry Sylvie, I love Paris, I love Quebec, but Italia Italia! Here we are nearer Mexico, so there was a lot of enthusiasm for the Mexico-France game. The French team on the field clearly thought Hernandez was offsides on that one disputed goal, but I think it was a fair call. I have trouble understanding the offsides rules, and the referees are so often capricious in applying them, as in the U.S.A. tie with Slovenia.
What an excellent practice we had today, probably one of the best we've ever had! The Rocketeers were working together like a well-oiled machine. There was some excellent dialogue and collaboration today, and some really excellent leadership by Andrew.
Yes, we did get the 3202 going and it works just fine. **It's amazing how well it works Sylvie, when I connected the wires from the ADC pins to p0, p1, p2, and p3...!. Duh. Thanks for leading me into discovering it on my own and not simply telling me: "Hey dummy: connect the wires!" We actually had five 3202's going at once. Each Rocketeer was asked to read the schematic, find and download the sample ADC program code on the Parallax website, wire their BOE proto-board, and then run the program to show that it worked. Then they were asked to make the voltage and "4096" appear on the computer screen from CH1 instead of CH0, and to make words appear on the (DEBUG) screen explaining what was happening.
Next, Andrew and Dylan worked with the ASP program and reassigned some of the pin-outs while Mike, Sean, and I ran wires from the ADC mounted on the stand-alone breadboard to a clone of the ASP. I won't go into great detail about which wires went where except to say that we ran them to the available pin-outs on the BOE and to Vss and Vdd, the same way they were wired on the ADC/ BOE proto-boards.
**They actually got the voltage and its corresponding "4096" digital number** to read out from the ADC onto the ASP's data screen... progress in our book! They were giddy with excitement! I was too, not simply because "it worked" (albeit marginally), but because the Rocketeers UNDERSTOOD and were able to explain what they had done to make it work.
Expect all the Rocketeers to post to the forum in a little while. I'm curious to see what they report and what questions they have. I'm going to clean up the carnage they left on the dining room table, then light the grill like Susanne asked. Tonight Tracy and Paul, we have celebratory STEAK! Happy Father's Day!
Today's practice was very productive! We accomplished a lot like we do at every all team practice because we come prepared. Each of us now has our own individual·BOE set up as follows:
CS to p0
Clock to p1
DataOut to p2
DataIn to p3
Vss uses a jumper wire to connect to the capacitor
Vss also connects with the Vss wire that connects the potentiometer to the BOE
The wiper is connected to Ch1 (but on the ASP it will be connected to Ch0)
A wire going from one of the leads on the potentiometer to Vss
A wire going from the other lead on the potentiometer to Vdd
All of us also have a potentiometer set up (thank you Mr. Kibler for the kit!) and an MCP3202 set up beside it (thanks for the ADCs Andrew!). Mr. Kibler will be uploading a picture of it.
We then set up a similar setup on the protoboard. This time we connected the wires from the protoboard into the p slots on the BOE on the robot with treads. This takes up pins 0, 1, 2, and 3. The only way we can think of to fit thosewires that were taken out would be to put them in pins·6, 12, 13 and 14 (6 and 14 were already free and 15 is for the MAWD).·This is what we did for testing purposes (but not permanent). CS went to p6, clock went to p14,·DataOut went to p13, and DataIn went to p12. We·shared pins for Din and Dout·afterwards. We also redefined the I/O definitions and commented out any refrences to the LED in the code·for the time being. We ran what will·eventually evolve into our·final program and the·output of data was scrambeled.·However, we did make good progress on the program today in addition to everything else.
Questions:
In the long run, we want to still·be able to have the LED on the BOE (which takes up pins 12 and 13). Is there any way we could do this AND have the rest of the other wires connected?
We are getting every closer to the end at good team meetings like this one. Dr. Allen, thank you for all your help, especially with the programming! Also, thank you to Sylvie, who has graciously helped us with the wiring!
Where to do suggest we go from here with the programming and wiring?
Attached is a scanned picture of the wiring layout for the analong-to-digital chip (ADC 3202) with a potientimeter. Also atatched is the sample code* (in Word and program code formats) that runs it. *From the Parallax website and from the ADC 3202 documentation sheet.
"If you build it, it will run..." - Field of Dreams
Today our practice was very productive, perhaps it was more productive than some of the practices last year! We had finished so many tasks that were needed to be done is such little time. We made execellent progress today. Each team member displayed much effort today staying on task and working.
First, we answered your questions from early(previous post). We did it such little time that what we had expected to do it in. After we set-up our homework boards and we wired our MCP3202 to it. We used the diagram of how the MCP3202 is set-up. For instance where the Vss is and the Vdd. On the same board, we wired a potentiometer to the breadboard also. We then used the sample program for the MCP3202·(please correct me if I am wrong) and this is how we knew that we wired the MCP3202 and the potentiometer correctly beacuse the program worked. Now, each team member should have those things on their homework·board or at least know how to set it·back up.·After that, we tried to recreate what we did on our homework boards to a single breadboard. We had set it up correctly and we had wires running from·the single breadboard to our prototype. We commented out the LED light and took it of our prototype so we could fit our neccesary wires on it to work.·We wired the wires up to all different pin#'s and redefining them in the program. We were "messing" with the program trying to get it all to work. But unfortunately, we ran out of time and we were not able to finish what we had started with the program.
We think it would be nice to keep the LED on the BOE. We like to have it there beacause it helps us knoe whether the program is running correctly or not. Is there a way that we could have the LED still on the BOE, and still be able to have everything needed on the BOE also?
Also, does it matter which pin # a specific wire goes into? For example, would it matter if the wire coming from CLK went it into P2 or say P9?
Where should we go from here. Thank you all for your time and help.
What an excellent practice we had today, probably one of the best we've ever had! The Rocketeers were working together like a well-oiled machine. There was some excellent dialogue and collaboration today, and some really excellent leadership by Andrew.
Yes, we did get the 3202 going and it works just fine. **It's amazing how well it works Sylvie, when I connected the wires from the ADC pins to p0, p1, p2, and p3...!. Duh. Thanks for leading me into discovering it on my own and not simply telling me: "Hey dummy: connect the wires!" We actually had five 3202's going at once. Each Rocketeer was asked to read the schematic, find and download the sample ADC program code on the Parallax website, wire their BOE proto-board, and then run the program to show that it worked. Then they were asked to make the voltage and "4096" appear on the computer screen from CH1 instead of CH0, and to make words appear on the (DEBUG) screen explaining what was happening.
Next, Andrew and Dylan worked with the ASP program and reassigned some of the pin-outs while Mike, Sean, and I ran wires from the ADC mounted on the stand-alone breadboard to a clone of the ASP. I won't go into great detail about which wires went where except to say that we ran them to the available pin-outs on the BOE and to Vss and Vdd, the same way they were wired on the ADC/ BOE proto-boards.
**They actually got the voltage and its corresponding "4096" digital number** to read out from the ADC onto the ASP's data screen... progress in our book! They were giddy with excitement! I was too, not simply because "it worked" (albeit marginally), but because the Rocketeers UNDERSTOOD and were able to explain what they had done to make it work.
Expect all the Rocketeers to post to the forum in a little while. I'm curious to see what they report and what questions they have. I'm going to clean up the carnage they left on the dining room table, then light the grill like Susanne asked. Tonight Tracy and Paul, we have celebratory STEAK! Happy Father's Day!
More later,
Mark
Team members and others,
Thanks for remembering to post your thoughts about today's team meeting! I'd like to elaborate on Mr. Kibler's previous post by uploading the program we worked on today, which is a mashup of the MCP3202 sample program and our functional MAWDBOE w/ CO2 program. It certainly isn't a work of art, but from what we were able to see on the debug window, it was rather promising! It isn't fully functional, nor did we have the time to try to integrate it with the data logger -- but it DOES read out a voltage, as Mr. Kibler said, which seems to be a big step forward in my opinion. I will take a second look at what we tried to accomplish today, and see if I can clean it up, but my knowledge of programming will only be useful to a certain extent.
Here's another picture of the same ADC 3202 wiring layout I just posted above, only in "GIF" format. Your computer may, or may not, run the previous file (in .tif format.)
So reading everybody else's posts, I don't really think there's much left to say about our meeting other than it was one of the best ones we've had so far!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ "Learn to obey before you command."
Dear Project Team,
Sorry for not posting during the Early evening. I was out and now is the first time that i have had access to the internet. I would of enjoyed to look back at our early practice, but it seems that every thing has been covered by my teammates. I'm not sure if I am getting ahead of myself, but I think our next logical step is to understand the wiring at a further extent. Our team this afternoon wired it in multiple ways. The first time we wired, numbers were outputting. What we did was change the points that the wire connected to. The wire being the one that connect to one of the chips pins to the wiper. The reason to be was that the wire that was connected to the wiper, was also connected to the chips pin that the program collected the voltage from. It seemed logical to connect the pin from the CO2 sensor that outputted the voltage depending on the current CO2 levels, to the pin on the chip that collected the voltage (If you think about it, the program outputted a number depending on the voltage that the potentiometer outputted after it went through the changes (In this explanation, a variable resistor), so your just replacing the potentiometer part that outputted the changed voltage, with the CO2 sensor's pin (CNTL) that outputted the changed voltage from the CO2 sensor. (In this case the,"Change" was the amount of CO2 in the air)).
I am not sure if any of you will understand this but I put it in the best way I can.
The second way we tried was with Sylvie's, "shorting" wiring scheme. I believe the same thing happened but I will need to confirmation on that. But then I remember that Mr. Kibler switched some wires around too quickly for me to record what was happening. (I remember Andrew giving me weird looks for we where both afraid what was going to happen.)
The things we will also have to work on is programming to make the CO2 number, (In this case the 4096) appear in columns. When we ran it the number pretty much went where they wanted to go (even before the initializing word!). Also we dearly need to mount all of the components (extra breadboard, CO2 sensor, etc..) on an secured location. A loose wire could cause ALOT of problems for us in this point.
It is late and I cannot remember if I am missing something very important. I just need guidance on how we will go about tackling our next steps.
It sounds like everyone had fun and you learned a lot yesterday.
I do have possibly very stupid question - I see a lot of references to "4096" in the posts. That is the reading you should get when the input voltage to the 3202 is 5 volts. More accurately, it should output 4096·whenever the voltage on pin 2 (input Ch 0) is equal to the voltage on pin 8 (Vdd/Vref - the reference voltage). But you should get a different number from the 3202 when the voltage on Ch 0 is less than Vref. For example, if you're sending 2.5 volts to Ch 0, you should get a reading of about 2048, and if you're sending 1.25 volts to Ch 0 you should get a reading of about 1024 (and so on). If Ch 0 is connected to ground (0 volts), you should get a reading of 0.
Is that happening? When you adjust the pot, do you get different readings on the output from the sample program?
Have you measured the voltage on the 3202's pin 2 to see that it is roughly the same as the voltage reading on the screen?
Your wiring diagram looks right to me (except it's upside-down <grin>).
If all of that is working, then you've achieved something that I find very exciting - the ability for your BS2 to know the value of a voltage from the outside world. A LOT of sensors send variable voltages to reflect the sensor reading, and with this setup, you now are able to read those sensors (with some minor adjustments, perhaps).
==================================
(** See my next post below - I was mistaken about some of this, I think. I'll leave it up, though.·**)
I do still wonder if Dr. Allen plans for you to do something to change the range of the 3202 in order to make better use of the output range of the CO2 sensor. I can't get to the product page right now - something seems to be down at the Parallax site - but I seem to remember that the sensor sends out voltages from 0 to 3.2 volts, or somewhere around there. In other words, it won't put out voltages as high as 5 volts, so everything from 3.2 volts (or whatever that upper end number is) to 5 volts is simply wasted range.
(* Note added: Okay, I can get to the product page now. It's 3.35 volts. *)
Right now the resolution of the 3202 is 5 volts/4096 steps = 1.22 mV per step. If you were to run the 3202 with a 3.2 volt power source on the Vdd/Vref pin, that resolution would be increased to 3.2 volts/4096 steps = .78 mV per step. Once you convert that to ppm of CO2, you'd have significantly more precise numbers to work with as compared with using the 5 volt reference. The cost of doing this would be a couple of extra parts to change the voltage on the Vdd pin - I think just a couple of resistors would do it. On the other hand, it might be the case that it would introduce inaccuracies because of the tolerances of the additional parts, and perhaps because their values would change with the temperature. This is really Dr. Allen's area, and it might have been better for me to not even mention it, though I don't think it's too confusing.
In addition, it may be that the upper end of the CO2 sensor's scale represents CO2 values that you simply won't come across during your flight. If that's the case, then you could go with an even lower Vref, cutting off those upper values that you'll never see anyway, and getting even better resolution (though of course again there's going to be a limit to how good it'll get because of the various sources of error in the wiring, in the stability of the power source, and so on). For example, it's possible that the only way you'll ever get voltages over 2 volts out of the CO2 sensor is to blow pure CO2 onto the sensor. If that's the case, you could use a Vref of 2 volts on the 3202, knowing that you're simply not going to exceed that value in flight, and have a resolution of 2 volts/4096 steps = .488 mV per step, or 2-1/2 times better resolution than you're getting with the 5 volt Vref. If you made these changes, you would of course have to change your program slightly where it calculates the CO2 content from the voltage reading. It'd be a matter of simply using a different number in that mod (**) step, to divide by .78 or .488 or whatever instead of by 1.22.
I imagine this this is part of why Dr. Allen started off with some questions to you about the values of CO2 that you could expect to find during your flight.
Let me add that it will work just fine without any of this - there's nothing wrong with sticking with the 5V reference, and it's quite possible that making the changes I'm talking about here won't get you anything worthwhile. I'd like to hear Tracy's input.
Post Edited (sylvie369) : 6/21/2010 10:23:12 AM GMT
It sounds like everyone had fun and you learned a lot yesterday.
·if you're sending 2.5 volts to Ch 0, you should get a reading of about 2048, and if you're sending 1.25 volts to Ch 0 you should get a reading of about 1024 (and so on). If Ch 0 is connected to ground (0 volts), you should get a reading of 0.
Is that happening? When you adjust the pot, do you get different readings on the output from the sample program?
Have you measured the voltage on the 3202's pin 2 to see that it is roughly the same as the voltage reading on the screen?
Yes, we really had a good practice yesterday and the Rocketeers seemed to enjoy it... especially Susanne's fresh molasses cookies!
You asked:
1) When you adjust the pot, do you get different readings on the output from the sample program?
ANSWER: Yes, on both channels, but not at the same time. As an exercise I had the Rocketeers rewire and reprogram the BOE so that voltage and numbers (0-4096) appeared from either channel 0 or 1. When they changed the voltage (dialed the potientiometer) the number changed correspondingly from 0-4096 on whichever channel they had wired/ programmed. Good learning. · 2) Have you measured the voltage on the 3202's pin 2 to see that it is roughly the same as the voltage reading on the screen? ANSWER: Yes they did measure the voltage output and the voltages were comparable on the voltmeter and the computer.
Christopher and I are in Washington, DC right now and he has freshman orientation the next three days at George Washington University. And of course I brought the·ASP/ BOE·with me! I carried it through security in its red plastic toolbox so things wouldn't get jounced around inside my luggage. No problems. Everyone was curious about it and Christopher did a thorough job of explaining what it was. The Rocketeers fly out of Manchester often enough and they're in the news enough that one of the TSA folks actually remembered·us. That was helpful.
Speaking of which, how did your launch go?·How did your electronics work?·Did you get all the parts back intact? If the weather was anything like it is here, it was glorious. DC is 90 degrees and hot and muggy. Have you ever see a tornado out there?
Les francaises sont desoles! Qu'est-ce se passe avec·"les Bleues"...? Quel horreur! I see too, that Tracy is a a fan of "Italia! Italia!" Boy, if the French team ends up playing Italy, who do we cheer for? Not that I'm an avid "football" fan, but I have taken an interest in the World Cup·of late . I saw that Tracy's team, Italy had a draw (tie? null set match?) with New Zealand, who they were expected to beat hands down (or rather, feet down.) All this excitement, and Brazil beat even Cote d'Ivoire. How is the American team faring?
Tracy, see attached photo.
Our plan with the AD converter is to get it wired on the stand-alone breadboard (now that we know it works) and then connect wires from the AD converter·pins to the·ASP/ BOE. We did this right at the end of practice yesterday·and the we ran Andrew's "mashup" cut-and-paste program (which he uploaded below.) It actually output the voltage and the corresponding "numbers" (0-4096) on the ASP's data screen when we ran it, though the data and column heading were scrambled. This seems like progress and it seems like the program is headed in the right direction. It was amazing how well the AD converter worked when we connected the wires to BOE pins p0, p1, p2, and p3... Duh!
Andrew has a complete ASP-BOE-CO clone set-up at home and I have mine here in DC. So we'll wire and program via the forum, as will the other Rocketeers. Right guys?
Tracy and Paul, what do you see as·the next logical steps? Do you think we're on the right track? More later
I do still wonder if Dr. Allen plans for you to do something to change the range of the 3202 in order to make better use of the output range of the CO2 sensor. I can't get to the product page right now - something seems to be down at the Parallax site - but I seem to remember that the sensor sends out voltages from 0 to 3.2 volts, or somewhere around there. In other words, it won't put out voltages as high as 5 volts, so everything from 3.2 volts (or whatever that upper end number is) to 5 volts is simply wasted range.
Right now the resolution of the 3202 is 5 volts/4096 steps = 1.22 mV per step. If you were to run the 3202 with a 3.2 volt power source on the Vdd/Vref pin, that resolution would be increased to 3.2 volts/4096 steps = .78 mV per step. Once you convert that to ppm of CO2, you'd have significantly more precise numbers to work with as compared with using the 5 volt reference. The cost of doing this would be a couple of extra parts to change the voltage on the Vdd pin - I think just a couple of resistors would do it. On the other hand, it might be the case that it would introduce inaccuracies because of the tolerances of the additional parts, and perhaps because their values would change with the temperature. This is really Dr. Allen's area, and it might have been better for me to not even mention it, though I don't think it's too confusing.
In addition, it may be that the upper end of the CO2 sensor's scale represents CO2 values that you simply won't come across during your flight. If that's the case, then you could go with an even lower Vref, cutting off those upper values that you'll never see anyway, and getting even better resolution (though of course again there's going to be a limit to how good it'll get because of the various sources of error in the wiring, in the stability of the power source, and so on). For example, it's possible that the only way you'll ever get voltages over 2 volts out of the CO2 sensor is to blow pure CO2 onto the sensor. If that's the case, you could use a Vref of 2 volts on the 3202, knowing that you're simply not going to exceed that value in flight, and have a resolution of 2 volts/4096 steps = .488 mV per step, or 2-1/2 times better resolution than you're getting with the 5 volt Vref. If you made these changes, you would of course have to change your program slightly where it calculates the CO2 content from the voltage reading. It'd be a matter of simply using a different number in that mod (**) step, to divide by .78 or .488 or whatever instead of by 1.22.
I imagine this this is part of why Dr. Allen started off with some questions to you about the values of CO2 that you could expect to find during your flight.
Let me add that it will work just fine without any of this - there's nothing wrong with sticking with the 5V reference, and it's quite possible that making the changes I'm talking about here won't get you anything worthwhile. I'd like to hear Tracy's input.
Paul,
· This makes perfect sense and I recall where Dr. Allen talked about this is a previous post. I suspect that we may find the answer, or part of it, in the "mV voltage-CO2 concentration" graph he asked us to look at, way back on some of the first pages in the forum. It shows the inverse realtionship between the two but more importantly it shows which millivoltage = what CO2 concentration. He also explained and then had the Rocketeers calculate us how much CO2 by volume = how much CO2 ppm (5% - 50,000 ppm or thereabouts, as I recall.) Together, this information should help us derive an answer about·the maximum·voltage/ millivoltage we will need to measure the·range of CO2 we're likely to encounter. Though I don't have the·"other" graph handy-- the one·that we posted that shows·the correlation between altitude and CO2--·I seem to recall a CO2 concentration of somewhere around 300-400 ppm up through 20,000 feet or so. I wouldn't think that CO2 concentration would exceed 1,000 ppm (1% by volume) between ground level and 20,000 feet, the highest altitude we'd encounter (last year the ASP went to 14,000 feet +/-) I would expect it to be much less than this but I think the ASP-2 should be designed and programmed to record a higher concentration... just in case we happen to encounter a higher concentration of "outlying data."
Does this makes sense and seem on-track?
Tracy, what's the research you're doing with wolf spiders? Why the interest in correlating temperature and wolf spider habitat (with 49 temperature probes nonetheless...!)·Are·they an indicator species?
Rocketeers, please check and see what CO2 concentration we're "likely" to encounter at ground level up through 25,000 feet altitude.
Okay, first of all, a "doh!" on my part. I wrote that the only thing that would need to be added to use a lower voltage (and get more resolution) on the 3202 would be a couple of resistors. But I completely forgot that the 3202 still has to communicate with a 5V BoE, which means that there would, I think, also need to be some kind of level shifting for the three data lines (CS, CLK, Din/Dout). That right there probably means it's not worth the trouble.
I'd spent about ten minutes writing my previous post, and within 30 seconds of getting up to so something else, I realized it was a mistake. The nonconscious mind can do wonders sometimes. And then, of course, I managed to get distracted and forget to come back and post about what I'd realized.
=====================
Okay, good to hear that you have the 3202s working completely. Isn't it exciting to see that stuff work? As a base for sensors, you've just made your BoE twice as useful.
I think that you should sit tight for now as far as the device goes, until you heard from Tracy about the next steps. The obvious thing is to connect the CO2 sensor's TP1 to the 3202's Ch0 (pin 2), and see if you can get voltage readings that change with the CO2 content. It might be interesting for you to see what kinds of numbers you get in normal air and exhalation.
=====================
I don't think there's much chance that France will face Italy this time around. Italy might advance, but France is done. Serait necessaire quelques miracles. Espana tambien, y entonces, quisiera ver los equipos de Chile y Mexico.
I did not wind up flying a rocket yesterday. The wind direction was good, but it was far too strong for me to put all that electronics at risk. Instead I supervised a Junior Level 1 certification and an adult Level 1. The Junior one was quite the adventure. The young man flew a Wildman Jr. fiberglass rocket on a Cesaroni I800 VMax. Blink and it's gone - 4500 feet up. It opened the chute just fine - at apogee (no electronics allowed for junior certs), and drifted far off to the east. He had a tracker in it, but was getting only·a sporadic signal at best. Someone helping him track it decided that the rocket was moving - that someone had picked it up and carried it off somewhere (ranger station? back to the range? simply stole it?). So the young man eventually gave up looking, but set up another rocket with a Cesaroni I470. We redid the paperwork, and put it on the pad, and I started walking down towards the recovery area, and on a whim, turned my radio to the frequency of the first rocket's beacon - and got a strong signal. Seconds later I heard the kid yelling "Paul!", and I turned around, and there he was holding the Wildman Jr., in perfect condition. One of our fellow rocketeers had gone back out to look for it, and found it.
So he decided to fly the I470 anyway. It lit, but cut off FAR too quickly - even from my distance, there was obviously something wrong, and I wasn't at all surprised to see it arc over and lawn dart right into the ground. I never did see the remains, but I have to believe the motor CATO'd. However, the young man got his L1, because the Wildman Jr. was just fine. Quite a roller coaster ride for him.
Rocketry sounds very exciting! It is not something allowed close to where I live, a population center famous for dry summers and big fires. Black Rock Desert is about 10 hours distant!
Now I think it is time to look more closely at the Parallax CO2 module, to try to understand its schematic diagram. You can find the schematic diagram in the Parallax documentation, and I encourage all of you to find it there on page 4 and to read the description, "theory of operation", on page 3.
27929-GasSensor-v1.0.pdf
The module consists (as is usual with electronic circuits) of several blocks that each serve a specific function. Here is a picture with the functional blocks numbered. If you have a module in front of you, you can identify the different parts on the actual circuit board, and Parallax makes it easy to find the parts by looking at the diagram that they provide on page 2 of the PDF.
1) A voltage regulator, it takes Vin in the range of 6.5 to 15 volts that you supply from the battery, and in turn supplies an output of exactly 6.0 Volts. That is the power that is needed to run the heater in the CO2 sensor. (Remember the heater?!)
2) A second voltage regulator that provides 3.3V output to operate the other active components, 5 & 6, everything but the heater. Note that the input for the 3.3V regulator comes from the 6V regulator.
3) This is a transistor that acts like a switch. When the CNTL from the BOE is at zero Volts, the switch is open, non-conducting, and when CNTL is high (5V) under program command, the switch is on and current can flow to operate the heater in the sensor.
4) The actual CO2 sensor. See how one side of the heater is connected to +6V, and the other side is connected to the switch? The NASICON ceramic tube that surrounds the sensor is attached to ground on one side. See the ground symbol, three horizontal lines in a kind of triangle? The right hand side of the NASICON is where the sensor output voltage appears when the heater is hot.
5) The sensor output is applied to an input of part number LMC6035, designated U2A. It serves two purposes. One is to act as an electrical buffer, to sense the voltage from the NASICON without drawing any electrical current that would disturb the equilibrium of the chemistry. The LMC6035 input draws only a few femtoAmperes (10-15, well below what the sensor manufacturer says is allowed. The LMC6035 is also set up to amplify the Voltage output of the sensor.
The amplification factor is set by the two resistors, one of value 150 kΩ and the other of value 1 MΩ. The gain factor is,
1000000/150000 + 1 = 7.67
Any Voltage that you measure on the test point, TP1, will be 7.67 times the actual output of the CO2 sensor.
6) The output of U2A is passed to the test point, TP1, and also to an input of U2B. U2B is actually in the same package as U2A. The LMC6035 is a dual operational amplifier (op-amp for short), two independent op-amps in one 8-pin package, a miracle of integrated electronics. It is an "analog computer" that can be configured for many different purposes depending on how you hook it up with outside components. Part U2A is hooked up as a buffer/amplifier, and part U2B is hooked up as a comparator. The output from U2A is compared with the Voltage you set on the potentiometer dial R3. See the (+) and (-) inputs on the left side of the triangle U2B? The output of U2B at the tip of the triangle is high (5V) when voltage on the (+) input is greater than the voltage on the (-) input. The LED turns on and the alarm on the BOE shows a "1".
7) You know by now, that is the 4-pin connector from the sensor module to the outside world. Also, you know that in order to get the signal you want, you are going to have to bring an extra wire over the the test point, TP1.
We are going to have to address the problem that Sylvie brought up. Is there going to be a change in the CO2 level that we can detect well with this setup, or are we going to need further modifications?
2) A second voltage regulator that provides 3.3V output to operate the other active components, 5 & 6, everything but the heater. Note that the input for the 3.3V regulator comes from the 6V regulator.
Question: Is that because stepping down from 6V to 3.3V produces less heat than stepping from whatever the input voltage was to 3.3V?
I have a device I made that uses both 5V and 3.3V, and I run a 9V battery to a 5V regulator, and then from there to a 3.3V regulator. It draws a fair amount of current, but runs nice and cool.
I have a similar device that draws about the same amount of current, but it's all going through just a 3.3V regulator. When I power that from 9V, the regulator gets pretty warm.
Tracy Allen said...
5) The sensor output is applied to an input of part number LMC6035, designated U2A. It serves two purposes. One is to act as an electrical buffer, to sense the voltage from the NASICON without drawing any electrical current that would disturb the equilibrium of the chemistry.
Question: Is that what the word "buffer" means in this context, then? That it senses voltage without drawing current? That was on my list of questions to ask when we got back around to this. I never would have guessed that, but in retrospect, of course, it makes perfect sense.·This gives me all sorts of ideas.
Tracy Allen said...
We are going to have to address the problem that Sylvie brought up. Is there going to be a change in the CO2 level that we can detect well with this setup, or are we going to need further modifications?
Excellent. I think I know how to answer this question now. I'll be very interested to see·what the Rocketeers come up with here.
Rocketry IS exciting, and often, so is teaching. In the last few days I learned that a colleague for whom I wrote a recommendation was awarded the annual Undergraduate Teacher of the Year award by the Society for the Psychological Study of Social Issues. Today I learned that one of my former undergraduate students, one who took my physiological psychology course as well as my research methods course has been accepted into the post-graduate program in Brain Science at Seoul National University. I am told that's the most selective placement in Korea. I cannot begin to say how happy I am about all of this, and how proud I am to be able to work with people of this quality (you guys as well!).
·· I'm not even sure where to begin. So many curious things happen when I·wire the ADC chip pins to the proto-ASP/BOE·(which still has servomotors mounted) and then·run·Andrew's·"mashup" program. The ADC chip is·installed on a stand-alone breadboard just as it was at Sunday's practice, not on a BOE.
Here is where the wires go from the ADC pins on the breadboard:
Vss pin to········· Vss slot on BOE
Vdd/VREF to······ Vdd on BOE
1) When·the program runs,·the P14 servo activates and·the servomotor/ tread moves...·a·short, rhythmic·pulse·at time. The "movement" subroutine that runs the servos is commented out and so the servos "shouldn't" run. I unplugged both servo motors #14 and #15.
2) When the program runs,·elasped time, sample number, CO2 alarm status (1 or 0), temperature, and humidity do appear. They continue to change accurately as the program runs but they appear in different location on the screen when the two voltage readings appears. Though the data is accurate, it no longer appears in columns. It appears on on line only, but still changes accurately.
3) Two voltages appear, along with two corresponding sets of numbers between 0 and 4096. They change somewhat rhythmically but the changes are NOT in sync with the servomotor pulses (which stop when the servos are disconnected.)
4) When I remove the wire from the ADC's Vdd pin to Vdd on the BOE, the program still runs and outputs data. I left it disconnected for now.
5) When I remove the wire from the ADC's Vss pin to Vss on the BOE, we lose all voltage and the corresponding '0-4096'·numbers.·I left it connected.
I think I understand some of the many things that have to happen·(*that's our first problem... "Cogito ergo confuso".) But there are so many "layers" of things that have to happen, and in the correct sequence, that I'm not sure where to begin.
1) We have to wire the ADC chip into the ASP/BOE correctly, and permanently, so that everyone has the same "clone" at home. I'm not sure that things are wired correctly right now (point: the wire from CH0 isn't connected to anything...!) But we are getting voltage on the screen and last year's program still runs and outputs data despite the "bad wiring".
2) We have to assign ADC pins and pin numbers in the program. This assumes that step 1 is correct. Pins/ pin numbers are assigned now, but are they correct? Recall that that the Vdd wire and the CH0 wires are NOT connected and the program still runs.
3) We have to write·a program to convert CO2 millivoltage input into the ppm. I'm not even close to being there.
So what are the next (baby) steps? Too many factors are·superimposed on one another and I can't wrap my head around things. I having a tough time seeing where to logically begin, what to do first.
Out of excitement and frustration I·find myself·disconnecting and connecting wires to see what happens. But that is not good science, nor is it logical. It is simply trial-and-error. Although it might eventually succeed it isn't the best approach, and it's time-intensive. I read somewhere that if you put an infinite number of monkeys, at an infinite number of computers keyboards, for·an infinite amount of time, one of them will "eventually" write Shakespeare's 'Hamlet'. Anyone have a banana...?!
Now·about those wolf spiders. So why are you studying their habitat and temperature (with 79 sensors!) anyway?
Do we·connect the (currently) unconnected CH0 wire to TP1...?
When it is time to run a wire to TP1 (from wherever), will the wire simply be soldered to the TP1 hole on the CO2 sensor? That's a smaaaaaaaaaaaaaaaaaaaaal hole!
Should we include a resistor in the wire?
So many questions, so little time!··
ROCKETEERS, where are you...? What questions do·YOU have?
When the two voltages and their corresponding numbers output to the computer screen, the output IS rhythmic. It changes approximately every 1 second. There seems to be no correlation with the CO2 sensor's "alarm" status (1 or 0.) That is, when the alarm status changes the voltage data doesn't change. When voltage changes each second, the alarm status remains the same. Conclusion: voltage is being measured from somewhere else. The RTC chip...?
Mark in NH said...
When the two voltages and their corresponding numbers output to the computer screen, the output IS rhythmic. It changes approximately every 1 second. There seems to be no correlation with the CO2 sensor's "alarm" status (1 or 0.) That is, when the alarm status changes the voltage data doesn't change. When voltage changes each second, the alarm status remains the same. Conclusion: voltage is being measured from somewhere else. The RTC chip...?
Mark
I'm going to leave all of the rest of this to Tracy, because of the "too many cooks spoil the soup" effect (if two of us tell you how to write your program, you're doomed).
As for soldering to TP1, I think you'll be surprised at how easy it is. That being said, my soldering life changed forever when I discovered solder wick:
It's available at Radio Shack, at an elevated price, of course, or from Sparkfun (though they're currently sold out), or from one of the general distributors (Mouser.com, Digikey.com) at a good price. If you plan to solder things, you should plan to have some on hand. These days I find myself soldering pins that are 1/3rd the size of that little TP1 hole without too much difficulty.
Question: Is that because stepping down from 6V to 3.3V produces less heat than stepping from whatever the input voltage was to 3.3V?
Yes, the waste heat produced is proportional to the voltage dropped across the regulator. The FAN250033 that supplies the 3.3 volts is a small part, an SOT23 package, and it is not capable of dissipating much heat. If it gets to hot, it will shut itself down (which is preferable to going up in smoke!!).
Hypothetical:
Input of FAN250033 supplied from 6V at the input, 3.3V at the output, and 0.01 Amp to light the LED and run the LMC6035.
Power = (6 - 3.3) * 0.01 = 0.027 Watt.
Temperature rise is calculated (approximately) a the parameter thetaJA that can be found in the FAN2500 data sheet.
temperature_rise = 220 °C/W * 0.054 Watt = 5.9 °C
So if ambient is 25°C then the inside of the chip rises up not even to human body temperature. That's safe! The internal chip temperature has to rise up to 150°C before it shuts itself off. I leave it to the reader to estimate what would happen if the input to the FAN2500 came from a 12V supply instead of 6V and if the current required were 0.05A instead of 0.01A.
Question: Is that what the word "buffer" means in this context, then? That it senses voltage without drawing current?
Yes. It is impedance transformation. At the input, (finite Volts)/(zero Amps) = (infinite Ohms). At the output, (no change in Volts) / (any change in Amps) = (zero Ohms). You have to get into the math analysis to justify that, but that is what it amounts to. In general, a buffer isolates one circuit from direct connection to another. Buffers are common in digital electronics too, with a somewhat different meaning. For example, a "tri-state" digital buffer is one that can either connect or disconnect two circuits. In one state the buffer has infinite ohms resistance to isolate the circuits, and in the other state it has zero ohms to connect them.
As for soldering to TP1, I think you'll be surprised at how easy it is. That being said, my soldering life changed forever when I discovered solder wick:
So it looks like we'll be ultimately soldering a wire to TP1? And it looks like "solder wick" is the solder of choice. I've never heard of it and I've only used thick rosin-core solder with flux. It always seems to "run" where you don't want it to, and it large blobs. That's why I asked about soldering such a tiny hole. And then I think about all those tiny holes on the perfboard. How do you keep solder from·going every which way, into hole you don't want it to go into? Do you use a really tiny soldering tip (and a magnifying glass...?)
Tracy, speaking of "waste heat", we're looking into a solar (photovoltaic) array for our roof. If we don't feed excess electricity back into the grid, and if the storage batteries are fully charged, where does excess heat-energy from a PV array go? Is there·a "left-over electricity" device that dissipates or discharges(?) left-over voltage as heat-energy? (*not that we have excess sunlight·and solar·energy in New Hampshire. We don't. It's a theoretical question more relevant to Arizona I suppose.
Even though data on·the computer screen is scrambled when we run andrew's "mashup" program, the data still outputs nicely to teh flash drive/ Data Logger. No CO2 data (of course.)
Even though the code for the servos is commented out, the servos if they are connected will hear any activity on pin p14. The servos don't know that those clock pulses are meant for the ADC. They are just pulses.
I'll take a look at the program after work. My inclination is to connect as follows:
On the stand-alone breadboard, connect a 1 kΩ resistor from Dout to Din.
Connect a wire from Din to p6 on the BOE.
Connect a wire from CS\ to p12 on the BOE
Connect a wire from CLK to p13 on the BOE
Connect a wire from Vss onto Vss
Connect a wire from Vdd/ref to Vdd.
That leaves p14 free for the servo. Does the ASP-2 ultimately use the servo to wander around after touchdown?
There is not a separate wire for Dout. Now Din and Dout are serviced by the same pin, p6. This may or may not work. We'll see.
For the CO2 sensor, I suggest that you solder a length of wire to TP1 and another to TP3. (Chris?! Maybe they have a lab at GWU! There used to be lots of electronics surplus places around Washington DC)
Connect a wire from Vss on the CO2 sensor to Vss on the BOE
Connect a wire from Vin on the sensor to Vin (9 Volts) on the BOE
Connect a 1 kΩ resistor from ADC channel 0 to a free node on the breadboard and from there to TP1.
Connect a 1 kΩ resistor from ADC channel 1 to a free node on the breadboard and from there to TP3.
The connection to TP3 will let you measure the setting of the calibration pot on the CO2 module
NOTE: Power for the CO2 sensor needs to be at least 6.5 Volts!!! Connect to Vin (=9+Volts), not Vdd (=5.0Volts).
Even though the code for the servos is commented out, the servos if they are connected will hear any activity on pin p14. The servos don't know that those clock pulses are meant for the ADC. They are just pulses.
I'll take a look at the program after work. My inclination is to connect as follows:
On the stand-alone breadboard, connect a 1 kΩ resistor from Dout to Din.
Connect a wire from Din to p6 on the BOE.
Connect a wire from CS\ to p12 on the BOE
Connect a wire from CLK to p13 on the BOE
Connect a wire from Vss onto Vss
Connect a wire from Vdd/ref to Vdd.
That leaves p14 free for the servo. Does the ASP-2 ultimately use the servo to wander around after touchdown?
There is not a separate wire for Dout. Now Din and Dout are serviced by the same pin, p6. This may or may not work. We'll see.
For the CO2 sensor, I suggest that you solder a length of wire to TP1 and another to TP3. (Chris?! Maybe they have a lab at GWU! There used to be lots of electronics surplus places around Washington DC)
Connect a wire from Vss on the CO2 sensor to Vss on the BOE
Connect a wire from Vin on the sensor to Vin (9 Volts) on the BOE
Connect a 1 kΩ resistor from ADC channel 0 to a free node on the breadboard and from there to TP1.
Connect a 1 kΩ resistor from ADC channel 1 to a free node on the breadboard and from there to TP3.
The connection to TP3 will let you measure the setting of the calibration pot on the CO2 module
NOTE: Power for the CO2 sensor needs to be at least 6.5 Volts!!! Connect to Vin (=9+Volts), not Vdd (=5.0Volts).
Dr. Allen,
Thank you for your detailed suggestion. I would like to clarify that the ASP-2 will not be roaming around after touchdown, as our NASA SLI robot, "Phidippides", did. After touchdown, the ASP-2 will remain collecting (baseline) data at a stationary spot, as well as at the launchpad before launch, until we locate and recover it. I'm headed off to hot and sunny Florida tomorrow morning, with Phidippides, soon to become our newest ASP-2 clone.
To Andrew was say, "Bon voyage" and "We'll see you and Phidippides on the forum."·Have a great time in Florida, be sure to wear your sun screen... and be careful you don't end up in Brazil! Thanks for all you fine work.
Tracy, I'll wire the apparatus like you suggested and report back. I already had the Din-Dout 2-into-1 wire, ready. Freshman/ parent orientation was chocker-block full and I was ready to be back here at the hotel midway through the afternoon. I'd forgotten home hot DC gets! But it was an excellent orientation and Christopher is really excited to be on campus.
I'll have to track down the electronics surplus places you mentioned. Maybe they have "solder wick" and a miniature soldering tip. I knew I'd fogotten something back in New Hampshire: the soldering iron!
If you're only 10 hours from Black Rock desert you should join Sylvie and I and all the other Rocketeers for the ARLISS·launch in September. We'll buy the gas... and the rocket fuel. We already have·the hotel rooms reserved and I'm sure the three of us would manage just fine. How far are you from Reno? Do they have wolf spiders in Nevada?
Comments
I'm not familiar with the kind of pot you have there, but I assume you were able to identify the right leads from the packaging, so you have the Vdd and Vss going to the two "outer" leads (as I described them) and the yellow wire is coming from the wiper. It looks to me as though you have four wires going to the pot - I'm not sure why that would be.
I have to pack up and get out to the launch, and probably won't be able to check in again until tomorrow. I hope you have a productive day - I look forward to seeing the progress tomorrow.
Paul
Yes, the two 'outer' wires on the pot go to Vss and Vdd on the BOE. Vss is GND and Vdd is +5 volts. That gives power right? The yellow wires goes from the wiper to CH-0
We'll (that is, THEY) will get it figured out at 1 PM if I don't.
Thanks for the early morning tutorial! (yawn) Go launch rockets, and keep the pointy end up as they say! I'm eager to hear all the details of today's launch.
Mark and the Rocketeers
Before I leave for practice, could you tell me the format in which the numbers for the voltages came up. Where they in labeled columns, rows, or just number separated by commas?
Dylan Landry
This is a response to your question that you asked us a few days ago, we are very sorry we are late. We copied and pasted the program you posted, which was fairly straightforward, and when we ran the program on our homework boards, we got these results:
Stamp:
-1000*1.2207=1220
-10000*1.2207=12207
-4096*1.2207=5000
-40960*1.2207=50000
-53687*1.2207=65535
-53688*1.2207=Overflow Error
-100000*1.2207(34464*1.2207)=42070
Calculator:
-1000*1.2207=1220.7
-10000*1.2207=12207
-4096*1.2207=4999.9872
-40960*1.2207=49999.872
-53687*1.2207=65535.7209
-53688*1.2207=65536.9416
-100000*1.2207=122070
When we multiplied 100,000 by 1.2207 we were surprised to get an answer instead of an overflow error. However, it multiplied 34464 by 1.2207 instead of 100,000 by 1.2207. We understand that it probably gave us an answer instead of an error because the STAMP used simple logic, and it computed 1*1.2207 and added the zeroes on afterwards. However, we are wondering why it decided to use 34464*1.2207 instead of 100000*1.2207? We realized that the last equation (100,000*1.2207) was different on the STAMP then the calculator. We are very confident that the calculator was correct, but why wasn't the STAMP? We also noticed that the STAMP rounded the numbers. All of our answers on the STAMP were rounded up to the next number except when it would go above 53687, which resulted in an overflow error.
Mike and Sean
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"Learn to obey before you command."
-Solon
@Mike and Sean,
I tried to explain the overflow error and why the Stamp turns 100000 into 34464 a couple of days ago, in this message:
http://forums.parallax.com/showthread.php?p=915797
I know it is hard to keep track of all the posts in this thread! PBasically, it comes down to the fact that the BASIC Stamp can only work with numbers up to 16 bits, which is numbers in the range of 0 to 65535. You know that 216 = 65536?! Memorize that number! If you want to continue to work with computers, it is a number you should learn to recognize instantly, and the whole series, 1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024, 2048, 4096, 8192, 16384, 32768, 65536. You should learn to write down that series quickly from memory. See the number 4096 there? That is the binary resolution of the MCP3202 analog to digital converter. 212=4096
The program had no trouble to identify the numbers between 53688 to 65535 as overflow (when multiplied times 1.2207), but it actually takes a more complicated PBASIC program to detect that someone enters a larger number, such as 100000, and flag it as an error.
Did you get the MCP3202 running?
I'm still busy today, setting up a system with 49 temperature probes to monitor the grassland habitats of wolf spiders.
As to world cup. My wife is Italian by birth, and her brothers expound the minutiae of the play and of the players. One of them is married to a Brazilian, so guess that there might be a divided household there. Sorry Sylvie, I love Paris, I love Quebec, but Italia Italia! Here we are nearer Mexico, so there was a lot of enthusiasm for the Mexico-France game. The French team on the field clearly thought Hernandez was offsides on that one disputed goal, but I think it was a fair call. I have trouble understanding the offsides rules, and the referees are so often capricious in applying them, as in the U.S.A. tie with Slovenia.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
What an excellent practice we had today, probably one of the best we've ever had! The Rocketeers were working together like a well-oiled machine. There was some excellent dialogue and collaboration today, and some really excellent leadership by Andrew.
Yes, we did get the 3202 going and it works just fine. **It's amazing how well it works Sylvie, when I connected the wires from the ADC pins to p0, p1, p2, and p3...!. Duh. Thanks for leading me into discovering it on my own and not simply telling me: "Hey dummy: connect the wires!" We actually had five 3202's going at once. Each Rocketeer was asked to read the schematic, find and download the sample ADC program code on the Parallax website, wire their BOE proto-board, and then run the program to show that it worked. Then they were asked to make the voltage and "4096" appear on the computer screen from CH1 instead of CH0, and to make words appear on the (DEBUG) screen explaining what was happening.
Next, Andrew and Dylan worked with the ASP program and reassigned some of the pin-outs while Mike, Sean, and I ran wires from the ADC mounted on the stand-alone breadboard to a clone of the ASP. I won't go into great detail about which wires went where except to say that we ran them to the available pin-outs on the BOE and to Vss and Vdd, the same way they were wired on the ADC/ BOE proto-boards.
**They actually got the voltage and its corresponding "4096" digital number** to read out from the ADC onto the ASP's data screen... progress in our book! They were giddy with excitement! I was too, not simply because "it worked" (albeit marginally), but because the Rocketeers UNDERSTOOD and were able to explain what they had done to make it work.
Expect all the Rocketeers to post to the forum in a little while. I'm curious to see what they report and what questions they have. I'm going to clean up the carnage they left on the dining room table, then light the grill like Susanne asked. Tonight Tracy and Paul, we have celebratory STEAK! Happy Father's Day!
More later,
Mark
Today's practice was very productive! We accomplished a lot like we do at every all team practice because we come prepared. Each of us now has our own individual·BOE set up as follows:
CS to p0
Clock to p1
DataOut to p2
DataIn to p3
Vss uses a jumper wire to connect to the capacitor
Vss also connects with the Vss wire that connects the potentiometer to the BOE
The wiper is connected to Ch1 (but on the ASP it will be connected to Ch0)
A wire going from one of the leads on the potentiometer to Vss
A wire going from the other lead on the potentiometer to Vdd
All of us also have a potentiometer set up (thank you Mr. Kibler for the kit!) and an MCP3202 set up beside it (thanks for the ADCs Andrew!). Mr. Kibler will be uploading a picture of it.
We then set up a similar setup on the protoboard. This time we connected the wires from the protoboard into the p slots on the BOE on the robot with treads. This takes up pins 0, 1, 2, and 3. The only way we can think of to fit thosewires that were taken out would be to put them in pins·6, 12, 13 and 14 (6 and 14 were already free and 15 is for the MAWD).·This is what we did for testing purposes (but not permanent). CS went to p6, clock went to p14,·DataOut went to p13, and DataIn went to p12. We·shared pins for Din and Dout·afterwards. We also redefined the I/O definitions and commented out any refrences to the LED in the code·for the time being. We ran what will·eventually evolve into our·final program and the·output of data was scrambeled.·However, we did make good progress on the program today in addition to everything else.
Questions:
In the long run, we want to still·be able to have the LED on the BOE (which takes up pins 12 and 13). Is there any way we could do this AND have the rest of the other wires connected?
We are getting every closer to the end at good team meetings like this one. Dr. Allen, thank you for all your help, especially with the programming! Also, thank you to Sylvie, who has graciously helped us with the wiring!
Where to do suggest we go from here with the programming and wiring?
An excited new team member,
Justin
"If you build it, it will run..." - Field of Dreams
Today our practice was very productive, perhaps it was more productive than some of the practices last year! We had finished so many tasks that were needed to be done is such little time. We made execellent progress today. Each team member displayed much effort today staying on task and working.
First, we answered your questions from early(previous post). We did it such little time that what we had expected to do it in. After we set-up our homework boards and we wired our MCP3202 to it. We used the diagram of how the MCP3202 is set-up. For instance where the Vss is and the Vdd. On the same board, we wired a potentiometer to the breadboard also. We then used the sample program for the MCP3202·(please correct me if I am wrong) and this is how we knew that we wired the MCP3202 and the potentiometer correctly beacuse the program worked. Now, each team member should have those things on their homework·board or at least know how to set it·back up.·After that, we tried to recreate what we did on our homework boards to a single breadboard. We had set it up correctly and we had wires running from·the single breadboard to our prototype. We commented out the LED light and took it of our prototype so we could fit our neccesary wires on it to work.·We wired the wires up to all different pin#'s and redefining them in the program. We were "messing" with the program trying to get it all to work. But unfortunately, we ran out of time and we were not able to finish what we had started with the program.
We think it would be nice to keep the LED on the BOE. We like to have it there beacause it helps us knoe whether the program is running correctly or not. Is there a way that we could have the LED still on the BOE, and still be able to have everything needed on the BOE also?
Also, does it matter which pin # a specific wire goes into? For example, would it matter if the wire coming from CLK went it into P2 or say P9?
Where should we go from here. Thank you all for your time and help.
Awaiting your reply,
Sean
Team members and others,
Thanks for remembering to post your thoughts about today's team meeting! I'd like to elaborate on Mr. Kibler's previous post by uploading the program we worked on today, which is a mashup of the MCP3202 sample program and our functional MAWDBOE w/ CO2 program. It certainly isn't a work of art, but from what we were able to see on the debug window, it was rather promising! It isn't fully functional, nor did we have the time to try to integrate it with the data logger -- but it DOES read out a voltage, as Mr. Kibler said, which seems to be a big step forward in my opinion. I will take a second look at what we tried to accomplish today, and see if I can clean it up, but my knowledge of programming will only be useful to a certain extent.
Keep up the great work team!
Andrew
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"Learn to obey before you command."
-Solon
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"Learn to obey before you command."
-Solon
Sorry for not posting during the Early evening. I was out and now is the first time that i have had access to the internet. I would of enjoyed to look back at our early practice, but it seems that every thing has been covered by my teammates. I'm not sure if I am getting ahead of myself, but I think our next logical step is to understand the wiring at a further extent. Our team this afternoon wired it in multiple ways. The first time we wired, numbers were outputting. What we did was change the points that the wire connected to. The wire being the one that connect to one of the chips pins to the wiper. The reason to be was that the wire that was connected to the wiper, was also connected to the chips pin that the program collected the voltage from. It seemed logical to connect the pin from the CO2 sensor that outputted the voltage depending on the current CO2 levels, to the pin on the chip that collected the voltage (If you think about it, the program outputted a number depending on the voltage that the potentiometer outputted after it went through the changes (In this explanation, a variable resistor), so your just replacing the potentiometer part that outputted the changed voltage, with the CO2 sensor's pin (CNTL) that outputted the changed voltage from the CO2 sensor. (In this case the,"Change" was the amount of CO2 in the air)).
I am not sure if any of you will understand this but I put it in the best way I can.
The second way we tried was with Sylvie's, "shorting" wiring scheme. I believe the same thing happened but I will need to confirmation on that. But then I remember that Mr. Kibler switched some wires around too quickly for me to record what was happening. (I remember Andrew giving me weird looks for we where both afraid what was going to happen.)
The things we will also have to work on is programming to make the CO2 number, (In this case the 4096) appear in columns. When we ran it the number pretty much went where they wanted to go (even before the initializing word!). Also we dearly need to mount all of the components (extra breadboard, CO2 sensor, etc..) on an secured location. A loose wire could cause ALOT of problems for us in this point.
It is late and I cannot remember if I am missing something very important. I just need guidance on how we will go about tackling our next steps.
Dylan Landry
I do have possibly very stupid question - I see a lot of references to "4096" in the posts. That is the reading you should get when the input voltage to the 3202 is 5 volts. More accurately, it should output 4096·whenever the voltage on pin 2 (input Ch 0) is equal to the voltage on pin 8 (Vdd/Vref - the reference voltage). But you should get a different number from the 3202 when the voltage on Ch 0 is less than Vref. For example, if you're sending 2.5 volts to Ch 0, you should get a reading of about 2048, and if you're sending 1.25 volts to Ch 0 you should get a reading of about 1024 (and so on). If Ch 0 is connected to ground (0 volts), you should get a reading of 0.
Is that happening? When you adjust the pot, do you get different readings on the output from the sample program?
Have you measured the voltage on the 3202's pin 2 to see that it is roughly the same as the voltage reading on the screen?
Your wiring diagram looks right to me (except it's upside-down <grin>).
If all of that is working, then you've achieved something that I find very exciting - the ability for your BS2 to know the value of a voltage from the outside world. A LOT of sensors send variable voltages to reflect the sensor reading, and with this setup, you now are able to read those sensors (with some minor adjustments, perhaps).
==================================
(** See my next post below - I was mistaken about some of this, I think. I'll leave it up, though.·**)
I do still wonder if Dr. Allen plans for you to do something to change the range of the 3202 in order to make better use of the output range of the CO2 sensor. I can't get to the product page right now - something seems to be down at the Parallax site - but I seem to remember that the sensor sends out voltages from 0 to 3.2 volts, or somewhere around there. In other words, it won't put out voltages as high as 5 volts, so everything from 3.2 volts (or whatever that upper end number is) to 5 volts is simply wasted range.
(* Note added: Okay, I can get to the product page now. It's 3.35 volts. *)
Right now the resolution of the 3202 is 5 volts/4096 steps = 1.22 mV per step. If you were to run the 3202 with a 3.2 volt power source on the Vdd/Vref pin, that resolution would be increased to 3.2 volts/4096 steps = .78 mV per step. Once you convert that to ppm of CO2, you'd have significantly more precise numbers to work with as compared with using the 5 volt reference. The cost of doing this would be a couple of extra parts to change the voltage on the Vdd pin - I think just a couple of resistors would do it. On the other hand, it might be the case that it would introduce inaccuracies because of the tolerances of the additional parts, and perhaps because their values would change with the temperature. This is really Dr. Allen's area, and it might have been better for me to not even mention it, though I don't think it's too confusing.
In addition, it may be that the upper end of the CO2 sensor's scale represents CO2 values that you simply won't come across during your flight. If that's the case, then you could go with an even lower Vref, cutting off those upper values that you'll never see anyway, and getting even better resolution (though of course again there's going to be a limit to how good it'll get because of the various sources of error in the wiring, in the stability of the power source, and so on). For example, it's possible that the only way you'll ever get voltages over 2 volts out of the CO2 sensor is to blow pure CO2 onto the sensor. If that's the case, you could use a Vref of 2 volts on the 3202, knowing that you're simply not going to exceed that value in flight, and have a resolution of 2 volts/4096 steps = .488 mV per step, or 2-1/2 times better resolution than you're getting with the 5 volt Vref. If you made these changes, you would of course have to change your program slightly where it calculates the CO2 content from the voltage reading. It'd be a matter of simply using a different number in that mod (**) step, to divide by .78 or .488 or whatever instead of by 1.22.
I imagine this this is part of why Dr. Allen started off with some questions to you about the values of CO2 that you could expect to find during your flight.
Let me add that it will work just fine without any of this - there's nothing wrong with sticking with the 5V reference, and it's quite possible that making the changes I'm talking about here won't get you anything worthwhile. I'd like to hear Tracy's input.
Post Edited (sylvie369) : 6/21/2010 10:23:12 AM GMT
You asked:
1) When you adjust the pot, do you get different readings on the output from the sample program?
ANSWER: Yes, on both channels, but not at the same time. As an exercise I had the Rocketeers rewire and reprogram the BOE so that voltage and numbers (0-4096) appeared from either channel 0 or 1. When they changed the voltage (dialed the potientiometer) the number changed correspondingly from 0-4096 on whichever channel they had wired/ programmed. Good learning.
·
2) Have you measured the voltage on the 3202's pin 2 to see that it is roughly the same as the voltage reading on the screen?
ANSWER: Yes they did measure the voltage output and the voltages were comparable on the voltmeter and the computer.
Christopher and I are in Washington, DC right now and he has freshman orientation the next three days at George Washington University. And of course I brought the·ASP/ BOE·with me! I carried it through security in its red plastic toolbox so things wouldn't get jounced around inside my luggage. No problems. Everyone was curious about it and Christopher did a thorough job of explaining what it was. The Rocketeers fly out of Manchester often enough and they're in the news enough that one of the TSA folks actually remembered·us. That was helpful.
Speaking of which, how did your launch go?·How did your electronics work?·Did you get all the parts back intact? If the weather was anything like it is here, it was glorious. DC is 90 degrees and hot and muggy. Have you ever see a tornado out there?
Les francaises sont desoles! Qu'est-ce se passe avec·"les Bleues"...? Quel horreur! I see too, that Tracy is a a fan of "Italia! Italia!" Boy, if the French team ends up playing Italy, who do we cheer for? Not that I'm an avid "football" fan, but I have taken an interest in the World Cup·of late . I saw that Tracy's team, Italy had a draw (tie? null set match?) with New Zealand, who they were expected to beat hands down (or rather, feet down.) All this excitement, and Brazil beat even Cote d'Ivoire. How is the American team faring?
Tracy, see attached photo.
Our plan with the AD converter is to get it wired on the stand-alone breadboard (now that we know it works) and then connect wires from the AD converter·pins to the·ASP/ BOE. We did this right at the end of practice yesterday·and the we ran Andrew's "mashup" cut-and-paste program (which he uploaded below.) It actually output the voltage and the corresponding "numbers" (0-4096) on the ASP's data screen when we ran it, though the data and column heading were scrambled. This seems like progress and it seems like the program is headed in the right direction. It was amazing how well the AD converter worked when we connected the wires to BOE pins p0, p1, p2, and p3... Duh!
Andrew has a complete ASP-BOE-CO clone set-up at home and I have mine here in DC. So we'll wire and program via the forum, as will the other Rocketeers. Right guys?
Tracy and Paul, what do you see as·the next logical steps? Do you think we're on the right track? More later
Happy Father's Day guys!
Mark
· This makes perfect sense and I recall where Dr. Allen talked about this is a previous post. I suspect that we may find the answer, or part of it, in the "mV voltage-CO2 concentration" graph he asked us to look at, way back on some of the first pages in the forum. It shows the inverse realtionship between the two but more importantly it shows which millivoltage = what CO2 concentration. He also explained and then had the Rocketeers calculate us how much CO2 by volume = how much CO2 ppm (5% - 50,000 ppm or thereabouts, as I recall.) Together, this information should help us derive an answer about·the maximum·voltage/ millivoltage we will need to measure the·range of CO2 we're likely to encounter. Though I don't have the·"other" graph handy-- the one·that we posted that shows·the correlation between altitude and CO2--·I seem to recall a CO2 concentration of somewhere around 300-400 ppm up through 20,000 feet or so. I wouldn't think that CO2 concentration would exceed 1,000 ppm (1% by volume) between ground level and 20,000 feet, the highest altitude we'd encounter (last year the ASP went to 14,000 feet +/-) I would expect it to be much less than this but I think the ASP-2 should be designed and programmed to record a higher concentration... just in case we happen to encounter a higher concentration of "outlying data."
Does this makes sense and seem on-track?
Tracy, what's the research you're doing with wolf spiders? Why the interest in correlating temperature and wolf spider habitat (with 49 temperature probes nonetheless...!)·Are·they an indicator species?
Rocketeers, please check and see what CO2 concentration we're "likely" to encounter at ground level up through 25,000 feet altitude.
Off to dinner with my son,
Mark
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"Learn to obey before you command."
-Solon
I'd spent about ten minutes writing my previous post, and within 30 seconds of getting up to so something else, I realized it was a mistake. The nonconscious mind can do wonders sometimes. And then, of course, I managed to get distracted and forget to come back and post about what I'd realized.
=====================
Okay, good to hear that you have the 3202s working completely. Isn't it exciting to see that stuff work? As a base for sensors, you've just made your BoE twice as useful.
I think that you should sit tight for now as far as the device goes, until you heard from Tracy about the next steps. The obvious thing is to connect the CO2 sensor's TP1 to the 3202's Ch0 (pin 2), and see if you can get voltage readings that change with the CO2 content. It might be interesting for you to see what kinds of numbers you get in normal air and exhalation.
=====================
I don't think there's much chance that France will face Italy this time around. Italy might advance, but France is done. Serait necessaire quelques miracles. Espana tambien, y entonces, quisiera ver los equipos de Chile y Mexico.
I did not wind up flying a rocket yesterday. The wind direction was good, but it was far too strong for me to put all that electronics at risk. Instead I supervised a Junior Level 1 certification and an adult Level 1. The Junior one was quite the adventure. The young man flew a Wildman Jr. fiberglass rocket on a Cesaroni I800 VMax. Blink and it's gone - 4500 feet up. It opened the chute just fine - at apogee (no electronics allowed for junior certs), and drifted far off to the east. He had a tracker in it, but was getting only·a sporadic signal at best. Someone helping him track it decided that the rocket was moving - that someone had picked it up and carried it off somewhere (ranger station? back to the range? simply stole it?). So the young man eventually gave up looking, but set up another rocket with a Cesaroni I470. We redid the paperwork, and put it on the pad, and I started walking down towards the recovery area, and on a whim, turned my radio to the frequency of the first rocket's beacon - and got a strong signal. Seconds later I heard the kid yelling "Paul!", and I turned around, and there he was holding the Wildman Jr., in perfect condition. One of our fellow rocketeers had gone back out to look for it, and found it.
So he decided to fly the I470 anyway. It lit, but cut off FAR too quickly - even from my distance, there was obviously something wrong, and I wasn't at all surprised to see it arc over and lawn dart right into the ground. I never did see the remains, but I have to believe the motor CATO'd. However, the young man got his L1, because the Wildman Jr. was just fine. Quite a roller coaster ride for him.
Now I think it is time to look more closely at the Parallax CO2 module, to try to understand its schematic diagram. You can find the schematic diagram in the Parallax documentation, and I encourage all of you to find it there on page 4 and to read the description, "theory of operation", on page 3.
27929-GasSensor-v1.0.pdf
The module consists (as is usual with electronic circuits) of several blocks that each serve a specific function. Here is a picture with the functional blocks numbered. If you have a module in front of you, you can identify the different parts on the actual circuit board, and Parallax makes it easy to find the parts by looking at the diagram that they provide on page 2 of the PDF.
1) A voltage regulator, it takes Vin in the range of 6.5 to 15 volts that you supply from the battery, and in turn supplies an output of exactly 6.0 Volts. That is the power that is needed to run the heater in the CO2 sensor. (Remember the heater?!)
2) A second voltage regulator that provides 3.3V output to operate the other active components, 5 & 6, everything but the heater. Note that the input for the 3.3V regulator comes from the 6V regulator.
3) This is a transistor that acts like a switch. When the CNTL from the BOE is at zero Volts, the switch is open, non-conducting, and when CNTL is high (5V) under program command, the switch is on and current can flow to operate the heater in the sensor.
4) The actual CO2 sensor. See how one side of the heater is connected to +6V, and the other side is connected to the switch? The NASICON ceramic tube that surrounds the sensor is attached to ground on one side. See the ground symbol, three horizontal lines in a kind of triangle? The right hand side of the NASICON is where the sensor output voltage appears when the heater is hot.
5) The sensor output is applied to an input of part number LMC6035, designated U2A. It serves two purposes. One is to act as an electrical buffer, to sense the voltage from the NASICON without drawing any electrical current that would disturb the equilibrium of the chemistry. The LMC6035 input draws only a few femtoAmperes (10-15, well below what the sensor manufacturer says is allowed. The LMC6035 is also set up to amplify the Voltage output of the sensor.
The amplification factor is set by the two resistors, one of value 150 kΩ and the other of value 1 MΩ. The gain factor is,
Any Voltage that you measure on the test point, TP1, will be 7.67 times the actual output of the CO2 sensor.
6) The output of U2A is passed to the test point, TP1, and also to an input of U2B. U2B is actually in the same package as U2A. The LMC6035 is a dual operational amplifier (op-amp for short), two independent op-amps in one 8-pin package, a miracle of integrated electronics. It is an "analog computer" that can be configured for many different purposes depending on how you hook it up with outside components. Part U2A is hooked up as a buffer/amplifier, and part U2B is hooked up as a comparator. The output from U2A is compared with the Voltage you set on the potentiometer dial R3. See the (+) and (-) inputs on the left side of the triangle U2B? The output of U2B at the tip of the triangle is high (5V) when voltage on the (+) input is greater than the voltage on the (-) input. The LED turns on and the alarm on the BOE shows a "1".
7) You know by now, that is the 4-pin connector from the sensor module to the outside world. Also, you know that in order to get the signal you want, you are going to have to bring an extra wire over the the test point, TP1.
We are going to have to address the problem that Sylvie brought up. Is there going to be a change in the CO2 level that we can detect well with this setup, or are we going to need further modifications?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Post Edited (Tracy Allen) : 6/21/2010 6:43:06 PM GMT
I have a device I made that uses both 5V and 3.3V, and I run a 9V battery to a 5V regulator, and then from there to a 3.3V regulator. It draws a fair amount of current, but runs nice and cool.
I have a similar device that draws about the same amount of current, but it's all going through just a 3.3V regulator. When I power that from 9V, the regulator gets pretty warm.
Question: Is that what the word "buffer" means in this context, then? That it senses voltage without drawing current? That was on my list of questions to ask when we got back around to this. I never would have guessed that, but in retrospect, of course, it makes perfect sense.·This gives me all sorts of ideas.
Excellent. I think I know how to answer this question now. I'll be very interested to see·what the Rocketeers come up with here.
Rocketry IS exciting, and often, so is teaching. In the last few days I learned that a colleague for whom I wrote a recommendation was awarded the annual Undergraduate Teacher of the Year award by the Society for the Psychological Study of Social Issues. Today I learned that one of my former undergraduate students, one who took my physiological psychology course as well as my research methods course has been accepted into the post-graduate program in Brain Science at Seoul National University. I am told that's the most selective placement in Korea. I cannot begin to say how happy I am about all of this, and how proud I am to be able to work with people of this quality (you guys as well!).
Paul
·· I'm not even sure where to begin. So many curious things happen when I·wire the ADC chip pins to the proto-ASP/BOE·(which still has servomotors mounted) and then·run·Andrew's·"mashup" program. The ADC chip is·installed on a stand-alone breadboard just as it was at Sunday's practice, not on a BOE.
Here is where the wires go from the ADC pins on the breadboard:
CS pin to·BOE··· PIN·6························
DataIn·to········· PIN···· 12
DataOut·to······· PIN···· 13
Clock·to··········· PIN···· 14
*CH 0 pin/ wire is not connected to anything...
Vss pin to········· Vss slot on BOE
Vdd/VREF to······ Vdd on BOE
1) When·the program runs,·the P14 servo activates and·the servomotor/ tread moves...·a·short, rhythmic·pulse·at time. The "movement" subroutine that runs the servos is commented out and so the servos "shouldn't" run. I unplugged both servo motors #14 and #15.
2) When the program runs,·elasped time, sample number, CO2 alarm status (1 or 0), temperature, and humidity do appear. They continue to change accurately as the program runs but they appear in different location on the screen when the two voltage readings appears. Though the data is accurate, it no longer appears in columns. It appears on on line only, but still changes accurately.
3) Two voltages appear, along with two corresponding sets of numbers between 0 and 4096. They change somewhat rhythmically but the changes are NOT in sync with the servomotor pulses (which stop when the servos are disconnected.)
4) When I remove the wire from the ADC's Vdd pin to Vdd on the BOE, the program still runs and outputs data. I left it disconnected for now.
5) When I remove the wire from the ADC's Vss pin to Vss on the BOE, we lose all voltage and the corresponding '0-4096'·numbers.·I left it connected.
I think I understand some of the many things that have to happen·(*that's our first problem... "Cogito ergo confuso".) But there are so many "layers" of things that have to happen, and in the correct sequence, that I'm not sure where to begin.
1) We have to wire the ADC chip into the ASP/BOE correctly, and permanently, so that everyone has the same "clone" at home. I'm not sure that things are wired correctly right now (point: the wire from CH0 isn't connected to anything...!) But we are getting voltage on the screen and last year's program still runs and outputs data despite the "bad wiring".
2) We have to assign ADC pins and pin numbers in the program. This assumes that step 1 is correct. Pins/ pin numbers are assigned now, but are they correct? Recall that that the Vdd wire and the CH0 wires are NOT connected and the program still runs.
3) We have to write·a program to convert CO2 millivoltage input into the ppm. I'm not even close to being there.
So what are the next (baby) steps? Too many factors are·superimposed on one another and I can't wrap my head around things. I having a tough time seeing where to logically begin, what to do first.
Out of excitement and frustration I·find myself·disconnecting and connecting wires to see what happens. But that is not good science, nor is it logical. It is simply trial-and-error. Although it might eventually succeed it isn't the best approach, and it's time-intensive. I read somewhere that if you put an infinite number of monkeys, at an infinite number of computers keyboards, for·an infinite amount of time, one of them will "eventually" write Shakespeare's 'Hamlet'. Anyone have a banana...?!
Now·about those wolf spiders. So why are you studying their habitat and temperature (with 79 sensors!) anyway?
From DC,
Mark
When it is time to run a wire to TP1 (from wherever), will the wire simply be soldered to the TP1 hole on the CO2 sensor? That's a smaaaaaaaaaaaaaaaaaaaaal hole!
Should we include a resistor in the wire?
So many questions, so little time!··
ROCKETEERS, where are you...? What questions do·YOU have?
Mark
As for soldering to TP1, I think you'll be surprised at how easy it is. That being said, my soldering life changed forever when I discovered solder wick:
http://www.sparkfun.com/commerce/product_info.php?products_id=9327
Near-instant forgiveness of sins.
It's available at Radio Shack, at an elevated price, of course, or from Sparkfun (though they're currently sold out), or from one of the general distributors (Mouser.com, Digikey.com) at a good price. If you plan to solder things, you should plan to have some on hand. These days I find myself soldering pins that are 1/3rd the size of that little TP1 hole without too much difficulty.
Yes, the waste heat produced is proportional to the voltage dropped across the regulator. The FAN250033 that supplies the 3.3 volts is a small part, an SOT23 package, and it is not capable of dissipating much heat. If it gets to hot, it will shut itself down (which is preferable to going up in smoke!!).
Hypothetical:
Input of FAN250033 supplied from 6V at the input, 3.3V at the output, and 0.01 Amp to light the LED and run the LMC6035.
Temperature rise is calculated (approximately) a the parameter thetaJA that can be found in the FAN2500 data sheet.
So if ambient is 25°C then the inside of the chip rises up not even to human body temperature. That's safe! The internal chip temperature has to rise up to 150°C before it shuts itself off. I leave it to the reader to estimate what would happen if the input to the FAN2500 came from a 12V supply instead of 6V and if the current required were 0.05A instead of 0.01A.
Question: Is that what the word "buffer" means in this context, then? That it senses voltage without drawing current?
Yes. It is impedance transformation. At the input, (finite Volts)/(zero Amps) = (infinite Ohms). At the output, (no change in Volts) / (any change in Amps) = (zero Ohms). You have to get into the math analysis to justify that, but that is what it amounts to. In general, a buffer isolates one circuit from direct connection to another. Buffers are common in digital electronics too, with a somewhat different meaning. For example, a "tri-state" digital buffer is one that can either connect or disconnect two circuits. In one state the buffer has infinite ohms resistance to isolate the circuits, and in the other state it has zero ohms to connect them.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Tracy, speaking of "waste heat", we're looking into a solar (photovoltaic) array for our roof. If we don't feed excess electricity back into the grid, and if the storage batteries are fully charged, where does excess heat-energy from a PV array go? Is there·a "left-over electricity" device that dissipates or discharges(?) left-over voltage as heat-energy? (*not that we have excess sunlight·and solar·energy in New Hampshire. We don't. It's a theoretical question more relevant to Arizona I suppose.
Even though data on·the computer screen is scrambled when we run andrew's "mashup" program, the data still outputs nicely to teh flash drive/ Data Logger. No CO2 data (of course.)
Mark
·
Even though the code for the servos is commented out, the servos if they are connected will hear any activity on pin p14. The servos don't know that those clock pulses are meant for the ADC. They are just pulses.
I'll take a look at the program after work. My inclination is to connect as follows:
On the stand-alone breadboard, connect a 1 kΩ resistor from Dout to Din.
Connect a wire from Din to p6 on the BOE.
Connect a wire from CS\ to p12 on the BOE
Connect a wire from CLK to p13 on the BOE
Connect a wire from Vss onto Vss
Connect a wire from Vdd/ref to Vdd.
That leaves p14 free for the servo. Does the ASP-2 ultimately use the servo to wander around after touchdown?
There is not a separate wire for Dout. Now Din and Dout are serviced by the same pin, p6. This may or may not work. We'll see.
For the CO2 sensor, I suggest that you solder a length of wire to TP1 and another to TP3. (Chris?! Maybe they have a lab at GWU! There used to be lots of electronics surplus places around Washington DC)
Connect a wire from Vss on the CO2 sensor to Vss on the BOE
Connect a wire from Vin on the sensor to Vin (9 Volts) on the BOE
Connect a 1 kΩ resistor from ADC channel 0 to a free node on the breadboard and from there to TP1.
Connect a 1 kΩ resistor from ADC channel 1 to a free node on the breadboard and from there to TP3.
The connection to TP3 will let you measure the setting of the calibration pot on the CO2 module
NOTE: Power for the CO2 sensor needs to be at least 6.5 Volts!!! Connect to Vin (=9+Volts), not Vdd (=5.0Volts).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Dr. Allen,
Thank you for your detailed suggestion. I would like to clarify that the ASP-2 will not be roaming around after touchdown, as our NASA SLI robot, "Phidippides", did. After touchdown, the ASP-2 will remain collecting (baseline) data at a stationary spot, as well as at the launchpad before launch, until we locate and recover it. I'm headed off to hot and sunny Florida tomorrow morning, with Phidippides, soon to become our newest ASP-2 clone.
Andrew
Tracy, I'll wire the apparatus like you suggested and report back. I already had the Din-Dout 2-into-1 wire, ready. Freshman/ parent orientation was chocker-block full and I was ready to be back here at the hotel midway through the afternoon. I'd forgotten home hot DC gets! But it was an excellent orientation and Christopher is really excited to be on campus.
I'll have to track down the electronics surplus places you mentioned. Maybe they have "solder wick" and a miniature soldering tip. I knew I'd fogotten something back in New Hampshire: the soldering iron!
If you're only 10 hours from Black Rock desert you should join Sylvie and I and all the other Rocketeers for the ARLISS·launch in September. We'll buy the gas... and the rocket fuel. We already have·the hotel rooms reserved and I'm sure the three of us would manage just fine. How far are you from Reno? Do they have wolf spiders in Nevada?
Mark