·· Can you check for us·and·find a clear, simple, and concise definition for·"intrinsic"? Then post the definition and explain what "noise intrinsic in the sensor" means. I'm eager to learn from you!·····
The main thing in your good graph is that is responded to the CO2 in your breath! The signal was obviously greater than the noise. There is not much use in trying to figure out where the noise is coming from in a graph where there is no response at all to CO2. Signal to noise ratio is a very important concept in electronics, whether it be your radio, your television, your stereo, or sensors like the MG811 or the SHT11 that you have on the ASP-2. There are lots of different kinds of noise. Some is like a hissssssssss, and some is like a pop!....pop!............pop! and some is like slowwowwwoowoooooooowwwwoo. Noise comes from the physics of the sensor itself (intrinsic), or from the circuit that amplifies the signal, or from the analog to digital converter, or from the rest of the system and the way it is wired up.
What should we expect for the resting level and sensitivity for this sensor? Given 100 of the sensors, how much variation would we see, compared to the graph in the data sheet? The data sheet does not tell us, does it? We can't test 100 sensors, but maybe 3 or 4. We want a graph like the good one that shows an unambiguous response to CO2. My plan for the data processing is to adjust the ASP-2 CO2 readings to fit the MG-811 typical curve. Say the ASP-2 says 2.5
Volts at a CO2 level where the MG-811 typical curve says 3 volts, that is our calibration multiplier, 3/2.5 = 1.2. I'll get into how to do that as we see reliable (clean response to breath CO2) data coming from the ASP-2.
Is it possible to install the sensor incorrectly in the socket?
Intrinsic means "belonging to a thing by its very nature". Noise can be defined as any unwanted signal in a system. Intrinsic noise is generated by the element itself. There are three types of noises: thermal, broadband and flicker noise. Thermal noise is generated by the random motion of electrons in a conductor.
Tracy Allen said... Signal to noise ratio is a very important concept in electronics, whether it be your radio, your television, your stereo, or sensors like the MG811 or the SHT11 that you have on the ASP-2. There are lots of different kinds of noise. Some is like a hissssssssss, and some is like a pop!....pop!............pop! and some is like slowwowwwoowoooooooowwwwoo. Noise comes from the physics of the sensor itself (intrinsic), or from the circuit that amplifies the signal, or from the analog to digital converter, or from the rest of the system and the way it is wired up.
Dr. Allen,
Ah yes! I've actually heard of that term before in terms of cable television. At last I know what all this noise we are seeing is called. As for what you said about installing the sensor incorrectly, as far as I know the sensor's pins aren't quite symmetrical, so it shouldn't fit if you turned it 180 degrees. I could be wrong though ...
ROCKETEERS from New Hampshire to California, and places beyond,
·· Mark today's date on your calendar: August 4, 2010. It's a banner day, the day the ASP-2 came to life! After spending all day yesterday at Dartmount Medical Center, Christopher and I spent most of today at the workbench. We soldered the resistors, capacitors, and wires directly to the ADC chip and perf board-- very tedious, focused, and time-consuming work for those who have never done it before (*you'll all get to do this at our next practice on August 14th.)
The soldering on our first perf board ADC chip looked "rough" but it worked OK. It gave us good, accurate CO2 millivoltage but the TP1 millivoltage was "stuck", as integers, between·4 and 8 (see attached graph #1.) However, the interference (intrinsic noise, or static) was virtually gone, the graph line showed little variation as it climbed beautifully to 3300 mV, and the sensor was very responsive to CO2 (our breath, which probably smelled like extra chunky peanut butter!) We weren't sure why to TP1 millivoltage stayed at 4-6 and we weren't content with the ratty soldering we did, so we made version 2 of the ADC chip perf board (we'llpost photos tomorrow.) This time the ASP-2 worked perfectly (see attached graph#2.)
We soldered all three resistors and the capacitor·directly to the top of the ADC's "legs" (pins) and we made sure that the Vss wire was in direct contact with the ADC chip pin,like Dr. Allen and some of the Rocketeers hypothesized and suggested. Loose wires were apparently the cause of the interference, as some of you suggested. Good teamwork, and thanks! Soldering the wires seems to have remedied the problem and I think we are almost there... August 4, 2010 was a banner day, thanks to our diligent team(work).
HOMEWORK QUESTION:·Rocketeers,what·should we do with the constant that you calcualted a few days ago? Where does·it fit into the project, the program, and the grand scheme of things? Remember that the sensor is only outputting millivoltage right now, NOT CO2 data...
·· Can you tell us how to get the program to write the TP1 millivoltage data to the flash drive, if it isn't going to complicate things? The program and the ASP-2 are·working too well·right now to jeaoparidze (jinx?) things.·It would give us yet another packet of flight data·to look at. Maybe it'll tell us something important, maybe not. Data is data and the more, the merrier! The TP1 millivoltage appears on the computer screen right now but it isn't written to the flash drive. I can't tell you ow giddy we were when the hardwired perf board ADC chip worked so well tonight!
Do you have any hikes or bike trips planned for the weekend? Being so close to the Northwest you could travel north and look for Sasquatch. I've heard that he's "out there", so be careful.·Some folks have seen him near Area 51 (but I'm not sure how credible the sources are.)·I spent yesterday at Dartmouth Medical Center going through a battery of tests and doctor's visits -- MRI, CAT scan, X-rays, getting poked with several different diameter needles-- as a follow-up to my cancer treatment. In the grand scheme of things, things are·good. That was relief! Did you see the aurora borealis last night?
...Is it possible to install the sensor incorrectly in the socket?
Tracy,
·· The CO2·sensor has six pins: 3 pins opposite another 3 pins, placed around the inside circumference of the round CO2·sensor. The pin spacing is symmetric and I've plugged the sensor in both ways with no adverse effects or 'bad' results.
It IS·possible to·attach the 4-pin (female) wire harness that we made·"backwards" onto the four CO2 sensor·module pins. But the four wires and our·harness·are well marked and that·didn't seem to be ·the problem.·Hypothesis rejected, but that's OK. It was certainly worth mentioning and checking. Ultimately I think it was simply loose wires; soldering everything on to the perf board remedied the problem(s)... knock on wood! We resume trials again tomorrow.
Mark, Can you please post the ASP-2 program that you used to obtain the above results with the CO2 sensor? Kudos on the success! There were results that Andrew posted earlier where the CO2 mV data showed up on the flash drive, but for some reason it is not on yours, so there is some minor difference between the programs and we have to resolve that.
I don't understand what you mean by "It gave us good, accurate CO2 millivoltage but the TP1 millivoltage was "stuck", as integers, between 4 and 8 (see attached graph #1.)" I see trial #4 and trial #5, and both of them seem to have good CO2 mV readings with good response to breath.
Beautiful. It's good to see that the sensor is responding as expected now, and that the interference problem is solved.
I'm surprised at how different all of these sensors are in terms of the voltages they're putting out. It looks to me as though yours, Tracy's, and the one in the datasheet are all completely different. I believe that Tracy has a plan for taking care of that in order to calibrate things. The good news is that if you're reliably getting _a_ response (and it looks as though you are), you can fly the thing, and worry about what the data mean afterwards. Of course you may need the sensor in working condition to decipher your results, so it might be especially important to recover things in good condition.
Mark in NH said...
Mark today's date on your calendar: August 4, 2010. It's a banner day, the day the ASP-2 came to life! After spending all day yesterday at Dartmount Medical Center, Christopher and I spent most of today at the workbench. We soldered the resistors, capacitors, and wires directly to the ADC chip and perf board-- very tedious, focused, and time-consuming work for those who have never done it before (*you'll all get to do this at our next practice on August 14th.)
Mr. Kibler,
This is fantastic to hear, and not at all surprising after the amount of work the team has put into the ASP-2.
Mark in NH said...
HOMEWORK]Rocketeers,</FONT>what should we do with the <FONT color=red>constant that you calcualted a few days ago? Where does it fit into the project, the program, and the grand scheme of things? Remember that the sensor is only outputting millivoltage right now, NOT CO2 data...
As I understand it, the constant is the only thing standing in the way of converting the millivoltage to an actual carbon dioxide level. Team members, please correct me if I am mistaken, but I believe we tried to figure how to go about replacing the current (incorrect) constant with our new one. However, as I remember we didn't understand the format of which the current constant was stored.
...Can you please post the ASP-2 program that you used to obtain the above results with the CO2 sensor? Kudos on the success!
I don't understand what you mean by "It gave us good, accurate CO2 millivoltage but the TP1 millivoltage was "stuck", as integers, between 4 and 8."
Tracy, · ·· Success, thanks to you and an excellent team effort. Thank you!····· We really couldn't have done it without you and Sylvie’s generous help and guidance. Really. · The ASP-2, in last night's configuration,·is running right now and all the data looks really solid. CO2 millivoltage is rising nicely to the stabilization point and there is virtually no fluctuation in the "other" millivoltage", the mV that's set with the potientiometer. The graph I just printed out, and all of it's accompanying data, is·perfect! More trials to come later in the day.
When I said, "It gave us good, accurate CO2 millivoltage but the TP1 millivoltage was "stuck", as integers, between 4 and 8", I was referring to the millivoltage that is adjusted with the potentiometer. I wasn't sure what to call it or how to reference it. · Attached is the current program, which is essentially the same as what we've been using all through July. I just re-named it "August 4" so we know that it's the version we were using when the ASP-2 came to life. Andrew, you may want to use this version of the program if you're not already. I'll post pictures later today. I have to run Christopher to work right·now (10 minutes ago!) We all deserve a celebratory root beer and vanilla ice cream float for dessert at dinner tonight. MAybe we can share one at the ARLISS launch in Nevada...? · "Success is 1% inspiration and 99% perspiration." ~ Thomas Edison · Success! · Mark ·
As I understand it, the constant is the only thing standing in the way of converting the millivoltage to an actual carbon dioxide level. Team members, please correct me if I am mistaken, but I believe we tried to figure how to go about replacing the current (incorrect) constant with our new one. However, as I remember we didn't understand the format of which the current constant was stored.
Andrew,
That is also how I see it, there was a format to the·incorrect constant that we could not figure out. I think we tried find·a converter online and we checked our "What's a Microcontroller" books and we could not figure it out.
Sean
Well that was strange! It seems that our original forum is back and that we can add more posts to the same "old" forum. Looks like we don't have to start a new forum after all. But there seems to be a new web screen format. Does anyone else see a difference?
OK, looks likewe're back up and running. One thing you have to do now is to figure out what do we do with the constant you calculated?
Sylvie makes a good observation and a good point. When the CO2 millivoltage stabilizes after it warms up, the millivoltage appears to be different from sensor to sensor. On Dr. Allen's graph the mV stabilized around at a different level than on our sensor. This will effect the number we get when CO2 millivoltage is multiplied by the constant.[/COLOR]
QUESTIONS:
1) How will this effect the number?
2) How do we compensate for the difference in the numbers in the two graphs?
I'm posting a test picture to see if it uploads to the new forum. It seems that everything else is in HTML code... no graphics :-(
NOTE: The forum seems to limiting the file size to 20 KB and I can't seem to upload any pictures, graphics, etc. Can someone else please try?
Can anyone see the graphs, photos, etc. now? After the forum was 'fixed' I can't see any of them (and the "smilies" I try to post
> appear as HTML code.)
The ASP-2 is running right now with the ADC chip soldered to perf board and the CO2 sensor mounted on the ASP and ythe data is near-perfect...!
I'll send pictures as I promised when the forum is ready to receive them.
++++++++++++++++++++++++++++++++++++++
Sylvie made a good point several posts back. If there is variability in the CO2 millivoltage between each sensor, the same constant can't be used for each sensor (can it?)
1) How do we use the constant?
2) How do we factor in the difference between sensors so we can get a baseline CO2 reading. Otherwise the data is just millivoltage and tells us nothing about CO2 concentration, on the ground or in the air.
HOMEWORK: *Andrew, please set your team to work answering these two questions. We need the answers for the program and for practice.
Here's what Jessica Ulman at Parallax said in an e-mail to us:
"While you can't post new messages to this forum, you can use it as a reference for any old information.
All images that were attached to previous posts have been migrated to the new forum, but will appear as broken links since the URLs refer to the old forums site. We will be running a script hopefully during the weekend to update all image URLs. For now, please refer to the image attachments."
I'm not sure if we can post graphs and photos yet, but she said that we should be able to soon.
Attached (I hope) is a graph from the ASP-2 with the CO2 sensor and the ADC chip perf board permanently installed on the ASP. The data is perfect! There is virtually no fluctuation in the potientometer millivoltage and it only ranges from 1001 to 1004. Once the CO2 sensor heats up and stabilizes there is also no fluctuation in the data until CO2 is applied.
Now we need to figure out what to do about the constant...! Where and how do we insert it into the program so that the flash drive reads out "ppm" instead of just CO2 millivoltage?
Tracy, can you help us figure out how to write the program/code so that the potentiometer voltage also appears on the flash drive? Since it's there we may as well record it as flight data.
It looks like the new forum accepted the graph. Enjoy!
When I looked at the graph a realized something that might pose a problem. When I looked at the graph, the first thing I noticed is when the sensor is warming up, (when mV slowly increases as time passes). If we are are in Nevada and make a successful launch, then look at the data, how can we differ the data from when it was warming up, from when it is actually taking reliable data?
I will put it in simpler forms. (This is as if it was reading out mV and not ppm) Lets say the ASP is running inside the rocket and warming up, (on the graph it would have the slowly increasing mV as time passes). Then we launch. Lets say that the CO2 levels are decreasing as altitude increases. After we make a successful retrieval and look at the data, we would encounter a problem. We would see when the ASP is warming up, (the slowly increasing mV as time passes), then a sudden rise in mV, (when it was increasing in altitude after launch). How could we tell if it was done warming up and providing reliable data? Maybe it wasn't done warming up and was giving out data that was lets say 0.2 mV off what the actual reading were.
I have to go soon for mountain biking, this was just a quick idea of what we might encounter in Nevada as a problem.
Please let me know your thoughts on this.
Aiiii. New forum software--This is going to take some getting used to. Many old posts are discombobulated. A transition, a pita. Parallax has gone through this several time before, and I appreciate that it is sometimes necessary. It is a fact of our modern technological life. I can't count the number of times I have had to dump well oiled way of doing things in order to keep up, or to advance and lost things in the process.
So, are the CO2 mV readings now successfully recorded to the flash drive?
I now have two of the sensors and will hook them up to run side by side. We now know that there is variation in the resting level among different MG811 sensors. The question I want to answer is, "Does the variation also apply to the sensitivity". Suppose one sensor comes to equilibrium at 3 volts and another comes to equilibrium at 2 volts. They are side by side in the same CO2 environment, say, 390 ppm CO2. Now we expose them both at the same time to an increased CO2 level that brings the first one from 3000mV down to 2500 mV (500mV decrease, 16.7%). Will the second one decrease from 2000 down to 2500 (500mV) or from 2000 down to 1670 (16.7%), or will it do something else? That is, is the sensitivity absolute mV or a percentage of the resting level?
Question #1: What do you suppose the sensor would read if you seal it in the air space inside a bottle half full of soda water? Would that reading change with temperature?
Question #2: do you expect the mV to go up, or down, as the ASP-2 ascends to 10000 feet? (you looked at this question before, but it is time to think about it again!)
"When I looked at the graph a realized something that might pose a problem. When I looked at the graph, the first thing I noticed is when the sensor is warming up, (when mV slowly increases as time passes). If we are are in Nevada and make a successful launch, then look at the data, how can we differ the data from when it was warming up, from when it is actually taking reliable data?"
Dylan,
This is a good question with a simple answer. Remember the 'MAWD', the Mini-Altimeter with Deployment feature? It reads '0' the whole time the ASP-2 is on the ground, inside the rocket, and again after it lands. But the millisecond that the rocket takes off the altimeter starts reading and recording data to the flash drive... right beside the CO2 level, temperature, and humidity. As the ASP-2 ascends inside the rocket this is happening fast. But when the ASP is deployed from the rocket it descends on a parachute, reading and recording temperature, humidity, CO2 level, and altitude, all the way down until it lands. See the attached graph and you'll get a better idea. The only thing that's not on the graph is altitude. This brings up another important point (see below).
==============================
Sylvie and Tracy,
The ASP-2 continues to give us good, 'clean' data. Check out the graph from one of today's trials. However (malheureusment, Sylvie!) NO ALTITUDE DATA is being written to the flash drive. Sacre bleu!
Here's what we did, here's what we tried, and here's what happened.
1) We put the ASP, in it's present configuration, in a vacuum chamber like last year. This triggers the MAWD (altimeter). The MAWD beeped out the maximum pressure altitude but NO DATA WAS WRITTEN to the flash drive.
2) Same procedure as step #1 except we had the ASP connected directly to a computer so we could see the data read out live time. Same result: we got good CO2, temp,and humidity, the pressure altitude beeped out when we finished, but no altitude appeared on the computer screen nor was any written to the flash drive.
3) We made a new wire harness for the MAWD with an in-line resistor (like last year.) The blue wire goes to Vss and the brown wire top P15 (same as last year; same as in the program.)
4) We made a third wire harness without a resistor and got the same result. The altitude "beeps" out but no data appears on the computer screen or on the flash drive.
I really, REALLY hope this is a simple program glitch that Tracy can help us fix! The ASP-2 is virtually done so far as building and wiring it and we sure as heck don't want to work backwards... too close to the launch date and the start of the school year.
+++++++++++++++++++++++++++++++
Tracy and Sylvie, what do you think? Does this look like a hardware or software problem? What doyou suggest we do to fix it? (Please tell me it's an easy program fix! Please.)
"So, are the CO2 mV readings now successfully recorded to the flash drive?"
Yes, we're getting excellent CO2, temperature, humditity, and elpased time data written to the flash drive (see the graph below in my reply to Dylan.) But the altitude data-- which is critical in order to correlate CO2 with altitude-- is NOT being written to the flash drive. Again, see my post to Dylan. It explains what happened, what we did to try and fix it, and where things are as recently as 5 minutes ago...
I really, REALLY hope it's a simple program fix, an overlap between old and new code that's causing the glitch. The ASP/hardware is "right there" (visualize me making an "OK" sign with my finger and thumb) and I really hope we don't have to dismantle it. I'm sooooooooooooo afraid of breaking something critical and going backwards...!
Thanks for the homework questions! They're very relevant to Dylan's question on the forum. You make a good point about the variation between each sensor and how to compensate for the difference between each sensor. If one reads 2200 mV at 300 ppm and another reads 3500 at 300 ppm, that needs to be factored in and compensated for... before we launch.
So what do you think about the altitude not being written to the flash drive? Is it a hardware glitch or a simple program fix? (Please say it's a simple program fix!)
"Aiiii. New forum software--This is going to take some getting used to. Many old posts are discombobulated... I can't count the number of times I have had to dump well oiled way of doing things in order to keep up, or to advance and lost things in the process."
Tracy,
Yes, SURPIRSE! A new forum with a new look! That's why we back up programs and data, right? Jessica Ulmann at Parallax e-mailed a link where we can access the pages from our previous forum:
I'm not certain whether all the graphs and program attachments will be there but most appear to be. Whew! (wiping make-believe sweat off of my brow with the back of my hand... where's THAT smiley face icon gone to?!)
We have altitude data on the flash drive! And CO2 (too!), and temperature, humidity, and "flight time." See the attached graph
>
It's late, we're tired, and we're exuberant! Silly us: we had the (ONLY) two wires going from the MAWD to the BOE reversed. How simple of a fix can that be? So we fixed it, as silly as it was for us to make such a rookie mistake. I guess we're not experts until next week, huh?
The ASP-2 performed nicely inside the vaccum chamber, cranking out reliable, interference-free (pressure) altitude, CO2 millivoltage, temperature and humidity. A few more tweaks and we may be there.
Speaking of "there", I'm off to bed. Christopher checked out 45 minutes ago while I stayed to put the graph together. Another eventful day at the Rocket Tree farm. Tee hee. 2 AM and I'm making double entendres, Sylvie!
Glad to see you found and fixed the problem, and even before I had a chance to pipe in with my thoughts. I'll tell you what I would suggested anyway. My first thought was "is this a different MAWD? Maybe they never turned on the telemetry function".
Obviously that wasn't it, but if you do try to use a different one, don't forget that you need to do that.
Question #1: What do you suppose the sensor would read if you seal it in the air space inside a bottle half full of soda water? Would that reading change with temperature?
Question #2: do you expect the mV to go up, or down, as the ASP-2 ascends to 10000 feet? (you looked at this question before, but it is time to think about it again!)
Question #1: If I'm not interpreting this the wrong way, I believe that the CO2 reading would significantly change with temperature because as temperature increases, the water becomes less soluble which drives more CO2 out of the liquid soda into the air space (which the ASP can sense).
Question #2: I know talked about this a lot many pages back, but to recap, the millivoltage decreases when CO2 decreases. Logically speaking, with an increase in altitude, there is a decrease in air pressure. With a decrease in air pressure there should also be less CO2. However, I could not find any reliable sources to confirm this.
QUESTION #1: "...the CO2 reading would significantly change with temperature because as temperature increases, the water becomes less soluble which drives more CO2 out of the liquid soda into the air space (which the ASP can sense).
COMMENT: Andrew, Could you explain HOW the CO2 reading would change? Would it increass or decrease? Please explain which and explain why. *Recall the the ASP-2 on measures and records milliviltage right now, noy CO2 concentrationin parts per million (ppm.)
QUESTION #2: "...to recap, the millivoltage decreases when CO2 decreases."
COMMENT: Andrew, refer to the questions above. What happens to the millivoltage reading (CO2 level) when you breathed CO2 on Phidippides' CO2 sensor (before I borrowed it...!) Did it it increase or decrease? Please look at some of the recent graphs and disucssions on the old forum and report back. On my ASP-2 clone, when I breath CO2 on the sensor the mV decreases every time. Do I have it wired wrong...?! Would that change as altitude increases even if the CO2 concentration stays the same?
We had a long (2:30 AM!) and productive day-- and night-- yesterday. The ASP now records altitude, too!
The woman from Sant Bani Academy (Rose?) e-mailed yesterday about the rocketry camp we're doing the week after next. I'll forward the e-mail to you with the details. Our sessions are in the afternoon.
What do you think about us contacting Mr.and Mrs. Moran at the Gerlach, Nevada High School and offering to speak to their students about ARLISS/Parallax like we did last year? I think it would be great to re-connect with them. Maybe we could even invite Mr. Beaver and some of his science class students to join us at the launch? What do you think?
"Question #1: What do you suppose the sensor would read if you seal it in the air space inside a bottle half full of soda water? Would that reading change with temperature?"
Answer: I agree with Andrew that more CO2 would be driven out of the water soda and into the air space in the bottle. Since more CO2 would be in the air space where the sensor is, the CO2 reading would increase and the amout of millivolts would decrease.
Question #2: do you expect the mV to go up, or down, as the ASP-2 ascends to 10000 feet? (you looked at this question before, but it is time to think about it again!)
CO2 decreases as millivolts increase (and vise versa). CO2 increases as altitude increases. Therefore, millivolts should decrease at 10,000 ft., right?
The curve for CO2 in your vacuum chamber is flat at about 3000 mV. That is interesting and begs a question. From the raw data, can you tell us if it is 3000 mV or closer to 3300 mV? It is hard to read from the graph.
Partial pressure (pp) of CO2 decreases as you run the vacuum pump. Partial pressures decrease along with total pressure. But the ratio of different gases does NOT change as you run the vacuum pump, so the parts-per-million of CO2 remains the same even though there is a lot less of it per unit of volume.
Now, has your experiment shown that CO2 is proportional to pp, or to ppm? Andrew's answer to question #2 went to the heart of this issue. The graph shows mV as a function of ppm of CO2, but the main thing that happens with altitude is what Andrew pointed out, less CO2 (meaning per unit volume).
The reason I asked about the 3000mV vs 3300 is that the electronics can only go up to a certain maximum somewhere below 3300. Might there be other things going on with the electronics or with extrapolation on the graph in Andrew's last post?
The CO2 concentration above soda water in a bottle should be quite high. Pressure causes CO2 to dissolve in water, more pressure, more CO2, and that is why when you take the cap off and release the pressure, bubbles instantly form in the soda as CO2 comes out of solution. Cold water can hold more CO2 than hot water, so given a set amount of CO2 in the bottle, the CO2 pressure should be greater over hot soda. Also water vapor, saturated, but much less partial pressure than CO2. Total pressure is the sum of the partial pressures of all the different gasses that are present.
After finally figuring out how to log in to Parallax, with much help from Justin, Andrew, and Mr. Kibler, here are my answers to your latest questions:
"Question #1: What do you suppose the sensor would read if you seal it in the air space inside a bottle half full of soda water? Would that reading change with temperature?"
The sensor would read CO2 because like Justin and Andrew said, CO2 would come from the soda water and go into the air space and thats where the sensor is so it will read CO2.
Question #2: do you expect the mV to go up, or down, as the ASP-2 ascends to 10000 feet? (you looked at this question before, but it is time to think about it again!)
Like Mr.Kibler and Justin said,CO2 decreases as millivolts increases and CO2 increases as altitude increases. So as I understand it, mV would decrease as the ASP-2 ascends to 10,000 ft.
Comments
·· Can you check for us·and·find a clear, simple, and concise definition for·"intrinsic"? Then post the definition and explain what "noise intrinsic in the sensor" means. I'm eager to learn from you!·····
Thanks,
Mr. Kibler
Signal to noise ratio is a very important concept in electronics, whether it be your radio, your television, your stereo, or sensors like the MG811 or the SHT11 that you have on the ASP-2. There are lots of different kinds of noise. Some is like a hissssssssss, and some is like a pop!....pop!............pop! and some is like slowwowwwoowoooooooowwwwoo. Noise comes from the physics of the sensor itself (intrinsic), or from the circuit that amplifies the signal, or from the analog to digital converter, or from the rest of the system and the way it is wired up.
What should we expect for the resting level and sensitivity for this sensor? Given 100 of the sensors, how much variation would we see, compared to the graph in the data sheet? The data sheet does not tell us, does it? We can't test 100 sensors, but maybe 3 or 4. We want a graph like the good one that shows an unambiguous response to CO2. My plan for the data processing is to adjust the ASP-2 CO2 readings to fit the MG-811 typical curve. Say the ASP-2 says 2.5
Volts at a CO2 level where the MG-811 typical curve says 3 volts, that is our calibration multiplier, 3/2.5 = 1.2. I'll get into how to do that as we see reliable (clean response to breath CO2) data coming from the ASP-2.
Is it possible to install the sensor incorrectly in the socket?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Intrinsic means "belonging to a thing by its very nature". Noise can be defined as any unwanted signal in a system. Intrinsic noise is generated by the element itself. There are three types of noises: thermal, broadband and flicker noise. Thermal noise is generated by the random motion of electrons in a conductor.
Here is the link to the website I got it from: http://www.analogzone.com/avt_0904.pdf
Sean
Dr. Allen,
Ah yes! I've actually heard of that term before in terms of cable television. At last I know what all this noise we are seeing is called. As for what you said about installing the sensor incorrectly, as far as I know the sensor's pins aren't quite symmetrical, so it shouldn't fit if you turned it 180 degrees. I could be wrong though ...
Andrew
·· Mark today's date on your calendar: August 4, 2010. It's a banner day, the day the ASP-2 came to life! After spending all day yesterday at Dartmount Medical Center, Christopher and I spent most of today at the workbench. We soldered the resistors, capacitors, and wires directly to the ADC chip and perf board-- very tedious, focused, and time-consuming work for those who have never done it before (*you'll all get to do this at our next practice on August 14th.)
The soldering on our first perf board ADC chip looked "rough" but it worked OK. It gave us good, accurate CO2 millivoltage but the TP1 millivoltage was "stuck", as integers, between·4 and 8 (see attached graph #1.) However, the interference (intrinsic noise, or static) was virtually gone, the graph line showed little variation as it climbed beautifully to 3300 mV, and the sensor was very responsive to CO2 (our breath, which probably smelled like extra chunky peanut butter!) We weren't sure why to TP1 millivoltage stayed at 4-6 and we weren't content with the ratty soldering we did, so we made version 2 of the ADC chip perf board (we'llpost photos tomorrow.) This time the ASP-2 worked perfectly (see attached graph#2.)
We soldered all three resistors and the capacitor·directly to the top of the ADC's "legs" (pins) and we made sure that the Vss wire was in direct contact with the ADC chip pin,like Dr. Allen and some of the Rocketeers hypothesized and suggested. Loose wires were apparently the cause of the interference, as some of you suggested. Good teamwork, and thanks! Soldering the wires seems to have remedied the problem and I think we are almost there... August 4, 2010 was a banner day, thanks to our diligent team(work).
+++++++++++++++++++++++++++++++++++++++++++++++++++
HOMEWORK QUESTION:·Rocketeers, what·should we do with the constant that you calcualted a few days ago? Where does·it fit into the project, the program, and the grand scheme of things? Remember that the sensor is only outputting millivoltage right now, NOT CO2 data...
+++++++++++++++++++++++++++++++++++++++++++++++++++
Tracy,
·· Can you tell us how to get the program to write the TP1 millivoltage data to the flash drive, if it isn't going to complicate things? The program and the ASP-2 are·working too well·right now to jeaoparidze (jinx?) things.·It would give us yet another packet of flight data·to look at. Maybe it'll tell us something important, maybe not. Data is data and the more, the merrier! The TP1 millivoltage appears on the computer screen right now but it isn't written to the flash drive. I can't tell you ow giddy we were when the hardwired perf board ADC chip worked so well tonight!
Do you have any hikes or bike trips planned for the weekend? Being so close to the Northwest you could travel north and look for Sasquatch. I've heard that he's "out there", so be careful.·Some folks have seen him near Area 51 (but I'm not sure how credible the sources are.)·I spent yesterday at Dartmouth Medical Center going through a battery of tests and doctor's visits -- MRI, CAT scan, X-rays, getting poked with several different diameter needles-- as a follow-up to my cancer treatment. In the grand scheme of things, things are·good. That was relief! Did you see the aurora borealis last night?
Off to bed. Success! Thank you!
Mark
·· The CO2·sensor has six pins: 3 pins opposite another 3 pins, placed around the inside circumference of the round CO2·sensor. The pin spacing is symmetric and I've plugged the sensor in both ways with no adverse effects or 'bad' results.
It IS·possible to·attach the 4-pin (female) wire harness that we made·"backwards" onto the four CO2 sensor·module pins. But the four wires and our·harness·are well marked and that·didn't seem to be ·the problem.·Hypothesis rejected, but that's OK. It was certainly worth mentioning and checking. Ultimately I think it was simply loose wires; soldering everything on to the perf board remedied the problem(s)... knock on wood! We resume trials again tomorrow.
Mark·
I don't understand what you mean by "It gave us good, accurate CO2 millivoltage but the TP1 millivoltage was "stuck", as integers, between 4 and 8 (see attached graph #1.)" I see trial #4 and trial #5, and both of them seem to have good CO2 mV readings with good response to breath.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
I'm surprised at how different all of these sensors are in terms of the voltages they're putting out. It looks to me as though yours, Tracy's, and the one in the datasheet are all completely different. I believe that Tracy has a plan for taking care of that in order to calibrate things. The good news is that if you're reliably getting _a_ response (and it looks as though you are), you can fly the thing, and worry about what the data mean afterwards. Of course you may need the sensor in working condition to decipher your results, so it might be especially important to recover things in good condition.
Congratulations.
Mr. Kibler,
This is fantastic to hear, and not at all surprising after the amount of work the team has put into the ASP-2.
As I understand it, the constant is the only thing standing in the way of converting the millivoltage to an actual carbon dioxide level. Team members, please correct me if I am mistaken, but I believe we tried to figure how to go about replacing the current (incorrect) constant with our new one. However, as I remember we didn't understand the format of which the current constant was stored.
·
·· Success, thanks to you and an excellent team effort. Thank you!···
·
The ASP-2, in last night's configuration,·is running right now and all the data looks really solid. CO2 millivoltage is rising nicely to the stabilization point and there is virtually no fluctuation in the "other" millivoltage", the mV that's set with the potientiometer. The graph I just printed out, and all of it's accompanying data, is·perfect! More trials to come later in the day.
When I said, "It gave us good, accurate CO2 millivoltage but the TP1 millivoltage was "stuck", as integers, between 4 and 8", I was referring to the millivoltage that is adjusted with the potentiometer. I wasn't sure what to call it or how to reference it.
·
Attached is the current program, which is essentially the same as what we've been using all through July. I just re-named it "August 4" so we know that it's the version we were using when the ASP-2 came to life. Andrew, you may want to use this version of the program if you're not already. I'll post pictures later today. I have to run Christopher to work right·now (10 minutes ago!) We all deserve a celebratory root beer and vanilla ice cream float for dessert at dinner tonight. MAybe we can share one at the ARLISS launch in Nevada...?
·
"Success is 1% inspiration and 99% perspiration." ~ Thomas Edison
·
Success!
·
Mark
·
That is also how I see it, there was a format to the·incorrect constant that we could not figure out. I think we tried find·a converter online and we checked our "What's a Microcontroller" books and we could not figure it out.
Sean
Andrew
Well that was strange! It seems that our original forum is back and that we can add more posts to the same "old" forum. Looks like we don't have to start a new forum after all. But there seems to be a new web screen format. Does anyone else see a difference?
Good morning! More later...
Mark
OK, looks likewe're back up and running. One thing you have to do now is to figure out what do we do with the constant you calculated?
Sylvie makes a good observation and a good point. When the CO2 millivoltage stabilizes after it warms up, the millivoltage appears to be different from sensor to sensor. On Dr. Allen's graph the mV stabilized around at a different level than on our sensor. This will effect the number we get when CO2 millivoltage is multiplied by the constant.[/COLOR]
QUESTIONS:
1) How will this effect the number?
2) How do we compensate for the difference in the numbers in the two graphs?
I'm posting a test picture to see if it uploads to the new forum. It seems that everything else is in HTML code... no graphics :-(
NOTE: The forum seems to limiting the file size to 20 KB and I can't seem to upload any pictures, graphics, etc. Can someone else please try?
Thanks,
Mr. Kibler
Can anyone see the graphs, photos, etc. now? After the forum was 'fixed' I can't see any of them (and the "smilies" I try to post
> appear as HTML code.)
The ASP-2 is running right now with the ADC chip soldered to perf board and the CO2 sensor mounted on the ASP and ythe data is near-perfect...!
I'll send pictures as I promised when the forum is ready to receive them.
++++++++++++++++++++++++++++++++++++++
Sylvie made a good point several posts back. If there is variability in the CO2 millivoltage between each sensor, the same constant can't be used for each sensor (can it?)
1) How do we use the constant?
2) How do we factor in the difference between sensors so we can get a baseline CO2 reading. Otherwise the data is just millivoltage and tells us nothing about CO2 concentration, on the ground or in the air.
HOMEWORK: *Andrew, please set your team to work answering these two questions. We need the answers for the program and for practice.
Thanks for the cool Aeropac mouse pad!
Mr. Kibler
Here's where all the pictures, graphs, programs, and attachments are. It's a link the old forum, sent to us by Parallax:
http://forums.parallaxinc.com
Here's what Jessica Ulman at Parallax said in an e-mail to us:
"While you can't post new messages to this forum, you can use it as a reference for any old information.
All images that were attached to previous posts have been migrated to the new forum, but will appear as broken links since the URLs refer to the old forums site. We will be running a script hopefully during the weekend to update all image URLs. For now, please refer to the image attachments."
I'm not sure if we can post graphs and photos yet, but she said that we should be able to soon.
Mr. Kibler
It seems that the maximum file size for attachments is seriously limited...
Mr. Kibler
Attached (I hope) is a graph from the ASP-2 with the CO2 sensor and the ADC chip perf board permanently installed on the ASP. The data is perfect! There is virtually no fluctuation in the potientometer millivoltage and it only ranges from 1001 to 1004. Once the CO2 sensor heats up and stabilizes there is also no fluctuation in the data until CO2 is applied.
Now we need to figure out what to do about the constant...! Where and how do we insert it into the program so that the flash drive reads out "ppm" instead of just CO2 millivoltage?
Tracy, can you help us figure out how to write the program/code so that the potentiometer voltage also appears on the flash drive? Since it's there we may as well record it as flight data.
It looks like the new forum accepted the graph. Enjoy!
Good morning,
Mark
I will put it in simpler forms. (This is as if it was reading out mV and not ppm) Lets say the ASP is running inside the rocket and warming up, (on the graph it would have the slowly increasing mV as time passes). Then we launch. Lets say that the CO2 levels are decreasing as altitude increases. After we make a successful retrieval and look at the data, we would encounter a problem. We would see when the ASP is warming up, (the slowly increasing mV as time passes), then a sudden rise in mV, (when it was increasing in altitude after launch). How could we tell if it was done warming up and providing reliable data? Maybe it wasn't done warming up and was giving out data that was lets say 0.2 mV off what the actual reading were.
I have to go soon for mountain biking, this was just a quick idea of what we might encounter in Nevada as a problem.
Please let me know your thoughts on this.
So, are the CO2 mV readings now successfully recorded to the flash drive?
I now have two of the sensors and will hook them up to run side by side. We now know that there is variation in the resting level among different MG811 sensors. The question I want to answer is, "Does the variation also apply to the sensitivity". Suppose one sensor comes to equilibrium at 3 volts and another comes to equilibrium at 2 volts. They are side by side in the same CO2 environment, say, 390 ppm CO2. Now we expose them both at the same time to an increased CO2 level that brings the first one from 3000mV down to 2500 mV (500mV decrease, 16.7%). Will the second one decrease from 2000 down to 2500 (500mV) or from 2000 down to 1670 (16.7%), or will it do something else? That is, is the sensitivity absolute mV or a percentage of the resting level?
Question #1: What do you suppose the sensor would read if you seal it in the air space inside a bottle half full of soda water? Would that reading change with temperature?
Question #2: do you expect the mV to go up, or down, as the ASP-2 ascends to 10000 feet? (you looked at this question before, but it is time to think about it again!)
"When I looked at the graph a realized something that might pose a problem. When I looked at the graph, the first thing I noticed is when the sensor is warming up, (when mV slowly increases as time passes). If we are are in Nevada and make a successful launch, then look at the data, how can we differ the data from when it was warming up, from when it is actually taking reliable data?"
Dylan,
This is a good question with a simple answer. Remember the 'MAWD', the Mini-Altimeter with Deployment feature? It reads '0' the whole time the ASP-2 is on the ground, inside the rocket, and again after it lands. But the millisecond that the rocket takes off the altimeter starts reading and recording data to the flash drive... right beside the CO2 level, temperature, and humidity. As the ASP-2 ascends inside the rocket this is happening fast. But when the ASP is deployed from the rocket it descends on a parachute, reading and recording temperature, humidity, CO2 level, and altitude, all the way down until it lands. See the attached graph and you'll get a better idea. The only thing that's not on the graph is altitude. This brings up another important point (see below).
==============================
Sylvie and Tracy,
The ASP-2 continues to give us good, 'clean' data. Check out the graph from one of today's trials. However (malheureusment, Sylvie!) NO ALTITUDE DATA is being written to the flash drive. Sacre bleu!
Here's what we did, here's what we tried, and here's what happened.
1) We put the ASP, in it's present configuration, in a vacuum chamber like last year. This triggers the MAWD (altimeter). The MAWD beeped out the maximum pressure altitude but NO DATA WAS WRITTEN to the flash drive.
2) Same procedure as step #1 except we had the ASP connected directly to a computer so we could see the data read out live time. Same result: we got good CO2, temp,and humidity, the pressure altitude beeped out when we finished, but no altitude appeared on the computer screen nor was any written to the flash drive.
3) We made a new wire harness for the MAWD with an in-line resistor (like last year.) The blue wire goes to Vss and the brown wire top P15 (same as last year; same as in the program.)
4) We made a third wire harness without a resistor and got the same result. The altitude "beeps" out but no data appears on the computer screen or on the flash drive.
I really, REALLY hope this is a simple program glitch that Tracy can help us fix! The ASP-2 is virtually done so far as building and wiring it and we sure as heck don't want to work backwards... too close to the launch date and the start of the school year.
+++++++++++++++++++++++++++++++
Tracy and Sylvie, what do you think? Does this look like a hardware or software problem? What doyou suggest we do to fix it? (Please tell me it's an easy program fix! Please.)
Mark
"So, are the CO2 mV readings now successfully recorded to the flash drive?"
Yes, we're getting excellent CO2, temperature, humditity, and elpased time data written to the flash drive (see the graph below in my reply to Dylan.) But the altitude data-- which is critical in order to correlate CO2 with altitude-- is NOT being written to the flash drive. Again, see my post to Dylan. It explains what happened, what we did to try and fix it, and where things are as recently as 5 minutes ago...
I really, REALLY hope it's a simple program fix, an overlap between old and new code that's causing the glitch. The ASP/hardware is "right there" (visualize me making an "OK" sign with my finger and thumb) and I really hope we don't have to dismantle it. I'm sooooooooooooo afraid of breaking something critical and going backwards...!
Thanks for the homework questions! They're very relevant to Dylan's question on the forum. You make a good point about the variation between each sensor and how to compensate for the difference between each sensor. If one reads 2200 mV at 300 ppm and another reads 3500 at 300 ppm, that needs to be factored in and compensated for... before we launch.
So what do you think about the altitude not being written to the flash drive? Is it a hardware glitch or a simple program fix? (Please say it's a simple program fix!)
Pasta for dinner. Chow, and ciao!
Mark
"Aiiii. New forum software--This is going to take some getting used to. Many old posts are discombobulated... I can't count the number of times I have had to dump well oiled way of doing things in order to keep up, or to advance and lost things in the process."
Tracy,
Yes, SURPIRSE! A new forum with a new look! That's why we back up programs and data, right? Jessica Ulmann at Parallax e-mailed a link where we can access the pages from our previous forum:
http://forums.parallaxinc.com
I'm not certain whether all the graphs and program attachments will be there but most appear to be. Whew! (wiping make-believe sweat off of my brow with the back of my hand... where's THAT smiley face icon gone to?!)
Mark
We have altitude data on the flash drive! And CO2 (too!), and temperature, humidity, and "flight time." See the attached graph
>
It's late, we're tired, and we're exuberant! Silly us: we had the (ONLY) two wires going from the MAWD to the BOE reversed. How simple of a fix can that be? So we fixed it, as silly as it was for us to make such a rookie mistake. I guess we're not experts until next week, huh?
The ASP-2 performed nicely inside the vaccum chamber, cranking out reliable, interference-free (pressure) altitude, CO2 millivoltage, temperature and humidity. A few more tweaks and we may be there.
Speaking of "there", I'm off to bed. Christopher checked out 45 minutes ago while I stayed to put the graph together. Another eventful day at the Rocket Tree farm. Tee hee. 2 AM and I'm making double entendres, Sylvie!
Mark
Obviously that wasn't it, but if you do try to use a different one, don't forget that you need to do that.
Question #1: If I'm not interpreting this the wrong way, I believe that the CO2 reading would significantly change with temperature because as temperature increases, the water becomes less soluble which drives more CO2 out of the liquid soda into the air space (which the ASP can sense).
Question #2: I know talked about this a lot many pages back, but to recap, the millivoltage decreases when CO2 decreases. Logically speaking, with an increase in altitude, there is a decrease in air pressure. With a decrease in air pressure there should also be less CO2. However, I could not find any reliable sources to confirm this.
QUESTION #1: "...the CO2 reading would significantly change with temperature because as temperature increases, the water becomes less soluble which drives more CO2 out of the liquid soda into the air space (which the ASP can sense).
COMMENT: Andrew, Could you explain HOW the CO2 reading would change? Would it increass or decrease? Please explain which and explain why. *Recall the the ASP-2 on measures and records milliviltage right now, noy CO2 concentrationin parts per million (ppm.)
QUESTION #2: "...to recap, the millivoltage decreases when CO2 decreases."
COMMENT: Andrew, refer to the questions above. What happens to the millivoltage reading (CO2 level) when you breathed CO2 on Phidippides' CO2 sensor (before I borrowed it...!) Did it it increase or decrease? Please look at some of the recent graphs and disucssions on the old forum and report back. On my ASP-2 clone, when I breath CO2 on the sensor the mV decreases every time. Do I have it wired wrong...?! Would that change as altitude increases even if the CO2 concentration stays the same?
We had a long (2:30 AM!) and productive day-- and night-- yesterday. The ASP now records altitude, too!
The woman from Sant Bani Academy (Rose?) e-mailed yesterday about the rocketry camp we're doing the week after next. I'll forward the e-mail to you with the details. Our sessions are in the afternoon.
What do you think about us contacting Mr.and Mrs. Moran at the Gerlach, Nevada High School and offering to speak to their students about ARLISS/Parallax like we did last year? I think it would be great to re-connect with them. Maybe we could even invite Mr. Beaver and some of his science class students to join us at the launch? What do you think?
Yawn!
Mr. Kibler
"Question #1: What do you suppose the sensor would read if you seal it in the air space inside a bottle half full of soda water? Would that reading change with temperature?"
Answer: I agree with Andrew that more CO2 would be driven out of the water soda and into the air space in the bottle. Since more CO2 would be in the air space where the sensor is, the CO2 reading would increase and the amout of millivolts would decrease.
Question #2: do you expect the mV to go up, or down, as the ASP-2 ascends to 10000 feet? (you looked at this question before, but it is time to think about it again!)
CO2 decreases as millivolts increase (and vise versa). CO2 increases as altitude increases. Therefore, millivolts should decrease at 10,000 ft., right?
Justin
The curve for CO2 in your vacuum chamber is flat at about 3000 mV. That is interesting and begs a question. From the raw data, can you tell us if it is 3000 mV or closer to 3300 mV? It is hard to read from the graph.
Partial pressure (pp) of CO2 decreases as you run the vacuum pump. Partial pressures decrease along with total pressure. But the ratio of different gases does NOT change as you run the vacuum pump, so the parts-per-million of CO2 remains the same even though there is a lot less of it per unit of volume.
Now, has your experiment shown that CO2 is proportional to pp, or to ppm? Andrew's answer to question #2 went to the heart of this issue. The graph shows mV as a function of ppm of CO2, but the main thing that happens with altitude is what Andrew pointed out, less CO2 (meaning per unit volume).
The reason I asked about the 3000mV vs 3300 is that the electronics can only go up to a certain maximum somewhere below 3300. Might there be other things going on with the electronics or with extrapolation on the graph in Andrew's last post?
The CO2 concentration above soda water in a bottle should be quite high. Pressure causes CO2 to dissolve in water, more pressure, more CO2, and that is why when you take the cap off and release the pressure, bubbles instantly form in the soda as CO2 comes out of solution. Cold water can hold more CO2 than hot water, so given a set amount of CO2 in the bottle, the CO2 pressure should be greater over hot soda. Also water vapor, saturated, but much less partial pressure than CO2. Total pressure is the sum of the partial pressures of all the different gasses that are present.
After finally figuring out how to log in to Parallax, with much help from Justin, Andrew, and Mr. Kibler, here are my answers to your latest questions:
"Question #1: What do you suppose the sensor would read if you seal it in the air space inside a bottle half full of soda water? Would that reading change with temperature?"
The sensor would read CO2 because like Justin and Andrew said, CO2 would come from the soda water and go into the air space and thats where the sensor is so it will read CO2.
Question #2: do you expect the mV to go up, or down, as the ASP-2 ascends to 10000 feet? (you looked at this question before, but it is time to think about it again!)
Like Mr.Kibler and Justin said,CO2 decreases as millivolts increases and CO2 increases as altitude increases. So as I understand it, mV would decrease as the ASP-2 ascends to 10,000 ft.