CO2 + temperature logger project

[HarveyH42]

Is there a reasonably affordable, accurate CO2 detector that's easy to read with a microcontroller (AVR)? Was just thinking about this while over reading the 'Climategate' thread... With twitter, and a cheap simple device, seems like we could generate our own global warming data, independently of politicians and crazed lunatics. Or, more realistically, we could monitor and graph the progression or progress locally. Might make a nice desktop toy, or even commercial product.

I've got most everything, graphic LCD, AVR, temperature sensors, but don't recall having seen a CO2 detector. Wonder if a carbon monoxide detector would be close enough, it's only off by one oxygen atom.

But, yes, I am seriously interested in building this project, and trying to keep it cheap and easy, with the hope of others doing the build and sharing their local readings online someday.

 

[tytower]

I'd be interested in joining in and supplying data from Australia .So put me on the list.
Don't know of a CO2 sensor directly but isn't it read from car exhausts along with carbon monoxide?

Have you noticed when you see photos and video ]shots of Glaciers , Ice and Snow at the caps and other places it always has a dark dirty layer of carbon dust on top. It does not evaporate and as the sun hits it it warms up . The ice melts below but the layer just sinks down on top.

Now I believe this is caused by burning coal for power . The Governments of many countries are doing this and profiting hugely from it, selling electricity to us .
It is clear from that, that they will do anything to cloud the issue and move the blame from themselves. I don't believe global warming exists but I do think the ice caps are melting.

We are not likely to change anything in time to arrest that melting . Our only hope is to clean up coal burning emmissions and any other contributor to that black coat and maybe then fresh falls of heavy snow will reverse the phenomena.

 

[DirtyLude]

Yes, that would resolve the issue. Amateurs using non laboratory calibrated equipment to put data together.

(^that was sarcasm)

Oxygen is measured in car exhaust sensors, not CO2. The level of sensitivity to measure CO2 in the atmosphere to be useful would be crazy. It would also be useless due to the non-scientific placement of the sensors. Urban centre's, farming areas, local conditions...

I am telling you with 100% certainty and I'm not trying to be condescending, but you don't have the ability to conduct a scientific study that is worth anything more than having fun trying to do it. Actual scientific organizations have a hard time putting forward statistics that aren't eaten by anyone who wants to object to it.

 

[HarveyH42]

Yes, that would resolve the issue. Amateurs using non laboratory calibrated equipment to put data together.

(^that was sarcasm)

Oxygen is measured in car exhaust sensors, not CO2. The level of sensitivity to measure CO2 in the atmosphere to be useful would be crazy. It would also be useless due to the non-scientific placement of the sensors. Urban centre's, farming areas, local conditions...

I am telling you with 100% certainty and I'm not trying to be condescending, but you don't have the ability to conduct a scientific study that is worth anything more than having fun trying to do it. Actual scientific organizations have a hard time putting forward statistics that aren't eaten by anyone who wants to object to it.

No offense taken, you kind of got the whole point of the project. It's doubtful the scientist are using much better methods of gather data. Don't recall ever seeing a map of all the log sites for their data collection, or reason for choosing a particular site. The data varies depending on location and elevation.

This way, the sampling locations would be random, the device I'm thinking, just records the average daily readings. It's not intended to predict the end of mankind. I think it would be interesting to see if the CO2 rises, falls, or basically stays at the same level over a period of time. The actual numbers can be left to climate scientist, whom I would expected to test calibrate their equipment before each reading, so the are accurate and precise. Instead of entrusting such critical work to some pot smoking under-graduate.

Anyway, it's just a gadget idea, might not be entirely practical or politically useful.

 

[ke5frf]

There is a point that the equipment being used, in order to be acceptable as scientific data, would have to be traceable to a calibration standard. This wouldn't be as difficult as it seems, however. The only obstacle is expense (and that is relative). Everything else is easy.

The equipment would have to be NIST traceable. The standard for equipment calibration is typically annual in frequency. Major cities all over the US have calibration shops with NIST traceable references. My company sends equipment to a local company called JW Test Systems. They typically charge $30-200 for a certification depending on the nature of the equipment. A simple CO2 analyzer wouldn't be on the high end. You do realze that CO2 detection can be done with very simple equipment?

The method is NDIR absorption. The equipment is simple, an IR source lamp, an optical path, and a detector...I hapen to work with NDIR equipment made by a Japanese company called Shimadzu. The test method is Total Organic Carbon analysis but the theory is the same.

Two paths of travel for the gases exist, one for the sample, another for the reference gas. The IR beam is introduced at equal intensisty, which can be balanced with reflective mirrors to the detector with the reference gas going through both legs. This is "zero". On the other side is a CAPACITIVE diaphragm or a THERMAL CONDUCTIVITY DETECTOR. The path of travel for both gas tubes will apply the flow equally on both capacitive plates or both resistive elements. When all things are equal, the capacitor or resistors will have a measureable stable signal.

When CO2 is presented to the sampling tube, working against the reference gas, (a pure gas such as helium or nitrogen will work) the absorption rate of the CO2 on the reference side will be different than the absorption rate on the measured side. The more CO2 is present, the more absorption will take place. In the case of a capacitive diaphragm this will cause temperature differential on the plates. A delta T will cause a change in pressure, causing the diaphragm to contract or compress, thus changing the reactance and the signal. In the case of thermal conductivity, a Wheatstone Bridge circuit will measure the change in resistance between the sample and reference legs.

This is a very simple and basic system. There are inexpensive portable detectors with 4-20 mA current loop signals available.

I'm going to post a theory of operation drawing.

 

[ke5frf]

IR absorption theory of operation

Here's an illustration that offers the basic concept.

 

[HarveyH42]

Doesn't seem like something to be built by the average hobbyist. Was hoping there was some kind of module available like CO detectors, alcohol detectors, Radon detectors...

Didn't get much chance to do much searching yesterday, having computer and internet problems. The internet seems to be the Wifi (free city), went back to dial up this morning, seems to be working. Gotten the Windows Blue-screen of death twice, once last night, once this morning. Little afraid it will die before I get this sent.

I know obtaining accurate and precise measurements will be virtual impossible, just consistent readings should be fine to measure any changes, that can be displayed on the LCD in an interesting graph format.

 

[HarveyH42]

The city WiFi isn't working out too good, can't remember dial-up being this slow, of well. Found some $20 CO2 sensors, the seem like all the same internally, just a different case. The data sheet is from Futurlec. Not too sure about how useful these are yet, have to search the part number, and see what kind of projects people have for these. The price is a little high, but not prohibitive, might give it a shot.