Shop OBEX P1 Docs P2 Docs Learn Events
dissolved co2 sensing — Parallax Forums

dissolved co2 sensing

Erik FriesenErik Friesen Posts: 1,071
edited 2012-07-10 08:01 in General Discussion
I am stumbling into and attempting to understand co2 sensing. I need a way to sense dissolved co2 in the 0-30 ppm range. Typical air sensors read in the 0-5000ppm range. Can anyone point me to a sumbersible probe that can do something like this?

I would prefer to use a ndir probe if possible. Could this be done using a membrane or "diving bell" of some sort?

Any reference on this subject would be appreciated also.

There are chemical probes on the market, but their use is not suited to an automated process.

here is an ndir
www.alphasense.com/ndir_sensor_notes.html
«1

Comments

  • LeonLeon Posts: 7,620
    edited 2010-03-11 21:08
    Measure the pH? That should be proportional to the CO2 concentration (it's a weak acid when dissolved in water). I'm not sure if it would be sensitive enough, though.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Leon Heller
    Amateur radio callsign: G1HSM
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-03-11 21:14
    That is actually what the chemical sensor is, only it has a membrane in front of it. The problem is, that you have to adjust the ph of the sample to 4.
  • John R.John R. Posts: 1,376
    edited 2010-03-11 21:53
    In the world of aquaria, especially in live plant environments (fresh water), there is a known relationship between water hardness, dissolved CO2, and pH, assuming that the all pH buffering is done with carbonate based buffers. In fact, monitoring the pH is what controls the CO2 Injectors. High pH => Turn the injector on.

    What's the application?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-03-11 22:15
    Catfish farming is the application. It has to be something that can be automated though, and we can't test every day for all those variables.
  • John R.John R. Posts: 1,376
    edited 2010-03-11 22:41
    Erik;

    I assume you have continuous pH measurement now. I don't know much about catfish farming. Does the process of raising catfish create phosphates in the water? If not, have you done any testing to determine how stable the "kH" (Carbonate Hardness) is? If this is fairly stable, and can be "assumed" to be within an acceptable range, you could pull the CO2 level from a chart based on pH.

    If the kH is all over the place, and/or you have phosphor or other non-carbonate buffers involved, that's a different story.

    If you find that the kH follows closely with TDH/GH (Total Hardness), you may also be able to get a value that's "close enough".

    I'm not aware of anyone doing direct measurement of disolved CO2 at the hobby level. There are some inverted bell type indicators, but these are based on liquid pH indicators inside the bell, and again, are only useful if the kH is know, and there are no non-carbonate buffers.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-03-11 22:45
    We are doing no ph testing at this point. Phosphates are in there. I have my doubts we can assume anything on these ponds, thus my hesitancy with the whole ph based idea. I am still getting into this co2 thing, trying to get a better understanding of it.

    aercon.net/index.html
  • John R.John R. Posts: 1,376
    edited 2010-03-11 23:09
    Ahhh yes... THAT kind of pond...

    As soon as you have any non-carbonate type pH buffering (including but not limited to phosphorus), the pH to CO2 relationship is toast.

    In the hobby of aquarium keeping, we face other fun stuff to measure, for example Iron. If you use the same kit to measure the same water 5 times, you're just as likely to get 5 significantly different answers as not. What it takes to accurately/meaningfully measure Iron is not practicable at this level, either in complexity of chemical testing, and/or cost (chemical reagents and/or equipment).

    What we end up resorting to is a combination of "assumptions" and/or secondary measurements for stuff like this.

    Some questions to possibly guide the discussion:

    * What is the purpose of measuring the CO2? Or what are the positive/negative effects of too much/too little CO2?

    * Is there some other variable that will change with CO2 that can be measured/observed?

    * Do we know how fast the CO2 level will change, and how much change is "significant"?

    * How accurate do we really need to know the CO2 level?

    * What causes the CO2 level to change?

    Some of these you may not know, and I don't know that even if we knew all the answers, we'd come up with a workable solution. I know in my planted fish tanks, the answers go something like this:

    We want CO2 about 30 PPM for optimal plant growth. CO2 levels, left to their own, will cycle daily, with higher levels at night, and lower during the day because of photosynthesis. CO2 will also "bleed off" through the surface of the water.

    If CO2 levels are too low, our plants won't grow as fast. If they are too high, we risk algae bloom and even fish die off.

    We can inject CO2 into the water. We control this by monitoring the pH. There is a very specific relationship between kH, pH and dissolved CO2.

    We need an alarm if the pH gets too low, as this may indicate the CO2 injector solenoid is "stuck open". This will crash the tank. As a result, pH needs to be constantly monitored.


    Obviously, you answers will be somewhat different, and the implications of the highly variable environment will have significant impact. An aquarium is a (relatively) controlled environment. What causes CO2 to change in your ponds is much more variable, and can also involve things like how acidic the rain is, what kinds of substrate you have, types of plants, etc.

    Are you after CO2 measurement for plant health or fish health? Is there something more easily measured? Perhaps Dissolved O2? (still not easy, but at least there are readily available measurement systems).

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-03-11 23:23
    Dissolved 02 is a piece of cake relatively. We are using galvanic sensors, and pretty much have that down. We are after co2 for fish health. Interestingly enough, aeration is almost never needed during the day. Algae has the capability of supersaturating a pond greater than 20ppm oxgyen.

    The co2 in these ponds will be mostly fish byproduct. The industry is starting to see a need in this area, partly to pinpoint problem ponds. $70,000 in single pond fish loss in not unheard of in this industry, and this can happen in a matter of an hours.

    Catfish can live comfortably in 4-5 ppm oxygen , and if the aerators are on, the farmers tell me they can survive down to 1ppm and less. By the way, trout require somewhere around 10ppm just to stay alive.
  • John R.John R. Posts: 1,376
    edited 2010-03-11 23:44
    Is the CO2 problem tied to Nitrites/Nitrates? In the miniature glass ponds, monitoring the nitrogen cycle is important. I don't know if this is any easier to measure electronically, and don't know if this is a factor at all in the fish ponds. Possibly just something easier to measure as an indicator.

    If you have a local "brew and grow" type shop (hydroponics), it might be worth a trip. The typical fish store does not get into the whole CO2 thing but the growers do. They might not have much more than the pH measurement thing happening, but that would be one place to start. They are also more likely to deal with the whole "non-carbonate buffering" thing.

    I'll be interested to see what you come up with. I'm in the process of growing from a couple of fish tanks to about 50 (breeding of aquarium fish), and need to automate my testing...

    I'd really LOVE to find an affordable direct CO2 reading system, but I'm not holding my breath.

    This is an interesting problem, and I can relate to the loss. Smaller scale on my end, but several hundred in fish can go poof very quickly. For others watching the thread, part of the problem is that the difference between not quite too much and too much is very small, and there are typically no secondary indications (the fish don't look sick first, they just die).

    In the case of your ponds, a thought just hit me. You may be able to "cheat" in your measurements. You really don't care about the level of CO2 par se, but that you are below a certain threshold. This may make measurement easier, and testing may show that pH or another secondary measurement could be used. Even if just to say, you need to check out the South East pond. I don't know if that gets the farmer what he wants, but it might be a first step. We're also looking at "changes" as opposed to "steady states", and this can also make secondary measurements more useful.

    In essence, can we change the problem from "what is the ppm of dissolved CO2" to something like "The value of XXX is changing in direction Y rapidly! Get your butt down to the pond and and check things out!"

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log
  • John R.John R. Posts: 1,376
    edited 2010-03-11 23:47
    Another thought: Can you measure the atmospheric CO2 level in the air, just above the pond surface to get a correlation? This is a dissolved gas, and as the level rises in the pond, some of this is going to be escaping through the surface. Possibly a spike in the air could be the warning.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-03-12 00:14
    I am still thinking on that, an oxygen sensor works similar in that it uses a 2 mil membrane that oxygen can cross, but not the water.
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-03-12 14:35
    Does anyone know offhand, what the relationship is between co2 in air and dissolved co2? For example, if you place a volume of air with a specific ppm in a sealed space with water, lets just say a sealed glass jar, when they reach equilibrium, what will the corresponding ppm be in both?
  • John R.John R. Posts: 1,376
    edited 2010-03-12 15:16
    I don't think this will not be a constant ratio/relationship.

    The ability of the water to hold CO2 is variable, depending on the temperature for certain, possibly/probably pH and other chemistry factors as well. The less CO2 the water will hold, the higher the PPM in the air at equalibrium.

    That aside, it leads to interesting thoughts. How signification are the non-temperature values? What happens if the atmosphere in the sealed chamber is something like nitrogen (Virtually 0 PPM CO2)

    I might have to order me a couple of the parallax CO2 sensors to do some experimenting with.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-11-06 18:25
    continuing this thread...

    How is the parallax mg811 co2 sensor as far as drift and calibration? What is the zero point? Does the output zero at 350ppm, or is it simply not tested below that?

    It seems a little unclear to me.

    This is essentially what I want to do - http://onlinelibrary.wiley.com/doi/10.1002/eco.95/pdf

    I can see no reason why equilibrium won't happen across a membrane. The principle seems similar to these http://www.greenleafaquariums.com/co2-drop-checkers/gleaf-drop-checker.html

    If equilibrium can happen across an air barrier, that tells me that the air at transfer will be quite high, for 15-30 ppm in water, I guesstimate 3000-5000 ppm in the air. My algebra is a little weak, I don't fully understand the formula's I need to calculate this, however, it seems to me that the available, or free co2 measured by this method is really what is important to fish health, anyway.

    The real problem with PH based measurement, is that a tenth point of ph makes a large difference in the ppm, thus making calibration and maintenance of automated ph measurement more headache than it is worth.
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2010-11-08 00:42
    There was an effort in the Stamps-in-class forum in the Arliss project thread to evaluate a batch of MG811 CO2 sensors. I can't vouch for exactly how the data were obtained, but you can see that there is a lot of variation in the equilibrium level as well as the warmup time and the noise level:
    attachment.php?attachmentid=72381&d=1281400986
    That is from Mark Kibler's post #737 on page 37 of the thread. There are also graphs with response to breath. The response is okay to higher concentrations, but it seems to bottom out and does not respond to concentrations less than about 350ppm.
    These do use a lot of power relatively speaking and become quite hot in operation. That might be a problem in an underwater cavity, if only for the reason that solubility of CO2 in water decreases with temperature. An NDIR type sensor as used in the papers you've referenced would probably be more satisfactory in terms of both accuracy and practicality.

    CO2 solubility in equilibrium with water is governed by Henry's law. At room temperature, divide partial pressure of CO2 by 29.41 to find the moles/liter dissolved (at equilibrium). In mass units CO2 is 44 grams per mole. The wiki on Henry's law has a pretty good description, complicated by the temperature dependence and by the fact that about 1% of the CO2 reacts to form the carbonate.
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-11-08 04:41
    The wiki on Henry's law has a pretty good description, complicated by the temperature dependence and by the fact that about 1% of the CO2 reacts to form the carbonate.

    I am still getting a grasp on the co2 thing, but if I understand it, figuring the ppm concentration in water from the air will be a bit tricky, without some external water testing for Kh?? periodically. However, it seems to me that the real number you get from the air is as important or more, as any ppm number directly from the water, not?
  • Tracy AllenTracy Allen Posts: 6,664
    edited 2010-11-08 11:55
    I don't know from a practical standpoint, but am interested in the subject. I've never had an aquarium, much less a fish farm. I did build instrumentation for a facility to grow algae (for fiber, paper) at the site of a former fish farm, but we were not monitoring or controlling CO2(aq) or Kh.

    John mentioned a chart. I did find this one interesting:
    http://www.thekrib.com/Plants/CO2/kh-ph-co2-chart.html

    I should have said in reference to Henry's law, "at room temperature, divide partial pressure (in atm) of CO2 by 29.41 to find the moles/liter dissolved (at equilibrium), then a) to convert to ppm mole fraction, divide that by 55.6 (moles of H2O in one liter), or to convert to mg/l multiply times 44000 (CO2 at 44 grams per mole". How likely is the system is to be in equilibrium, given the rate constants of CO2(air)<-->CO2(aq) and the associated carbonate reactions? It does seem that measuring the CO2 in air in a bell or cavity could be indicative of changes if not absolute values.
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-11-20 12:41
    Here is something interesting in regard to dissolved oxgyen - http://www.likeafish.biz/

    Anyway, thanks tracy, I finally am to the point where I am starting to get a grasp of this thing. Do you know of any spreadsheet formulas that include temperature?

    What about mol conversion to ppm? I am a little unclear. Can you recommend some starter text books on this stuff?

    Here is what I have so far.
    1446 x 1200 - 132K
    1156 x 1200 - 90K
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-11-20 13:52
    Check these formulas out, see if they look right.
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2011-01-06 15:11
    Yet another question -

    Do the total co2 molecules (in equilibrated air/water by volume) for a given concentration in the water, change with depth? I think they don't. I think the air sensor reading may stay the same, as it is reading whats left of the absorbed light. As pressure goes up, the equilibrium goes down, IE( at 25C - 1 atm, 30ppm water = 19563ppm air 1.2 atm 30ppm water = 16300ppm air.)
  • $WMc%$WMc% Posts: 1,884
    edited 2011-01-07 17:03
    Erik Friesen:
    '
    Have you looked at ORP. This is a form of PH but with oxygen in mind, it should be right up your alley.
    '
    ORP = Oxygen Reduction Potential
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2011-01-07 20:07
    We have a working oxygen measurement system though. One thing this method does is cut out the uncertainty of ph based measurement. Any ad that advertises trouble free ph measurement with no long term drift, etc. is suspect.

    Look around at all the trouble fish tank owners (with plants) go to figure out their carbon dioxide. You about have to be a chemist to keep everything calibrated.
  • $WMc%$WMc% Posts: 1,884
    edited 2011-01-08 10:26
    We have a working oxygen measurement system though. One thing this method does is cut out the uncertainty of ph based measurement. Any ad that advertises trouble free ph measurement with no long term drift, etc. is suspect.

    Look around at all the trouble fish tank owners (with plants) go to figure out their carbon dioxide. You about have to be a chemist to keep everything calibrated.
    '
    Any thing that advertises "Trouble free PH measurement" is a lie.
  • RobotBrianRobotBrian Posts: 1
    edited 2011-02-02 10:41
    Check out this sensor from the oil and gas industry. I think they are pricey, but compared to the $70k loss you talked about, it should a deal!

    http://www.contros.eu/products-hydroC-CO2-OG.html

    Applications In-situ measurement of dissolved CO2 motivated by:
    Air- sea gas exchange; Ocean Acidification; Limnology; Climate Studies; Agriculture/ Fish
    Farming; Fresh water control; Carbon Capture and Storage (CCS)
    Detector High-precision optical analysing NDIR system ● Maintenance free
    Principle Dissolved CO2 molecules diffuse through a silicone membrane into the patented detector
    chamber, where their number is determined by means of IR absorption spectrometry.
    Concentration dependent IR light intensities are converted into output signals.
    Dimension/ Weight 90 d x 526 mm corrosion-free titanium/ 5,9 kg (2,6kg in water)
    Operation depth 2000, 4000, 6000 m version available
    Temperature range 3 … +30°C (Arctic version: -4 … +30°C)
    Measuring range 0… 3000 ppm (other ranges available) ● Standard calibration is 200… 1000ppm ● Units
    selectable: µatm, ppm, etc.
    Equilibration time first signal after 5s, T63 ~ 1min (with external pump)
    Resolution < 1 ppm
    Accuracy ± 1% of upper range value (as the total sum of all errors)
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2011-02-02 10:50
    That is industrial grade, I would guess they are around $5000. I am close to having something ready to go here, I'll have a fraction of that cost into it. Mine only has to withstand a couple feet of head pressure, that one ??? Its a nice looking unit, similar in form to the one I am building. I hadn't seen that one.
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2011-03-01 13:36
    Here is what I have so far, it seems to be working nicely. I am struggling a little try to figure out a proper potting compound, it tends to separate from the cast acrylic machining. I would like a clear polyurethane, but I can't find any with a waterproof rating, only water resistant.
    867 x 2048 - 132K
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2011-06-13 13:23
    Thought I'd update this, this is what we have so far.
    528 x 1312 - 200K
    pdf
    148K
  • elefixelefix Posts: 1
    edited 2012-07-09 23:17
    Curious how this project went. Were you successful in producing an accurate aqueous CO2 probe? If so, can you share details? If not, what issues did you run into?
  • Erik FriesenErik Friesen Posts: 1,071
    edited 2012-07-10 05:31
    I have had one in my fish tank since the last post. I had one sensor go bad because a sucker fish latched onto the face and ruined the membrane, and sucked water inside. Since then I double layered, and haven't had any issues. As far as stability, this appears to be quite stable, even over a years time.

    I have been busy with other things, so we haven't been building these lately. One problem I have run into is with the potting, if the parts are not perfectly installed, they tend to pull of the board, and then they can't be fixed.

    We do intend to build more of these though.

    This sure beats the ph system of measuring co2. In fact, we intend to do a estimated ph reading based off the co2. As you probably know, ph is a headache, for sure if you try to automate it.

    The graph is monitoring the fishtank in my office, you can see how it follows day and night, affected mostly by human respiration.

    If you want one, let me know, I shouldn't really release complete plans here, as this is a paid project for www.aercon.net
    952 x 585 - 103K
  • BitsBits Posts: 414
    edited 2012-07-10 07:24
    Couldn't you simply use a laser spectrometer? I think a crude one could be made using the propeller and some other optics.
Sign In or Register to comment.