Thank you so much for taking the time and your detailed reply
@CFassett .
There is a lot to unpack here, and I hope my post will be less confusing for readers if I address your remarks in the same order as in your post. However, I believe it is helpful to make one exception and start with my reply on your last remark as an introduction.
D: What I really want someone to experiment with, is the new Hamilton CO₂NTROL Dissolved Carbon Dioxide Sensor, which is a solid-state optical sensor that measures the CO₂ directly in the water and is supposed to be maintenance-free. I can't tell what the cost is though, you have to have a company and formally request a quote from Hamilton. Xiaozhuang Wong said that he was going to request a quote and if it was under $3k, he would get one. Never heard anything else about it though, I just left a follow up comment today. Even if it's crazy expensive and out of the range of the average hobbyist, I think it would be useful to have some people experimenting with it to provide *real* numbers about what's going on with CO₂ inside our tanks.
My initial inspiration for this thread was
@Art experimenting with a pH probe in a drop checker, using a membrane. I reviewed how professional dissolved CO2 sensors work, like the Hamilton sensor, and that led me to new insights how some progress for our hobby could be made, yet at affordable price levels for hobbyists.
Professional dissolved CO2 sensors measure the partial CO2 pressure in a gas pocket that is in equilibrium (following Henry’s law) with water. This is the graph from the Hamilton’s site:
The advantage of these methods is that one measures CO2 partial pressure in a gas pocket (the above CO₂NTROL Optical Measurement), rather than using pH as a proxy. When measuring CO2 in a gas pocket, we rely only on the gas equilibrium following Henry’s law, and will be independent on water KH, and all other chemistry.
Professional dissolved CO2 probes have always been expensive, and not affordable for hobbyists. These are complicated instruments, with a membrane and CO2 measurement technology, integrated into a portable housing that needs to be water proof, resistant to some level of water pressure for in situ measurements and serviceable. Then, these are not consumer products, so generally less volume efficiency for manufacturing and distribution.
I was really excited discovering how MEMS technology has brought state of the art CO2 measurement (Photoacoustic spectroscopy, but also other techniques) within a very low price range (37 USD) and small package. This is one piece of the puzzle for us to imitate what professional dissolved CO2 sensors, including Hamilton's, do. But then, why use a membrane, why build it in a portable measurement device, and why don’t we use the air pocket above the tank for our measurement? When combining these pieces of the puzzle we can re-create the benefits of professional dissolved CO2 meters but at a much lower price level for our hobby.
1. have you looked at
Hamza's Reef Calculators? I'm not suggesting that you actually use the calculators, but he provides 3 methods of measuring CO₂ in water with sources. The sources may be of interest. Additionally, the creator, Hamza Muhammad Arain, seems to have expertise in the area and may be of use as a consultant. He provides a direct contact email (
hamzasreef@gmail.com) as well as a means to contact him through the Reef Central forum at the bottom of the site
Thank you for this reference, I had not seen it before. The science papers he references to all use pH as a proxy for dissolved CO2. This is indeed a rather intricated science area, hundreds if not thousands of posts on aquarium fora, however with our method we can avoid the pitfalls as we measure CO2 directly in the gas phase and don’t need any knowledge about water chemistry or pH.
2. I've looked at Henry's Law probes in the past, and the way I've typically seen them implemented is essentially as digital drop checkers. I think mounting them within a canopy or under a lid would be slow and inaccurate, simply because of the volume of air involved and the lack of an airtight seal.
I would agree in principle if we would inject CO2 with a diffuser or reactor in the water, and use a probe above the water to control our injection. It would have to be tested though, as it is hard to estimate the relevant timings in the feedback loop.
It is important to note that in traditional methods we inject in the water and measure in the water, but we now change that to injecting above the water and measuring above the water.
For our control mechanism, where we stabilise the gas pocket above the water and let the water follow in equilibrium there is no longer this concern of slow feedback loops. I can elaborate if this is not clear to readers.
3. I'm not confident that a CO₂ injected tank is able to maintain itself on plant-produced O₂ alone.
It is important to note that we only supplement the air pocket above the tank from 0.04% to 1% CO2 or (much) less. This means that oxygen and nitrogen would be at nearly ambient levels (20%, 80%). With just small supplement of CO2 above the tank, we do not need the tank to be fully airtight. If we are worried about O2, then add an air pump for a guaranteed minimum level or O2 replenishment. Blowing in some air will give some loss of the <1% supplemented CO2 in the air that escapes, but according to my estimations these CO2 losses are far less than what we have in a CO2 injected open top tank where the CO2 is free to outgas into the open ambient air. My estimation is that this method can save up to 80-90% on CO2 consumption, even if the tank is not perfectly sealed
A: So, I've seen self-contained probes that have a gas permeable membrane and a sensor that sits inside the tiny air pocket. The size of the air pocket seems to be the biggest contributor to the speed of the detection. I believe these units typically cost more, but they also require frequent maintenance to stay accurate.
These are what I described as “professional dissolved CO2 probes”. These are the state of the art tools, as they measure CO2 directly and not using pH as proxy. With the new MEM’s technology and a membrane we could do the same, but I believe the method proposed in this thread will work better for our particular application (we don’t need to do field work in rivers, lake or oceans).
B: Another option would be to build a small reactor-type device with the sensor inside
Agree, it is not easy to design this so that the sensor will be guaranteed to no get submerged in aquarium water.
C: A design that would have a lot of promise would be essentially a "diving bell" design at the surface of the aquarium.
This could definitely work, but would need to be tested to verify if reaction time is good enough for the control loop. It is an idea that I like, no membrane and can use open top tank. Could use the MEM’s technology, and eliminate all water chemistry from the measurement.
Again thank you for your thoughts
@CFassett , and happy to follow up the dialogue if you’re not convinced with the above.
Note: This seems an attractive project for a hobbyist or science student. I may be wrong, but forgive me believing that this can be a really interesting innovation for our hobby, at low cost. A physics and electrical engineering student could make a inexpensive prototype, test it on a tank with water, and get to the bottom of this in a 1-2 months project. I am happy to be mentor, please reach out if I can help.
Yes, didn't mean to hijack the post's topic.
especially for your pending tube-in-a-box or tube-in-a-tube no-bypass reactor design!
