[microtonal]

I think the issue is that the common tech requires sensors in an air-chamber. E.g. NDIR works by firing IR at a frequency that is absorbed by CO2. A sensor on the other side either measures the amount of IR light that got through (optical NDIR) or pressure/sound waves (photoacoustic NDIR). I guess that it's hard to use any existing sensors, because they are relatively large and probably water could easily get into the chamber.

Would be extremely cool if Apple, Samsung, and others can crack this, though I think they'd have done it already if it was easy.

[londons_explore]

Oxygen sensors used in car catalyst systems use a different effect based on electrochemistry. I see no reason that couldn't be minuaritized to grain-of-sand size.

The question is if oxygen levels are as good an indicator as CO2 levels... I suspect not.

[GuB-42]

Based on numbers, O2 concentration is probably not a good indicator.

Clean air contains about 20.9% O2 and 0.04% CO2. At 2000 ppm CO2, which according to the author is bad enough to impair judgement, that's 0.2% CO2, it that CO2 is the result of respiration, it means that about 0.2% O2 was consumed, so that's a drop from 20.9% to 20.7%, a very small difference. 20.7% is not low enough to have a significant effect, the CO2 itself is the problem, not the drop in O2. And using O2 concentration as a proxy for CO2 doesn't look reliable to me: the difference is small and other things, like humidity can affect O2 concentration.

As for the sensor, O2 sensor in cars compare the O2 concentration between the outside air and exhaust gases, it needs outside air as a reference, but what you are measuring is the outside air itself, you don't have that reference.

[kryogen1c]

>that's 0.2% CO2, it that CO2 is the result of respiration, it means that about 0.2% O2 was consumed,

I dont know anything about human respiration, but I know a little about chemistry and theres no reason to assume this is true. Basic stoichiometry.

According to a random article on the internet[1], nominal co2 production is 80% of oxygen consumption.

Your point appears broadly correct, just wanted to point out some faulty reasoning that could lead to incorrect results in the future.

[1] https://societymechanicalventilation.org/wp-content/uploads/...

[GuB-42]

CO2 concentration doesn't start at zero, and by coincidence, if CO2 production is 80% of oxygen consumption, consuming 0.2% oxygen results in 0.16% CO2, add it to the base 0.04% and you get 0.2%.

[jmb99]

> O2 sensor in cars compare the O2 concentration between the outside air and exhaust gases, it needs outside air as a reference

Source?

[GuB-42]

https://en.wikipedia.org/wiki/Oxygen_sensor#Automotive_appli...

It looks that some O2 sensors that don't require a reference have been used (titania sensors) but even though they have some advantages, they are less precise and mostly obsolete.

[roland35]

I had a project miniaturizing nasa tech for detecting hypoxia with o2 and CO2 sensors. It used a phosphorescent dye that changed a delay flash (ie you blinked a light, the dye would absorbed and blink back after a delay) based on temp and o2.

CO2 was measured with infrared but water also absorbed it, so you need to heat things up enough to not have water. It can be small, but not watch small.

All and all interesting stuff!

[derefr]

> CO2 was measured with infrared but water also absorbed it, so you need to heat things up enough to not have water. It can be small, but not watch small.

Can't you just measure CO2 "naively"; but then also, separately, measure rH; and then use the rH value to grab a research-calibrated LUT to pass the raw CO2 value through?

(I presume this is why all the little standalone CO2-sensor boxes you can buy also have rH displays. They're measuring it anyway to normalize the CO2 value, so they may as well make it a feature and display it.)

[roland35]

You could maybe, but water tended to collect on the surface (this was exhaled breath, so pretty humid).

Maybe if it was ambient air and not breath the humidity might not be as bad?

[picture]

Electrochemical pile style oxygen sensors continuously deplete themselves whether actively measured or not. Common smart home oxygen piles have a fixed lifetime of a few years, and they're quite sizable (probably about as much volume as a whole smartwatch). Putting the same chemistry in an even smaller package would likely result in lifetime measured in hours

[pfdietz]

I assume this is because of diffusion of materials at elevated temperature. The sensor would, I think, require a lower temperature than an electrolyzer, since the current would be much lower. But it would be best if lower temperature solid oxide electrolytes could be discovered.

[Lwerewolf]

The ones in cars need to be heated up quite a bit in order to work, and you still need reference air. Otherwise, I'm pretty sure that CO2 isn't a problem but rather an indication of a lack of oxygen in the first place, so it technically could work... just not if you're measuring the environment itself.

[XorNot]

This is in theory not a problem: getting an oxygen sensor to 700 degrees if it's a tiny spec on a chip is not necessarily hard or would even require a lot of power.

But...oxygen concentration is essentially indepedent of CO2. We measure CO2 at part per million levels, whereas O2 is 20% of the air.

(In that context CO2 is surprisingly toxic given that 1000 ppm can impair mental acuity).

[coryrc]

> whereas O2 is 20% of the air.

The goal of a gasoline engine's sensor is to accurately and precisely measure the point where O2 concentration reaches zero, so ambient air levels are not quite as relevant.

[OutOfHere]

No, it's nonsense to assert that CO2 is due to a lack of oxygen.

[noja]

I don’t think we actually care about co2 levels. I think we use them as a proxy for o2 levels (same as our bodies do). So your idea would be great.

[XorNot]

This is extremely wrong: CO2 impairment kicks in around 1000 ppm[1] possibly lower.

You can hit this breathing by yourself in an unventilated 3x3m room (literally measured in my house).

1 https://pmc.ncbi.nlm.nih.gov/articles/PMC4892924/

[spockz]

You hit it even easier when driving in a car with the internal circulation turned on to keep nasty fumes out.

[uxhacker]

In a car with Recirculation Mode on Levels routinely spike between 1,500 and 4,000 ppm within 20 to 30 minutes.

I wonder how many driving accidents can be saved by having a co2 monitor in the car.

[eurleif]

I had to drive a rental car for a bit recently, after getting rear-ended, and I was shocked to discover that it defaulted to recirculation mode every time it was turned on, regardless of whether you'd turned it off the previous time. I felt light-headed in it a few times before I realized I needed to manually turn recirculation off every time. Horrible behavior, and I don't doubt that it's responsible for many accidents.

[teravor]

there is some ambiguity here. submariners are exposed to really high co2 levels and are doing fine. it's possible that regular high co2 is just a proxy for bad air, and that you can acclimate to purely high co2 pretty well.

i have felt bad in high co2 environments before but i have never been in a controlled high co2 environment.

[XorNot]

I'm not implying permanent effects. The studies involved show a marginal reduction in cognitive reasoning on standardized tests.

Which is the point: if you're in a room trying to do deep work, that's likely a problem. I suspect submarines are all over the place here: you might be exposed to higher levels routinely but you also have regular access to chemical scrubbers and decent ventilation. I'd be fascinated to know what levels are tracked as normal.

[teravor]

apparently those scrubbers are less efficient the lower the co2 is (also need to move much more air which ruins stealth). iirc submarines are routinely exposed to multiple thousands.

[lelanthran]

> I think we use them as a proxy for o2 levels (same as our bodies do).

Probably. ISTR that depriving a body of oxygen results in a different response than overloading the body on CO2. It's why if you completely displace all air in the room with CO2, people choke, panic, etc, but if you use Nitrogen, people just keel over dead without realising it.

[mathgeek]

The evolved response to CO2 is part of the human body’s ability to filter and remove CO2 via the respiratory system. AFAIK we don’t have similar capacity for Nitrogen because it’s not a primary waste product of that system.

[wongarsu]

Dissolving CO2 in water creates some carbonic acid (H2CO3) that will decompose back to water and CO2 when the CO2 concentration drops. Blood has a fair bit of water, and carbonic acid is much easier to detect than oxygen or nitrogen gas

We evolved to detect CO2 because that's by far the easiest thing to detect that's still a reasonable proxy for the performance of our respiratory system

[mathgeek]

We evolved to detect CO2 because it was an evolutionary advantage over those who didn’t.

[vintermann]

I'm pretty sure that in a room where we replaced nitrogen with co2, we would be dead even if O2 concentrations were the same. Something about partial pressure. I notice AI explanations agree with me (not going to copy and paste them).

[foobarbecue]

EDIT: ignore this; I was confused / misinformed

It's about pH. CO2 creates carbonic acid when it dissolves in water. Your blood pH, in turn, controls how much you feel like you need to breathe. So with high CO2, your respiration rate slows down, and that can lead to low oxygen levels.

Note that the physiology and biochemistry of this is complicated (e.g. blood is a very good pH buffer and it's actively regulated by kidneys etc) and it's very much a nascent field of research, so I think AI will be overconfident and hallucination-prone.

Source: I worked in high-co2 caves for my PhD so have read about this a lot. I always carried a CO2 monitor. Our rule was to get out if we saw 20,000 ppm or greater. I spent thousands of hours above 10,000ppm.

[jijijijij]

You are right about the pH implications, but respiratory acidosis leads to hyperventilation, not hypoventilation. CO2 will kill you regardless of oxygen supply.

https://en.wikipedia.org/wiki/Hypercapnia

[ahartmetz]

My medical student flatmates were talking a lot about acidosis and alcalosis :)

It was the first time that I heard about them. These basically never happen if your body and environment are halfway decent, but they are important in exceptional situations.

[dummydummy1234]

Wouldn't high CO2 make you breath faster?

[foobarbecue]

Interesting, the linked article does say that.

Pretty sure I learned the effect was the opposite (high CO2 --> slower respiration). Note that that was ~15 years ago when I would have read that. Maybe I just misunderstood, or thinking has changed.

edit: reading now I see I was wrong about this. Thanks for the correction!

[dgellow]

What makes you think that?

[OutOfHere]

Huh. You don't know that, and are making it up. It's almost certainly false.

[noosphr]

We do.

[NathanielK]

Sensiron STCC4 uses thermal conductivity sensing thats very compact (4x4x1.2mm). It's pretty new to the market, but maybe in the future it'll happen.

[kingkawn]

Mesh with the other Apple/et al devices in the room to take multiple samples and aggregate the results for an overall picture of the ambient co2

[stonegray]

Just waiting for the followup post on HN: How I sent CO2 warnings to my entire office using an ESP32

[amelius]

What if a movie theater puts an Apple CO2 meter next to an air inlet? Everybody will think the air is safe.

[kingkawn]

If that’s the sole source and the application does thoughtful analysis it could determine that there are sections of the room that are better than others, yes

[amelius]

But realistically, using what sensors?

And (maybe less realistically) what if the theater puts 5 Apple sensors inside a sealed CO2-free chamber, spread around the room?

[kingkawn]

That’s the point of this thread that each device would have a small sensor that would sync and aggregate with others in the room

[amelius]

I think the thread established that CO2 sensors are too bulky for that.