CO2 monitoring (and the X-CCR)

[joshk]

Stuff like the NERD

Speak for yourself, I love my NERD The only thing I hate is that Shearwater stopped issuing updates for the NERD 1 and continued to sell it for years afterwards until the NERD 2 came out.

I track the RMS RCT and RST time in my dive log and have found the RMS scrubber time predictions to be reasonably consistent, for both short and longer dives.

Honest question here, how do you know the predictions are consistent? Do you mean the RMS provided scrubber duration is consistent? Or that the usage is consistent against some other benchmark not described?

Where I see RMS providing value is that the scrubber is active and it's activity is changing. If you run down to the last few predicted minutes does that mean you're minutes away from a breakthrough? Or minutes away from some unknown buffer?

 

[uwxplorer]

Honest question here, how do you know the predictions are consistent? Do you mean the RMS provided scrubber duration is consistent? Or that the usage is consistent against some other benchmark not described?

Where I see RMS providing value is that the scrubber is active and it's activity is changing. If you run down to the last few predicted minutes does that mean you're minutes away from a breakthrough? Or minutes away from some unknown buffer?

The latter. The "buffer" is defined by rEvo based on their unpublished testing and assumes steady state function. The predicted time until "cycling" or "dump it all" point works like the "expected range" function of a car.
What happens if you anticipate a larger than before exertion for the final part of the dive? I guess you have to fudge the numbers and hope for the best. This is where the notion of "prediction" shows its limits, compared to a mere "reaction front" indicator such as pioneered by AP (the fuel gauge, which is the thing to rely on - when it is not defective).
As far as telling you that the scrubber is functioning, again, read JohnnyC's post again. I would love to talk with the guy who experienced a CO2 hit DESPITE of the rMS... Clearly those guys do long dives where a scrubber monitor might be helpful to push the limits.
Don't get me wrong, I was a very happy adopter of the rMS (despite the serial failures of my sensors), but in the end, I just couldn't stand the random drop or increase in predicted scrubber life from dive to dive, in dives which were pretty much exertion-free and in similar conditions.
I believe a tinge of skepticism is good to have when you are talking about a device you rely on to survive until the next dive.

 

[joshk]

The "buffer" is defined by rEvo based on their unpublished testing and assumes steady state function.

“Unpublished” shouldn't be one of the reasons we don't know something.

 

[Tassi Devil Diver]

Honest question here, how do you know the predictions are consistent? Do you mean the RMS provided scrubber duration is consistent? Or that the usage is consistent against some other benchmark not described?

I keep track of the RMS information in my dive log (RCT, RST, temperature and depth), for my local diving with water temperatures between 12-16C I am consistently getting 4 to 4.5hrs of scrubber time before I have to cycle the canisters (ie dump the top canister move the bottom canister into the top position and put a fresh canister into the bottom position). I have experienced no wild fluctuations or inconsistencies, I have not observed any unexplainable "random drop or increase in predicted scrubber life from dive to dive".

If you run down to the last few predicted minutes does that mean you're minutes away from a breakthrough? Or minutes away from some unknown buffer?

When I say regularly run the RMS down to the last few minutes of scrubber time, I am referring to cycle time (that is scrubber time remaining in the top canister). The RMS provides two remaining scrubber time readings RCT (remaining cycle time) and RST (remaining scrubber time). RCT is the time remaining before the top scrubber needs to be dumped, RST is the time remaining before both scrubbers need to be dumped. RST is considered for emergencies only. So I will happily run my RMS down to the last few minutes of RCT knowing I still have typically between 1-1.5hrs of RST remaining.

 

[uwxplorer]

“Unpublished” shouldn't be one of the reasons we don't know something.

rEvo has argued that they could not expose their trade secrets for fear that their competitors would steal them. That is their right, since they did not patent their method (AFAIK).

 

[uwxplorer]

I keep track of the RMS information in my dive log (RCT, RST, temperature and depth), for my local diving with water temperatures between 12-16C I am consistently getting 4 to 4.5hrs of scrubber time before I have to cycle the canisters (ie dump the top canister move the bottom canister into the top position and put a fresh canister into the bottom position). I have experienced no wild fluctuations or inconsistencies, I have not observed any unexplainable "random drop or increase in predicted scrubber life from dive to dive".



When I say regularly run the RMS down to the last few minutes of scrubber time, I am referring to cycle time (that is scrubber time remaining in the top canister). The RMS provides two remaining scrubber time readings RCT (remaining cycle time) and RST (remaining scrubber time). RCT is the time remaining before the top scrubber needs to be dumped, RST is the time remaining before both scrubbers need to be dumped. RST is considered for emergencies only. So I will happily run my RMS down to the last few minutes of RCT knowing I still have typically between 1-1.5hrs of RST remaining.

Listen, I am really glad for you. My average in similar conditions was around 3.5 hrs when it "worked" (I am lean diver, so I would have expected to be on the high end of scrubber duration, but who knows how the sex/weight information is handled - and useful- in the algorithm). Similarly shaped divers reported the same issues (not systematically, mind you).
What part of this is unit-to-unit variation, diver-to-diver variation? Maybe the upcoming paper will address some of this. I honestly doubt it, as this would be logistically pretty daunting to set up and carry out.

 

[joshk]

I keep track of the RMS information in my dive log (RCT, RST, temperature and depth), for my local diving with water temperatures between 12-16C I am consistently getting 4 to 4.5hrs of scrubber time before I have to cycle the canisters

Are these scrubber durations across the same workloads and depths or independent?

rEvo has argued that they could not expose their trade secrets for fear that their competitors would steal them. That is their right, since they did not patent their method (AFAIK).

Sure, it's their right. I disagree that it's the right choice and think we should push for open access to the data and methods to calculate information we depend on for diving.

 

[uwxplorer]

Are these scrubber durations across the same workloads and depths or independent?

They are dependent on many more parameters: temperature, basal metabolism, packing, sorb, etc.

Sure, it's their right. I disagree that it's the right choice and think we should push for open access to the data and methods to calculate information we depend on for diving.

I agree that this would be preferable. Better be realistic though...

 

[joshk]

They are dependent on many more parameters: temperature, basal metabolism, packing, sorb, etc.

Of course, but workload and depth are two I thought contributed most to scrubber duration and would be good benchmarks.

 

[Dr Simon Mitchell]

That's the problem with current CO2 sensing and temperature-based breakthrough estimations, they're just not that accurate or reliable. I know a couple people who have had CO2 hits despite zero warning whatsoever from either their RMS system or the temp stick. Again, zero indication that there was any CO2 issue and they (fortunately) bailed out and survived.

Hi again Johnny,

Just for completeness, if you read my long earlier post there may be a perfectly good reason for this. These events may have been due to CO2 retention. Remember that this can happen in the complete absence of any CO2 rebreathing, and so you would not expect any warning from the RMS or temp stick. You might ask why bailing out would help in that setting. Well, the divers probably bailed out onto a lower work of breathing system which would definitely help in a retention situation, and the fact that they bailed out indicated that they were well aware they had a problem and would have been consciously responding to their perceived need to breathe more (rather than sub-consciously not responding which is what leads to retention). I am not saying you are wrong, but rather I'm reminding you that there is a perfectly plausible explanation why a diver could have a CO2 hit with no warning from a temp stick or the RMS.

I know several others who have had alarms go off on a freshly packed scrubber 5 minutes into the dive with no other indicators that there was any kind of CO2-related issue.

This must have been with one of the infra-red CO2 monitors because a temp stick would not do this and the rationalisation you illustrate here exemplifies one of the problem with these CO2 monitors. If the diver is not going to believe the monitor, then there is not much point in having it. If this occurred in an inspo (for example) it could mean that the diver left the o ring and spacer out of the centre section during the assembly leading to bypass. If, however, the diver tells himself that the scrubber is new so the monitor must be wrong, then the point of having it is completely lost. Such rationalisations are all the more likely if there is a known incidence of false positives, which make it easier for the diver to dismiss alarms.

For both Johnny and UWExplorer,

The primary purpose of the tempstick study was to establish whether the temp stick (or RMS) would reliably warn the diver prior to true CO2 break through (to a dangerous level) during simulated dives involving low or moderate exercise. Although we found that this was essentially the case, there were aspects of the results that will interest you. I really can't say more before the paper is published.

Simon M