CNS Calculation and Verification Model

AD_ward9

Bar and ATA are almost the same: 1 ATA is 1.013 bar (the mean pressure at sea level).

The papers on which the CNS limits are determined, are in ATA, so we use ATA as the reference. The reason ATA is used is because deco algorithms are generally in ATA, because everything is reference to the surface pressure.

Plotted below is the effect of adding the CNS margins being proposed here, in ATA.

Cheers,

Alex

Attached Images

CNS_margins.gif (7.1 KB, 112 views)

Drmike

Hi Alex,

What validation has ever been done on these CNS limits?

I ask because it seems to me that many divers exceed them regularly and some divers (myself included) routinely do dives with huge CNS% with no ill effect.
I know its hardly subjective but when you have divers doing dives with huge% CNS it does make a bit of a mockery

__________________
Get a girlfriend you sad twat - a Rebreather is an unfaithful mistress - dont blind yourself to her faults just because she goes down on you

Genesis

What if the "clock" model is just plain wrong?

What if the actual model is one of risk per hour of exposure, with the risk changing with increasing PO2, but there is no clock?

That is, what if at 1.6 the risk is, say, 1:1,000
At 1.5 its 1:2000
At 1.4 its 1:3000
at 1.3 its 1:10,000
at 1.2 its 1:100,000
at 1.1 its 1:1,000,000
at 1.0 its 1:infinite

Or something like this

In other words, it decays more rapidly as you go lower, but the risk is there from the first minute of exposure.

This would likely LOOK like a clock most of the time, but it explain why you "get away with" blowing the CNS clock numbers, because there is in fact no clock!

Something to think about..... (yes, I know that view is directly contrary to what is taught - but it sure fits with what people are doing, and what is actually happening, doesn't it?)

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

AD_ward9

There are a lot of papers measuring CNS on humans, rats and pigs. There is an accepted ratio between incidence in rats and humans, as there is also for DCS (it is about 4:1 if I remember, rats being 4 times more tolerant). These give the incidence of O2 tox: the 75% level causing O2 tox symptoms in 1% of the study group. Unfortunately this gives reasonable data on risk with high CNS loadings, but not how the curve reduces with lower loadings, because the sample group is small in each study.

There is concensus that there is a wide range in susceptibility. There is only very weak statistical correlation between specific individuals in a study group and O2 tox susceptibility. This means you can get away with it one day, and be hit the next.

I would point to the fact that 6 incidents of apparent CNS convulsions on this forum would indicate a risk factor of between 1 in 10,000 and 1 in 100,000 hours. Most divers do not run CNS loadings above 50% on many dives, so anecdotal evidence you mention needs to be treated very carefully indeed.

There are quite a few papers looking at the risk of He or CO2 in O2 tox, and these seem to indicate a down rating is needed of around 0.15 in PPO2 levels. Moving the risk from 1% of a group at 75% to 1 in 10^-9 at 100% requires a further reduction in PPO2. The most optimistic figure I get to, is 0.1: this is simply because the different studies give such a spread. Add the two together and we get 0.25 as the PPO2 downrating. This issue is dealt with quite nicely by Genesis above in principle, but the risk does seem to reduce sharply with reducing PPO2. Once the risk is below the 1 in a billion hour figure, it can be ignored for practical purposes.

There is a lot of experience in commercial saturation diving, where a PPO2 of 0.7 is used, and a lot of research has been carried out on behalf of different companies to arrive at this figure.

Cheers,

Alex

Drmike

I am aware that He and CO2 are contributors, I am aware there is wide spread variance between individuals what I am not sure of is if the CNS% clock means anything in real terms as it doesnt take one important factor into account - what the diver is doing. Resting on deco or swiming against a current.

Has there been any published studies on the validity of the 'CNS clock' on humans in rest/work situations.

Can anyone point to the original research? I am not convinced that you are not just extrapolating ontop of flawed data (itself extrapolated)

Obviously higher PPO2s mean higher risk of toxing but what is the recover period really like?

I am informed that there are many reasons someone can convulse on dry land as well as underwater - just because they convulse underwater doesnt automatically mean they had an O2 convulsion

__________________
Get a girlfriend you sad twat - a Rebreather is an unfaithful mistress - dont blind yourself to her faults just because she goes down on you

bgpartri

Who are these six people? Where are they reported?

__________________
http://www.rebreather.ca

Good judgement comes from experience, and experience comes from bad judgement.

AD_ward9

See post 26 on the Guide about setpoint selection thread.

Have you had a look at the papers we cited? It is in our doc describing the CNS formal model. Those papers contain a lot of further citations, and there are lots more. I spent a whole morning recently just sieving through papers on this subject. The oil industry has funding a huge amount of work in this area due to their use of saturation diving.

Each study is limited in size. However, making the assumption the population is distributed in a Gaussian manner then the data provides enough points for an engineer to decide how many sigma should be enclosed by the distribution and use that. This is what we have done. What may be helpful is for us to publish the exact data points we have used, with reference to the original papers. You are raising very valid points, so we will do that as part of the safety case documentation for the O.R. rebreather.

On your point about not every convulsion underwater is O2 tox, we are treating all 6 reported rebreather convulsions as O2 tox because the incidence is just far too high otherwise (compared to O.C. SCUBA). However, I accept your point: there are other risks that result in convulsions.

Cheers,

Alex

bgpartri

Thanks, I found them.

I was thinking that there were 6 deaths, but they weren't deaths, and your paper didn't say they were. And close calls are certainly valuable in researching problems. But I also note that it is very difficult, if not impossible, to determine the actual PPO2 levels for these divers. Are we talking about current limited sensors?

I've looked at two of the studies you quote and both of them refer to PPO2 of 1.61 to 2.52. I don't think there is any argument that a 4 hour exposure at 1.77 PPO2 is dangerous.

There is nothing there about PPO2 ranges in common use by rebreather divers.

Did I miss something?

I would hate to think that divers start shifting their perception of risk to the wrong places and solving the wrong problem.

__________________
http://www.rebreather.ca

Good judgement comes from experience, and experience comes from bad judgement.

AD_ward9

I hope this reply helps explain the issue.

First, for safety engineering we must look at "incidents". The term has a special meaning in safety, encompassing either an accident or something that can be an accident with a small change in circumstances. The 6 incidents reported were all actually accidents. For the purpose right now, there is no point distinguishing between fatal accidents and non-fatal accidents, because both must be treated as a "Critical Failure".

Second, the purpose here is not scientific research. It is engineering a product with a CNS exposure indicator based on research already carried out. The research establishes data. There is a large body of this data. In engineering this Open Revolution rebreather, we need to determine how to express 100% exposure being a 1 in 10^9 hour risk. Others may want to limit the PPO2 set point they use based on some of this data. Our problem in meeting this target is that the data consists of points, or rather a spread of points, from different studies. We must decide carefully what points to use.

The data in summary are:
a. There is a 1% O2 tox incidence at 75% CNS loading, using nitrox (see original CNS paper quoted).
b. Helium reduces O2 tolerance considerably. Papers cited.
c. Small amounts of CO2 does the same. Papers cited.
d. Incidence in the rebreather population is between 1 in 10K and 1 in 100K based on reports on this forum. Yet the average dive gets nowhere near 100% CNS loading.
e. There is a large body of papers on effects of O2 toxicity under pressure, giving other data points. A lot of the work is on rats, so you have to divide the PPO2 levels by a correction factor to get human data. The researchers involved do this in their studies, for the size of rat they use, and state how in their paper. If a researcher uses a PPO2 of 4.0 for 5 minutes to make a rat convulse, a human can do the same with a PPO2 of 1.3, for example, with a longer exposure. This is similar to the way rats can tolerate decompression 4 times better than people.

With the NOAA CNS calculation you can clock up quite high CNS exposures at what may seem to be modest PPO2 set points. A diver does not need a PPO2 of 2.5 to kill himself, he can do it quite nicely with a PPO2 of 1.3: it just takes a bit longer, and the incidence in the dive population is lower.

The issue you mention of sensors limiting is a quite different one to that being discussed here. A rebreather that can allow PPO2 to rise significantly above the set point when its sensors have current, or more often voltage limits, is a bad design: by that I mean it does not meet the current European standards (BS EN 61508, stipulated by EN14143:2003), nor does it meet accepted minimum standards for safety critical design. If it is carrying the CE mark, it is doing so incorrectly and can be challenged. This project is about designing a compliant rebreather, with better than 1 billion hours mean time between critical failure, in an open manner that may trigger others in the industry to look at their methods, and help us all learn from each other to get safer products. We have disclosed algorithms for managing the sensors, and are disclosing a full formal model for this.

Cheers,

Alex

Drmike

Thanks Alex,

PM sent

__________________
Get a girlfriend you sad twat - a Rebreather is an unfaithful mistress - dont blind yourself to her faults just because she goes down on you