Info Deeply Safe Labs: A website for dive computer testing

[tursiops]

What they originally did instead was tell people to "sit every 6th day out". I

What they actually said (from the 1994 RDP Report) was:

"From the test results we further conclude that performing multiple dives over multiple​

days with the RDP is acceptable, and can be done with no greater risk than is encountered in many​

common practices of recreational divers. Even so, based on this evidence and the suggestions of​

other experts, we recomme d limiting the number of full-time dives per day to 3 or at most 4, and​

suggest including a day with a reduced level of diving ( or none) every 2 or 3 days."​

 

[inquis]

My understanding is they just use M0 all the way down as the model was not designed for decompression diving in the first place.

I thought about that possibility, but that would make the first deco stop occur at the max depth which is rather useless.

 

[dmaziuk]

I thought about that possibility, but that would make the first deco stop occur at the max depth which is rather useless.

No, I meant what you said: the slope is parallel to the ambient pressure line. It's `Pamb + M0`.

 

[scubadada]

The slope is not given in Baker's paper either. A case can be made for assuming 1.0, as that seems to be the asymptotic value in other sets.

My understanding is they just use M0 all the way down as the model was not designed for decompression diving in the first place.

I don't know about DSAT, Buhlmann delta M varies from 1.54 at 8 min to 1.15 at 54.3 min.

 

[scubadada]

Maybe a smarter person than me can explain the behavior of DSAT vs. Buhlmann ZH-L16C on first vs. repetitive dives.

Thanks

 

[inquis]

@scubadada - the higher M-values of DSAT make it more liberal than Buhlmann (GF 100/100) when faster tissues are controlling -- i.e., a clean squarish profile. Conversely, DSAT is more conservative when slower tissues are controlling -- i.e., repetitive dives. This much I'm sure about.

The bit that has me confused is that a GFhigh of 95 causes the NDL times to match fairly well. On a clean NDL dive, a faster tissue should be controlling, and the GF will make Buhlmann even more conservative (and DSAT even more liberal by comparison). However, DSAT is *already* more liberal when faster tissues are controlling. My speculation is that there is some interplay with the different slopes (assuming DSAT is using a slope of 1.0 or a 1:1 increase of its M-values for each foot of depth), but OTOH, NDL shouldn't be impacted by anything other then the surface M-values. So yeah, someone smarter than me needs to explain that as well.

 

[dmaziuk]

I'm not sure DSAT M-values given in e.g. Powell are what the DSAT computers are using: they may be the ones used for creating PADI tables but tables come with built-in safety margin: you seldom spend all your planned bottom time at planned max depth.

Buhlmann's numbers were apparently adjusted from "tables" B set to "computers" C set that is more conservative from the 27 min TC on.

 

[scubadada]

I'm not sure DSAT M-values given in e.g. Powell are what the DSAT computers are using: they may be the ones used for creating PADI tables but tables come with built-in safety margin: you seldom spend all your planned bottom time at planned max depth.

Buhlmann's numbers were apparently adjusted from "tables" B set to "computers" C set that is more conservative from the 27 min TC on.

Yes. C was made more conservative than B in compartments 5, 6, 7, 8, 9, 10, 11, 12, and 14.

 

[Umuntu]

sure, but like him, having been involved in dive computers at the manufs level, I know that a LOT of thought goes into it, and the intended use for the computer is an important factor in implementation decisions. Much of what he is doing is going WELL outside the design use case for many units and ignoring what the actual manuals say. His argument that it's a valid means to test/expose is.. wrong. There are many good reasons why that is, some algorithm, some programming, some hardware, and some design decisions. But, push some units far past design envelope and if the manuf didn't choose to just lock them out, you will see..exactly what he found. Is the problem the computer or the user not using as designed? If I try and do all the aerobatics in a Cessna 150 that a 150 aerobat can do, when the wings fold on me it's my fault, no matter how much I scream on the way down "BUT OTHER CESSNA 150's CAN DO IT, why didn't this one?"

I'm ok with him doing that, IF he CLEARLY annotated with what the manufs recommended use for the unit was, and any specific things regards settings from the manuals. Plus, having like you read many, many papers on deco, some of what he is claiming is.. cherry picking at best and some is just wrong. So, supply references or don't make unsupported claims.

I’m not aware of any computers that explicitly explain their “design limitations” in or on the box they come in. Should’nt these “design limitations” be disclosed before purchasing on the box ?

If the computer doesn’t lock you out, the assumption must be that the dives are within the “design limitations.”

 

[JMarc]

If the computer doesn’t lock you out, the assumption must be that the dives are within the “design limitations.”

That's NOT a safe assumption.

More generally, the assumption that profile suggested by a dive computer have all the same risk for a given hardening is false.

The assumption that we know how to provide such a thing is also false.