Reconciling deco models with no-fly recommendations.

[Mr Carcharodon]

Unfortunately there's no model of which I'm aware that will tell you what's "enough" deco for ascending to X thousand feet after Y hours of SI.

Is'nt the whole point of m-values to say how much is enough? After all an ascent X thousand of feet after Y hour of SI is going from one ambient pressure to another. Why is calculating how long to spend between 10 fsw and the surface any different than calculating the time to spend between the surface and 8000 feet above sea level? The driving force is pressure is it not? That is certainly what Buhlmann thought.

It seems that the DAN recommendations are based on worst case enveloping models and the dissolved gas or bubble models take into account actual dive times and pressures. As such DAN is usually overly conservative but unless you can do a calculation what are you going to do?

Nitrogen v. Helium diffusion times seem to be based on molecular weight. That assumption flows down to half times. Buhlmann and VPM use the same half times.

 

[Dr. Lecter]

Is'nt the whole point of m-values to say how much is enough? After all an ascent X thousand of feet after Y hour of SI is going from one ambient pressure to another. Why is calculating how long to spend between 10 fsw and the surface any different than calculating the time to spend between the surface and 8000 feet above sea level? The driving force is pressure is it not? That is certainly what Buhlmann thought.

It seems that the DAN recommendations are based on worst case enveloping models and the dissolved gas or bubble models take into account actual dive times and pressures. As such DAN is usually overly conservative but unless you can do a calculation what are you going to do?

Nitrogen v. Helium diffusion times seem to be based on molecular weight. That assumption flows down to half times. Buhlmann and VPM use the same half times.

So why aren't we all using Buhlmann with GF to calculate this?

 

[kr2y5]

This can be modeled as a saturation dive at 8000 feet (there is about 9 fsw worth of pressure increase going from 8000 feet down to the surface), followed by the dive proper. I am having some trouble getting V-Planner to generate sensible results (it seems to hang forever if you give it a 10000-minute interval in the dive profile @ 9 fsw to simulate "saturation" at the surface), but I will try to post them here shortly, once I find deco software that works... if nothing else, then just to throw in a data point into the discussion.

---------- Post added June 5th, 2014 at 05:48 PM ----------

There it is, finally... with a 10,000-minute (about a week) of a "saturation dive" to 9 fsw @ 8000 feet altitude, followed by 30 min. dive to 100+9 fsw, we get somewhere around 9-10 hours of a "surface interval" (the last planned deco stop is at 10 fsw @ 8000 feet, but if we surface at that altitude, we'll be at 9 fsw, so in reality we'll then be offgassing a little bit faster). I hope this time around I didn't end up making any silly typos. This result is at least on the same order of magnitude as DAN's recomendations (which for deco dives, as far as I can remember, was 24-48 hours). I certainly wouldn't follow this plan and bet my own life or health on its correctness, but I think it'd be very interesting to see how much more aggressive these predictions typically are compared to DAN's guidelines across a wider range of scenarios, and whether DAN's guidelines (which, AFAIK, are based on purely statistuical analysis of DCS occurrences) can be "explained" in the framework established by the popular deco models.

Decompression model: VPM - B
Surface interval = 5 day 0 hr 0 min.
Elevation = 8,000ft (s)
Conservatism = + 3
Dec to 9ft (0) Air 50ft/min descent.
Level 9ft 9999:39 (10000) Air 0.27 ppO2, 9ft ead
Dec to 109ft (10002) Air 50ft/min descent.
Level 109ft 28:00 (10030) Air 0.90 ppO2, 109ft ead
Asc to 70ft (10031) Air -30ft/min ascent.
Stop at 70ft 0:42 (10032) Air 0.65 ppO2, 70ft ead
Stop at 60ft 2:00 (10034) Air 0.59 ppO2, 60ft ead
Stop at 50ft 5:00 (10039) Air 0.53 ppO2, 50ft ead
Stop at 40ft 7:00 (10046) Air 0.46 ppO2, 40ft ead
Stop at 30ft 12:00 (10058) Air 0.40 ppO2, 30ft ead
Stop at 20ft 20:00 (10078) Air 0.34 ppO2, 20ft ead
Stop at 10ft 590:00 (10668) Air 0.27 ppO2, 10ft ead
Surface (10668) Air -30ft/min ascent.

 

[rjack321]

My god you make it so friggen complicated. People have been deco diving in Lake Tahoe for eons. Its at 6200 ft. Also in Colorado at >8,000ft (I think some of the highest lakes are at 13,000ft). And you don't need to friggin agonize over every last molecule. None of it exactly follows the models anyway. Every model is wrong, some are just more useful than others.

Of course you won't get to pontificate if unless you made it "sound" very very complicated and mathematically obtuse.

 

[kr2y5]

My god you make it so friggen complicated. People have been deco diving in Lake Tahoe for eons. Its at 6200 ft. Also in Colorado at >8,000ft (I think some of the highest lakes are at 13,000ft). And you don't need to friggin agonize over every last molecule. None of it exactly follows the models anyway. Every model is wrong, some are just more useful than others. Of course you won't get to pontificate if unless you made it "sound" very very complicated and mathematically obtuse.

Perhaps I'm misunderstanding your point, but it's not clear to me why you can compare this scenario to a regular dive at altitude, if that's what you're suggesting. During ascent to altitude, it's the slowest tissues that are controlling, as exemplified by the tissue bar from my Petrel taken at the beginning of the flight, shown below. Compared to a regular altitude dive, such as that at Lake Tahoe, we start a dive at sea level with slowest tissues already loaded by the sea level pressure, and then load them even further during the dive. Are you suggesting that one can dismiss that factor in modeling off-gassing? In my regular dives during a surface interval, it's always intermediate tissues that are loaded the most.

 

[Lorenzoid]

Others more knowledgeable than I have given detailed responses, but I'll take a stab at this from my simplistic point of view. It seems to me that there is no deco model that has been thoroughly tested to ensure it works the same way throughout all ranges of pressure change from the bottom of the sea up to high altitudes. A model only "works" to the extent that actual dive data has validated that it keeps most people from getting bent most of the time. For any given model, there are some well-tested ranges, dive profiles, etc., in which the results seem to fit the model--that is, very few people have gotten bent--but outside of that it is mainly untested extrapolation. DAN's guidelines are, in effect, saying "we just can't be certain," and therefore err on the side of conservatism.

 

[Red Sea Shadow]

What's stopping you from applying SI O2 and how long can you stay on it before flying/ascending (great article on that in the new tech mag issue)?

This article is in the 15th issue of Tech Diving Mag. There's also some info on the same topic in both the 1st and the 4th issues.

 

[rossh]

If we calculate tissues pressures for a typical 60 minute deco dive, then very roughly, you need about one hour surface interval for every 1000ft / 300m of elevation rise. This is the kind of thing you test each time you drive home from the dive site or boat trip.

Airplane cabins are a different story. You ascend to 8000 to 9000 ft and stay there for many hours. The air is dry, and you sit in uncomfortable chairs in a fixed position. And if a cabin should depress and oxygen mask drop from the ceiling, then you are at much higher risk of injury than the rest of the passengers and well outside the tested range of tolerable limits for passengers. The mask oxygen is not enough to treat or offset a DCI injury. Also in an airplane cabin, you are not likely to get much help in mid flight if a DCI arises, and a 747 is unlikely to divert for you while over the middle of the Pacific.

All those things put together, require that you have a significant surface interval before flight, to get your tissue levels back down to normal levels like all the other passengers, with a safety margin to spare.

 

[Mr Carcharodon]

So why aren't we all using Buhlmann with GF to calculate this?

That's a good question. In part is seems Dan and Buhlmann are answering different questions. Buhlmann was trying to find what was just enough to get a significant, but non-zero, bends rate for specific combinations of depth, time and mix. DAN is trying to get a worst case recommendation for someone who is pushing the limits. It seems pretty reasonable to calculate the time needed to reach m-values for your expected cabin pressure based on the details of your dives. From memory the difference in pressure from sea level to 8000 feet above sea level is about 4 fswa. The quirk is needing to consider the slowest compartments which have relatively low m-values and which do not usually come into play at sea level.

Buhlmann did most of his experimental work using hyperbaric chambers in Zurich. Since he was working at altitude he had to work out a way to take altitude into account. I have no idea what the seats were like inside his chamber but they may not have been all that more comfortable than airplane seats. A m-value with a bit of gradient factor hedging seems very reasonable.

 

[Dr. Lecter]

A m-value with a bit of gradient factor hedging seems very reasonable.

Yes, it does, if you're simply returning to a given ambient pressure and wish to know what delta between that pressure and your tissue inert gas pressures won't result in DCS.

I suppose you could find a way to run the math for an m-value where the saturation levels at the start of the dive are consistent with sea level, then there's the loading of the dive, then you perform sufficient decompression to reach an m-value plus hedging that's safe for surfacing at sea level...and then you do the math to figure out how much SI (with or without O2) is necessary to get the same m-value plus hedging for an ascent to, say, 10,000 feet. That nobody who's made a commercial implementation of Buhlmann seems to have offered that kind of planning feature says something to me, though.