Why plan decompression with a Gf (lo)?

[Jay]

I have a q which I'd like to air given it's similar to the OP's ultimate question; it's about the numerical difference between Gf-lo and Gf-hi.

If Gf-hi=Gf lo then the latter has no meaning except that the lower the absolute number is, the more of a slight (clockwise - pic below) twist from the M-value line (given the M-value and Ambient lines are not parallel). And having a lower Gf-lo obviously means a stronger clockwise twist; deeper stops than otherwise. Given that obvious comment, is there a feeling / logic / evidence etc as to the appropriate difference between these numbers?

From Gradient Factors | Dive Rite

 

[stuartv]

Given all that, my question is: what is the reason for continuing to employ a pair of gradient factors, instead of scraping the Gf Lo altogether (or, put it another way, set it the same as Gf Hi)? Has any study been done that shows that using any Gf (lo) produces less DCS cases, or at least less intravenous bubbles, than not using it? Has it been shown that using a Gf (lo) reduces the calculated integral of supersaturation?


edited for clarity

I asked @Dr Simon Mitchell that very question a while back. MY interpretation of his response is/was that there is no data to suggest that 50/80 is any better than 80/80. But, there is no data to suggest that 80/80 is better than 50/80, either. But, 80/80 is a bigger departure from historical best practices than 50/80 is, so it's probably better to go with 50/80, at least for now. I.e. better to make a small change than a big change. People used to use 30/70 a lot. Going to 50/80 is a jump. Going to 80/80 would be a bigger jump.

My additional spin is that 50/80 versus 80/80 makes a pretty dang small difference in the ascent for the diving I am doing. So, definitely no strong motivation to make that bigger jump to using 80/80.

 

[Dr Simon Mitchell]

I have a q which I'd like to air given it's similar to the OP's ultimate question; it's about the numerical difference between Gf-lo and Gf-hi

Hello Jay,

Stuart has answered it pretty well.

RossH has often claimed, and correctly in my view, that the early 2000s saw the use of bubble models, or GF choices made to emulate bubble models (with a very low GF-Lo to force deep stops), become very widespread. Indeed, it is probably true to say that just about everyone was incorporating a degree of deep stopping in their decompression approaches. As you will have gathered from countless discussions on line, there is a growing body of evidence which suggests that we have probably over-emphasised deep stops, and our decompression is less efficient as a result.

There is impetus for moving away from deep stop emphasis, but this is where things become problematic because none of the studies have defined optimal decompression. This makes those of us who are often asked about it a little cautious. It has also become reasonably clear that padding shallow stops reduces decompression stress and so we think that backing away from GF 100 as the GF-Hi is probably sensible. Putting all this together, common sense suggests we should reduce deep stops and pad shallow stops to some extent, but there are no data that guide this objectively. That is why we have ended up talking about paradigms like 50-80, or even 50-70. These represent a reduction in deep stopping (particularly on very deep dives) with some padding of shallow stops. Is it optimal? We don't know. Is it the correct reduction in deep stopping? We don't know. It may be that 80-80 is good, or 70-70 is good, but since just about everyone has been deep stopping for a while, it makes sense to de-evolve from deep stopping in smaller steps than some are advocating.

Sorry, probably creates more questions than it answers.

By the way, that GF diagram is incorrect. The GF-Lo scale at the top of the graph should be vertical like the GF-Hi one down at the bottom left.

Simon M

 

[tursiops]

By the way, that GF diagram is incorrect. The GF-Lo scale at the top of the graph should be vertical like the GF-Hi one down at the bottom left.

??
Why do you say this? What relevance does the positioning of the scale have? Is it not just a way to have the GF line intersect something and give it a numerical value?

 

[Celt]

over the years that i have worked deep wrecks on air using us.navy tables, I've personally seen 3 cases of bends that needed treatment in a chamber and 2 suspect that were not treated, of the 3, 2 were serious and left damage, a burning sensation in the feet and buttocks in one, and a loss of feeling down the back of the legs in the other. All 3 divers completed decompression but 2 admitted a rapid assent, the other to a large amount of alcohol the night before, After repeat deep dives on air we all suffered small blood blisters when in a bath or shower and achy joints. Should you change your deco for repeat deep air dives and to what.

 

[Jay]

??
Why do you say this? What relevance does the positioning of the scale have? Is it not just a way to have the GF line intersect something and give it a numerical value?

I'm guessing it's (merely) because it makes the representation of the ascent w.r.t. Gf-lo graphically clearer; on my pic above it's off the chart / doesn't tie them together. The pic below shows that more clearly:

 

[KenGordon]

??
Why do you say this? What relevance does the positioning of the scale have? Is it not just a way to have the GF line intersect something and give it a numerical value?

Take a look at the supposed 30/80 line. Consider when it is at 30. In the diagram it is at 30 at some pressure higher than the pressure at the first stop. By the time we get to the first stop it is some way to 80, so not 30.

 

[EFX]

At the risk of sounding dense and totally off the wall what would be the problem with a higher GF early in the ascent with a lower GF for surfacing? Say 80/50 so that you get shallow quicker but surface with a theoretical lower level of supersaturation. Keeping your ascent below 80% of maximum and continuing to deco out to a lower level of saturation would more closely replicate the profile in the NEDU study which had extremely long final stops as well as maximize the time spent at shallower depths which could provide more time with less gas consumption

Reversing the GF's for low and high will add deco time. You can experiment with this if you download my spreadsheet dive_040318_v24_0.xls. To reverse the values of GFLo and GFHi you need to disable the error checking on the dive sheet. You can do this by opening the VB editor by pressing Alt-F8, in the right window page down to the check_profile() routine, go to the bottom and comment out the if block which checks that GFHi >= GFLo. Then save the code.

For your convenience I compared three versions of a dive to 180 ft for 20 minutes on air, descent = 60 ft/min, ascent = 30 ft/min, with no deco gases. GF's are listed in this order: Lo/Hi. DT = deco time, RT = run time, FSD = first stop depth:

50/80, DT=63, RT=86, FSD=70
80/50, DT=95, RT=118, FSD=50
30/66, DT=95, RT=118, FSD=80

The 30/66 dive shows an equivalent dive to the 80/50 profile. As you can see the 80/50 does get you shallower quicker than the 50/80 but at the expense of greater deco time.

 

[tursiops]

I'm guessing it's (merely) because it makes the representation of the ascent w.r.t. Gf-lo graphically clearer; on my pic above it's off the chart / doesn't tie them together. The pic below shows that more clearly:

View attachment 453469

Maybe, although the graph also says, "Divers began using Gradient Factors to force gas content models to impose deeper stops." This is incorrect,. of course. The reason for the GFs was to stay away from the M-values. That some choices of GFs do in fact impose deeper stops was a consequence, not a motivation.

 

[tursiops]

Take a look at the supposed 30/80 line. Consider when it is at 30. In the diagram it is at 30 at some pressure higher than the pressure at the first stop. By the time we get to the first stop it is some way to 80, so not 30.

Yes, that's true. The horizontal Lo scale ought to be down at the same position as the "ascent starts" line. I just don't see why it has to be a vertical scale.