GFHi - practical meaning?

[Jay]

A GFHi of 60 means if your NDL reaches zero when you're at depth and if you surface then, you'll be at 60% of the (unmodified) m_value (for your leading tissue). Ditto for 90%. 30/60=50% difference. Neither affect the ascent rate, but the recommended ascent rate is used in the computation.

GFHi doesn't appear to work that way. That is, if you have a GFHi of 60 that you'll reach the surface with the leading TC at or under 60% of the surfacing m-value (Mo in the literature). The reason I say this is Baker's sample deco program (see his paper "Decolessons") for a GF of 30/75 has the diver surfacing at an m-value of 92% of Mo. In my spreadsheet under the column %AoM which presents the same data as Shearwater's SurfGF, but not in real time, the ascent to surface segment most of the time shows a value greater than GFHi. Common sense tells you that GFHi should work the way you describe it above. In the algorithm in Baker's program and in my spreadsheet GFHi does not modify Mo directly; It modifies the a and b coefficients in the equation that determines the decompression ceiling or NDL time (the same equation is used).

Just for reference here are the equations for calculating a and b:

a = Mo - s(Ps)
b = 1/s

Where s = the slope, Ps = the absolute pressure at the surface at sea level

I replicated baker's dive profile from "Decolessons" in my ss. His total DT adding in the initial ascent to the first stop is 96 minutes. My spreadsheet shows 106 minutes using the starting depth for ascents and descents in the calculation for inspired inert gas pressure. One reason for the greater DT in the ss is I switch to the (2) deco gases at a shallower depth (the switch depth is non-selectable). There are some other minor differences as well that affect the overall DT. I will be posting a major revision soon. In the new version I added EAD and a choice to show the surface GF or current GF as Baker calculates it. Baker came up with a GF of 91.2% upon surfacing. The ss came up with 91%. Selecting Surface for the %AoM column produced 63% which comes under the GFHi of 75%.

From my limited Rec diving experience, I 'know' that at depth when NDL=0, SurfGF ~~=GFHi (~~ bec of difference in ascent rate assumptions), then post ascending (at the computer's given rate), at the surface GFHi = current GF (GF99) = SurfGF.

Because I never ascend directly to the surface (and especially when NDL = 0) I don't know for sure it all holds true. By definition it seems it should. I could set a low GFHi on a dive and give it spin ...

From my limited Tec/deco knowledge, when stops are cleared, one should be surfacing at GFHi. The stops are designed to land you at GFHi on the surface right?

But as you see in the quotes above (from another thread Recreational Ascent Rate in the last 15 feet ), Baker's 90m deco dive is replicated by @EFX, uses a GFHi of 75, and surfaces with a GF of 92%. How is that possible?

The SS takes ~15 seconds to calculate.

 

[ChuckP]

Something ain’t right. I dive 45/80 - when ndl gets to zero and my first stop is shown at 30’ or 20’, when my deco is clear, my surfgf equals 80

 

[rsingler]

Yes, that bothered me. But I trusted @EFX knowledge of the math to accept that statement implicitly.
@EFX , care to explain in further detail what you meant when you evaluated Baker's sample deco program? How do you think he prescribed GFHi of 75, yet had a SurGF of 92%?

 

[EFX]

Baker doesn't calculate SurfGF in his sample program. He calculates, using Shearwater's terminology, GF99. You can look at his program output yourself. I've included the file Decolessons which is his paper. Go to the end of that paper to see the output. The pertinent column is labeled Max %M-value. Here is an excerpt from his paper:

"Next, we call the subroutine MVCALC (for M-value calculation) to determine the maximum Percent M-value (PMVMAX) across all 16 compartments UPON ARRIVAL AT THE TRIAL DEPTH. This represents a worst-case condition in terms of
proximity to an M-value. This procedure is done as a second safety check in the process. The maximum Percent M-value is written to the output file for each segment during the ascent so that the user (the diver) will easily spot any discrepancy. In addition, this is valuable information for divers who wish to tailor their profiles according to individual disposition and physiology (i.e. set personal deco limits on a Percent M-value basis). The Percent M-value parameter is also used to gauge the effect of various Gradient Factors when using the Gradient Factor Method for conservatism."

This comment, for me, indicates that the Max %m-value will not be the same upon arrival at the surface as the TC calculation at GFHi = 75% upon surfacing.

BTW, I use Baker's formula in my spreadsheet to calculate %AoM using the Current option and I get the same results he does (92%) when I simulated his dive.

 

[rsingler]

Hmmm. Well, a GF99 of greater than GFHi might be of concern to any diver during the dive. I figure the diver's presumption (or at least MY presumption) is that when I specify GF 50/70, my maximum %M-value for the leading compartment at least, will never exceed 70% at any time during the dive.
Is this an unreasonable assumption?

And at the surface, my GF99 should = my SurGF, which I thought was <= GFHi.

 

[EFX]

Finishing the dive with GF99 <= GFHi is a very reasonable assumption and makes perfect sense. In my trials of the spreadsheet for both NDL and deco dives I found some dives finish at or below GFHi and some finish higher, but all dives finish below GFHi = 100% (M0). Why? I don't know other than the calculation that uses GF computes the ceiling from which the stop time is determined which then determines the TC pressures. To get SurfGF (%AoM with Surface selected) I divide the CTC pressure by its M0 and multiply by 100 to get the percentage.

 

[Jay]

Thanks @EFX. In that table of Baker's at the end of the dive at a depth of 0, he computes a GF of 91% (Segment 41). The SS computes the same in row 60 (0 depth also). Both are for 0.3 of a minute. Changing the SS to "surface" it then computes 63%. But since we're at the surface then those numbers should be equal (ignoring the 0.3min of time), or very close to equal. Am I missing something? If I'm not, then it would appear to be a computational issue?

 

[Subcooled]

A GF/Hi of 85 means that a friend of mine ends up in the recompression chamber because he was 15% off the maximum and succumbed to other factors that the computer was not aware of (being cold, overweight, 40+, in bad aerobic shape, carrying heavy gear and so on).

In theory the GF Hi means how close to the theoretical limit you dare to go (in %), but there are other factors too.
GF Low sets the deep stops. I know that a value of 30 is preset on some computers.

Dive computers only do an educated guess but the diver must know better.

70 was the default, 85 recommended by friends to avoid "excessive decompression stops" (avoiding decompression stops is a stupid idea, IMO), 90 used by at least one older guy who has not been too many times to the chamber...

I am sure that there are more knowledgeable people on this topic than me but I do know for sure that a GF/Hi=85 can get you into a hospital if you ignore all the best practices. Not everybody is young and fit.

 

[michael-fisch]

High GFHigh numbers are an excellent way of displaying how big your brass balls are.
20 years ago I was reduced to diving a GFHigh of 125% since doing a GFHigh of 150% like Buchaly and Waldbrenner were doing wore me out.
Now I use a GFHigh of 85% for single dives and drop it to 80% for multiple dives over a few days.

 

[Fastmarc]

I know the replies here are more referencing deco dives when suggesting GF85 as being 'ballsie', but it seems that on basic rec computers, at least for the 1st dive they mimic GF95 (my observation with my Perdix vs my Mares RGBM. Not sure what it's like with repetitive dives yet). With this being used by people mostly ignorant of deco stress ( eg. riding NDLs), why are not more divers heading to the chambers? As was mentioned in another thread, rec divers are often experiencing more dec stress than deco divers, so shouldn't these computers be even more conservative of at least GF85?.