On a NDL dive, which computers' NDLs are not affected by GFLo?

[huwporter]

OK, lets have a play with some deco software to see how we might interpret this request.

We obviously can't use GF-Hi and GF-Lo to calculate any stops, since this is no-stop diving - "No Deco Limit" - by definition, duh.

GF-Hi is used to control the maximum gradient experienced on surfacing. So we have to set our NDL so that we can ascend from our current depth directly to the surface - with no stops on the way - without exceeding it.

GF-Lo is used to control the first stop depth. We can't actually stop (see above) so we also have to set our NDL so that we can ascend directly from current depth to the last (and only) stop depth without exceeding GF-Lo when we arrive there, at which point we can smoothly continue to the surface.

So how would this work for, say, an 30 metre / 100 feet dive on Nitrox 32? Let's use common gradient factors 30/85 and a last stop depth of 6m. (20 ft)

Our GF-Hi means we can't exceed GF85 on direct ascent to the surface. We can stay at 30m on nitrox 32 for 23 minutes before exceeding 85% of our permissible supersaturation in any tissue on direct ascent to the surface, hence the 'conventional' NDL for Nitrox 32 at 30m is 23 minutes:

=========================================================================
 FIRST DIVE:
 30m for 23 mins OC on 32/0

 Phase                Depth   Time       RunTime   Mix O2/He Ceiling GF
 -----                -----   ----       -------   --------- ------- --
 Descent To:          30m     1.7 min    1:40      32/0      0m      0
 Bottom Phase:        30m     21.3 min   23:0      32/0      0m      0
 Ascent To:           0m      3.3 min    26:20     32/0      0m      84
 Surface:                                26:20                       84

 Deco Time (21-6) = 0
 Estimated Total Runtime (including switches and gas breaks): 26 mins.
=========================================================================

But what about our GF-Lo? If we do this dive above, we actually need to insert a first deco stop at 9m as well as a stop at 6m so as not to exceed 30% of our maximum supersaturation at 'first stop depth'. (If we'd gone directly to 6m, we'd have briefly experienced 37% of permissible supersaturation in one or more tissue, which is above our limit.):

=========================================================================
 FIRST DIVE:
 30m for 23 mins OC on 32/0, GF 30/85

 Phase                Depth   Time       RunTime   Mix O2/He Ceiling GF
 -----                -----   ----       -------   --------- ------- --
 Descent To:          30m     1.7 min    1:40      32/0      0m      0
 Bottom Phase:        30m     21.3 min   23:0      32/0      0m      0
 Ascent To:           9m      2.3 min    25:20     32/0      0m      21
 Stop:                9m      1.0 min    26:20     32/0      0m      21
 Stop:                6m      1.0 min    27:20     32/0      0m      34
 Surface:                                28:00                       80

 Deco Time (21-6) = 2
 Estimated Total Runtime (including switches and gas breaks): 28 mins.
=========================================================================

Well, this cannot be, since this is no-stop diving, so we must set our NDL shorter, to just before we exceed Gf30 at our last stop depth. This turns out to be 17 minutes, with a surfacing gradient of 73:

=========================================================================
 FIRST DIVE:
 30m for 17 mins OC on 32/0

 Phase                Depth   Time       RunTime   Mix O2/He Ceiling GF
 -----                -----   ----       -------   --------- ------- --
 Descent To:          30m     1.7 min    1:40      32/0      0m      0
 Bottom Phase:        30m     15.3 min   17:0      32/0      0m      0
 Ascent To:           6m      2.7 min    19:41     32/0      0m      29
 Stop:                6m      0.0 min    19:41     32/0      0m      29
 Surface:                                20:21                       73

 Deco Time (21-6) = 0
 Estimated Total Runtime (including switches and gas breaks): 20 mins.
=========================================================================

...so sure, you could interpret GF Hi and GF Lo this way, so long as you accepted the corresponding reduction in NDL - 23 minutes to 17 minutes in this example. Or you could just set your surfacing gradient lower.

For shallower dives with last stop depth 6m, it makes less difference. For 24m (all on Nitrox 32), the 'NDL' for GF85 would go from 40 minutes to 37 minutes for '30/85'.

If you use 3m as your last stop, your NDL will be excessively reduced, and it actually gets worse for shallower, longer dives - for 30m, your NDL needs to be reduced to 10 minutes so as not to exceed GF30 at 3m (with surfacing gradient of 52), and for 24m it needs to be reduced all the way to 15 minutes:

=========================================================================
 FIRST DIVE:
 24m for 15 mins OC on 32/0

 Phase                Depth   Time       RunTime   Mix O2/He Ceiling GF
 -----                -----   ----       -------   --------- ------- --
 Descent To:          24m     1.3 min    1:20      32/0      0m      0
 Bottom Phase:        24m     13.7 min   15:0      32/0      0m      0
 Ascent To:           3m      2.3 min    17:20     32/0      0m      29
 Stop:                3m      0.0 min    17:20     32/0      0m      29
 Surface:                                17:40                       50

 Deco Time (21-6) = 0
 Estimated Total Runtime (including switches and gas breaks): 18 mins.
=========================================================================

Does it make any sense to do this? Personally I don't think so. Leaving NDL as solely controlled by GF-Hi and inserting optional 'safety' stops to taste makes more sense to me.

 

[scubadada]

Hi @huwporter

But how do dive computers work? See On a NDL dive, which computers' NDLs are not affected by GFLo?. Clearly, there are differences between, how, at least some, dive computers deal with GF Low and how deco planning software deals with GF Low. My Dive Rite Nitek Q gives the exact same NDLs at GF 90/90 as it does at GF 10/90 (it is the same for Shearwater). GF low is not implemented unless the surfacing GF will be exceeded. On the other hand, the planning software implements the GF Low during any ascent.

What dive computer(s) do you use? Run the 90/90, 10/90 example and let us all know what you find with regard to the NDLs

 

[dmaziuk]

But how do dive computers work

They have magic smoke inside their microchips, it's what makes them work. If the smoke comes out, them microchips don't work no more. That's how.

 

[scubadada]

They have magic smoke inside their microchips, it's what makes them work. If the smoke comes out, them microchips don't work no more. That's now.

Not the only place smoke is coming from

 

[dmaziuk]

 

[huwporter]

But how do dive computers work?

Who knows? Some (most?) dive computer manufacturers add all kinds of proprietary fudges to their programming (adaptive this, enhanced that, folded whatever) over and above the base model they are supposed to be running. I was examining solely how you would need to think logically about implementing more than just 'maximum surfacing gradient' for a no-deco plan using straight Buhlmann/GF, and the effect of that on the calculated NDL.

(edit to add) My conclusion is, that it doesn't make sense to use anything other than maximum surfacing gradient in no-stop diving, and I'm happy to see that the manufacturer of the computer I use agrees with me.

 

[dmaziuk]

More importantly, who cares? If you understand that two GFs don't make sense on a no-stop dive, you set your birdix to 90/90 for no-stop dives and know exactly which GF will be used to compute your NDL. It's the 90.

If you don't understand what you're doing, get a Zoop and whine about it being too conservative instead.

 

[Jay]

Back to the OP: Which (if any) computers' NDLs are affected by GFLo, GFHi ... (not: how are the computers affected / their inner workings / theoretical/ logic ... etc - that's a different thread). This thread has a simple single question ...

Reason for OP: It seems clear that only GFHi affects NDL on a NDL dive, but the only computers that this can be said for with certainty are Shearwaters and Dive Rite's (discontinued) Nitek-Q. At multiple times on SB the statements are categorically made about GFs and NDLs (re NDL diving), argued etc (ground hog days), but there's not yet been the effort to gather evidence of what computers output.

The answer is easy to find with computers with custom GFs: run a dive plan for a given depth, one at 10/90 and the other at 90/90 and see if the NDLs are different.

GFLo does not affect NDL on a NDL dive
Shearwater (all)
Dive Rite's discontinued Nitek - Q

GFLo does not typically affect NDL on a NDL dive and/or the effect is minimal
TDC-3 <-- very provisional (see post#69)

GF Lo affects NDL on a NDL Dive.
none yet

To yet know:

Unknown but does have custom GFs (i.e. easy to know if GFLo affects NDL)
Divecomputer.eu/Deep 6
Dive Soft
Garmin Descent
Heinrichs Weikamp
Mares - Genius
Ratio/SEAC (Tech+ has custom GFs)
SeaBear (consumed by SP)
Scubapro - HUD
Suunto EON Steel

Unknown but doesn't have custom GFs
Liquivision Lynx

 

[dmaziuk]

My conclusion is, that it doesn't make sense to use anything other than maximum surfacing gradient in no-stop diving, and I'm happy to see that the manufacturer of the computer I use agrees with me.

And by that you mean your computer generates you deco stops at 9 and 6 metres once your NDL counts down to 0?

 

[huwporter]

And by that you mean your computer generates you deco stops at 9 and 6 metres once your NDL counts down to 0?

Good question; from a quick play with the planner, it appears not. Although I'd argue that depending on the GFs set, perhaps it should. That could be one of those little differences of interpretation in exact detail of how to apply gradient factors on top of Buhlmann that mean minor profile differences do appear between planners, i.e: do you trigger the first stop at the last stop depth before exceeding GF-Lo in any tissue, or at the first depth after exceeding GF-Lo in any tissue?

Baltic dive planner gives you a first stop at 6m for 24@30m/ean32/GF20/85, and tells you that you hit GF50 at the first stop, so it looks like they interpret first stop depth the second way. Interpreting it the first way ought to give you two stops. ...But frankly, the difference between 1@9, 1@6 and just 1@6 in the water is so small as to be utterly discountable when dealing with messy biology.

...But that is a different conversation.

(Personally, honestly, I don't ever really pay any attention to the NDL. I tend to do either dives where I know I'm not going to get anywhere near deco and I have a dive plan anyway, or dives where I'm expecting to do significant deco and I have a dive plan anyway.)