Page 1 of 1

strange behavior of deco calculation

Posted: Saturday 19. October 2013, 22:54
by Tom
Hi !

Could somebody be so kind to explain the deco behavior of my ostc 2n:

I went down with Nx 26. At 30m @5mins change to 22/36. At 32 min change to Nx 26.

At 33.6 mins at 38 m Deco notice: 2mins at 38 m ?

Is there an explanation for this ?

Thanks a lot in advance

Re: strange behavior of deco calculation

Posted: Sunday 20. October 2013, 07:57
by TnT
Did your ostc show this decostop during the dive, or is it only the desktop application (subsurface) which shows it? In the latter case, it might be a bug in the subsurface (or even libdivecomputer). If you provide a memory dump of your device, I can have a look:

http://www.libdivecomputer.org/contribute.html#memory

Re: strange behavior of deco calculation

Posted: Sunday 20. October 2013, 12:13
by Tom
Yes, it showed my at my dive shortly after i changed the gas from Tx 22/36 back to Nx 26.

I was a little suprised when i saw the deco notice red letter of missing a deco stop in this situation, because it made no sense, doing a deepstop with 1min after changing gas from trimix to nitrox in this deep. And you see, I'm still alive :)

Okay, here we go: I took a memdump, read out the dives (it's dive 1 from the 19.10, which is the important one) an looged all.

The rest of the files in the next post due to restrictions...

Thanks a lot in advance !!!

Re: strange behavior of deco calculation

Posted: Sunday 20. October 2013, 12:14
by Tom
here are the rest of the files.

Re: strange behavior of deco calculation

Posted: Sunday 20. October 2013, 13:08
by Tom
I took a look at the .XML File.

Here is the sequence with the mistake attached as jpeg.

Re: strange behavior of deco calculation

Posted: Monday 21. October 2013, 07:51
by tiefunten
Hi,

just an idea and without a close look on your data: What are your GF-settings? Maybe the "jump" from 36% He to 0% He in almost 40m depth resulted in a too high tissue saturation with He for your lower GF - and therefore the OSTC calculated this deco-stop.

Regards,

Jörn

Re: strange behavior of deco calculation

Posted: Monday 21. October 2013, 10:59
by TnT
Tom Wrote:
-------------------------------------------------------
> I took a look at the .XML File.
>
> Here is the sequence with the mistake attached as
> jpeg.

As you already discovered, the 38m decostop is indeed stored in the ostc data. So it's not a bug in subsurface/libdivecomputer. They are just reporting what's in the ostc data.

Re: strange behavior of deco calculation

Posted: Monday 21. October 2013, 18:14
by danou
Hi Tom,

Are you aware of ``Isobaric counter diffusion''? the basic rule is :

Never let the partial pressure of Nitrogen rise (significantly)

When you switch from TX22/36 to EAN26 at 40m, you switch from 42% to 74% N2,
which is an increase of 76% in PPN2!!!

See f.i. the book "Deco for Divers" by MArk Powell (very good book IMHO).

For this dive to 44m. two better option would be either
-- the cheap one: use 2 gases (Air first then EAN50 for deco), or
-- as you can afford trimix: TX22/36 as first gas then EAN60 at 17m for deco.

Best wishes,
--
Daniel Flipo.

Re: strange behavior of deco calculation

Posted: Monday 21. October 2013, 20:38
by Tom
Hey Daniel,

you are right, but at this situation there was no danger of an isobaric counter diffusion.

I changed for TX 22/36 back to Nx 26. HE is the faster gas, N2 the slower diffusing gas. I had Argon in my drysuit, so that there was no danger of significant rise of Nitrogen.

Deco for divers ist a very good book.

The gas-setup was Nx 26, at 30m change to Tx 22/36; change at 40m after the Tx stage was empty back on Nx 26, ascending to 21m, change on EAN 50....

I think EAN 50 is better than EAN 60, because you can start the deco at a deeper level, and the pressure difference in the oxygen window is higher than at 17m, so deco is more effective-

But back to the issue: Why showed me the OSTC a deco level ?!?

My GF settings are GF lo: 20% and Gf high 85%.

My friend has the opinion, that the ostc planned the first deco level deeper to get the max. ppO2 of 1.6 for deco. I think this could be the solution of the issue...

Re: strange behavior of deco calculation

Posted: Monday 21. October 2013, 23:02
by kaheksa
It is because of "Isobaric counter diffusion"

When coming off trimix you shouldn't breath a mix that has a higher % of nitrogen than 1.5 (IANTD 1.4) times the % you had in trimix

So with 22/36 you have 42% of N2 and the next mix should be max 63% N2 (Nx37)

If you have a higher % of N2 then your partialpressure is so high that He cant come out and you are ongassing nitrogen.

So the computer is OK and trying to save your ass!

(maybe a quick look at the book or a tx course?)

Re: strange behavior of deco calculation

Posted: Tuesday 22. October 2013, 14:53
by Bardass
Tom Wrote:
-------------------------------------------------------
> Hi !
>
> Could somebody be so kind to explain the deco
> behavior of my ostc 2n:
>
> I went down with Nx 26. At 30m @5mins change to
> 22/36. At 32 min change to Nx 26.
>
> At 33.6 mins at 38 m Deco notice: 2mins at 38 m ?
>
> Is there an explanation for this ?
>
> Thanks a lot in advance


i think you set Gas Switch (CF55) at 2min
and you set MOD for Nx 26 at 38m

Re: strange behavior of deco calculation

Posted: Tuesday 22. October 2013, 15:50
by heinrichsweikamp
Bardass Wrote:
> i think you set Gas Switch (CF55) at 2min
> and you set MOD for Nx 26 at 38m

For those who follow the bitbucket code repository: I have now removed the CF55 stuff from the deco list and will show the gas change stops differently.

Isobaric counter diffusion is an issue with your gas setup here, but I think this was caused because you configured your travel gas also as a deco gas here. It's better to use two gases in the list (Both with Nx26, one as "First" and one as Decogas with change depth=38m) for a dive like this.

Regards,
Matthias

Re: strange behavior of deco calculation

Posted: Tuesday 22. October 2013, 19:10
by Rob
Hi Matthias,

heinrichsweikamp Wrote:
-------------------------------------------------------
>I have now removed the CF55 stuff from
> the deco list and will show the gas change stops
> differently.

so CF55 is not used any more with Beta 2.74 and the following?
Configure the Gasswitch stop with the decogas?

Regards
Rob

Re: strange behavior of deco calculation

Posted: Wednesday 23. October 2013, 13:09
by heinrichsweikamp
Hi Rob,

CF55 will still be used as before but the gas change stops will not be added to the decolist anymore.

Regards,
Matthias