Subsurface not working out SAC on sidemount dives
Andrew Trevor-Jones
andrew.trevor0jones at gmail.com
Sun Nov 6 11:59:48 PST 2022
Hi Michael,
Thanks for that.
I had also done exactly that (adding a second tank to a single tank dive) and saw the doubling of the SAC.
As I keep saying, the SAC is based on the total gas consumption over the dive and it doesn't matter when you used which tank.
Also, as further proof, I manually added the gas switches (when the gas changes actually happened) to the graph. The overall SAC for the dive DID NOT CHANGE. i.e. the gas switches are not being used in the calculation (or if they do, they make no difference). I also get SAC for each tank.
If I had BS gas changes (when they didn't actually happen), again the overall SAC for the dive does not change but the SAC for each tank is way off.
Andrew.
On 7 Nov 2022, at 12:45 am, Michael Andreen <harv at ruin.nu> wrote:
Hi,
Actually, I thought I had seen the same behavior as Andrew, so I did a simple test. Taking a single tank dive and just adding a 2nd tank with same size and similar pressure change results in double SAC for the dive, even if the new tank is not marked as used.
Same behaviour with 4.8.6 that I happened to have an AppImage of.
/Michael
On Saturday 5 November 2022 13:55:05 CET Willem Ferguson via subsurface wrote:
> Hi Andrew,
>
> I also dive sidemount, so I have had to work through these issues. You
> need to consider that the SAC rate that Subsurface calculated for you
> specifically are not representative of your dive. No problem with the
> Subsurface software at all: it might be a case of garbage in, garbage
> out. Let me explain. For Subsurface to calculate SAC correctly, it needs
> to know when you are using each of the two cylinders. Without that
> information it cannot do the SAC calculation. Since you are not
> recording the change from using one cylinder to the other cylinder, it
> assumes you are diving using the first cylinder for the whole dive. So
> it records your total gas volume used from that cylinder, taking into
> account the whole duration of the dive. Obviously this is not accurate.
> The fact that the second cylinder in your cylinder table is not striked
> out with a red line means that Subsurface did not record you using that
> cylinder and considers it as an unused cylinder. Therefore the SAC
> calculated is only based on the one cylinder you used and is not
> accurate at all. It is one of the prices to pay when diving with more
> than one cylinder: if you do not provide the appropriate information
> Subsurface cannot calculate the gas usage or SAC. Many dive computers
> allow recording cylinder changes during a dive. On my Galileo it means
> pressing one of the buttons twice for each change, a rather trivial
> activity once you get used to diving with a dive computer. The changes
> are loaded as part of the upload from the dive computer to Subsurface.
> For quite a few years I did this recording by writing the times of
> changes on a wrist slate and logging the changes by hand. If you record
> cylinder changes then the SAC calculation will be accurate. So your
> choice is to take no extra trouble and understanding that the SAC rate
> value is totally inaccurate, or to take the trouble and record the
> cylinder changes. Once you progress to using a stage cylinder or a deco
> cylinder you will be faced with the same problem. So this is not unique
> to sidemount at all. Enjoy your dives.
>
> Kind regards,
> willem
>
>
>
> On 2022/11/04 21:50, Andrew Trevor-Jones wrote:
> > Sorry, my point was that subsurface is ALREADY calculating SAC for my
> > dives with two tanks. I wasn't making a suggestion for how it should
> > do it.
> >
> > I'm not diving with stages. I am effectively diving sidemount but the
> > tanks are on my back. Both tanks have the same mixture.
> >
> > Here are the cylinders:
> >
> > Here is the log showing the calculated SAC:
> >
> > And yes, I fully understand that Subsurface takes the compressibility
> > of the gases into consideration in the calculation and wrote it that
> > way: "total gas used" takes this into account.
> >
> > Or are you saying it is not already calculating this way and I have
> > some kind if weird version of Subsurface?
> >
> > Andrew.
> >
> > On 4 Nov 2022, at 10:11 pm, 'Robert C. Helling' via Subsurface Divelog
> > <subsurface-divelog at googlegroups.com> wrote:
> >
> > What you are missing is that Subsurface does not know you are side
> > mount diving. When diving with stages which contain different gases it
> > does not make much sense to add everything together. Rather you want
> > to compute the SAC rate per gas. To this end, Subsurface needs to know
> > during which part of the dive which gas was used and this is the point
> > of the gas switches.
> >
> > You would at least need to tell Subsurface somehow it should consider
> > the two independent cylinders as one combined reservoir of gas. One
> > might argue that it makes sense as a default to compute SAC and
> > gas-use per gas rather than per cylinders, so to add all cylinders
> > containing the same (or sufficiently similar) gas together.
> >
> > BTW, to compute the SAC, Subsurface uses a slightly more sophisticated
> > formula than the one you suggest (one that takes gas compressibility,
> > actual surface pressure and water density into account).
> >
> > Best
> > Robert
> >
> > andrew.tr <http://andrew.tr>... at gmail.com <http://gmail.com> schrieb
> > am Freitag, 4. November 2022 um 06:47:19 UTC+1:
> >
> > I often use 2 independent tanks on a dive (back mount but no
> > manifold so two independent regulators).
> >
> > Subsurface always calculates to overall SAC/RMV for the dive
> > correctly.
> >
> > It should notneed to know about gas switches or anything like
> > that. All it needs to know is the starting/ending pressures and
> > size of each tank. From there it calculates the total gas used
> > and divides it by (average depth in metres/10)+1
> >
> > Or am I missing something?
> >
> > Andrew.
> >
> > On 4 Nov 2022, at 4:30 pm, Gmail im Auftrag von Martin Gröger
> > <megr... at gmail.com> wrote:
> >
> > hm...
> >
> > IF you could set the gases to 21/0 for tank1 and 21/1 for tank2
> > you SHOULD have a perdix an it should have a sidemount-mode.
> > use this mode and you be happy....
> >
> > a workaround for the tank change could be not to change cylinders
> > by pressure but to change cylinders by time e.g. changing
> > cylinders every 10 minutes. so you can set the cylinder changing
> > in subsurface by hand very easy.....
> >
> >
> >
> > keep on howling
> >
> > grey
> >
> > Linus Torvalds <torv... at linuxfoundation.org> schrieb am Do., 3.
> > Nov. 2022, 18:26:
> >
> > On Thu, Nov 3, 2022 at 6:00 AM Oliver Stutz <ols... at gmail.com>
> > wrote:
> > >
> > > Is there a fix for this? Would appreciate the help as it was
> > always a
> > > nice quick reference without needing to calculate this
> > myself. I have
> > > attached some pics of a previous dive to show.
> >
> > So subsurface should do the right thing if you just make it
> > obvious
> > that both tanks are used.
> >
> > And you'd think "I have different pressures" would be that, and it
> > *could* be, but subsurface doesn't really use that kind of
> > implicit
> > knowledge.
> >
> > That's largely just because the input is so noisy: you'd need
> > to do
> > more "real data analysis" to use the pressure differences as
> > "now the
> > person is breathing using this cylinder".
> >
> > You'd think it's trivial, but sensor values are noisy, and
> > even if
> > they were perfect they have other real reasons for pressure
> > changes:
> > sensors possibly aren't actually "absolute pressure" but
> > relative (aka
> > "gauge pressure"), and most notably temperature ends up
> > changing the
> > cylinder pressure too.
> >
> > So while it looks trivial for a human to say "look, I see the
> > pressure
> > line dropping, I'm clearly breathing using this cylinder",
> > it's rather
> > messy in the details.
> >
> > End result: without explicit cylinder change events, subsurface
> > doesn't really think you used that cylinder.
> >
> > Now, I realize that when you do alternating sidemount diving, you
> > would not do a cylinder change as you switch. And that's not
> > even what
> > I'm suggesting. You literally need just *one* gas change event for
> > subsurface to go "ok, you're using that other cylinder", and
> > you can
> > add that later by hand too.
> >
> > And at that point subsurface _should_ start taking that other
> > cylinder
> > into account for your global SAC rate.
> >
> > I say "should", because who knows. It's not the kind of diving
> > I do,
> > so it's not like I've tested much of it. But with a gas change
> > event,
> > I think subsurface should do all the same things it does for just
> > stage bottles etc, and that has worked at some point (also not the
> > kind of diving I do, but I *have* done).
> >
> > Linus
> >
> >
> >
> >
> >
> >
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20221107/36e2da86/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PastedGraphic-13.png
Type: image/png
Size: 432775 bytes
Desc: not available
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20221107/36e2da86/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PastedGraphic-14.png
Type: image/png
Size: 446627 bytes
Desc: not available
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20221107/36e2da86/attachment-0003.png>
More information about the subsurface
mailing list