Fwd: Re: [PATCH] Fix up SAC calculations for ATM/bar confusion

[Jukka]

On 25.02.2013 23:04, Dirk Hohndel wrote:
> Chris Lewis <chrislewis915 at gmail.com> writes:
>
>>> As an example one real dive: Double7/300, pressure from 260 to 10 bar.
>>> Subsurface says I used 3500 litres but Suunto Dive Manager 3176,4 litres.
>>> It's a 9 % error.
> But that assumes that the Suunto DM number is correct.
> And I have absolutely no reason to believe that it is any better than
> ours...
Actually Suunto's model is far from perfect ! After each full 50 bar 
theres a flat. The bigger pressure, clearer flat. See 300 and 310 bar.

140    139.581
150    149.551
160    155.642
170    165.37
180    175.097
190    184.825
200    194.553
210    198.113
220    207.547
230    216.981
240    226.415
250    235.849
260    237.01
270    246.126
280    255.242
290    264.357
300    273.473
310    273.369
320    282.187
330    291.005
340    299.824
350    308.642
This information collected via testing DM3.1.0 each point at a time.

>>> I'd like to see some correction, but I agree using gas-mix and
>>> temperature sounds too much. But, one day, someone will send a patch . . .
>> The reality is that the temperature in the cylinder is a mystery to us.
>> Sure we can assume local ambient temperature at the start of the dive
>> (assuming no solar heating effects) but then what is the temperature of the
>> gas in the cylinder during the dive.
>> We would have to account for both cooling due to immersion of the cylinder
>> in water and adiabatic effects of dropping pressure inside the cylinder.
>>
>> All seems more hassle than its worth. I would suggest some simple
>> correction factor over a certain pressure.
> But that's the fundamental problem here. There are so many moving
> pieces. I am not sure we will ever get it right.
>
> So let's assume we take whatever temperature reading we have at the time
> during the dive into account, plus the non-linearity of compression. And
> the mixture of gas in the tank. Is that all it takes to get the correct
> volume of gas inside the tank?
Even a simple correction factor would give a better result.

Jukka
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.hohndel.org/pipermail/subsurface/attachments/20130225/276f7e36/attachment.html>