different tissue definiton in OSTC and subsurface

[Lubomir I. Ivanov]

On 11 January 2013 00:26, Robert C. Helling <helling at atdotde.de> wrote:
>
> On Jan 10, 2013, at 8:37 PM, Jan Schubert wrote:
>
> Hi,
>
>> I just stumbled over the fact that the OSTC seems to use different
>> definitons for halftime, a and b for N2 and He for the fastest tissue
>> (#1) then the definitions in subsurface. Some implementations for
>> Bühlmann different between Tissue 1a and 1b, might be that this might be
>> a reason, just a guess.
>
> OK.
>
>> Anyway, changig the values to the one used in
>> the OSTC and also ignoring the precompiled values for 1sec
>
> You realize those are used in the deco calculation, not the half-times? In any case, for a half-time t12 you can compute the precompiled value for one second as
>
> 1 - exp(-ln(2) / 60 / t12)
>
> which should be well approximated by ln(2) 60 / t12 (the error being about the square of this number).

here is a slightly better 1/x (even-like) approximation for:
1 - exp(-ln(2) / 60 / x)
----------

ln(2) / 60 / (x + 0.0060)

near zero for x, it should converge with at least half the error of:
ln(2) / 60 / x

(error in the lines of 5e-3)

i can't say much else, as i have no idea what are you doing here and
this might not be needed _at all_.
it looks like exponential decay, of sorts..

lubomir
--