[FLASH-USERS] how to Implement Isothermal EOS

[Robert Fisher]

To add to what others have said, the "adiabatic" approximation of setting 
gamma to a small value is a commonly-used one which should work well. 
There is, however, a very substantial caveat.

The key constraint is that because the gas still remains weakly 
adiabatic, compressional heating will eventually cause the flow to deviate 
substantially from isothermality. What one needs is to identify the peak 
compression for a given problem, and to set the gamma small enough such 
that the gas will remain isothermal through that compression.

As Kevin says, running with small deviations from 1 should not require any 
modifications. It is possible to run with values much closer to unity with 
some minor modifications to the Riemann solver as Tomek has also 
indicated.

Best wishes,

Bob

On Thu, 7 Feb 2008, Kevin Olson wrote:

> I have implemented both things Tomasz suggested and they seem to work fine.
> The easiest is to set gamma = 1.01 or so (too much smaller and things are 
> unstable)
> or to simply add a cooling/heating term that keeps the temperature fixed. 
> The riemann solver does not need to be modified for either of these 
> approaches since you are still using the same EOS.
>
> Kevin Olson
>
> On Feb 7, 2008, at 8:08 AM, M.A. Latife wrote:
>
>> Hi all,
>> I am using FLASH2.5. Is there any way to implement isothermal EOS. should i 
>> use heating and cooling source terms for this purpose or is there any other 
>> way to do this?
>> kind Regards
>> M.A.Latife
>> 
>