[FLASH-USERS] Help for EOS ERROR (Newton-Raphson failed in multitemp)

[李芳]

dear klaus，
thanks for your reply. I solved this problem by changing the cfl and the solver theme. I will also try your suggestions. 


best wishes
-------------
Fang Li
On 09/29/2016 01:36, Klaus Weide wrote:
On Tue, 27 Sep 2016, Fang Li wrote:

I'm using FLASH to simulate the HEDP problem. I'm using two materials (Al+CH) in the simulation.
The job aborted with the prompt "Newton-Raphson failed in multitemp, multitype Eos.F90" as seen below.
I have succeeded in other simulation using two material. I just changed the target shape.
Can anyone help me to figure out what's the problem? Thanks very much.

Dear Fang Li,

It is hard to say much without additional information. There could be
a variety of reasons that could cause this error.  Running your code
in a debugger may be ncessary to narrow down the problem.

As a first step, you should write a checkpoint or plotfile immediately
before this problem occurs, and check whether all variables show
reasonable behavior at that point.

Changing the following runtime parameters MAY help:

 D eos_tolerance  tolerance for the Newton-Raphson iterations
 PARAMETER eos_tolerance REAL    

 D eos_maxNewton  maximum number of Newton-Raphson iterations to try.
 PARAMETER eos_maxNewton INTEGER

 PARAMETER eos_maxFactorUp   REAL
 PARAMETER eos_maxFactorDown REAL


I noticed that in your message, your species mass fractions do not exactly add up to one:

 mass fractions:
   cham   9.934238787156949E-01
   poly   6.578875272006729E-03
   targ   1.000000000000000E-99

I get a sum of about 1.00000275; the difference from 1 is small but I
would expect better agreement. Is this difference already present in
your initial conditions?  If not, maybe setting irenorm=1 helps?



Klaus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20160929/2990b163/attachment.htm>