I use CFX5.5.1 I add a body force model in the domain. It has a problem of calculation. An error usually appears in the process of calculation and stops the process: error #001100279 has occured in subroutine ErrAction. Message: Floating point exception:Type Unknown Details of error:- Error detected by routine MAKDIR CDRNAM=CopyDs CRESLT=ILEG Details of error:- Error detected by routine POPDIR CRESLT=ILEG An error has occurred in cfx5solve

I tried to change the advection scheme, local timescale factor,better initial guess,different boundary condition. It is still there. Anyone knows how to solve it ?

thank you very much!

Hi cfxbeginer,

You have crashed the solver with a floating point exception. This means it has done an illegal mathematical operation, probably dividing by zero.

Possible causes for the problem are (in order you should test):

1) A non-physical setup, eg in compressible flow, specifying a mass flow rate greater than that physically possible in choked flow;

2) Too coarse a grid;

3) Too coarse time steps;

4) Poor initial conditions.

I hope that helps. If you specify what your simulation is a bit more we may be able to help you more.

Regards, Glenn

Hi, I agree, what Glen has said. Its always easiest to check timestep first. I had the same problem yesterday, running a steady state simulation of buoyant air flow with the automatic timestepping turned on. I changed this to include a max time step value of 0.5 sec (which was a reduction from 3sec which the solver came up with), and this solved the problem. The other suggestions of Glens are all good but can be more difficult to track down ie you set up a problem how you think it should be setup, however what you have done may result in unrealistic physics - how do you know that ? I guess that one is just an experience thing ? (If only there were some way to check things like that automatically, checking BC's etc ).