Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

Negative rho without element distorsion

Hello,

My computation crashed with a "negative rho" error termination, but the two 3D elements considered are not distorted. Also, the time step did not decreased at all before the crash. I noticed an added mass (still very low) and a fluctuating "ERR". (see attached file)

Could you tell me more i) about this negative rho error, and ii) how is computed this "ERR"?

Thanks in advance.

node4-nt4-np32-2245738.log

Find more posts tagged with

English

HyperMesh

Radioss

v2022

Accepted answers

All comments

Mathis

Hi Blaisem,

I have attached a pdf with a set of slides explaining the different points you are asking for. I hope it will help you.

I see you have 99% error at the begining of your simulation. I would recommand to check the first h3d frame to check if your initial conditions are correct. T01 can also help you check energies.

Regards,

Mathis.

Error.pdf

Blaisem

Hi Mathis,

Thank you. The energy error starts indeed at 99% and goes down to -99% for similar calculations I ran.

I saw in the pdf that the negative rho error can be introduced by the EOS.
I model some water: I used to use the Stiff-Gas EOS, but in my current model it makes the fluid elements stiffer. So I changed to the Polynomial EOS with ALE formulation. Since I used it, the energy error starts at 99% and goes down to -99; it was always at zero with the Stiff Gas EOS.

I also noticed that the added mass (DM/M) was null in fully Lagrangian formulation, but is not anymore with ALE.

Would you have any recommendations for parameters to use in LAW6 with Polynomial EOS to model some water? I still don't know why the results are so different with the Polynomial and the Stiff-Gas EOS.

Regards, MB

Mathis

For the EoS, is it possible for you to share your input? My general recommandation would be to plot your EoS.For that purpose, I have attached a compose script that will allow you to check if your EoS seems to do what you want to do.

About the added DM/M, in can be due to the inlet or oulet of your model. Some mass will either be added or removed during your simulation.

You have some correspondance between the stiffgas EoS and the polynomial EoS (it is detailed in the Radioss documentation of Stiffgas EoS). If you follow the formula, I would expect the results to be the same.

Regards,

Mathis.

VERIFICATION_polynomial_EOS.oml

VERIFICATION_Stiffened_gas_EOS.oml

Blaisem

Thanks. Parameters of both EoS are attached.

The whole model is fully closed, so that no mass is supposed to be added or removed...

Could you also help me to find how the different energies we can plot with the T01 file (external forces work, DTE etc.) are calculated?

fluid_eos.JPG