Thunderhead Engineering Forum

Please login or register.

Login with username, password and session length
Advanced search  

News:

Forum moved to https://forum.thunderheadeng.com

Author Topic: Numerical instability  (Read 6003 times)

Senid

  • Member
  • **
  • Posts: 40
    • View Profile
Numerical instability
« on: March 06, 2014, 05:09:32 am »

How can I solve this problem. Picture of my computer is attached, also the problem that I have is attached, if I could with some kinds of some settings in Pyrosimu this fix.
Best regards
Logged

Charlie Thornton

  • Thunderhead
  • *****
  • Posts: 851
    • View Profile
Re: Numerical instability
« Reply #1 on: March 06, 2014, 08:54:07 am »

Can you post your model?

In section 4.2 "Common Error Statements", the FDS User Guide has the following to say about numerical instability errors:

Quote
It is possible that during an FDS calculation the flow velocity at some location in the domain can increase due to numerical error causing the time step size to decrease to a point1 where logic in the code decides that the results are unphysical and stops the calculation with an error message in the file CHID.out. In these cases, FDS ends by dumping out one final Plot3D file giving you a hint as to where the error is occurring within the computational domain. Usually, a numerical instability can be identified by fictitiously large velocity vectors emanating from a small region within the domain. Common causes of such instabilities are mesh cells that have an aspect ratio larger than 2 to 1, high speed flow through a small opening, a sudden change in the heat release rate, or any number of sudden changes to the flow field. There are various ways to solve the problem, depending on the situation. Try to diagnose and fix the problem before reporting it. It is difficult for anyone but the originator of the input file to diagnose the problem.
Logged

Senid

  • Member
  • **
  • Posts: 40
    • View Profile
Re: Numerical instability
« Reply #2 on: March 06, 2014, 09:06:25 am »

Model attached.
Logged

Wojciech_Stigo

  • Member
  • **
  • Posts: 10
    • View Profile
Re: Numerical instability
« Reply #3 on: March 10, 2014, 05:05:35 am »

I think that your fire parametres are not set properly. You have very big value of fire and You set ramp-up time as 1 s. So after 1s od simulation your fire will reach 25000 kW multiplied by 8 meters squared (surface of vent) So after 1s your fire will reach 200 000 kW. This is not real situation. And in my opinion this is a reason of error...

1. You should change area of fire (for example 5x5m)
2. Set HRR as 1000 kW/m2       (25m2 * 1000kW/m2 = 25000kW)
3. Set ramp - up time to T^2 and set this time for example as 900s.

Your mesh is not correct too.. Try to get 0,5m cell size in all directions
Logged

Senid

  • Member
  • **
  • Posts: 40
    • View Profile
Re: Numerical instability
« Reply #4 on: March 10, 2014, 07:38:58 am »

Many thanks for these tips. I need to simulate fire ignition tank with diesel fuel where the total heat liberated is about 344 000 MJ (10 m3 tank with diesel fuel). If surface sources put the 5x5 (25m2) how much to put HRRPUA and ramp up time because the total heat liberated should be about 344 000 MJ.
Logged

Senid

  • Member
  • **
  • Posts: 40
    • View Profile
Re: Numerical instability
« Reply #5 on: March 13, 2014, 02:07:40 am »

And under these parameters as you write reports to me a mistake.
Logged