ViewFactors

Numerical methods and mathematical models of Elmer
Post Reply
wmiller
Posts: 20
Joined: 09 May 2017, 18:05
Antispam: Yes

ViewFactors

Post by wmiller »

Hi

I have a problem with ViewFactor calculations.

I got the message:

ViewFactors: Number of rowsums below 0.5 is: 2 (out of 1673)

Afterwards Normalization failed.

This error occurs in a sequence of computations for a growing crystal in Czochralski process.

Geometry and meshing remain the same only the crystal is a little bit longer and crystal holder is moved upwards but without any qualitative change of geometry.

Does anybody have an idea were to dig for solving the problem ???

Variation of the grid density at radiation boundaries does not really helps .....
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: ViewFactors

Post by raback »

Hi

Is you geometry partially open? Are you sure that the normals point to the right direction? The easy way is to define emissivity only at the side of the materil radiating outwords. Then the code knows how to orient the normals.

By construction we know that viewing angles should for closed systems sum up to 2Pi (half space). The code only complains if this is missing by 50% so there could be something fishy going on. The nice thing with closed geometries is that this fact may be used to increase accuracy such that this sum will be exact. Then the radiating heat is better conserved.

-Peter
wmiller
Posts: 20
Joined: 09 May 2017, 18:05
Antispam: Yes

Re: ViewFactors

Post by wmiller »

Hi

we checked that the geometry is not open.
We use the trick with no emissivity on the side of the cave.

In certain cases the crystal is just 1 mm longer and the View Factor calculation fails.
I also have the suspicion that very fine grid might cause the problem. When the grid in the crystal is finer the View Factor calculation fails even for cases when for a coarser grid it is successful.
May be this is the reason to separate the grid for bulk calculation and for View factors in CGsim (but then you have the issue of interpolation ....)

In any case it needs some further checks.

Wolfram
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: ViewFactors

Post by raback »

Hi Wolfram,

If you could produce a minimal case where this occurs that could help.

-Peter
Post Reply