dummy density bug

Submitted by rqwang on
cgoniva

Hi!

I think you are right, with version "CFDEM coupling version 2.1.0 beta" a bug came into the code, sorry for that!
We are currently working on the fix.

Thx,
Chris

Tue, 8 Nov 2011 - 12:25 Permalink
rqwang

In reply to by anonymous_stub (not verified)

Hi Chris,

I find there is still a dimension problem with the new code.

I got an error message like

[3] --> FOAM FATAL ERROR:
[3] Different dimensions for =
dimensions : [1 -3 -1 0 0 0 0] = [0 0 -1 0 0 0 0]

Please help me.

Thanks.

rq

Wed, 9 Nov 2011 - 23:39 Permalink
cgoniva

In reply to by anonymous_stub (not verified)

Hi!

When do you get this error?
The units of Ksl now changed (see version history), did you account for that?

(your suggested bugfix would not have checked the correctness of the dimensions)

In the momentum balance Ksl*(Urel)/rho is used and should have the units [m/s 1/s], so
Ksl [kg/m3 1/s] * Urel [m/s] / rho [kg/m3] --> [m/s 1/s]

Further Ksl composes of the forces on the particles Fparticle in [N] divided by ( cell volume [m3] * Urel [m/s]),
thus Ksl = Fparticle [N = kg*m/s2] / (Vcell [m3] * Urel [m/s]) --> [kg/m3 1/s].

Cheers,
Chris

Thu, 10 Nov 2011 - 07:58 Permalink
rqwang

In reply to by anonymous_stub (not verified)

Hi Chris,

This is my mistake. I only changed the Ksl dimension in my case.

After I changed KslNext and KslPrev, the simulation can continue.

Thanks for all the help.

rq

Thu, 10 Nov 2011 - 19:37 Permalink

Hi Chris,

I think the dummy density bug is still in the solver code, i.e.

Line 81 of cfdemSolverPiso.C should be

Ksl.internalField() = particleCloud.momCoupleM(0).impMomSource()/rho;

Please check it.

Thanks.

rq