High particle volume fractions in CFDEM

Dear Evan,

I think you should have a look at the CFD timestep. I had similar problems with simulating the settling of particles. With time the voidfraction decreased an therefore the velocity and the Courant number increased. With reducing the CFD timestep the simulation got stable. Another option was to smooth the voidfraction which is exchanged by DEM to CFD during the coupling procedure (see CFD/constant/couplingProperies), to avoid single cells which have by change very low voidfractions and therefore high velocities.

I tried also to alter the coupling intervall, but this didn't help much (a coupling intervall of 100 resulted in the most stable simulations in my case, but I think this depends on the ratio between particle size and CFD mesh size). Would be interesting to hear about your findings.

Best regards
Philipp

Hi all,
I'm simulating the sediment transport with my solver(interFoam+liggghts). But I did not succeed yet!
The velocity in inlet and outlet of channel in vicinity the particles became very high!
I guess some subjects:
- properties of particles OR different properties of particles before and after insertion!
- I don't know why Solving for p_rgh didn't stop in terminal after timeStepFraction()=1 !?(Please see below)

[
timeStepFraction() = 1
update Ksl.internalField()
smoothSolver: Solving for Ux, Initial residual = 1, Final residual = 5.99123e-05, No Iterations 1
smoothSolver: Solving for Uy, Initial residual = 1, Final residual = 0.0140736, No Iterations 1
smoothSolver: Solving for Uz, Initial residual = 1, Final residual = 0.0123131, No Iterations 2
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 1, Final residual = 0.0084426, No Iterations 6
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 0.00156927, Final residual = 9.0544e-06, No Iterations 6
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 7.90966e-05, Final residual = 2.70149e-07, No Iterations 7
time step continuity errors : sum local = 5.42811e-10, global = 1.00011e-10, cumulative = 1.00011e-10
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 2.63875e-05, Final residual = 1.17216e-07, No Iterations 5
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 1.1953e-06, Final residual = 9.64691e-09, No Iterations 4
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 6.55706e-08, Final residual = 3.72289e-09, No Iterations 1
time step continuity errors : sum local = 7.50865e-12, global = 2.5975e-12, cumulative = 1.02609e-10
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 4.17602e-07, Final residual = 7.54834e-09, No Iterations 3
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 4.16853e-08, Final residual = 3.48123e-09, No Iterations 1
suppressing ddt(voidfraction)
GAMG: Solving for p_rgh, Initial residual = 4.44297e-09, Final residual = 4.44297e-09, No Iterations 0
time step continuity errors : sum local = 8.96096e-12, global = 1.0151e-12, cumulative = 1.03624e-10
smoothSolver: Solving for epsilon, Initial residual = 1.20493e-05, Final residual = 9.51893e-10, No Iterations 1
smoothSolver: Solving for k, Initial residual = 1, Final residual = 5.11388e-05, No Iterations 1
ExecutionTime = 40.24 s ClockTime = 125 s

Starting time loop

Courant Number mean: 1.71469e-05 max: 0.000278115
Interface Courant Number mean: 0 max: 0
Time = 4e-06

MULES: Solving for alpha1
Phase-1 volume fraction = 0.501333 Min(alpha1) = 0 Max(alpha1) = 1
MULES: Solving for alpha1
Phase-1 volume fraction = 0.501333 Min(alpha1) = 0 Max(alpha1) = 1
MULES: Solving for alpha1
Phase-1 volume fraction = 0.501333 Min(alpha1) = -1.10441e-28 Max(alpha1) = 1
Starting up LIGGGHTS
Executing command: 'run 100 pre no post no'
run 100 pre no post no
Step Atoms KinEng 1 Volume
11000102 50000 4.7159825e-11 1.992485e-13 0.0015
CFD Coupling established at step 11000110
CFD Coupling established at step 11000120
CFD Coupling established at step 11000130
CFD Coupling established at step 11000140
CFD Coupling established at step 11000150
CFD Coupling established at step 11000160
CFD Coupling established at step 11000170
CFD Coupling established at step 11000180
CFD Coupling established at step 11000190
CFD Coupling established at step 11000200
11000202 50000 1.9498214e-10 2.7142563e-13 0.0015
Loop time of 13.6989 on 4 procs for 100 steps with 50000 atoms
LIGGGHTS finished

Coupling...

timeStepFraction() = 2
update Ksl.internalField()
smoothSolver: Solving for Ux, Initial residual = 0.270672, Final residual = 2.73495e-05, No Iterations 1
smoothSolver: Solving for Uy, Initial residual = 0.337075, Final residual = 2.41268e-05, No Iterations 1
smoothSolver: Solving for Uz, Initial residual = 0.332952, Final residual = 3.69354e-05, No Iterations 1
smoothSolver: Solving for epsilon, Initial residual = 3.67666e-05, Final residual = 2.0772e-09, No Iterations 1
smoothSolver: Solving for k, Initial residual = 0.522607, Final residual = 3.25428e-05, No Iterations 1
ExecutionTime = 48.34 s ClockTime = 150 s

]

Thanks
Elham

Hello elham,

you should share your solver to have some information regarding what could be wrong with the timestep fraction,
you should share the input of you simulation to to guess what could be wrong with velocity,
however you have very high residual, you should take a look at that

Best regards,
Riccardo