Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

Fluent-EDEM Coupling Problem

I am studying the motion of long flexible bodies in a rotating flow field using coupled Fluent and EDEM calculations with an Eulerian coupling interface. The long flexible bodies are modeled using the fibers bonding model in EDEM2025, but some parameters are highlighted in red and I'm unclear about the cause. The figures below show the parameter settings for the long flexible bodies and the coupled calculation time step settings for Fluent-EDEM.

The simulation results include residual plots and a drag force plot. The coupled calculation proceeds normally from 0 to 0.00065s, but the calculation suddenly diverges after 0.00065s. Do you have any good suggestions for the cause of the divergence?

Find more posts tagged with

EDEM

Accepted answers

Stephen Cole

Typical reasons could be:

EDEM time-step too high, particles have exploded due to contact forces and caused divergence.
Ratio between EDEM and Fluent time-step too high, particles have received too much force from the fluid. Typically fluent time-step is 10x —100x greater than EDEM. Solution is to decrease the Fluent time-step so the ratio between EDEM and Fluent is lower
Fluid mesh is too small - ideally with Fluent the mesh cells are larger than the particles. There isn't a good resolution to this other than to increase mesh cell size. You can reduce the under-relaxation factors in Fluent for momentum and volume fraction, but I'd recommend checking the Fluent docs on that

I'd try checking the EDEM and Fluent cases separately to see if there is any obvious instabilities in either uncoupled which may help.

Regards

Stephen

PrasadAvilala

Hi Bai,

In addition to Stephen Cole suggestion, i would also suggests adding damping co-efficient as non-zero.

Also, just want to understand what makes in EDEM the time step is too low, is it shear modulus is high if yes can you try reducing and stabilize other bonding model parameters.

Also, when you say divergences what is the meaning do you see particles are instable/ explosion in EDEM or divergence is observed in fluent?

Thanks,

Prasad A

PrasadAvilala

Hi Bai,

I feel you no need to provide such a large shear modulus instead you can use as 3e7 pa and fiber properties please play.

i would suggest first fix the properties of fiber model in EDEM alone by dropping fiber on to ground and see the flexibility.

Once the fiber model stable in EDEM alone then please couple with CFD run then you wont see any instabilities.

Thanks,

Prasad A

All comments

Stephen Cole

Typical reasons could be:

EDEM time-step too high, particles have exploded due to contact forces and caused divergence.
Ratio between EDEM and Fluent time-step too high, particles have received too much force from the fluid. Typically fluent time-step is 10x —100x greater than EDEM. Solution is to decrease the Fluent time-step so the ratio between EDEM and Fluent is lower
Fluid mesh is too small - ideally with Fluent the mesh cells are larger than the particles. There isn't a good resolution to this other than to increase mesh cell size. You can reduce the under-relaxation factors in Fluent for momentum and volume fraction, but I'd recommend checking the Fluent docs on that

I'd try checking the EDEM and Fluent cases separately to see if there is any obvious instabilities in either uncoupled which may help.

Regards

Stephen

PrasadAvilala

Hi Bai,

In addition to Stephen Cole suggestion, i would also suggests adding damping co-efficient as non-zero.

Also, just want to understand what makes in EDEM the time step is too low, is it shear modulus is high if yes can you try reducing and stabilize other bonding model parameters.

Also, when you say divergences what is the meaning do you see particles are instable/ explosion in EDEM or divergence is observed in fluent?

Thanks,

Prasad A

bai

Hi Stephen,

Thank you for your valuable feedback. I'll go through your suggestions one by one until the issue is resolved.

Best regards,

Bai

bai

Hi PrasadAvilala,

Thank you for your valuable suggestions. In my simulation, the long flexible bodies represent cotton fibers. The shear modulus was set to 3e+9 Pa, which resulted in the time step in the EDEM simulation is too low. When the simulation results diverged, I observed instability in the fiber model, with bonds formed between particles disengaging. As shown in the figure below, blue represents particles and red represents bonds.

Best regards,

Bai

bai

Additionally, particle explosion phenomena occasionally occur.

PrasadAvilala

Hi Bai,

I feel you no need to provide such a large shear modulus instead you can use as 3e7 pa and fiber properties please play.

i would suggest first fix the properties of fiber model in EDEM alone by dropping fiber on to ground and see the flexibility.

Once the fiber model stable in EDEM alone then please couple with CFD run then you wont see any instabilities.

Thanks,

Prasad A