|
[Sponsors] |
Time stepping issue when running a simulation for fluid flow in a tube |
![]() |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
![]() |
![]() |
#1 |
New Member
Join Date: Aug 2023
Posts: 4
Rep Power: 3 ![]() |
Hello,
This is my first time on this website. I have been working on an FSI simulation on ANSYS. My model is a pretty basic tube flow. The tube is 60mm long, 6mm diameter, and 1mm thick. The fluid domain is non-Newtonian with inlet velocity condition and outlet pressure. the velocity is time dependent. When running my simulation at 0.01 timestep size i face no issues, however i would like to reduce the time step size to 0.001. this is where my problems starts as the model becomes highly unstable and i get excessive deformation in the structural domain. I am not quiet sure of the possible reasons behind this error and what areas should i explore to fix the issue. Thank you in advance. |
|
![]() |
![]() |
![]() |
![]() |
#2 |
New Member
Swapnil
Join Date: Apr 2022
Posts: 3
Rep Power: 4 ![]() |
I am facing the same issue. Have you figured it out?
|
|
![]() |
![]() |
![]() |
![]() |
#3 |
New Member
Join Date: Aug 2023
Posts: 4
Rep Power: 3 ![]() |
Hello,
Yes I have. In fluent, you go to dynamic meshing and then you double click on the system coupling region (the fluid-solid interface region you have created). you double click on it and in the window you head over to the "solution" tab if i am not mistaken, in fluent 19 it will be the last tab. activate it and set it on "coefficient based stabilization". set the coefficient (I suggest starting with a value of around 0.5 and working your way up). Eventually you will hit an optimum value. Things to look out for: 1- When you add stabilization your system will require more iterations to achieve convergence at the interface. So make sure you increase the number of iteration to an acceptable value that would achieve convergence. 2- Careful for overstabilization of the system, as it could kill off some interesting features for you (I've come to this realization after doing sensitivity tests experimenting with increments of the stabilization coefficient). Unfortunately, I am yet to understand the exact effect of this coefficient as i am fairly novice in the field of FSI, let alone FSI within ANSYS. However this issue is not uncommon and is faced by many experts in the field as i have interacted with a few of them after bringing up this issue. I hope you manage to resolve the issue and please let me know of any other problems you might face moving further. I also suggest looking into articles that tackle such problem as it will give you better insight. |
|
![]() |
![]() |
![]() |
![]() |
#4 |
New Member
Swapnil
Join Date: Apr 2022
Posts: 3
Rep Power: 4 ![]() |
Hi, Thank you for your response.
I have used both coefficient and volume based scale factors previously. I tried with the coefficients within range of 0.1 to 0.5. But solution is diverging. Geometrically, my configuration looks similar to yours. But I need even lower time-step of 10^-4 to capture initial wave propagation. I will try with your suggestion, will try with coefficients above 0.5. In your case, How much iterations it took to converge? Which mesh you are you using, structured/unstructured ? Thanks. |
|
![]() |
![]() |
![]() |
Tags |
ansys, fsi 2-way coupling |
|
|
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Adaptive vs fixed time stepping and simulation time | LynnCFD | FLUENT | 0 | March 7, 2023 19:21 |
[Other] Contribution a new utility: refine wall layer mesh based on yPlus field | lakeat | OpenFOAM Community Contributions | 58 | December 23, 2021 02:36 |
AMI speed performance | danny123 | OpenFOAM | 21 | October 24, 2020 04:13 |
Superlinear speedup in OpenFOAM 13 | msrinath80 | OpenFOAM Running, Solving & CFD | 18 | March 3, 2015 05:36 |
How to write k and epsilon before the abnormal end | xiuying | OpenFOAM Running, Solving & CFD | 8 | August 27, 2013 15:33 |