Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
fluid flow in a 3D pipe
Posted Jun 22, 2016, 7:31 a.m. EDT Computational Fluid Dynamics (CFD) Version 5.2 3 Replies
Please login with a confirmed email address before reporting spam
Hello,
I´m just starting with Comsol and I tried to model the fluid flow (laminar flow) through a straight pipe, directly in 3D. I managed to do it in 2D axisymetric, but when I try to do it in 3D, the velocity profile far away from the velocity inlet (so when the flow is fully developped) does not correspond to the analytic solution: the maximum velocity should be twice the average velocity, yet it is significantly inferior to this value. I am absolutely certain that the tube is long enough for the flow to be fully developed.
My only hypothesis is that Comsol does not recognise the geometry as a tube, but as something else (parallel plates for example?). I have tried to implement the geometry in many differnt ways, even by importing it from SolidWorks.
Does anyone have an idea of what could be the problem?
Thank you very much
I´m just starting with Comsol and I tried to model the fluid flow (laminar flow) through a straight pipe, directly in 3D. I managed to do it in 2D axisymetric, but when I try to do it in 3D, the velocity profile far away from the velocity inlet (so when the flow is fully developped) does not correspond to the analytic solution: the maximum velocity should be twice the average velocity, yet it is significantly inferior to this value. I am absolutely certain that the tube is long enough for the flow to be fully developed.
My only hypothesis is that Comsol does not recognise the geometry as a tube, but as something else (parallel plates for example?). I have tried to implement the geometry in many differnt ways, even by importing it from SolidWorks.
Does anyone have an idea of what could be the problem?
Thank you very much
3 Replies Last Post Jun 24, 2016, 7:18 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Mesh quality is very important in obtaining reasonable results in 3-D. I had similar problem.
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Besides manipulating the mesh, try to use P2+P1 instead of P1+P1 and use higher geometry shape orders. This will certainly change the results as it did in my case.
Amin,
Amin,
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Thank you Amin, it indeed changed significantly the results.
Derek
Derek