Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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 Tube around a radially growing/shrinking cylinder

Please login with a confirmed email address before reporting spam

Good Morning,

I guess I have a rather simple problem, but am not sure wether or not I am able to implement it at all and how exactly to do it. So I present you the problem and my solution afterwards.

I have a cylinder of diameter 10mm in a tube of diameter .2 and 1m length, the cylinder is placed at .5m/.1m. A fluid is flowing through the tube around the cylinder. For some reason the cylinder grows or shrinks with time at a defined velocity and I want to know how that influences the flow field. Over the cylinder boundary there is also mass transfer of a tracer substance that helps me to understand the influence of cylinder growth or shrinkage on mass transfer.

Now my idea is to define (in 2D) my geometry, which I did. Also the flow is laminar so I use the Single Phase-Fluid-Flow Physics. For the tracer transport I use the appropriate Physics for Transport of Diluted Species. The growth of the cylinder I implement via the Deformed Mesh Node and first define a constant normal velocity (for simplicity) on the cylinder boundary. When the cylinder grows the surronding fluid adjacent to the wall must have the velocity of the wall, so I use a moving wall boundary condition on the cylinders surface. Here is a problem in my eyes. I can define a moving wall in the fluid flow node, but must specify a velocity. I would expect, that somehow the deformed mesh speed and this boundary should be connected in some way, but they are not, and I cannot do it manually by entering the speed of the deformed mesh on the boundary.

Is there a mistake in my idea? Did I forget something or need to do something differently? Any help is greatly appreciated.

André

2 Replies Last Post 20 août 2015, 00:41 UTC−4

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 9 févr. 2015, 07:50 UTC−5
Hello again,

the upper problem I figured out by prescribing a constant velocity in the geometry deformation node. Afterwards I also prescribe this velocity on the moving wall node in the Fluid Flow Physics node and it worked perfectly.

Now I have a problem. This prescribed constant mesh speed is unfortunately not a constant but a function of mass transfered over the boundary. If I try to define the prescribe velocity in the geometry deformation node and the moving wall verlocity in the fluid flow node via the total flux through the boundary (surface of the cylidner) I encounter either one out of two errors:

1) 'Pointwise constraints for time derivatives are not supported.' This happens when I compute the mesh speed in the Geometry Deformation node via the total molar flux of the transfer substance and when I use the resulting mesh speed (xTIME, yTIME) to define account for the mesh motion on the moving wall boundary node in the fluid flow physics and for the changed mass transfer in the Transport of Diluted Species node.

2)'Circular variable dependency detected.
- Variable: mod1.chds.cfluxx_csolvent
- Geometry: 1
- Domain: 2'
This occurs when I use the flux term to directly define the deformation speed in all physics nodes instead of the mesh velocity as done above.

I am quite sure, that the error is my fault, because I do something according to my understanding instead of how COMSOL needs the information, but I cannot figure out what exactly is wrong. I also didn't find some helpful information in the manual, which is either caused by a lag of information in the manual or the fact that I have the right information found, but still do not understand it.

Maybe some of you can help me with my problem. This time I also attach the model file to this post, which may help you to understand what exactly my model is about.

Short Model Summary:
1) Laminar flow of a fluid in a cylindrical tube arround a cylinder in the flowpath.
2) The fluid contains a tracer substance solvent with a fixed inlet concentration that is transfered to the cylinder. Due to that mass transfer the cylinder begins to gain size.
3) The velocity of the growth, meaning the velocity at which the outer boundary of the cylinder moves, is the total mass flux across the boundary divided by the concentration at the boundary. So I defined it in the Deformed Mesh Geometry Node.

Best thanks for any advice in advance and I hope that the attached model file will help you to understand what my problem is.

André
Hello again, the upper problem I figured out by prescribing a constant velocity in the geometry deformation node. Afterwards I also prescribe this velocity on the moving wall node in the Fluid Flow Physics node and it worked perfectly. Now I have a problem. This prescribed constant mesh speed is unfortunately not a constant but a function of mass transfered over the boundary. If I try to define the prescribe velocity in the geometry deformation node and the moving wall verlocity in the fluid flow node via the total flux through the boundary (surface of the cylidner) I encounter either one out of two errors: 1) 'Pointwise constraints for time derivatives are not supported.' This happens when I compute the mesh speed in the Geometry Deformation node via the total molar flux of the transfer substance and when I use the resulting mesh speed (xTIME, yTIME) to define account for the mesh motion on the moving wall boundary node in the fluid flow physics and for the changed mass transfer in the Transport of Diluted Species node. 2)'Circular variable dependency detected. - Variable: mod1.chds.cfluxx_csolvent - Geometry: 1 - Domain: 2' This occurs when I use the flux term to directly define the deformation speed in all physics nodes instead of the mesh velocity as done above. I am quite sure, that the error is my fault, because I do something according to my understanding instead of how COMSOL needs the information, but I cannot figure out what exactly is wrong. I also didn't find some helpful information in the manual, which is either caused by a lag of information in the manual or the fact that I have the right information found, but still do not understand it. Maybe some of you can help me with my problem. This time I also attach the model file to this post, which may help you to understand what exactly my model is about. Short Model Summary: 1) Laminar flow of a fluid in a cylindrical tube arround a cylinder in the flowpath. 2) The fluid contains a tracer substance solvent with a fixed inlet concentration that is transfered to the cylinder. Due to that mass transfer the cylinder begins to gain size. 3) The velocity of the growth, meaning the velocity at which the outer boundary of the cylinder moves, is the total mass flux across the boundary divided by the concentration at the boundary. So I defined it in the Deformed Mesh Geometry Node. Best thanks for any advice in advance and I hope that the attached model file will help you to understand what my problem is. André


Please login with a confirmed email address before reporting spam

Posted: 9 years ago 20 août 2015, 00:41 UTC−4
Hi Andre,

Did you managed to solve that "Pointwise constraints for time derivatives are not supported". I'm getting the same error on a 'Moving wall' element in 'Free and Porous Media Flow' physics. I would appreciate any advice here.

Thanks!

Andre
Hi Andre, Did you managed to solve that "Pointwise constraints for time derivatives are not supported". I'm getting the same error on a 'Moving wall' element in 'Free and Porous Media Flow' physics. I would appreciate any advice here. Thanks! Andre

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.