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.
ERROR: nonlinear solver did not converge, when simulating ball movement in meso-size chamber.
Posted 5 gen 2013, 21:41 GMT-5 Modeling Tools & Definitions, Parameters, Variables, & Functions, Studies & Solvers Version 5.0 9 Replies
Please login with a confirmed email address before reporting spam
Hi all,
I have seen several threads about this error and tried to figure it out myself using the sample cases but I still fail to simulate this. Any help will be greatly appreciated.
I am trying to simulate the movement of a 1 mm diameter ball inside a chamber. There is a water flow on the top of chamber, from left to right. I am using fluid structure interaction (FSI).
I keep getting the same error message that says
Nonlinear solver did not converge.
Time : 4.6875e-005
Segregated group 2
Attempt to evaluate real square root of negative number
Function: sqrt ... (and so on)
I also get a list of warning saying that "Inverted mesh element near coordinates ..., ...) that are located inside the ball (which is linear elastic solid material). I am also confused because it supposedly have less deformation compared to fluid area.
Any help or idea will be greatly appreciated. Thanks in advance.
I have seen several threads about this error and tried to figure it out myself using the sample cases but I still fail to simulate this. Any help will be greatly appreciated.
I am trying to simulate the movement of a 1 mm diameter ball inside a chamber. There is a water flow on the top of chamber, from left to right. I am using fluid structure interaction (FSI).
I keep getting the same error message that says
Nonlinear solver did not converge.
Time : 4.6875e-005
Segregated group 2
Attempt to evaluate real square root of negative number
Function: sqrt ... (and so on)
I also get a list of warning saying that "Inverted mesh element near coordinates ..., ...) that are located inside the ball (which is linear elastic solid material). I am also confused because it supposedly have less deformation compared to fluid area.
Any help or idea will be greatly appreciated. Thanks in advance.
Attachments:
9 Replies Last Post 10 nov 2015, 06:04 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Anybody? I am really sorry about this, but I am just a newbie who still trying to figure stuff on the COMSOL.
Any help or comment would be greatly appreciated. Thanks!
Any help or comment would be greatly appreciated. Thanks!
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
no reason to be sorry, but starting COMSOL directly with FSI (a 3 physics combination case) is setting the bar quite high up! ;)
The best is to start step by step, one physics at the time, and repeat a few of the model library examples first.
There could be many reasons why the solver fails: mesh, mesh density, boundary mesh missing or not complete, BC values missing or incomplete, and more critical initial: values too far from the solution, difficult to say like that
Do not give up, you will see after some time you will become a FEM master ;)
--
Good luck
Ivar
no reason to be sorry, but starting COMSOL directly with FSI (a 3 physics combination case) is setting the bar quite high up! ;)
The best is to start step by step, one physics at the time, and repeat a few of the model library examples first.
There could be many reasons why the solver fails: mesh, mesh density, boundary mesh missing or not complete, BC values missing or incomplete, and more critical initial: values too far from the solution, difficult to say like that
Do not give up, you will see after some time you will become a FEM master ;)
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Ivar,
thank you for your advice. I already did some of the tutorials on laminar flow, heat transfer, and FSI. I also already tried to modify the FSI/microfluidic tutorial from a filament that stand perpendicular to the flow to be a free moving sphere inside a cavity, which also does not work.
Some things I notice is that when I use the same FSI tutorial (filament stand against moving fluid in microfluidic channel), but make the filament free to move (take out the constraint on the base), I got the similar error message. Even if I modify the channel to be a well, I still got error message. Do you have any advice for modeling buoyant solid sphere that is moved by fluid?
Thanks in advance.
thank you for your advice. I already did some of the tutorials on laminar flow, heat transfer, and FSI. I also already tried to modify the FSI/microfluidic tutorial from a filament that stand perpendicular to the flow to be a free moving sphere inside a cavity, which also does not work.
Some things I notice is that when I use the same FSI tutorial (filament stand against moving fluid in microfluidic channel), but make the filament free to move (take out the constraint on the base), I got the similar error message. Even if I modify the channel to be a well, I still got error message. Do you have any advice for modeling buoyant solid sphere that is moved by fluid?
Thanks in advance.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
FSI and large motions are tricky for the ALE, re-meshing sometimes helps ...
have you looked at the level set method ?
--
Good luck
Ivar
FSI and large motions are tricky for the ALE, re-meshing sometimes helps ...
have you looked at the level set method ?
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Hi
I changed your uploaded program with following modifications in the study section.
1) Segregated to Fully Coupled.
2) Free to Strict in time stepping with Initial step 0.001, Maximum step 0.01.
Based on this, I got attached figures. It looks buoyant solid sphere blocks right flow channel after 0.0924 seconds. I hope this helps you.
Thanks,
Myungwu
I changed your uploaded program with following modifications in the study section.
1) Segregated to Fully Coupled.
2) Free to Strict in time stepping with Initial step 0.001, Maximum step 0.01.
Based on this, I got attached figures. It looks buoyant solid sphere blocks right flow channel after 0.0924 seconds. I hope this helps you.
Thanks,
Myungwu
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Hi Myungyu,
thank you so much for help solving this problem! I have yet found the solution until now and your help is greatly appreciated.
I followed your suggestion, but I got this error message instead. Attached are the figure of steps that I believed following you, the error message, and the error *.mpg file (let me know if you need the comsol file with results).
One thing that change compared to before is that i now use version 4.4b. Does it change anything?
Could you tell me what I did wrong please? Thank you again! I really appreciate it!
Sofie
thank you so much for help solving this problem! I have yet found the solution until now and your help is greatly appreciated.
I followed your suggestion, but I got this error message instead. Attached are the figure of steps that I believed following you, the error message, and the error *.mpg file (let me know if you need the comsol file with results).
One thing that change compared to before is that i now use version 4.4b. Does it change anything?
Could you tell me what I did wrong please? Thank you again! I really appreciate it!
Sofie
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Hi Sofie,
Your result is exactly same as mine. If my understanding is correct, the circle moves to right-up direction and finally blocks the fluid flow after 0.0934 second. Please draw the fluid flow profile from the Results with 2D at time 0 and 0.09 second. You can see the circle is moved. I did not attaching figures because, for some reason, "Attach File" button of this website did not work at this time. I hope this is the result you want, otherwise please let me know.
Thanks,
Myungwu
Your result is exactly same as mine. If my understanding is correct, the circle moves to right-up direction and finally blocks the fluid flow after 0.0934 second. Please draw the fluid flow profile from the Results with 2D at time 0 and 0.09 second. You can see the circle is moved. I did not attaching figures because, for some reason, "Attach File" button of this website did not work at this time. I hope this is the result you want, otherwise please let me know.
Thanks,
Myungwu
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Hi all,
I have this problem and I think it could be because of boundary condition. When the free objects arrive the boundary of fluid, it must recognize it as a boundary condition, but it doesn't.
Can any body tell me how to solve this problem?
I want my solid model deform and pass through the channel.
I really appreciate this regard.
bests.
I have this problem and I think it could be because of boundary condition. When the free objects arrive the boundary of fluid, it must recognize it as a boundary condition, but it doesn't.
Can any body tell me how to solve this problem?
I want my solid model deform and pass through the channel.
I really appreciate this regard.
bests.
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi
I am facing same error while solving a problem where a solid ball is kept at the interface of 2 fluids and the change in interface has to be measured.
I am facing same error while solving a problem where a solid ball is kept at the interface of 2 fluids and the change in interface has to be measured.