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.

Simulating pipe flow

Natasha Systems biology

Please login with a confirmed email address before reporting spam

Hello ,

I am trying to simulate fluid flow and couple to species transport in a pipe section of diameter(d) ~5 micrometers and it branched into two pipes of d/2 and merger to form a pipe of dia, d again.

Inlet velocity = 0.00018 m/s and pressure boundary condition is 10Pa.

While trying ot simulate, the fluid flow simulation fails with the following error

Undefined variable. - Variable: timestep - Geometry: geom1 - Boundaries: 2, 4 Failed to evaluate Jacobian of expression. - Expression: (comp1.dsp.stab_ds_c3)*(dvol)

I am also attaching the mph file. Any suggestions on how to resolve this error?



5 Replies Last Post 10 gen 2020, 02:17 GMT-5

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 8 gen 2020, 23:52 GMT-5

I have simulated similar problems and my best bet would be to use finer mesh sizes. Let us know if that helps!

Moin

-------------------
Moin
I have simulated similar problems and my best bet would be to use finer mesh sizes. Let us know if that helps! Moin

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 9 gen 2020, 03:46 GMT-5

I had a quick look at your model. Items 'Wall Mass Transfer 1' and 'Continuity 1' are not assigned to any boundary/domain.

I had a quick look at your model. Items 'Wall Mass Transfer 1' and 'Continuity 1' are not assigned to any boundary/domain.

Phillip COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 9 gen 2020, 03:51 GMT-5
Updated: 5 years ago 9 gen 2020, 03:52 GMT-5

Hello Natasha,

the error occurs because set the equation form of the Transport of Diluted Species in Pipes interface was set to Time dependent. This setting introduces time dependent variables that the stationary solver can not handle.

The model will converge if you change this setting back to Study controlled or Stationary.

Success with your simulation project!

-------------------
Best Regards
Phillip
Hello Natasha, the error occurs because set the equation form of the *Transport of Diluted Species in Pipes interface* was set to *Time dependent*. This setting introduces time dependent variables that the stationary solver can not handle. The model will converge if you change this setting back to *Study controlled* or *Stationary*. Success with your simulation project!

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 9 gen 2020, 06:29 GMT-5

I did nothing else but assigned 'Wall Mass Transfer 1' to all boundaries; thereafter the model converged.

I did nothing else but assigned 'Wall Mass Transfer 1' to all boundaries; thereafter the model converged.

Natasha Systems biology

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 10 gen 2020, 02:17 GMT-5

Hello Everyone, Thanks a lot for the suggestions and inputs. I've uncoupled the transport dynamics. I tried to solve for fluid flow.

I'm facing two problems, 1. Defining pressure boundary conditions In the documentation of pipe flow module it is mentioned, " The pressure condition can only be applied to a point which is connected to exactly one edge (inlet or outlet of a system)". However, by default in "Pressure 1" tab of pipe flow module in my file , the pressure BC is set to the outlet node. (If this is not possible in the pipe flow module , is there a possibility to use laminar flow module and simulate fluid flow in the same geometry using inlet pressure BC)

However, I'd like to change this to inlet (i.e. apply pressure BC to inlet). Any suggestions on how to do this will be really helpful.

  1. Regarding the type of friction flow model and convergence. I observed that if the Friction model ( in flow resistance tab of pipe properties 2,3,4) is set to "Churchill" the solver ocnveregs to a solution, if "User defined" and fD = 0 is set the solver doesn't converge. I'd like to ask for explanation on why this occurs.
Hello Everyone, Thanks a lot for the suggestions and inputs. I've uncoupled the transport dynamics. I tried to solve for fluid flow. I'm facing two problems, 1. Defining pressure boundary conditions In the documentation of pipe flow module it is mentioned, " The pressure condition can only be applied to a point which is connected to exactly one edge (inlet or outlet of a system)". However, by default in "Pressure 1" tab of pipe flow module in my file , the pressure BC is set to the outlet node. (If this is not possible in the pipe flow module , is there a possibility to use laminar flow module and simulate fluid flow in the same geometry using inlet pressure BC) However, I'd like to change this to inlet (i.e. apply pressure BC to inlet). Any suggestions on how to do this will be really helpful. 2. Regarding the type of friction flow model and convergence. I observed that if the Friction model ( in flow resistance tab of pipe properties 2,3,4) is set to "Churchill" the solver ocnveregs to a solution, if "User defined" and fD = 0 is set the solver doesn't converge. I'd like to ask for explanation on why this occurs.

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.