CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Error in AMG Solver (http://www.cfd-online.com/Forums/fluent/67440-error-amg-solver.html)

RyGuy August 14, 2009 13:24

Error in AMG Solver
 
Hello All,

I am using coupled-implicit realizable k-epsilon viscous model with four boundary conditions to model a transonic compressible single phase flow.

The boundary conditions are main_inlet, powder_inlet, outlet and nozzle_walls.

I am using Intensity and Hydraulic diameter option for internal compressible flows along with the ideal-gas law for the density of the gas (air) in the compressible flow and the sutherland law for temperature affected viscosity values.

I get the following error:

Error: Internal error at line 947 in file '..\..\src\rp_mstage.c'.
divergence detected in AMG solver
Error: Object()

Reducing the Courant number does not seem to help...

Has anyone ever seen this error? How do I resolve this issue?

Anxiously awaiting a reply...

Pfalcon August 14, 2009 14:01

Is that in the first iteration?

RyGuy August 14, 2009 14:22

Thanks for the quick reply. No one usually ever replies......

The iterations start out good for the first 5 approximately, but quickly starts to diverge and before about 20 iterations (as well as Fluent automatically reducing the Courant number many times) I receive that error message.

Have you seen the likes of this before?

If you would like more information regarding my process I can gladly make it available to you!

Thanks for your time and effort,
Ryan

Pfalcon August 14, 2009 14:50

Seen this before. Can be difficult to start a calculation. Check all boundary conditions again and make sure that they are defined in the right way. Make sure the mesh has reasonable quality. Also it probably helps to start out with a simpler more robust turbulence model (SA for example), and check the solution strategies in the FLUENT manual.

RyGuy August 14, 2009 15:12

Hello Again,

I have checked my boundary conditions and they seem reasonable to me.

Main_inlet - is a pressure inlet corresponding to the proper edge.
Powder_inlet - is a pressure inlet corresponding to the proper edge.
Nozzle_walls - all edges needed are chosen.
Outlet - is a pressure outlet corresponding to the proper edges.

It seems when I switch to the SA 1-eqn model the residuals do not have as much trouble converging as before. If I can reach convergence with the 1-eqn model I can just switch it back to, say the k-epsilon model and iterate again correct?

I will continue to search in the manual about this topic as well....
Thanks for your time
Ryan

Pfalcon August 15, 2009 13:26

just switch back to k-e after the flow has settled on SA, with a bit of luck k-e runs stable and you reach nice convergence

shahed.malekipour February 8, 2014 11:31

Quote:

Originally Posted by Pfalcon (Post 226404)
Is that in the first iteration?



in my case, I get the error in the first iteration! what should I do?

shahed.malekipour February 8, 2014 11:45

Quote:

Originally Posted by shahed.malekipour (Post 474057)
in my case, I get the error in the first iteration! what should I do?

thank u everyone, I found the reason.

federicodauria May 21, 2014 06:21

internal error at line..
 
hi, I've the same problem.
I am using a density based type to study an ideal gas flow trough a nozzle. k-omega model.
The boundary conditions are pressure farfield inlet, pressure outlet.

I get the following error after about 50000 step:

Error: Internal error at line 1549 in file '..\..\src\rp_mstage.c'.
divergence detected in AMG solver.
what's the problem? what should I change in the settings?

thanks for your attention

prabha007 January 26, 2015 23:56

how rectifed the Error that you mentioned in your blog,i am facing the same error?
 
i am facing the following error. my problem type are
supersonic flow through constant cross section. which has one pressure inlet and three pressure outlet. pleas help me, i cannot able to solve this one.My id- prabha.aero@gmail.com

Internal error at line 1549 in file '..\..\src\rp_mstage.c'.
Divergence detected in AMG solver iter continuity x-velocity y-velocity z-velocity energy k epsilon time/iter
1 1.0000e+00 1.0000e+00 1.0000e+00 1.0000e+00 1.0000e+00 4.2860e-01 6.9502e-01 0:23:18 999
2 5.9241e-01 9.3721e-01 9.7271e-01 9.4691e-01 5.9003e-01 2.6348e-01 5.1220e-01 0:25:17 998
3 7.3231e-01 8.7290e-01 9.0639e-01 8.8105e-01 7.2436e-01 1.7387e-01 5.4305e-01 0:30:10 997
4 9.7003e-01 7.9767e-01 8.2500e-01 8.0102e-01 9.4848e-01 1.3600e-01 3.7917e-01 0:30:45 996
5 1.0000e+00 7.3452e-01 7.6763e-01 7.3059e-01 1.0000e+00 1.2598e-01 4.1529e-01 0:34:32 995
6 1.0374e+00 6.7770e-01 7.2468e-01 6.7018e-01 1.0205e+00 1.1571e-01 3.2431e-01 0:34:13 994
# Divergence detected in AMG solver: Coupled -> Decreasing coarsening group size!
# Divergence detected in AMG solver: Coupled -> Increasing relaxation sweeps!

Divergence detected - temporarily reducing Courant number to 0.5
and trying again...

Divergence detected - temporarily reducing Courant number to 0.05
and trying again...

Divergence detected - temporarily reducing Courant number to 0.005
and trying again...

Divergence detected - temporarily reducing Courant number to 0.0005
and trying again...

Divergence detected - temporarily reducing Courant number to 5e-05
and trying again...

Error: Internal error at line 1549 in file '..\..\src\rp_mstage.c'.
Divergence detected in AMG solver
Error Object: #f

chenxiaohu December 30, 2015 02:45

Quote:

Originally Posted by prabha007 (Post 529146)
i am facing the following error. my problem type are
supersonic flow through constant cross section. which has one pressure inlet and three pressure outlet. pleas help me, i cannot able to solve this one.My id- prabha.aero@gmail.com

Internal error at line 1549 in file '..\..\src\rp_mstage.c'.
Divergence detected in AMG solver iter continuity x-velocity y-velocity z-velocity energy k epsilon time/iter
1 1.0000e+00 1.0000e+00 1.0000e+00 1.0000e+00 1.0000e+00 4.2860e-01 6.9502e-01 0:23:18 999
2 5.9241e-01 9.3721e-01 9.7271e-01 9.4691e-01 5.9003e-01 2.6348e-01 5.1220e-01 0:25:17 998
3 7.3231e-01 8.7290e-01 9.0639e-01 8.8105e-01 7.2436e-01 1.7387e-01 5.4305e-01 0:30:10 997
4 9.7003e-01 7.9767e-01 8.2500e-01 8.0102e-01 9.4848e-01 1.3600e-01 3.7917e-01 0:30:45 996
5 1.0000e+00 7.3452e-01 7.6763e-01 7.3059e-01 1.0000e+00 1.2598e-01 4.1529e-01 0:34:32 995
6 1.0374e+00 6.7770e-01 7.2468e-01 6.7018e-01 1.0205e+00 1.1571e-01 3.2431e-01 0:34:13 994
# Divergence detected in AMG solver: Coupled -> Decreasing coarsening group size!
# Divergence detected in AMG solver: Coupled -> Increasing relaxation sweeps!

Divergence detected - temporarily reducing Courant number to 0.5
and trying again...

Divergence detected - temporarily reducing Courant number to 0.05
and trying again...

Divergence detected - temporarily reducing Courant number to 0.005
and trying again...

Divergence detected - temporarily reducing Courant number to 0.0005
and trying again...

Divergence detected - temporarily reducing Courant number to 5e-05
and trying again...

Error: Internal error at line 1549 in file '..\..\src\rp_mstage.c'.
Divergence detected in AMG solver
Error Object: #f


hi,do u solve this question???? I have same question...:confused::confused::confused:
If you have konwn the reason, please tell me ,THX in advance.

ashwingm June 5, 2016 15:22

How I solved the problem
 
I also had the following error, "Error: Internal error at line 1549 in file '..\..\src\rp_mstage.c'. "

If you are doing an axi-symmetric(example:CD nozzle, nose cone, anything that will be rotated along an axis), then make sure that, under BOUNDARY CONDITIONS, your axis of symmetry is set as "AXIS" and not "SYMMETRY".

My solver started solving, once I made this change. Hope this helps.:D


All times are GMT -4. The time now is 23:40.