Problem with Residual plot raising up to hige numbers na do not converge

fardinkhalili · March 14, 2014, 16:22

Hi, I am working ono my thesis about CFD and Aeroacoustics of wind turbine with rotating blades. I got perfect results when I ran my simulation in steady-state condition for Curle and Proudman noise Models. For example I followed the first tutorial for aeroacoustics and the results were OK and I got convergence near to 10^-4 for residual plot. But, when I change the physics to unsteady, and change the motion to rotating, either using Reynolds-Averaged Navier-Stokes or DES or LES, everything goes wrong and the residual plot rises to hige numbers up to 10^7!!!! I change the motion to stationary or rotation but nothing got better. What is your opinion and how can I solve this problem.

fardinkhalili · March 14, 2014, 16:26

pressure corrections limited on 9 cells in STATIC
pressure corrections limited on 60083 cells in ROTATING
Temperature corrections limited on 327 cells in STATIC
Temperature corrections limited on 34752 cells in ROTATING
Minimum Temperature limited to 100 on 439 cells in ROTATING
Maximum Temperature limited to 5000 on 405 cells in ROTATING
Maximum Temperature limited to 5000 on 33 faces on Wind Turbine Rotor.Default
Temperature limited to maximum value 1270 times on model-part "ROTATING"
Breaking out of time step iteration.
Time solution is in an inconsistent state.

fardinkhalili · March 14, 2014, 16:28

And here is the residual plot for first iterations.

me3840 · March 19, 2014, 15:27

How are you initializing the transient simulation? Is the flowfield static and your turbine suddenly is spinning at maximum RPM? That will be very difficult to converge, you need to ramp its speed or start your solution off from your steady-state solution.

ping · March 25, 2014, 10:35

you should certainly be starting your unsteady solve from a reasonably converged steady run - very easy to do.
also check your timestep - i suggest one that gives you about 1 degree of rotation per timestep.

March 14, 2014, 16:22	Problem with Residual plot raising up to hige numbers na do not converge	#1
fardinkhalili New Member Fardin Khalili Join Date: Feb 2014 Posts: 10 Rep Power: 12	Hi, I am working ono my thesis about CFD and Aeroacoustics of wind turbine with rotating blades. I got perfect results when I ran my simulation in steady-state condition for Curle and Proudman noise Models. For example I followed the first tutorial for aeroacoustics and the results were OK and I got convergence near to 10^-4 for residual plot. But, when I change the physics to unsteady, and change the motion to rotating, either using Reynolds-Averaged Navier-Stokes or DES or LES, everything goes wrong and the residual plot rises to hige numbers up to 10^7!!!! I change the motion to stationary or rotation but nothing got better. What is your opinion and how can I solve this problem.

March 14, 2014, 16:26	Errors while running the program	#2
fardinkhalili New Member Fardin Khalili Join Date: Feb 2014 Posts: 10 Rep Power: 12	pressure corrections limited on 9 cells in STATIC pressure corrections limited on 60083 cells in ROTATING Temperature corrections limited on 327 cells in STATIC Temperature corrections limited on 34752 cells in ROTATING Minimum Temperature limited to 100 on 439 cells in ROTATING Maximum Temperature limited to 5000 on 405 cells in ROTATING Maximum Temperature limited to 5000 on 33 faces on Wind Turbine Rotor.Default Temperature limited to maximum value 1270 times on model-part "ROTATING" Breaking out of time step iteration. Time solution is in an inconsistent state.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Cannot run the code properly: very large time step continuity error	crst15	OpenFOAM Running, Solving & CFD	9	December 14, 2014 19:17
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 07:20
rhoSimplecFoam Mach0.8 no pressure values	CFDnewbie147	OpenFOAM Running, Solving & CFD	16	November 23, 2013 06:58
pisoFoam with k-epsilon turb blows up - Some questions	Heroic	OpenFOAM Running, Solving & CFD	26	December 17, 2012 04:34
Upgraded from Karmic Koala 9.10 to Lucid Lynx10.04.3	bookie56	OpenFOAM Installation	8	August 13, 2011 05:03

March 19, 2014, 15:27		#4
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 25	How are you initializing the transient simulation? Is the flowfield static and your turbine suddenly is spinning at maximum RPM? That will be very difficult to converge, you need to ramp its speed or start your solution off from your steady-state solution.

March 25, 2014, 10:35		#5
ping Senior Member Ping Join Date: Mar 2009 Posts: 556 Rep Power: 20	you should certainly be starting your unsteady solve from a reasonably converged steady run - very easy to do. also check your timestep - i suggest one that gives you about 1 degree of rotation per timestep.