BiCGStab

Astarta · May 4, 2012, 06:11

Hi!
I try to solve with BiCGStab solver, but preconditioners DIC and DILU doesn`t work.
Have anyone worked with it? which preconditioner should be used?
I use 1.6-ext version

Best regards,
Astarta

Linse · May 5, 2012, 08:17

Hi Astarta!

Unfortunately I do not have any experience with the -ext-branch. But in any case it would be helpful if you could be more specific on the symptom of the problem:
Does it say "oops, I did not find the solver"? Or is it "bad luck, this solver is not valid"? Or does the simulation start but in the end it blows up?

What you could try (no guarantee for that, as I am using that in OF-2.0) is to set as preconditioner "none" or "diagonal" (don't know at the moment if they are for symmetric or asymmetric matrices). Either it works for starting the simulation. Or it does not work and it should show you the preconditioners that are acceptable.
In case the simulation starts but blows up at some point, put in some gibberish for the preconditioner. Then it should show you the acceptable preconditioners...

Hope that helps a bit, but please give us more information on the symptoms of the problem! And if and how it gets solved by chance.

May 4, 2012, 06:11	BiCGStab	#1
Astarta New Member Sergey Zinkevich Join Date: May 2011 Location: Russia, Moscow Posts: 17 Rep Power: 15	Hi! I try to solve with BiCGStab solver, but preconditioners DIC and DILU doesn`t work. Have anyone worked with it? which preconditioner should be used? I use 1.6-ext version Best regards, Astarta

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
calculation stops and BICGStab did not converge	vandermessi	STAR-CCM+	1	March 6, 2012 04:19
TurbDyMFoam-Ggi	JulytoNovember	OpenFOAM Running, Solving & CFD	2	January 31, 2011 02:36
GGI in OpenFOAM	hjasak	OpenFOAM Running, Solving & CFD	59	April 30, 2010 09:30
are the cg or bicgstab method suitable for dense	ztdep	Main CFD Forum	8	September 6, 2008 10:41
[r] Need some help with bicgstab.	sandy	Main CFD Forum	1	March 3, 2008 12:10

May 5, 2012, 08:17		#2
Linse Senior Member Bernhard Linseisen Join Date: May 2010 Location: Heilbronn Posts: 183 Blog Entries: 1 Rep Power: 16	Hi Astarta! Unfortunately I do not have any experience with the -ext-branch. But in any case it would be helpful if you could be more specific on the symptom of the problem: Does it say "oops, I did not find the solver"? Or is it "bad luck, this solver is not valid"? Or does the simulation start but in the end it blows up? What you could try (no guarantee for that, as I am using that in OF-2.0) is to set as preconditioner "none" or "diagonal" (don't know at the moment if they are for symmetric or asymmetric matrices). Either it works for starting the simulation. Or it does not work and it should show you the preconditioners that are acceptable. In case the simulation starts but blows up at some point, put in some gibberish for the preconditioner. Then it should show you the acceptable preconditioners... Hope that helps a bit, but please give us more information on the symptoms of the problem! And if and how it gets solved by chance.