An strange situation:bigger maxCo leads to slower run

immortality · June 25, 2013, 14:47

Hi
I changed maxCo and observe that
maxCo=.1 then 1micro second in simulation time takes 32s in run time
maxCo=.2 run time=50s
maxCo=.15 run time=45s
how can such thing be possible?(it's possible because has occurred

but what may be the cause?)
my case is compressible and unsteady.
I thought maybe it was because of unsteadiness but it was 33s when I changed to .1s again.

immortality · June 25, 2013, 15:43

Hi again
I figured it out!
its because of tolerances.when timeStep is larger because solver should reach residuals to the specified tolerance MORE iterations should be done.then maybe time duration takes even more!
it depends on tolerances that specified too.
then:
"HIGHER timeStep is NOT necessarily equal to LOWER Time"( at least in unsteady simulations I know its so)
is there any other comment that I may have missed?

June 25, 2013, 14:47	An strange situation:bigger maxCo leads to slower run	#1
immortality Senior Member Ehsan Join Date: Oct 2012 Location: Iran Posts: 2,208 Rep Power: 26	Hi I changed maxCo and observe that maxCo=.1 then 1micro second in simulation time takes 32s in run time maxCo=.2 run time=50s maxCo=.15 run time=45s how can such thing be possible?(it's possible because has occurred but what may be the cause?) my case is compressible and unsteady. I thought maybe it was because of unsteadiness but it was 33s when I changed to .1s again. __________________ Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King. To Be or Not To Be,Thats the Question! The Only Stupid Question Is the One that Goes Unasked.

June 25, 2013, 15:43		#2
immortality Senior Member Ehsan Join Date: Oct 2012 Location: Iran Posts: 2,208 Rep Power: 26	Hi again I figured it out! its because of tolerances.when timeStep is larger because solver should reach residuals to the specified tolerance MORE iterations should be done.then maybe time duration takes even more! it depends on tolerances that specified too. then: "HIGHER timeStep is NOT necessarily equal to LOWER Time"( at least in unsteady simulations I know its so) is there any other comment that I may have missed? __________________ Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King. To Be or Not To Be,Thats the Question! The Only Stupid Question Is the One that Goes Unasked.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Strange Behavior of A Mesh Run at Two Dfferent Velocities Using Turbulence Model	zachnap	Main CFD Forum	0	May 11, 2011 15:22
Working directory via command line	Luiz	CFX	4	March 6, 2011 20:02