Backward Euler

smagmon · April 14, 2009, 12:02

Hi All, I am observing some disturbances in my simulation, and I would like to test the discretization scheme. The problem is that I don't know how to change the tolerances of the method or some other parameter. Can I only change between First order or Second Order BKW Euler? Or can I change the tolerances for each case?
I am trying to observe the wave propagation and its reflection using as input a displacement as a step function. when I apply such input as mesh displacement, some disturbances are generated and those run fast then the sound velocity.
I am using air as ideal gas in the domain, and as input I am using a step function with 10nm of amplitude.
Can anybody help me? Tips are also welcome...

Thanks

ghorrocks · April 14, 2009, 20:40

Hi,

More discretisation options are available in CFX than are shown on the GUI. You will need to use the command file to access them. To see everything available look in the RULES file, in (CFX ROOT)/etc, under the advection scheme heading.

Glenn Horrocks

smagmon · April 16, 2009, 03:03

Hi Mr. Horrock,
I looked in that file but I couldn't see information about tolerances. Is it dependent of some other parameter? My doubt is about the tolerance settings for the transient scheme.

Thanks for helping.

ghorrocks · April 16, 2009, 08:10

Hi,

There is no "tolerance" specifically for the temporal discretisation. The numerical tolerance of relevance is the equation residuals and for some cases the imbalances. You should always test the effects of different convergence levels to ensure you have adequate convergence anyway.

Glenn Horrocks

April 14, 2009, 12:02	Backward Euler	#1
smagmon New Member Join Date: Mar 2009 Posts: 13 Rep Power: 17	Hi All, I am observing some disturbances in my simulation, and I would like to test the discretization scheme. The problem is that I don't know how to change the tolerances of the method or some other parameter. Can I only change between First order or Second Order BKW Euler? Or can I change the tolerances for each case? I am trying to observe the wave propagation and its reflection using as input a displacement as a step function. when I apply such input as mesh displacement, some disturbances are generated and those run fast then the sound velocity. I am using air as ideal gas in the domain, and as input I am using a step function with 10nm of amplitude. Can anybody help me? Tips are also welcome... Thanks

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Blood Flow - Euler Euler	Ilias	FLUENT	0	January 12, 2009 10:16
euler model	CN	FLUENT	0	July 4, 2005 16:35
Forward and Backward Euler.	Question Man.	Main CFD Forum	2	March 29, 2004 13:19
Boundary Layer created by Euler Solvers	Jim	Main CFD Forum	31	November 17, 2001 23:18
Euler equations?	Jan Ramboer	Main CFD Forum	2	August 19, 1999 01:58

April 14, 2009, 20:40		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Hi, More discretisation options are available in CFX than are shown on the GUI. You will need to use the command file to access them. To see everything available look in the RULES file, in (CFX ROOT)/etc, under the advection scheme heading. Glenn Horrocks

April 16, 2009, 03:03		#3
smagmon New Member Join Date: Mar 2009 Posts: 13 Rep Power: 17	Hi Mr. Horrock, I looked in that file but I couldn't see information about tolerances. Is it dependent of some other parameter? My doubt is about the tolerance settings for the transient scheme. Thanks for helping.

April 16, 2009, 08:10		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Hi, There is no "tolerance" specifically for the temporal discretisation. The numerical tolerance of relevance is the equation residuals and for some cases the imbalances. You should always test the effects of different convergence levels to ensure you have adequate convergence anyway. Glenn Horrocks