Residuals Behaviour

Rajkool · April 28, 2020, 16:41

Hello all,

I am not familiar with CFD. I have applied some basic settings, boundary conditions, and started calculating.

The curves look like converging.

The question is, the starting point of the curve is not from "Zero".

Is it common or did I made anything wrong?

Any type of suggestion is highly appreciated.

Thank you
Raj

vinerm · April 29, 2020, 11:09

Why do you expect residuals to start from 0? Ideally, it should be 0 when the case is converged. Practically, it is never 0 because the system is numerical and discrete.

Rajkool · April 30, 2020, 07:47

Can you give me some more information, if possible?

And is the my solution according to your reply?

Thank you
Raj

vinerm · April 30, 2020, 08:20

The residuals look alright, as expected. However, the case setup may or may not be complete. First of all, flow most likely would be turbulence and you will have to enable a turbulence model. Secondly, you should always enable gravity for a multiphase case with appropriate values for operating density and reference pressure location. And it will take a lot of iterations, usually, 1000s for convergence.

Rajkool · April 30, 2020, 11:32

No, it is the laminar model. Yeah, I have given 1000 iterations. I just posted it while it was simulating. I am just confused about that, can I consider this convergence is correct or not?

vinerm · April 30, 2020, 11:42

Residuals are not good for checking convergence. You should create a few monitors, such as, velocity at outlet, and/or pressure or volume fraction at outlet or in the domain. Once these monitors stop oscillating, then the solution should be considered converged.

April 29, 2020, 11:09	Residuals	#2
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 35	Why do you expect residuals to start from 0? Ideally, it should be 0 when the case is converged. Practically, it is never 0 because the system is numerical and discrete. Rajkool likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

April 30, 2020, 08:20	Residuals	#4
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 35	The residuals look alright, as expected. However, the case setup may or may not be complete. First of all, flow most likely would be turbulence and you will have to enable a turbulence model. Secondly, you should always enable gravity for a multiphase case with appropriate values for operating density and reference pressure location. And it will take a lot of iterations, usually, 1000s for convergence. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

April 30, 2020, 11:42	Convergence	#6
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 35	Residuals are not good for checking convergence. You should create a few monitors, such as, velocity at outlet, and/or pressure or volume fraction at outlet or in the domain. Once these monitors stop oscillating, then the solution should be considered converged. Rajkool likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Residuals show oscillating behaviour !	anon_l	STAR-CCM+	1	October 30, 2016 14:38
rhoSimpleFoam - strange behaviour of the residuals	Werne	OpenFOAM Running, Solving & CFD	2	December 16, 2014 04:27
Strange residuals behaviour	xxxx	Main CFD Forum	1	July 13, 2013 14:40
kOmegaSST omega residuals strange behaviour	zordiack	OpenFOAM Running, Solving & CFD	0	August 8, 2012 16:23
residuals behaviour on steady state simulations	c120613	CFX	5	October 20, 2011 17:20

April 30, 2020, 07:47		#3
Rajkool Member Raj Join Date: Jan 2020 Posts: 92 Rep Power: 6	Can you give me some more information, if possible? And is the my solution according to your reply? Thank you Raj

April 30, 2020, 11:32		#5
Rajkool Member Raj Join Date: Jan 2020 Posts: 92 Rep Power: 6	No, it is the laminar model. Yeah, I have given 1000 iterations. I just posted it while it was simulating. I am just confused about that, can I consider this convergence is correct or not?