Persistent mass flux error

mczero57 · March 7, 2017, 19:27

Hello Cfders :

This is a somewhat open ended question - I ask it to have some good discussion on this topic.

I am running a steady 2d axisymmetric hypersonic case with a well validated flow solver. The case is the flow around a hypersonic body and there is some unsteadiness due to vortex shedding from the rear of the body. So, the residuals dont go down much. I have checked the y+ values on all walls and they are within the permissible range - I am using wall functions.

When I post-process, I see that the mass flux balance over the whole domain has a -2.8% error. I have tried finer grids, changing the bc to nonreflecting, running it out further, etc. all to no avail.

If this was your case, what else would you try ?

Thanks

Sam

mprinkey · March 8, 2017, 07:56

How are you post processing? Does the flow solver output face-based mass flux data or are you interpolating that from cell- or node-based data? This can make all of the difference. For cell-based FVM, I like to store the mass flux data on the faces as (somewhat) redundant information specifically for this reason. In principle, you can reproduce the mass fluxes from non-face data, but it will require duplicating, say, the gradient computation for velocity and density, limiting of that gradient, and interpolation to the boundary faces. Off the shelf post processing tools likely won't do exactly the same thing unless it was co-developed with the CFD code.

The second issue is that your problem is NOT steady if you have vortex shedding. The fact that "the residuals dont go down much" indicates to me that you don't have a solution yet.

March 7, 2017, 19:27	Persistent mass flux error	#1
mczero57 New Member sam Join Date: Jan 2011 Posts: 20 Rep Power: 15	Hello Cfders : This is a somewhat open ended question - I ask it to have some good discussion on this topic. I am running a steady 2d axisymmetric hypersonic case with a well validated flow solver. The case is the flow around a hypersonic body and there is some unsteadiness due to vortex shedding from the rear of the body. So, the residuals dont go down much. I have checked the y+ values on all walls and they are within the permissible range - I am using wall functions. When I post-process, I see that the mass flux balance over the whole domain has a -2.8% error. I have tried finer grids, changing the bc to nonreflecting, running it out further, etc. all to no avail. If this was your case, what else would you try ? Thanks Sam

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
UDF for DPM particle deposition modelingpa	haghshenasfard	Fluent UDF and Scheme Programming	10	September 15, 2019 03:03
Compile calcMassFlowC	aurore	OpenFOAM Programming & Development	13	March 23, 2018 08:43
error compiling modified applications	yvyan	OpenFOAM Programming & Development	21	March 1, 2016 05:53
[OpenFOAM] Native ParaView Reader Bugs	tj22	ParaView	270	January 4, 2016 12:39
How to install CGNS under windows xp?	lzgwhy	Main CFD Forum	1	January 11, 2011 19:44

March 8, 2017, 07:56		#2
mprinkey Senior Member Michael Prinkey Join Date: Mar 2009 Location: Pittsburgh PA Posts: 363 Rep Power: 25	How are you post processing? Does the flow solver output face-based mass flux data or are you interpolating that from cell- or node-based data? This can make all of the difference. For cell-based FVM, I like to store the mass flux data on the faces as (somewhat) redundant information specifically for this reason. In principle, you can reproduce the mass fluxes from non-face data, but it will require duplicating, say, the gradient computation for velocity and density, limiting of that gradient, and interpolation to the boundary faces. Off the shelf post processing tools likely won't do exactly the same thing unless it was co-developed with the CFD code. The second issue is that your problem is NOT steady if you have vortex shedding. The fact that "the residuals dont go down much" indicates to me that you don't have a solution yet.