A Problem with my OpenFoam animation

saeedi · August 4, 2015, 12:23

Hi there,

I am new to OpenFoam and just started working with that.

I have conducted a test on a 2D cylinder at Re=100 (laminar), just to make sure that the basics of my openfoam simulation are OK.

The simulation went well and I also made an animation for the flow evolution which can be downloaded from here:

https://www.dropbox.com/s/zqnvxg7hbt...-SLOW.avi?dl=0

There is some thing that puzzles me a bit. Although it is quite obvious from the contour structures and topology that the flow is from left to right, in the animation it seems that those high U contours are begin convected from the outflow boundary to the cylinder. It is not a visual mistake as I refined the time step and got the same result.

I suspect there might be some more subtleties related to OpenFoam. I checked whatever I knew but could not find the problem. If I ran the animation in the reverse order then it looks quite nice!

I'd appreciate your help.

FMDenaro · August 4, 2015, 13:55

Quote:

Originally Posted by saeedi

Hi there,

I am new to OpenFoam and just started working with that.

I have conducted a test on a 2D cylinder at Re=100 (laminar), just to make sure that the basics of my openfoam simulation are OK.

The simulation went well and I also made an animation for the flow evolution which can be downloaded from here:

https://www.dropbox.com/s/zqnvxg7hbt...-SLOW.avi?dl=0

There is some thing that puzzles me a bit. Although it is quite obvious from the contour structures and topology that the flow is from left to right, in the animation it seems that those high U contours are begin convected from the outflow boundary to the cylinder. It is not a visual mistake as I refined the time step and got the same result.

I suspect there might be some more subtleties related to OpenFoam. I checked whatever I knew but could not find the problem. If I ran the simulation in the reverse order then it looks quite nice!

I'd appreciate your help.

Are you sure that you have the correct order of the zones in Tecplot? It seems simply reversed ...

saeedi · August 4, 2015, 14:04

Yes,

I checked it.

It seems that the timestrands are correct and in the right order.

FMDenaro · August 4, 2015, 14:07

try to do the animation with a vector plot, it could be more useful

saeedi · August 4, 2015, 14:10

I did.

Here is the link:

https://www.dropbox.com/s/e3mvvvxnm0...ector.avi?dl=0

FMDenaro · August 4, 2015, 14:15

Quote:

Originally Posted by saeedi

I did.

Here is the link:

https://www.dropbox.com/s/e3mvvvxnm0...ector.avi?dl=0

it still seems the order of the zones is reversed....

saeedi · August 4, 2015, 14:20

But if you look at the flow development, from time=0 to the first couple of snapshots, it shows that the flow is developing correctly and I think that is an indication that the zone order is right. But suddenly we see that reverse behaviour around time=1.

Also, when I do it in tecplot frame by frame, the time associated with each snapshot and their orders are OK.

It is really strange.

saeedi · August 5, 2015, 13:15

No more idea?

FMDenaro · August 5, 2015, 13:22

Quote:

Originally Posted by saeedi

Hi there,

I am new to OpenFoam and just started working with that.

I have conducted a test on a 2D cylinder at Re=100 (laminar), just to make sure that the basics of my openfoam simulation are OK.

The simulation went well and I also made an animation for the flow evolution which can be downloaded from here:

https://www.dropbox.com/s/zqnvxg7hbt...-SLOW.avi?dl=0

There is some thing that puzzles me a bit. Although it is quite obvious from the contour structures and topology that the flow is from left to right, in the animation it seems that those high U contours are begin convected from the outflow boundary to the cylinder. It is not a visual mistake as I refined the time step and got the same result.

I suspect there might be some more subtleties related to OpenFoam. I checked whatever I knew but could not find the problem. If I ran the animation in the reverse order then it looks quite nice!

I'd appreciate your help.

I suspect that we are victim of some optical joke due to the movie...
I tried to see each frame of your movie and it seems to work fine

FMDenaro · August 5, 2015, 13:24

you know, depending on the sampling frequency we can have a sort of stroboscopic effect....

saeedi · August 5, 2015, 13:27

That came to my mind as well.

But when I refined the time-step and consequently the writing interval by 4 times, I still observed the same thing.

FMDenaro · August 5, 2015, 13:44

Quote:

Originally Posted by saeedi

That came to my mind as well.

But when I refined the time-step and consequently the writing interval by 4 times, I still observed the same thing.

did you see frame-by-frame?

saeedi · August 5, 2015, 13:46

Yes I did.

Same thing.

That's Why I always prefer to use my own code. It is way easier to track.

August 4, 2015, 12:23	A Problem with my OpenFoam animation	#1
saeedi Member Mike Join Date: Apr 2011 Location: Canada Posts: 83 Rep Power: 15	Hi there, I am new to OpenFoam and just started working with that. I have conducted a test on a 2D cylinder at Re=100 (laminar), just to make sure that the basics of my openfoam simulation are OK. The simulation went well and I also made an animation for the flow evolution which can be downloaded from here: https://www.dropbox.com/s/zqnvxg7hbt...-SLOW.avi?dl=0 There is some thing that puzzles me a bit. Although it is quite obvious from the contour structures and topology that the flow is from left to right, in the animation it seems that those high U contours are begin convected from the outflow boundary to the cylinder. It is not a visual mistake as I refined the time step and got the same result. I suspect there might be some more subtleties related to OpenFoam. I checked whatever I knew but could not find the problem. If I ran the animation in the reverse order then it looks quite nice! I'd appreciate your help. Last edited by saeedi; August 4, 2015 at 14:06.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Can OpenFoam solve this problem?	salazardetroya	OpenFOAM Running, Solving & CFD	1	July 29, 2015 22:34
Problem when installing Openfoam 2.0.x with mac	giovaniharyadi	OpenFOAM Installation	0	March 24, 2012 18:45
New OpenFOAM Forum Structure	jola	OpenFOAM	2	October 19, 2011 06:55
Cross-compiling OpenFOAM 1.7.0 on Linux for Windows 32 and 64bits with Mingw-w64	wyldckat	OpenFOAM Announcements from Other Sources	3	September 8, 2010 06:25
Problem installing OpenFOAM 1.5 installation on RHEL 4.	vwsj84	OpenFOAM Installation	4	April 23, 2009 04:48

August 4, 2015, 14:04		#3
saeedi Member Mike Join Date: Apr 2011 Location: Canada Posts: 83 Rep Power: 15	Yes, I checked it. It seems that the timestrands are correct and in the right order.

August 4, 2015, 14:07		#4
FMDenaro Senior Member Filippo Maria Denaro Join Date: Jul 2010 Posts: 6,769 Rep Power: 71	try to do the animation with a vector plot, it could be more useful

August 4, 2015, 14:10		#5
saeedi Member Mike Join Date: Apr 2011 Location: Canada Posts: 83 Rep Power: 15	I did. Here is the link: https://www.dropbox.com/s/e3mvvvxnm0...ector.avi?dl=0

August 4, 2015, 14:20		#7
saeedi Member Mike Join Date: Apr 2011 Location: Canada Posts: 83 Rep Power: 15	But if you look at the flow development, from time=0 to the first couple of snapshots, it shows that the flow is developing correctly and I think that is an indication that the zone order is right. But suddenly we see that reverse behaviour around time=1. Also, when I do it in tecplot frame by frame, the time associated with each snapshot and their orders are OK. It is really strange.

August 5, 2015, 13:15		#8
saeedi Member Mike Join Date: Apr 2011 Location: Canada Posts: 83 Rep Power: 15	No more idea?

August 5, 2015, 13:24		#10
FMDenaro Senior Member Filippo Maria Denaro Join Date: Jul 2010 Posts: 6,769 Rep Power: 71	you know, depending on the sampling frequency we can have a sort of stroboscopic effect....

August 5, 2015, 13:27		#11
saeedi Member Mike Join Date: Apr 2011 Location: Canada Posts: 83 Rep Power: 15	That came to my mind as well. But when I refined the time-step and consequently the writing interval by 4 times, I still observed the same thing.

August 5, 2015, 13:46		#13
saeedi Member Mike Join Date: Apr 2011 Location: Canada Posts: 83 Rep Power: 15	Yes I did. Same thing. That's Why I always prefer to use my own code. It is way easier to track.