[fschneider]

Hello,
I'm using InterFoam to simulate a bubble rise. Is there a good way of tracking the CG or velocity of a bubble in OpenFoam 9?

[dplthuy]

Hi,

I'm not sure about the tracking of the bubble through the domain, but you could have a look at the work of Spitzenberger (https://www.sciencedirect.com/scienc...64X?via%3Dihub). They 'fixed' the existing functionObject extractEulerianParticles so that it gives information on droplets/bubbles passing a faceZone in your domain. Position, volume and velocity are part of this information.

[Swirl]

Hi,


I wanted to ask, if you were able to implement the extractEulerianParticles FO? Because for me it reports completely incorrect data by making the flux through every face on the patch into a separate particle instead of recognizing the coherent structures.


I would greatly appreciate any feedback.

[piu58]

I used this case for comparing against theoretical results. I found the outline of the bubble from the alpha.water field. From this outline I calculated the centre of mass.

[Swirl]

I am not working on the bubble-case myself. I need to collect the droplet size distribution from a jet entering a crossflow.

Unfortunately, the extractEulerianParticles FO ist not working as advertised for me, so I hoped to find someone who is using it successfully. Identifying the drops individually "by hand" is impossible in my case.

[dplthuy]

Hi Swirl,

I recognize your problems with the ExtractEulerianParticles FO. As I already mentioned in my previous post in this thread, there is an updated version of this FO available through the paper by Spitzenberger (link in my previous post). I think this is written for OpenFOAM v1906.
I was able to use this in v2106 without many issues. Only thing I had to do was use the regionSplit2D functionality from v1906 instead of v2106 (located under $FOAM_SRC/meshTools/regionSplit2D).

Good luck!

[Swirl]

Hi,


thank you so much for your reply! Do I understand you correctly, that you made changes to the extractEulerianParticlesMod code to run it on a version newer than v1906? Because I tested extractEulerianParticlesMod with the same result as with the standard version.


Sorry if this is a dumb question, I am not very familiar with the OpenFOAM source code.

[dplthuy]

Hi,

I did not make changes to the extractEulerianParticlesMod code itself. However, in extractEulerianParticlesMod.C, there is an include statement for regionSplit2D.H. This includes the regionSplit2D functionality from your compiled Foam installation into the extractEulerianParticlesMod. I found that this regionSplit2D functionality is different in v2106 than it is in v1906, and causes bad performance of extractEulerianParticlesMod when used with v2106. Therefore, I compile the extractEulerianParticlesMod using the regionSplit2D functionality from v1906. And then it works perfectly when running with v2106!

Disclaimer: it has been a while since I did all of this, so I'm not sure if there were any other compilation issues when using the v1906 regionSplit2D in v2106 directly. But since I don't remember, they were probably minor!

Hope this is clear.

Kind regards,
Dennis

[Swirl]

Ok, never mind, I solved it


For anyone else trying to get extractEulerianParticlesMod to run and is just as bad with coding as I am:

1. Download the source code of the regionSplit2D function from Openfoam v1906 and recompile it (ideally under a new name) on your system.

2. Modify the code of extractEulerianParticlesMod to use the regionSplit2D function which you just compiled by swaping the function and header names to whatever you named it in Step 1. Also remember to add the libraries to your Make/files.

This worked perfectly for me and produces correct results.

Again thank you so much for the hint @dplthuy ! There is no way I would have figured this out on my own.


EDIT: didn't see you reply. But you are exactly right. Simply switching to the old regionSplit2D works perfectly

[zhoufan]

hi Swirl,

I have also encountered the same problem. I would like to ask for your advice on how to compile the regionSplit2D in v1906 into v2106. I followed your method but encountered many errors. Can you provide me with the code for the part you modified?

Thanks.

[Swirl]

Hey zhoufan,

please excuse the late reply. In case you haven't solved the problem yet, I have attached the modified source code. Please be aware, that the base function object is the modified version of extractEulerianParticles as published in
https://www.sciencedirect.com/scienc...64X?via%3Dihub I have compiled it in OF v2112, so you may encounter additional issues depending on the version you are working with.

Hope this helps.