I am using commercial software. CFX5.1 I want to simulate particle tracking in Unsteady flow

According to CFX-manual It is possible to calculate Particle tracking in Steady flow, but It is not to simulate particle tracking in unsteady flow.

In my opinion, If I use USER-fortan well that will be possible. CFX user, or anybody ....... help me please If you have example flie(.f, .fc, . geo) or comment...

help me help me... hel.me........... please

Hello Faithkim

I am not sure what you want to do with your model. If you want to use transient particle tracking to understand important fluid behavior in your system it can be difficult to extract the result (unless you want to make a visualisation movie - that would be great !!) Have you considered using the scalar transport/dispersion model available ?

However, there is several things that you can do:

1) you can implement it into the code via USER fortran as you suggest. I would think that you have to rewrite most of the code yourself.

2) You can write a separate program which takes the flowfield data from CFX and integrate the particle path. You can either write a few lines which dump all the relevant data in a specified file or use the output file ( *.fo)

The problem either way is that if you work with bodyfitted coordinates or unstructured grid - there is some geometry isues which has to be resolved.

The govering equations for particles can in most cases be reduced to something simple which is easy to solve (for example only include the forces from boyancy, drag and virtual mass ). The assumption is that there is no feedback from the particles on the fluid and this can in some cases be wrong (many electrical charged particles for example).

I cannot see any pratical problems for AEA that has prevented them in making this facility available - but maybe in the next version ?!

Regards

Michael

good luck with the project.

Hello Michael Thanks a lot for your comments I am interested in your second method. ------------------------------------------------------- 2) You can write a separate program which takes the flowfield data from CFX and integrate the particle path. You can either write a few lines which dump all the relevant data in a specified file or use the output file (.fo) ------------------------------------------------------- But I have many problems for using this method. I don't know file system and variable in CFX and ... well Have ever you tried to approach this problem? If possible May I get example files using simple governing equations for particle as you said or detail comments ?

Hello, faithkim!

[1] I have experiences in particle tracking for the st. st. only. by using CFX4.1 and 4.2 [2] There are full descriptions for the data structures of CFX dump file(solver reference book), for example header, variables, geometry and so on. [3] In 4.1, "rdump.f" source code may help your understanding for the dump file data structures [4] Recently i made a prototype fortran routine based on rdump.f for reading particle trajectory source in dump file ,and it add mass, momentum and heat source to the next fluid calculation via USRSRC.f. but i don't think it may help your "transient" problem. [5] Final and most important information; Lasse Rosendahl, professor of Aalbor univ., is the gru of separated particle integration with fluid result file. He developed "PCOMBUST" utility for the purpose. Any web searching site will find his site with key word, CFX and PCOMBUST.

I hope these informations are helpful. Regards, Hyun S. Lee