[walakaka]

Hi guys,

I am currently trying to model a fully developed flow profile for a simple pipe flow.

I have been looking at the PitzDailyExptInlet tutorial. The script under constant/boundaryData/inlet/points is confusing me slightly.

To my understanding, this is what the tute has done:

• The points are defined for min_z. This is the flow profile for the inlet cross section

• The points are defined for max_z.

It is the points defined for max_z that confuses me because it lies nowhere on the geometry. The script constant/boundaryData/inlet/points is attached with this post.

Could someone assist me in this matter please. Any help is greatly appreciated.

Kind regards
Shaq

[hulli]

Hi may be that helps.

In order to get a fully developed flow, I use a empty box + cyclic boundries + fvoptions with a target flow speed, and let it run until the turbulences have developed. This usually takes some time for a flow over a flat plate, so I have a supler high wirteIntervall.

In the next step this is the model that serves as the model from which I map the flow field. I restart the latest time let it run and sample a plane that I want as inflow for my inlet of my "final model". Therfore I followed the intructions on that link, which explaines you how you sample during runtime. It is important that you use a sampling rate at the timestep interval that you want to use in your inlet model.

surface sampling during runtime

Then you have to adapt the files in your postprocessing folder and copy them into the constant folder of you desired case. I dont remember exacly but you have to modify the points to centers and also modify some headers.

I hope that helps a bit
H

[walakaka]

Thanks Hulli!

From my understanding:
1) You ran your model and obtained a fully developed turbulent flow
2) You restart the latest time, and sample the plane that you would be using as the inflow field for your "final model". [The sample plane would give us a fully developed flow field]
3) You adapt your sampled fully developed flowfield (in post processing) and copied it into your initial flow field conditions at the inlet (in the constant folder)

Could I ask you which functions did you use to obtain the flowfield?
cellSource
faceSource
fieldAverage
fieldMinMax
probes
readFields
sets
surfaceInterpolateFields
surfaces

I will be modelling a 3D cylindrical pipeflow with no slip boundary conditions. Will this be okay? I noticed this is very different to your model.

Do you also have any recommendations as to where can I find some documentation that could help me fit the developed velocity flow field from the post processing format into the constant folder of my desired case.

[hulli]

Hey man,

sorry for the late answer. Well, thats more or leaas what I did. I cant find the files right now. The probem is that you need to modify the files of step 2 and change them so they seacly have the same format as shown in the pitz.. case. I would suggest that you trie that mannualy forst and then do it dynamically by a matlab script or sth. more suffisticated... I hope hat helps. Best H