[dhruv]

Hello Foamers,

I want your inputs on a problem I am simulating with OpenFoam.

I am simulating flow through a offset fin array using OF. For that, I have created the geometry, and using Channel Flow solver. I am using LES simulation and tried running the case by two different models: (a) Smagorinsky (b) oneEqEddy model. Both have given me similar results. The main aim right now is to compare the values of friction factor obtained from the simulations to the available values in literature (like Webb and Joshi, or Manglik Bergles, etc). I have given a uniform velocity field at the inlet. My courant no. is 0.5 and my time step is quite small (1e-5).

Now, I run this simulation for 1 sec. This is little more than 200 flow through times. I have the following questions regarding this simulation

1) While calculating the f (friction factor), I need to take the pressure drop. Which pressure drop value should I take? The pressure gradient written out at every time step during the Channel Foam run? Is this pressure gradient or pressure gradient divided by the density of the fluid? If I use this pressure gradient, I get severe underprediction of f @ particular Re, when it is compared to the literature.

2) I observe that even after about 100 flow through times, there occurs relaminarisation of flow. Meaning, at some time intervals, the flow is turbulent, but later, it becomes sort of laminarised again. I did not expect this kind of behaviour. Can anyone comment on this?

3) My aim of the experiment is to study the entrance region effects in this channel with offset fin arrangement. Is it possible with Channel Foam and with a uniform BC on velocity @inlet? I have this doubt, since I am using Channel Foam, which uses a cyclic bc.

[sachin]

Hi there...
Not much that I can help...
1. I havent gone through the latest channelFoam solver ...but if this for incompressible fluids.. this might be giving out delta p/density terms...still you need to go through the code to know what is the exact one.

2. The flow when entering from larger zone into smaller zone, due to different hydraulic diameters this might be correct. But if you solve turbulent model even in this region why should it be a laminar flow...and whether will it be correct.
You can set offset fin region as laminar zone and avoid turbulent calculations in this region...

I am interested to know and see the model you have developed for offset fins...

Can you send it to me or upload it here.

Best luck...

[gu1]

Can I run channelFoam for any other geometry than channel?
I am simulating the internal flow in a pipe... I would like to know if it is possible, or it is necessary to modify the solver.