[RedAdmiral]

I have some questions concerning turbulence models and simulating the behaviour of solid particles in a fluid stream.


My fluid stream has a free stream velocity of 1 m/s, in a pipe of DN26 and a few areas where my velocity drops down to about 0,2 m/s. The speed drops are not gradually. There are also zones where recirculation is possible.
My particles have a Ø with a range from 10 to 200 µm.

I'm not sure what turbulence model is usefull to simulating the particle behaviour as accurate as possible. As a start I used a fixed turbulence velocity model. Since it's usefull for preliminary results.

The other available models are:

• Mixing length
• K-epsilon
• K-omega
• SST

But the measured behaviour in a real situation does not fit with the results I get from simulating.



Which model would you suggest?
Is the fixed turbulence viscocity not usefull in this situation, not even for preliminary results?

Also I've calculated the wall distance to my viscous layer:
0,01 mm (pipe is DN26).

Now I wonder if it's necessary to simulate the friction with near wall meshing or a wall function in this kind of model.
I've used slip wall as a start. Because my fluid is water and the viscocity is quite low.

Is the behaviour of the viscous - and buffer layer on the turbulent layer to important to ignore?



I'm quite new to modeling turbulent flow but eager to learn.
If someone has other good suggestions or tips.
I'll be glad to read them