Simulating pressure drop of water through volume of solid pellets (filter)

[Joel79]

I have an interesting simulation challenge that I am working on. I am trying to determine the pressure drop that a flow of ~50 gpm of water through a volume of cylindrical pellets that are ~.25" in diameter. The pellets are not perfectly spherical and the surface is quite rough. The volume of pellets would be around 6" in diameter and 3 feet long. I have determined the volumetric ratio of pellets/fluid through a physical volumetric test (pack with pellets / measure water volume).

I'm trying to model the fluid flow in ANSYS but I'm not sure the my starter 3D model is going to steer me down a successful path. I am basically laying sheets of distributed balls on top of each other with some overlap to emulate a packed cylinder. The attached picture illustrated what I have in mind (does not currently reflect proper diameter or channel dimensions). The surfaces of this model are quite complex and the sims take a while to run (4 hours or so).

I've been searching for white papers on the subject but have not yet identified a good reference. Wondering if anyone has a clever approach that I could consider.

Thanks in advance!

 

[1503-44]

Well first of all you will not get a uniform body-centered-cubic distribution using "not perfectly spherical, rough surfaced spheres". Even if you do, the layers will shift to fill in the gaps in between the spheres in the layer below and above. A lot more spheres will fit in your box, even if they are perfect. I'd try some emperhical gravel filter models before messing with CFD, especially if you have non-spherical, very rough shapes. Googling "hydraulics of gravel filters" brings up quite a few leads.

“What I told you was true ... from a certain point of view.” - Obi-Wan Kenobi, "Return of the Jedi"

 

[Compositepro]

Packed beds are a common occurrence in the chemical industry. I know there are empirical equations to calculate what you want. I suggest you look in Perry's Handbook or similar.

 

[EdStainless]

There may be some empirical solutions available from water treatment also since softeners and ion exchange are basically beds of spheres.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[IRstuff]

Perhaps these are of some value

https://www.cpp.edu/~tknguyen/che435/Notes/P4-fluidized.pdf

http://thermopedia.com/content/46/

https://en.wikipedia.org/wiki/Fluidized_bed

http://ww2.che.ufl.edu/unit-ops-lab/experiments/FB/FB-manual.pdf

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Latexman]

Look up the Ergun equation. I’ve used it many times.

Good Luck,
Latexman
Pats' Pub's Proprietor

 

[Joel79]

Thanks for the great advice. I found some empirical equations in Perry's Handbook related to "Beds of Solids". The equations specified are basically the Ergun equation. The numbers that I'm coming up with are quite practical.

Side note - I went ahead and kicked off the ANSYS sim to help validate my equations. Results are not too far off (~20%) - so I'm certainly off to a good start.

Thanks again for the solid feedback!