Hi.

I have two connected parallel channels. In one channel there is flow of mixture having mass fraction of 0.01. I have define the common edge of channels as porous jump (which is actually a membrane). I want that only one component of the mixture flows in the other channel from the first one(through the membrane)whereas the other is stopped.

Thank you in advance for giving any ideas

Hello!

I have simulated simmilar process. I have solved the problem in the following way:

1. I have build a model (mesh) consisting of two channels (feed & permeate) that are divided by a wall(membrane). The wall had infinitesimal thickness (i.e. it was not made of cells but of faces). The mesh was built in such a way that the elements were highly condensed near the membrane.

2. I have written some macros that model sink on the feed side and sources on the permeate side in the cells adjacent to the membrane ( in this manner, the transport via the membrane has been simulated). The macros were done in such a way that they loop over the feed or pemreate channel's cells checking if they are adjacent to the membrane zone. If yes, then they compare fluid parameters (for example partial pressures or osmotic pressures) on both sides of the membrane and return a value of the sink or source (mass flux).

Hope it applies to your problem.

Olek

Thank you Olek for your response. I have two questions related to your reply. 1. What was the zone type for membrane; porous jump or porous

2. How can I write macros to model source and sink.

Please help me, I am actuallly new user of fluent.

Muhammad Shakaib

Hi!

The zone type for the membrane is WALL - so the construction of the model is: two fluid zones (permeate and feed) divided by a wall zone.

As I have written before - the macros loop over the fluid zones checking if a cell is adjacent to the membrane - if yes , then they produce sink or source in this cell. If you are completely new in the UDF programming you have to consult the UDF manual and , unfortunately, spend some time on it :| .

BR

Olek

Hallo! I also try to simulate a membrane, but like a fluid porous zone. I'm very interested, how is defined the source term which you use. How many components have you in the system? Do you use your membrane for separation or distribution?

Hello!

Temporarly, I have no time to wirte a longer letter and i will appear in my office in 4 days. You can expect my response then.

BR

Olek

[Rubblevg]

Hi all,

I'm having the same problem. I try to model a semipermeable membrane by setting a negative mass source for the permeating phase at one side and a positive mass source on the other side of the membrane. Only I am not sure what expression to use for the mass source so that it doesn't make the flow unphysical. Do you have any idea?

[miketsap]

Originally Posted by Olek
;125475

Hello!

I have simulated simmilar process. I have solved the problem in the following way:

1. I have build a model (mesh) consisting of two channels (feed & permeate) that are divided by a wall(membrane). The wall had infinitesimal thickness (i.e. it was not made of cells but of faces). The mesh was built in such a way that the elements were highly condensed near the membrane.

2. I have written some macros that model sink on the feed side and sources on the permeate side in the cells adjacent to the membrane ( in this manner, the transport via the membrane has been simulated). The macros were done in such a way that they loop over the feed or pemreate channel's cells checking if they are adjacent to the membrane zone. If yes, then they compare fluid parameters (for example partial pressures or osmotic pressures) on both sides of the membrane and return a value of the sink or source (mass flux).

Hope it applies to your problem.

Olek

Hello, can you post your UDF code for your situation? So we can use it as a guidance. Thank you