[adi959719]

Hello everyone,

I have set-up 2 user defined scalar to be solve in fluent. One of the UDS uses the default unsteady function and mass flow rate as the flux function. In other words, I have implemented a scalar transport equation with the scalar having the units of #/kg. (Note that in my case the scalar is space charge, thus its unit is Coulombs/kg (C/kg))

I want one of the wall in my transient simulation to act like a sink, where it depletes the UDS (space charge) at a specified rate. My current approach is by using Fluent's GUI to set the boundary condition, at the wall, as "specify flux". I assume the unit would be in C/(m^2.s). For example, if I set it to +1, I assume that it means the wall will depletes Coulombs of space charge per m^2 of the wall per second.

However, this does not seem to work and my solution keep diverging even at a timestep and under-relaxation parameter smaller than I would like it to be ideally. Note that the simulation does not diverge when i specify the flux boundary condition to be zero.

I have look at this post (Perfect sink for an UDS scalar concentration), and seems that there is another way to specify the sink wall. But I think that post only applies if the wall depletes all scalar that come in contact with it. In other words, it is the same as specifying a zero value BC for the scalar.

So my questions is:
1. How should I implement a sink BC that depletes the UDS at a specified rate? Is my current approach correct?
2. If my approach is correct, then is there a work around to avoid the diverging solution without decreasing timestep and the under-relaxation parameter?