Thanks for the help so far.
I have done some testing with different models etc. and I've found out that even though the Reynolds numbers are low, turbulence plays a big part in why the small inlet peforms better. There are other factors at play too, but not as much as the turbulence.
What type of turbulence model should I be using with these simulations? K-epsilon isn't stable but eddy viscosity and low Re k-eps seem to be stable. Also, is advection scheme 1 suitable for such a simulation or should I try others?
Apologies for the delay here, only just noticed page 2.
You could try low re-ke, ADV1 should be OK. Maybe try 5 also, but I suspect it would yield the same results