How is the velocity field at the boundary computed from a specified pressure outlet boundary condition?

Since I am not getting an answer to this question let me rephrase the question somewhat in the hope I will get an answer:

I am computing a cylindrical jet flow. Specyfing an outflow boundary condition at the boundary (which sets du/dx=0 at the boundary) gives inaccurate results. However specifying the pressure leads to better results in comparison with measurements. Now I am trying to find the reason why this is so and in connection with this I would like to know how the velocity is determined at a specified pressure outlet boundary. Is the SIMPLE scheme used for this (e.g. the Poisson eq.) or are the velocities extrapolated?

Any answer would be very welcome. Thank you.

Outflow boundary condition is used when you don't have any information of outlet of your system. That means an approximation is considered in your calculation. Usually, at outflow boundary, velocity magnituds are determined by continuity of your system(mass_in=mass_out) and flow directions are determined by the direction of adjacent grid points. These are kind of an indirect method of calculating outlet velocity. In case of pressure boundary conditon, velocity components at outlet are calculated directly.(Frankly, I don't know the exact way of calculation at pressure b.c.) So, there is a difference between outflow b.c. and pressure b.c..

As I know, in case of outflow b.c., ther are some restrictions to use it. (e.g. existence of back flow at outle, compressible flow etc.)

[samurai_01]

In outflow BC, the velocity components have their noumall gradients as 0 ie a neumann BC. In pressure outlet BC the pressure is computed through interpolation from internal solution of velocity.