I think this is just a general feature of plenums, I remember a professor telling me this once but I've not been able to recall the reason, although I've tried many times before.

Your case doesn't actually represent the real engine running though, because all four cylinders don't draw simultaneously. Therefore the flow through the plenum is actually less than what you've simulated. The intention of a plenum is to try and make the pressure in the plenum uniform so that each branch sees the same pressure and gets uniform flow. This is achieved by oversizing the plenum compared to the branches so the velocity in the plenum is "zero" or close to it. This reduces the pressure drop along the plenum and effectively allows no pressure difference between the inlet and the far end.

The plenum here isn't large enough compared to the branches to have very low velocity when all branches are flowing. In fact with the plenum is probably only 4 times the area of a branch, so the velocity will be the same, or even more in the plenum as it is in the branch.

This thread is old now so I'm sure you've finished your project, but it would have been good to see it with flow in only one branch at a time.

Really good job though, nicely done geometry. What sort of mesh did you use, did you put an inflated boundary on it to capture the boundary layer, what about the turbulence model? I would like to see your finished project results.

has someone this CAD file? or another one from the inlet manifold?