No it is pellet friction! In my model the pellet friction is one of the most dominant factors by far (apart from TP diameter). Probably because the transfer port flow equations assume flow based on an orifice plate model. 10N of (constant) piston seal friction is completely negated by an extra few mm of spring compression.
If the release friction for the pellet is too low (~1N) then the pellet will move before any significant pressure has built up in the chamber (or barrel).
For anything around 1 to 10N I can get fairly reaslistic behaviour, but it is effectively controlling the barrel expansion volume (flow is still limited by the TP port).
If the pellet friction is extremely high (~200N), then we have the equivalent of a closed piston and get damped simple harmonic motion of the piston.