Thanks for this, you are right of course that the M-Geschoss aircraft shells reduce the space allowed for propellant.
Another factor affecting gun performance is recoil: the MG-FF was the first Luftwaffe cannon to use M-Geschoss shells (from mid-1940) and it was found that the propellant + light shell combination did not generate enough recoil to operate the automatic gun mechanism, even when stuffing the case with as much propellant as possible. It was necessary to reduce the weight of the bolt and the strength of the recoil spring (creating the MG-FFM) which then meant that the gun could not use full-power (not M-geschoss) ammunition; when tracer shells were wanted, the heavy standard HE-T shells were used but reduced in weight by using light-alloy rather than brass fuzes, and downloading the propellant so that the MV was only 585 m/s.
Yes, as the MG-FF is a blowback-operated weapon, as per most such weapons it works on a momentum exchange.
As such you have the momentum of the projectile and the charge, and the momentum of the bolt assembly which will balance. The momentums of all of these are simply their masses times their velocities. For the projectile, it's its mass times its muzzle velocity. For the charge, it's its mass times its gas velocity (related to the type of propellant, its burning temperature and its pressure). For the bolt, it's simply its mass times its velocity.
If the original projectile was fired at 600 m/s and weighed 134 g, then the projectile’s momentum alone would be 8.2 N·s. For the 92 g M-Geschoß at 695 m/s its momentum is 6.5 N·s. For the 117 g projectiles at 585 m/s its momentum is 7.0 N·s. I’ve not added the propellant momentums, as the maths is a bit more complex and I don’t feel like doing it at the moment.
Anyway, it’s safe to say that when both the projectile and charge impulses are combined, the later loadings were made to produce similar impulses. So yes, the use of earlier higher impulse rounds would blow the bolt back far too early and far too quickly damaging the gun.
If you really want the bolt velocities calculated, that program should be able to estimate them by putting the mass of the bolt into the equation instead of the whole gun. The program normally calculates recoil force at the shoulder, which is the big gun equivalent, of trunnion pull.