I suggested that regeneration may be a cause as this is a well known factor to be dealt with in AC and DC drive/motor systems.
Regeneration resistors are switched onto the DC bus to absorb this motor momentum energy, provide braking for diesel locomotives, slow the decent of elevators and you will find such circuitry on almost every axis of multi-axis machine tool systems. When the regen resistor circuit fails the resulting condition is 'bus over-voltage'.
If your motor outputs a voltage when power is removed and the motor is still turning regeneration is occurring.
If your motor doesn't do that,,, well, you have designed a very unique motor. From your motor drawing I can say the motor is quite normal in that it will also act as a generator.
Considering blocking diodes, yes, depending on how the diode is wired any regeneration current should be blocked from charging the capacitor. Even when you consider diode junction capacitance it is very unlikely regeneration would charge the capacitor because unusually high frequencies would be the only thing passed by such capacitance.
So, I'll now assume a blocking diode is employed to block regen current. And that throws regeneration out the window for charging the cap.
I'll just sit back and keep quiet until I see enough detail to answer your questions intelligently.
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein
"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg