Added a support bearing to the primary gear to take some load off the motor. The gear is mounted at the end of 7/8" shaft extension, which meant the tangential and longitudinal forces from the primary gear on the motor shaft were all being carried by the internal bearings in the motor, putting a large stress on the shaft near the point where it exits the motor. The images below show the primary helical gear which is positioned about 3 inches beyond the end of the motor shaft on a shaft extension, coupled through the shaft adapter the gear is attached to. I believe this is what caused the shaft to shear off just at the motor housing, many months ago.
Original Design - No Support
New Configuration with Support Bearing
Minor Configuration DCF Update
Noticed a slight hesitation around 3,000 RPM. Changed D-axis current controller proportional gain (0x4641_21) to 0 (uses Q axis value for Kp) and D-axis current controller integral gain (0x4641_22) to 0 (also use Q axis value for Ki). This seems to resolve the hesitation, which I suspected was simply the D-Axis control loop overshooting the target field-weakening FW Id value.