The force that FF joysticks apply to the stick is both static and dynamic. It is therefore important that there is not much "give" in the motor drive. Direct coupling is required. To make the motor torque apply sufficient force, some sort of gearing is also needed. I used the 8mm motor shaft as a pulley, and coupled the force to the yoke via thin 0.5mm steel wire. (used in old copy machines to move the scanner bed)
The elevator force needs to be pretty strong, to simulate the "out of trim" force, so I used a bigger motor for that axis. This motor can be mounted on the wooden frame (shown below at the left) I drilled a hole in the motor shaft, and the steel wire goes a couple of turns around the shaft, then through the holes originally meant for the springs, one side fastened with the yoke T piece, then around a pulley, and back to the motor. Tension the steel wire.
The aileron force does not need to be that strong. A smaller motor (from
old copy machine) was
used. It needs to be mounted on the same moving plate as the aileron potmeter is mounted on. Again I used thin steel wire. To create some
gearing, a small pulley was used on the motor shaft, and a big wheel was added to
the yoke shaft, clamped in between the big nuts.
Notes on motor power dissipation: Since the motors are operated in
non-rotating condition, there is no cooling effect of the rotor windings.
Since there is at times some 10 watts of power dissipation, extra (fan)
cooling needs to be applied, especially for the small aileron motor.