The Futaba S3003 Servo Motor PWM Signal Testing

Today I assembled the circuit for the Pulse Width Modulation (PWM) to generate the signal to run the Futaba s3003 servo motor.

Here is the circuit for PWM,

and here is the Futaba S3003 servo just for information.
The circuit is asembled on the protoboard to shorten the testing period. A few parts changes had to be made due to unavailability of the accurate items. The 80.5kohm was replaced by 84kohm, the 36nF capacitor was replaced by 47.3 nF, and the two 47kohm & 560k ohm variable resistors were replaced by 500kohm and 1Mohm. Once the circuit is completely assembled, I add on one Single Pole Double Throw (SPDT) toggle switch as a signal trigger point. Since last time I bought 4 pieces so I tried all. Out of four, only one can fully turn 180 degree when the PWM signal is and return to actual location once the PWM signal is off. The other three, once received the PWM signal it turned a bit,but never return to actual position once the signal is off. For the time being I'm not sure whether the problem is due to faulty servos, or PWM signal incorrect. The best hypothesis that I can say now is there could be a problem with the servos as when I manually turn the three servos, they do turn 180 degrees once the PWM signal is applied. The only thing is that they cannot return to actual position once the signal is OFF.

This is the completed assembled circuit on the protoboard