I've had the hood up on a few servos and modified them for continuous rotation. I never measured the internal voltage across the pot but it's not surprising that it would be regulated to something different from the power supply.
Normally you would set the external pot (DAC in you case) so that you got no rotation when there is no PWM signal to the servo. Then the PWM signal just controls the speed and direction of the continuously rotating servo. Is that what happens in your setup?