I am trying to set up a servo controller that will automatically make the servo switch direction when it nears the end of its travel range.

I designed a solution that uses a quad comparator and a peak detector to produce a useful control signal. The circuit has one signal input and two outputs that control a signal to an H-bridge driver. As the voltage across the potentiometer approaches 5V, the first output is high and the second is low. As the voltage approaches 0V the first output is low and the second output is high. The signal flips when the voltage gets close to 5V and again near 0V. It works, but there are some drawbacks.

Is there a standard analog approach or IC that performs the same operation while using a potentiometer for the signal source?

  • \$\begingroup\$ Are you both 1) sensing position with the pot, and 2) letting the servo amplifier work normally with the pot, too? Is this possible? Or are you replacing the servo amplifier with your own one? \$\endgroup\$
    – tomnexus
    Mar 31 '15 at 1:03
  • \$\begingroup\$ I am replacing the servo amplifier with my own \$\endgroup\$ Mar 31 '15 at 4:12

The standard answer is to use a microcontroller to read the pot and make the decisions.

You don't say what the drawbacks of the current circuit are. A comparator with some positive feedback or hysteresis, and an inverter, will do the trick, but likely it sounds awful when reversing, and takes a huge spike of current.

The micro will be able to drive the H-bridge with some sort of PWM, and thus gently slow the motor down and reverse it again. You will define a voltage profile in software, or even make your own PD control loop and have the servo follow an internally generated position profile.

With some programming experience, the learning curve with an Arduino will be short. And if you need to produce several of them, you could use a small 8-pin microcontroller with internal clock, not very expensive.

  • \$\begingroup\$ Thanks. A comparator with hysteresis is the exact sort of thing I was looking for. My largest concern with my circuit is the issue of minimizing droop rate in the peak detector while maintaining fast acquisition. \$\endgroup\$ Mar 31 '15 at 19:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.