I am powering a small sg90 servo motor (see datasheet) directly using a 12V battery and a simple circuit to step down the voltage to the appropriate one for the servo. I noticed the servo has some coil windings and I suppose I could model it as an RLC circuit.

My question is, how can I know if a flyback protection diode is needed in cases like this? The manufacturer does not seem to care about the inductance since there is no mention of it in the datasheet...

I've already make some brief tests such as a few 90 degree turns and everything went well so far so I guess it is not vital to put such a diode (at least with short usage periods) however I would like to get some enlightment on this topic.


No diode is needed, but you should put both some bulk capacitance and high frequency bypass capacitance across the power leads.

Keep in mind that this is a "hobby servo" motor. You are not driving the motor directly. You are sending signals to a controller in the same package as the motor, and that controller drives the motor. Therefore, you should try to provide reasonably clean power and clean signals, but your circuit isn't driving a motor.

  • 1
    \$\begingroup\$ Yup. The flyback diodes would already be on the servo driver circuit board inside the servo. \$\endgroup\$ – Vince Patron Jul 12 '16 at 22:26
  • \$\begingroup\$ @VincePatron To sum it all up as far as my understanding goes: indeed the flyback diode IS needed since the servo is an inductive load, but it is already mounted inside with the electronics controller inside the motor, so it is NOT needed outside the servo since it's already built-in. \$\endgroup\$ – mickkk Jul 13 '16 at 12:13
  • \$\begingroup\$ @mic: That's basically right, except that in this case there won't be a single flyback diode since the drive to the motor is more complicated than a single switch. The motor must be driven in both directions, so there is probably a H-bridge in there. This either has 4 explicit diodes, or is maybe using the body diodes of the FETs being used as switches. \$\endgroup\$ – Olin Lathrop Jul 13 '16 at 15:29
  • \$\begingroup\$ Yes, exactly what you guys said! Also, @mic, I would follow Olin's advice and put good decoupling at the servo power leads. Motors put all sorts of noise into the power rails. Also, a bit off topic, but make sure you design and test for when the servo is stalled (i.e. set to move but mechanically stuck). Very high current loads. Power supply should either limit current gracefully (and other circuits on same rail must handle must gracefully) or power supply can maintain full voltage and let servo handle the high mechanical load and supply current. \$\endgroup\$ – Vince Patron Jul 13 '16 at 16:59

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.