2
\$\begingroup\$

Context: I've built a robotic arm using servo motors, and have been having an issue with powering my servos. The servos are all rated for 3A/6V, and I've been powering it with a modified computer power supply that is capable of outputting 17A at 5.2V. For the most part, the arm works beautifully with this power supply, and the lower voltage doesn't seem to affect its performance, except in the special case I will note below.

Problem: When the arm is at full extension, and the servos are really exerting themselves, they sometimes make very quick movements, which causes them to draw more current than their 3A rating. In essence, there are current spikes, and these spikes are tripping a switch in my power supply, which shuts it off.

Idea for Solution: To correct this problem, I'd like to keep the same power supply, but introduce a capacitor bank that can provide a charge/current reservoir to the servos in case they need an extra boost. I'm not sure if this is a good idea, or if there is anything I need to consider before pursuing this option. If anyone has any advice or helpful resources, I would be very appreciative.

Thank you!

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Don't design a solution that feeds more amperage than your motors are rated for. They'll fatigue over time. The underlying problem sounds to be that the motors are not suited for the environment, not that you can't drive enough power. \$\endgroup\$ Commented Mar 3, 2016 at 19:44
  • \$\begingroup\$ Although I don't generally suggest migration to another site, I think that you would get a muct faster and more comprehensive answer over on Electrical Engineering, so I'm going to migrate it, even though it is on-topic on Robotics. They can always send it back if they don't like it. \$\endgroup\$ Commented Apr 2, 2016 at 22:11

1 Answer 1

Reset to default
2
\$\begingroup\$

Just some ideas for things to check first:

If you're using something like an ATX power supply, I'd suggest you look for its specs, or if not available the ATX power supply specification, and check the section about capacitive loads. (I believe too high a value can cause instability in the supply itself.)

I would also recommend (if you're not doing so already) using completely independent wiring for the 0V and Power wires for each servo - that is, say three sets of thick wire, each going from one servo to the supply. Instead of having one thick pair of wires which feeds all of the servos. (Just in case there are any problems with resistance in sections of the wiring under heavy current.)

Finally you might want to fit small decoupling capacitors to suppress noise spikes, fitted close to the servos themselves. I'm talking about any reasonably low ESR (equivalent serial resistance) ceramic capacitor or similar, these are commonly used for motor noise suppression. (Surprisingly electrolytic capacitors aren't so good for this, they often have higher resistance.)

Actually... I wonder if https://electronics.stackexchange.com/ would be a good place to search for answers too as there are bound to be people there used to motors and power supplies...

\$\endgroup\$