0
\$\begingroup\$

We are building a robotic arm that requires the use of 6 servos. I'm planning to use one external power source to power all the servos. The problem is that each servo has a different current draw rating. Some servos draw up to 10A when stalled (which should not happen). In the same time, we are using small servos, which draw 2A of current if stalled. However, they all have the same voltage rating (~6V).

So, I was wondering how they can all be powered using only one power source?

Here are more details:

  • Total stall current: ~30A
  • Total required power: ~170W
\$\endgroup\$
6
  • 2
    \$\begingroup\$ This isn't really answerable until you decide what you want to have happen on stall - do you want to limit current to that motor, or limit the overall current, or declare a fault and disarm the entire system, or ? Beware that large heavy assemblies of RC hobby servos may not turn out to perform well - the system is likely to have a lot of load- and inertia-dependent angular flex, poor resolution, etc. \$\endgroup\$ – Chris Stratton Oct 10 '17 at 19:13
  • 1
    \$\begingroup\$ The motors will not be stalled. I have chosen servos that have more than the required torque. \$\endgroup\$ – Khaliiid1992 Oct 10 '17 at 19:33
  • \$\begingroup\$ Well then you need a 6V power supply capable of at least 30A (if you want it to not brown-out when they're all stalled at the same time) \$\endgroup\$ – user253751 Oct 11 '17 at 1:04
  • \$\begingroup\$ You could use something like 24V and then use step down DC converters for each servo. \$\endgroup\$ – Voltage Spike Oct 11 '17 at 16:19
  • \$\begingroup\$ That is a good idea using a DC-DC converters to step the voltage down. But would that be good having 6 servos, which mean more than one converter. \$\endgroup\$ – Khaliiid1992 Oct 11 '17 at 17:19
1
\$\begingroup\$

As long as the peak sum of the instantaneous current draw from all the servos is less than the rated output of the power supply and all the power wiring is a star from the power supply (+ and -) you should have no issues. If you are going to exceed the current rating for very short periods of time (<1/10 second or so) you might be able to add additional capacitance to the output of the power supply (at the supply or at the servos depending on the load pattern), but not all power supplies like large capacitive loads (it can affect the voltage regulation loop stability). Also remember that if you add capacitors, they draw current at startup proportional to the ESR of the capacitors. The power supply will probably have a slow start ramp to the output voltage which helps limit this current, but it does have to be taken into account in your design.

\$\endgroup\$
1
\$\begingroup\$

You stated in a comment that they will not stall because you've selected more torque than necessary. Based on that, I assume you will implement a fuse or PTC or something as a safety to cap overall current usage.

You then can select a power supply based on your intended usage scenario. If your robotic arm will run all six servos at once, you'll need a power supply capable of 12A (72W). If you restrict the number of servos operating simultaneously in your control software, you could perhaps use a lower-rated supply. For example if you only ever allow three servos to operate at a time, you could use a 6A (36W) supply.

Edit: You'll need to know which servos are allowed to operate at a time, since you mentioned they are all different. Sum the current usage of the servos that will operate simultaneously.

As a rule of thumb, plan for some overhead for inrush current, inefficiencies, etc. Good power supplies will list their constant current rating and may already have some overhead/overload handling.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks a lot. But I wasn't planning to use a fuse, should I? \$\endgroup\$ – Khaliiid1992 Oct 11 '17 at 17:17
  • \$\begingroup\$ @Khaliiid I had a project where I thought a servo would not stall. I was using a buck converter capable only of the current I expected. A piece of material got into the mechanics and caused it to stall. The extra current draw fried the buck converter and the servo started smoking. I scared the crap out of an actor wearing the rig... Yes, please use current protection of some kind to handle the unforeseen. :) \$\endgroup\$ – JYelton Oct 11 '17 at 18:34

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.