1
\$\begingroup\$

I'm using a Adafruit Servo Driver (PCA9685) and I have to solder a capacitor depending of the number of servos.

We have a spot on the PCB for soldering in an electrolytic capacitor. Based on your usage, you may or may not need a capacitor. If you are driving a lot of servos from a power supply that dips a lot when the servos move, n * 100uF where n is the number of servos is a good place to start - eg 470uF or more for 5 servos. Since its so dependent on servo current draw, the torque on each motor, and what power supply, there is no "one magic capacitor value" we can suggest which is why we don't include a capacitor in the kit.

Right now i'm using only 12 servos (I shoould use a 1200uF capacitor ) but in the future i'm going to use 14 servos (1400uF)

my question is: What happen if I use a bigger capacitor (1400uF) for 12 servos or less?

\$\endgroup\$

4 Answers 4

1
\$\begingroup\$

Agree with previous answers regarding practical effects. Here's some of the theory to help understand what happens when you use a bigger capacitor for this purpose.

The measure of capacitance is the capacity of the device to hold charge for a given voltage (C = Q/V). So if you picture your capacitor as two plates separated by a material that prevents current flowing, the bigger the capacitance, the same voltage held across the plates will result in a greater build up of charge on the plates (imagine that the effective area of the plates is larger).

As you are using your capacitor to act as a short term power source for when the supply dips, a bigger capacitor means a larger pool of charge to draw upon when these temporary dips occur. This is captured in the maths by the time constant tau (https://en.wikipedia.org/wiki/RC_time_constant) which indicates that increasing capacitance slows down the discharge rate. Therefore by having a larger capacitor connected to the supply rails close to the servo, it slows down the exponential decay of the voltage level by providing the current that the power supply has suddenly stopped giving. The more charge it has to give, the longer it can provide the current for.

This is why there is no golden rule for how large your capacitance needs to be. It is a direct measure of how much energy you require and for how long.

\$\endgroup\$
2
  • \$\begingroup\$ So if I use a super big capacitor, can the capacitor burn the circuit because is providing energy for too long? \$\endgroup\$
    – Ricardo
    Commented May 11, 2018 at 3:16
  • 1
    \$\begingroup\$ As the capacitor discharges, the greater capacitance means it has more charge to give, hence it's voltage gets depleted slower. As it is a passive discharge, nothing will burn by doing this (although the voltage will start to decrease, be careful of browning-out any chips it is powering by putting them into dangerously low voltage states for too long). If the capacitor is fully depleted though and power is suddenly restored, it will charge up very quickly initially as it behaves like a short circuit. You might need to find a way to limit this inrush current if it is too high. \$\endgroup\$ Commented May 11, 2018 at 6:41
1
\$\begingroup\$

The more torque you make the servos generate by moving or holding position the more current they require. The faster this torque demand changes (particularly increases in required torque), the larger this capacitor needs to be. The capacitor can provide power for very short-term demands, but mostly it reduces the change in current vs change in time for the power supply wiring.

If you have an oscilloscope, you can connect is across the capacitor leads and test your application. If you see significant, short term dips in the power supply voltage you need a bigger capacitor. If you see long-term dips in the power supply voltage you need to fix the power supply or power supply wiring.

\$\endgroup\$
0
\$\begingroup\$

Along with the above answer as an audio technician for many years the 'rule of thumb' for audio amplifiers was 2,000 uF per amp of current consumed. This was to make sure the ripple in the DC supply was filtered down to 100 mV or less.

I should point out that digital circuits consuming the same current could get away with 470 uF per amp, because digital circuit are able to some degree to 'ignore' mild noise on the digital lines.

There is no "one magic capacitor value".

OMG-an honest statement if I ever heard one. Your servo drivers would be ok with 1,000 uF per amp, but if you have dynamic torque issues where load can be light then suddenly very heavy I would consider at least 2,000 uF per amp of current consumed. This includes 'Peak' amps which seemed to be one of your main concerns. Also, except for high voltage types over 450 volts, capacitors are low cost.

If you measure voltage dips of 5% or more with 2,000 uF per amp under a sudden load, I would go with 10,000 uF capacitors and just make it not an issue anymore.

\$\endgroup\$
3
  • \$\begingroup\$ That means if I use a bigger capacitor nothing happen, right? \$\endgroup\$
    – Ricardo
    Commented May 11, 2018 at 3:10
  • 1
    \$\begingroup\$ @Ricardo The only possible problem I see from using a bigger capacitor than necessary could be inrush current. \$\endgroup\$ Commented May 11, 2018 at 3:14
  • \$\begingroup\$ Inrush current can cause a short-circuit right? \$\endgroup\$
    – Ricardo
    Commented May 11, 2018 at 3:20
0
\$\begingroup\$

Good answers. However one might look also on EMI and EMC,

You have yo calculate the current your devices consume. Once it is known ∆i, you also need to understand at what minimum voltage your devices could work ( consider load temperature, humidity....)

Once you know the min voltage, the Delta V can be driven.

Then C = (∆i/∆v) *∆t

**""During power up there is an inrush current to the circuit that needs to be limited. Otherwise the power supply could be demmaged !!!

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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