On the power connection page, it is suggested that we use zener diode for protection when using back up power supply. The connection suggested is to connect zener across the +5v and GND of the servo rails directly.



   However when I read about zener diode, then the circuit to control voltage is suggested as follows.


  Notice the resistor in series with the zener.  I understand that this resistor will consume the voltage which is excess to the required voltage. However in the connection suggested for PXHawk, there is no resistor.  In this case how is it able to regulate the voltage? My knowledge of electronics is limited, so I would appreciate an explanation of this.

Views: 12195

Reply to This

Replies to This Discussion

 The second circuit if for voltage control circuit. Note that here we are not using zener for voltage stabilization. If for some reason the voltage spikes above the safe limit then we want a pass circuit to prevent the PIXHawk from being damaged. This is the purpose of this zener. So you do not need resistor. Hope this helps.

Putting it another way, the zener will act as a very low value resistor if voltage gets too high and clamp the voltage, but it's not meant to do this for long periods of time since it will burn out if you send that amount of current through it for more than short spikes.

I have no idea how long my zener can withstand current before burning out though, whether it's milliseconds, seconds, or longer.

In the zener diode documentation the diagram is intended to drop the voltage from a higher voltage (say 10 volts) to 5 volts using a resisor and the zener, but in the pixhawk quick start guide, the zener is intended to cut off the voltage spikes that are for very short durations, and the voltage on average is alreay 5V...


If you look at the screen shot of the oscilloscope reading they provided:


The time frame is for one millisecond (that's the whole reading), and the spike is aproximatelly 1/10 of that, so this means that the spike duration is about 1/10,000 of a second, and a zenner diode can handle that with no problems.


I also suggest to put a capacitor in parallel with the zenner of any value between 100uF and 1,000uF (and the cap should be at least 6 volts of course), this way the zener will cut off the higher peaks, and the cap will make the voltage more smooth (more like a straight line)

By the way, the zener should not be exactly 5V, I think a 5.2V zener should be used... in the pixhawk quick start guide I see only one reference the zener voltage, and it says 5.6V, and then later on it says "digital servo causes the voltage on the rail to rise above the critical 5.7V level", so that's a little too close... only 0.1 voltage difference until it gets "critical"...

I just want to point out a mistake in the photo of the servo connector: it shows that the diode is connected between positive and signal, instead of ground.

You can probably flip a JR connector around in the pixhawk, so for safety sake, I'd recommend using a Futaba connector and the diode in the correct pins.

In relation to this..

"To complement the diode, it is also useful to add a capacitorin parallel to the diode. The capacitor will smooth out eventual voltage ripples. As advised for the diode, the capacitor should be connected with as short wires as possible. Do not oversize the capacitor."

1) what is the recommended capacitory?  Will a 25v 1000uf work or is that 'oversized'?

2) How do we parallel it.. should it be done this way?

--------(+)-----/[capacitor]----/[zener diode with rev polarity]

or this way:

--------(+)-----/[zener diode with rev polarity]----/[capacitor]

thanks for the help!

I use a 6.3V 200uF capacitor.

Actually, in my case when using a redundant backup 5V supply to power the ESC output rail, my PixHawk will only boot with the capacitor added.

If I don't use a capacitor, the Pixhawk just makes cricket-like sounds with crazy flashing of the LEDs instead of booting. See it happen after 40secs here:


I suspect that it has something to do with the sequence in which voltage rises at the power module input and the ESC connector during power up, and the power selection mechanism in the Pixhawk.

The capacitor in my case, serves to slow the down the initial voltage rise at the ESC input.

I've reported it a month ago to 3DR support, but haven't heard anything of it since.

thanks.. may I know how you soldered it up? is it this way:

A) capacitor then zener in reverse

--------(+)-----/[capacitor]----/[zener diode with rev polarity]


B) zener in reverse then capacitor

--------(+)-----/[zener diode with rev polarity]----/[capacitor]

thanks in advance!

If I'm interpreting your ASCII art correctly, then B makes no sense, because either your circuit is open, or you want to use the capacitor in series.

The capacitor must be connected in parallel, so it doesn't even matter in which ESC/servo plug you stick it. You can make one plug for the zener diode as shown in the documentation, and another one for the capacitor.

The only reason I'm even using a capacitor is simply to make it boot. Otherwise I wouldn't even bother, because I don't have noisy equipment connected to my Pixhawk. My ESCs are opto isolated too, so it's as clean as can be.

*slap on forehead*  =)  thanks for taking time to clarify that! I've been searching exactly what it means  by 'parallel' and I just never thought of doing it in a separate servo lead! 


I have the same issue on my Tricopter, I've had to power the servo rail via an external BEC.  Cant explain why the BEC from the ESC is not liked yet the external one is all good.

I've tried a 200uf capacitor but it didnt work.  I tried them 2 in parallel (for 400uf) and it didn't work, did you get a resolution in the end?


Hi guys,

Is this a suitable capacitor to use in parallel with the zener?




Reply to Discussion


© 2020   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service