Noise filter for Uno power supply?

I'm trying to power an Arduino from a model aircraft receiver that is powered by the battery eliminator circuit from an electronic speed controller; this gives me a supply voltage of 6V, the bare minimum that an Uno needs to function. I’ve found this does work however when I connect up servos on this same power supply the Arduino tends to brown out due to noise coming from the servos.

While it would obviously be easier to power the Arduino off of a separate power supply (which is what I'm currently doing) I'm still curious as to what it would take to power it from the receivers power supply. The noise is relatively small at around a 200mv drop during servo start-up as shown below but it is still causing the Arduino to freeze for a second when the servos first move.

Where would be a good place to get information on how to start building a noise filter circuit for the power supply? I’m wondering if a simple low pass resistor capacitor circuit might do the trick but analogue electronics isn’t my strongest skill.

Thanks in advance for any suggestions All the best

The first thing I'd try is a capacitor across the battery. Maybe 1000uF for starters. If a 1000uF capacitor works but it 's physically too big, you can experiment with smaller capacitors.

Or, a capacitor across the Arduino's power supply connectors, plus a diode in-between the battery and Arduino (with the motors connected directly to the battery). The diode prevents the capacitor from discharging into the motors*, but a regular silicon diode will have about a 0.6V drop across it. If the voltage drop is problem, a Schottky diode will have about 0.2V across it.

(You can't generally use an RC filter in a power supply because there is too much voltage drop across the resistor.)

  • The diode makes the capacitor more effective in filtering-out negative voltage spikes when the motor suddenly draws more current. But with positive spikes, it's no more effective than the capacitor alone. However, the Arduino's on-board regulator should be very effective in killing positive spikes.

So something like this?

Yes.

Using of course, a Schottky diode.