Voltage regulator problems

Hello,

I have a Due D00047562 that I am having problems with. Occasionally it makes a high pitched squealing noise and the supply voltages and analogue readings become erratic. The squealing seems to come from what looks like an inductor nears pins 40-42. I’m guessing this is some kind of switch mode regulator's transformer or inductor.

The minimum setup for this seems to be with an AdaFruit 1947 display attached (which draws <100mA), however it has also happened with a blue tooth module and a few LED’s – again around the 100mA mark. Quite often the thing will run fine for hours on end but then suddenly start playing up if it is reset or reprogrammed.

This occurs intermittently and happens when powered by USB and by 12Vdc external supply. Once it starts the only way to correct the problem is to power down completely.

Is this a common problem or does it sound like a fault with the board?

It doesn't sound right!

Yes its the inductor for the 5V switchmode buck regulator, it sounds like the
regulator has gone into repeated shutdowns. Is the inductor very hot
when this happens? Ferrite inductors become less efficient at high
temperatures, leading to thermal runaway - causing the regulator to
reset?

Using an external 5V regulator would take the extra load off the board
as a workaround. BTW I don't think that regulator does anything if under USB power.

Perhaps you have everything in a box? Add ventilation to prevent temperature
build up if so.

Its worth checking the supply to the board is up to scratch

Hi thanks for the reply,

The project isn't in a box and the regulator is cool - room temperature is about 18 degrees C today.

The same collection of boards and (similar) code run fine on a Leonardo, Uno and Netduino.

I'm sure its not the USB hub nor the 12V supply - they work fine with the other boards and I've scoped them and there is no noise nor brown outs.

When its running ok it is ok, however transients that can set it off at any point. It has been running fine for two days until 10 minutes ago when I reprogrammed it.

I really suspect that it's the board and that there's some problem with a dodgy capacitor....

D