Digital PSU resets when powering DC motor

[argintviu]

Hi all, beginner in electronics here.

I have built a digital power supply some time ago (created by Tuxgraphics) for use with general microcontroller development. I think you might be familiar with it, there have been some discussions about it on the forum. Recently I have tried to power a small drill with it to make some holes in a PCB and for some reason the MCU either reset or weird characters appeared on the screen as soon as the drill was turned on.

Another interesting thing with it while powering the motor is that the supply became unstable, e.g. the voltage and current were all over the place (it's like the MCU couldn't decide how to set them). The drift on both of them was around 20-30%. This doesn't happen with other load types.

I am new to electronics (programming background) but I suspect that there are some voltage spikes feeding back into the microcontroller via the ADC ports (as the drill motor is an inductive load). What I don't understand is how can this happen, as there is an antiparallel diode mounted on the output of the power supply so the spikes should go back to the motor. I tried to add an RC snubber on the output with 1uF non-polarized cap + 150R resistor, but it only reduced the occurrence of the problem.

Is there any way to improve this design and get rid of the issue?

Thanks for reading!

[Benta]

I won't comment on the tuxgraphics design.

But for a PCB drill supply, a transformer and a bridge rectifier is normally all that you need, not even a filter cap.
Cheap drills run at 5000...10000 rpm, which will accommodate all drills up to 2 mm.
For PCB drilling with 0.6...1.0 mm drills, 20,000 rpm would be better suited.

If you need variable speed, look at adding a filter cap and an LM338.

[argintviu]

Thanks for the answer.

I already used a transformer for the drill, it worked fine, just wanted to use the bench supply out of convenience (e.g. having an universal supply for everything).

It's strange how a complicated design like this can be made useless by a simple motor. I wonder how this problem is circumvented in commercial bench power supplies.

I'd like to build a new bench supply, one which has the ability to also run motors and be a bit more stable (I'll keep the tuxgraphics one for lighter work). Would this design be more suitable?

https://www.eevblog.com/forum/projects/amspire's-power-supply-in-a-pcb/

[Zero999]

It could be numerous things: perhaps the surge when the motor started or the noise from the motor interfered with the power supply?

You could try better RF filtering on the output of the PSU.

[Short Circuit]

Maybe check their latest firmware versions to start with?

mtime: 2016-12-08 Changes since last version:
- reduce the amount of LCD resets.
mtime: 2016-11-20 Changes since last version:
- make LCD init more robust against interference.

[free_electron]

that's not a power supply ... that is a DISASTER !

[Benta]

that's not a power supply ... that is a DISASTER !

Now you know why I did not want to comment on the tuxgraphics design.

[argintviu]

While tinkering with the resistor divider from the voltage sensing circuit I noticed that when the voltage on the output rail gets bigger than what the supply is set to provide, instability starts, e.g. the voltage drops randomly. Gotta read the code, this might be a software quirk or even some kind of protection. This also explains the instability when powering the motor (which raises the output voltage with those spikes).

I'll keep searching for a way to prevent the voltage feeding back into the loop. For starters I'll borrow the MOSFET + zener protection from  amspire. Hopefully I'm not beating a dead horse .

It could be numerous things: perhaps the surge when the motor started or the noise from the motor interfered with the power supply?

You could try better RF filtering on the output of the PSU.

Indeed, the motor is creating voltage spikes that make their way back in the supply via the ADC that is used for sensing voltage.

Maybe check their latest firmware versions to start with?

I just installed the latest firmware, but it doesn't change anything. Most likely a hardware problem.

that's not a power supply ... that is a DISASTER !

Heh, in a way it is, but it's good for beginners (e.g. as a first MCU project). I'd dare say it's the DSO138 of power supplies.