Buffering/Decoupling AC-DC 5V Power Line

[awef]

I'm powering my PCB from a USB power wall wart (220VAC -> 5VDC).  I believe the USB power supply (iPhone style) is a Switch Mode supply since it's not large/heavy.

When a high power motor on the same AC Circuit starts, my 5V line can vary +/-10-15V for 10-20ms.  It looks more like line noise/ringing than a "square wave pull down" when I look at my scope output.

Can AC noise really affect my DC power rail that much?

I've tried putting a 10mF capacitor across 5V/Ground, but it seems to make little difference.

Are there common ways to decouple my PCB/Power Supply from my nasty neighbor?

[David Hess]

The USB power wall wart was just cheap and has poor line regulation.  A better one would not have that problem.  You could try attaching a large aluminum electrolytic capacitor to the output.

[awef]

I tried another more expensive SMPS but had the same effect.

I'll try the capacitor idea.

Is there anything in a power supply specification that I could look for which would imply better line regulation?  Or perhaps to ask a different way... If I can't find a good power supply and need to design it myself, what should I definitely not miss?

I've made a few buck/simple smps boards before by following designs, but never had to address this particular problem in the past.

[David Hess]

Is there anything in a power supply specification that I could look for which would imply better line regulation?  Or perhaps to ask a different way... If I can't find a good power supply and need to design it myself, what should I definitely not miss?

It is an aspect of line regulation.  Current mode designs are better for this because of their feedforward negative feedback.  USB chargers and a lot of consumer gear is just incredibly cheap.

Another way to handle it is to add a shunt regulator to the output which might be as simple as a zener diode.

Double check that what you are measuring is real.  The Y class EMI suppression capacitors to ground might be coupling common mode noise into the ground of your test instrument.

[awef]

Thanks for your thoughts David! 

I'm not familiar with the concepts regarding Y-Class Capacitors/Common Mode Noise Coupling, but I'll try to look around and see if I can figure out the basics.