Why is ungrounded power supply ground at line ground?

The switcher probably has a floating output.

Verify the DC and AC voltage is zero, then measure the resistance from your Beaglebone ground to the power line ground. You will probably find it is an open circuit.

Proper grounding can help mitigate problems from ESD damage, EMI interference, line spikes from heavy machinery, random noise from switching normal loads, flourescent lighting noise and other line crud. But it can be a can of worms when dealing with poorly designed circuits, and changes can make things worse instead of better. So if you are not experiencing problems, about the only thing I would suggest is to ground the Beaglebone to the house water pipe ground using a capacitor and parallel resistor to help minimize ESD damage. Your mileage may vary....

The local power supply negative is probably referenced to mains neutral, which is usually held close to main ground. Hopefully its not actually tied to neutral, else if you reverse the active and neutral lines, the power supply negative will be at mains active. You probably see that that is not a good thing !

Well it turns out the ground on the processor is at 0V wrt power ground - that's the mystery - the resistance measures ~0 ohms (using ohmmeter (!))

The reason I'm keen on this is that I'm using this result as a figure of merit to detect anomalous conditions.

Actually, the 2-prong plug is reversible AND it works the same at 240V - no neutral.

It cant be 0 ohms to ground whichever way round its plugged in. Time to re-do your measurements.

NT

A typical DVM reads zero volts when measuring an open circuit.

And a scope shows ~100mV between grounds regardless of which way it's plugged in. When I measure an open circuit the scope shows a dead flat line. I'm just an EE - maybe I have to go back to school :-)

Don't recall a NEMA 2-prong 240V plug. Got a number? Most 2-prong

No - the device can plug into 240 or 120, so they made a 120 non-polarized plug. I have a plug converter, but for this I just wired a 120 receptacle for 240. Aaaanyway - that part is easy...

I think you have the wrong cord on the widget for 240V.

How are you measuring it?

what voltage do you measure between 5V and ground?

How are you measuring it?

what voltage do you measure between 5V and ground?

What else is connected to the beaglebone?

Is by any chance your BBB connected to a another device, for example a PC or a TV? Most consumer devices have the connector shields connected to mains earth, and if I remember correctly in the BBB the USB/HDMI connector shields are shorted to the digital ground.

Using scope; nothing else connected to BBB except Ethernet; there's 5V between 5V and ground; BBB is working.

Your 'scope is most likely grounded.

George H.

On Wednesday, January 7, 2015 at 1:41:36 PM UTC-8, snipped-for-privacy@gmail.com wrote :

gardless of whether I plug it into ungrounded 120V or 240V, the ground on t he processor is at 0V (or very close) to power line ground. I'd love to lea rn why.

I think I answered my own question - the key is virtual ground; the power s upply does an excellent job of building a virtual ground that is spaced fro m 5V in the same relation that the spacing of ac ground is relative to the ac power. You must not run a BBB without a grounded power supply, yet the B eagleboard site recommends power supplies similar to the one I'm using (2 p rongs, no ground prong.)

shielded ethernet cable?

-Lasse

Hmm, Not a "Schuko" plug by any chance?