Hi all,

I'm trying to understand the full mains waveform visible on my cro when just the probe tip is connected to a low voltage supply. The cro trace shows 324V p-p @ 50Hz, which is Australia's mains supply.


The photo below shows the double insulated SMPS is connected to the probe tip of a Rigol DS1054Z, with the cro probe ground clip unconnected and the other SMPS supply socket unconnected. Both cro probe and cro are on x10 setting. DC coupling.

Connecting the cro probe tip to either SMPS supply socket gives the same result.

According to a powerpoint tester, the powerpoint is wired correctly and earth is functioning as expected.

I plugged in a double-insulated 5VDC wall-wart instead of the SMPS and got a 10V p-p waveform instead of the mains waveform.

With the cro probe and probe ground unconnected to anything, there's noise around 25mV p-p. I know it's normal to see some mains pickup on a cro, so am fine with the 25mV, but am disconcerted by the apparent mains waveform... Have I got faulty equipment, or faulty understanding ?

Thanks in anticipation,
Dan

 

The way you have it set up you probably have got a spurious waveform because you have not connected the probe earth to the green lead, you are using the mains earth to both the scope and the psu so all sorts of strange things can be picked up like that. The power supply green lead may or may not be tied to earth and if there is an isolating transformer inside (probably not), strange galvanic coupling may happen, If you connect the probe earth to the green lead, it is then still possible to get a ground hum loop.
Disconnect the psu from the mains and measure for close to zero ohms from the green lead to the mains plug earth terminal. that will tell you whether it is resistively isolated, but not if there is some capacitive/ inductive coupling due to the transformer windings, or other sources.
A lot of smp's actually isolate the mains earth from the output and connect the -ve to earth via a small value high voltage capacitor.

 

Thanks recklessrog. I should have described that the psu doesn't have a ground pin.

Yes, it's got to be capacitive coupling in the psu. Without the green lead connected, there's no circuit so the capacitive coupling impedance isn't subject to a resistor divider as it would be if current flowed, so I see full coupled mains voltage. When I insert various high resistances between green lead and probe ground, the waveform reduces. I've calculated the capacitive coupling impedance is around 370 kOhms. So I get it mathematically now... am still trying to get my head around it electrically...

 

As mentioned above, you do need to hook the ground lead of the CRO to the power supply 0V.
Otherwise there is just a "capacitor" from the CRO probe input to the mains. This capacitor is via the switchmode supply transformer, and maybe mains RF bypass caps to the low volts DC side of the power supply.
What you are doing is not much different that hooking a length of wire acting as an antenna onto the CRO. But it will be worse (mains wise) as the "antenna" is capacitivly connected to the mains.
Having a power supply with no earth on the mains side is not real good. Just out of interest, connect a multimeter on ACV between the DC out and the CRO (mains) ground to see what volts you read. Often it can be substantial, and has been known to pop electronics connected.
A lot of cheap laptop power supplies do not have an earth ans some have caps connected from the active to the DC out for RF noise filters. While these caps are not very large, there can still be a fair current that can flow. And I have seen a spark jump as connections are made.

 

Thanks @dendad.

I just measured 112V AC (RMS) between DC power supply output (either terminal) and Cro probe ground. Multiplying by sqrt(2), this is 158V peak to peak, almost exactly half of the AC input voltage...

 

That is about what I expected. There will be a cap from Active to 0V out, and another from Neutral to 0V. So there is a capacitive voltage divider that is giving you half the mains voltage.
That is prety normal. The same is on most switch mode supplies that do not have a mains earth.