Floating voltages and bad GND connections — getting zapped and worse

Nothing good can come from a freak idea after mi…..


This time I actually learned something, or in more pleasant words: I was reminded that there was something about floating power supplies, measuring voltages and something called ‘common ground’. Only I don’t remember ever being taught that it can hurt too.

Here’s what I did:

I plugged an FTDI breakout board (BUB) into my PC’s USB port and connect that to the LED controller I keep telling about and boring you all the time. Then I used a new switching wall wart to power the Jyetech oscilloscope that behaves strange (even after replacing the ATmega64 cpu) to do some tests. Basically measure the PWM signals coming out of the AVR and see how long it takes to crash the scope. Should have been a no brainer, just connect both “GND” together and measure some PWM signals. Only it didn’t work properly. Strange slowly changing signals showed up on the scope and when I touched the wires I’d see some square waves from time to time. So I thought: “must be a bad RCA plug then” and tried to make it work by pulling here and there. “Pulling is futile” (quoth the scope). After some more messing with it and cursing I accidentally touched some metal parts of the scope and got zapped! Nothing serious (if you know what I mean), but nevertheless it hurt. First I thought the wall wart was sick and I tested another one, but that didn’t change much of the pain in my fingers.

This is what I measured between the USB’s GND and the “-” of the wall wart:


I used a 10x probe, so what you see (and I felt) are 100V/div!

Here’s a makeshift illustration of where the voltage came from:


And now a very very bad video:

Now I’m beginning to actively realize why those guys in helicopters repairing HV transmission lines always attach this “equalizing rod” or whatever you may call it. What annoys me most is that I’m pretty sure these 200Vpp could have wrecked my AVR board I was messing with at the same time. It was behaving very strange for some minutes after this incident and had to be reflashed several times.

The good thing coming from this little reminder of basic electronics is that I immediately checked the RCA plug for propers mechanical specs and boy it is truly a piece of shit !


The ID of the male plug should be 8.0mm, the OD of the female plug should be 8.3mm. Making good contact by expanding the 8.0mm to 8.3mm, all fully reversible like a spring. Fine. This magnificent piece of hardware has an ID of 8.58mm. Totally useless. Suffice to say I immediately cut it off the probe wire and trashed it and the other one still waiting to be used.


This entry was posted in Electronics., Fix me. and tagged , . Bookmark the permalink.

5 Responses to Floating voltages and bad GND connections — getting zapped and worse

  1. Helmut says:

    tja, da hilft nur Hornhaut an den Fingern. :-)

  2. Pingback: Arduino based passive analogue input selection & volume control - Page 7 - Head-Fi: Covering Headphones, Earphones and Portable Audio

  3. Justin Kenny says:

    Similar experience happened to me, although I was asking for it; I tried measuring the output of a Geiger tube with the Jyetech oscilloscope on AC coupling mode, and then got shocked twice. I guess I was charging the AC coupling capacitor to the 235 volts or so at the Geiger tube’s anode. Amazingly, the oscilloscope still works.

  4. robert says:

    Yup, seems likely. It’s good that small capacitors don’t store much energy. Enough to make you fingers twitch though. A friend of mine told me a story about touching a 400V capacitor. He dismantled his SLR’s external flash. Apparently after charging it up fully. It took him a while till he had full use of his right arm again ;-)

  5. robert says:

    Here’s a post with some more info regarding filtering using 2 class Y capacitors on the AC side.


    It seems to be quite common to have 2 class Y caps on the AC side, which also work as a 1:1 capacitive potential divider. The middle point is connected to the (-) pin on the DC side. Therefore it “floats” at about 1/2 the AC side’s voltage. The capacitance (mid to high pF range) is chosen such that currents are way way below what is lethal to humans.

Comments are closed.