Signal generator and oscilloscope earth clip issue

Thread Starter

Elerion

Joined Sep 11, 2017
125
Hello everyone.
I just realized something strange playing around with a simple voltage doubler.
doubler.png
The doubler works fine. I use a signal generator as input. A simple DMM measures the DC output.

The thing is: just cliping a scope's probe clip to the circuit can make it go crazy.

Attached, two scope captures.
First, probe earth clip is connected to signal generator's negative terminal. Ok.
SDS00002.png
Second, with earth clip swapped (clip to positive output, probe tip to signal generator negative terminal).
SDS00003.png
That disturbs the operation so badly! 90V peak!

The siggen is not earthed (it is powered from a two prong power cable) and I see no reason for this to happend...
The DMM can measure any two points in the circuit with no problem whatsoever.
There must be something importanted I'm missing. Any idea?

If I attach the scope directly to the siggen, nothing is wrong either polarity. The issue shows up with the voltage doubler...
The signal just seems more stable if I attach the earth to the signal generator negative terminal.

Thank you!
 

Danko

Joined Nov 22, 2017
1,828
Because of current leakage from scope and siggen power supplies,
voltage between scope and siggen housings is 87.2 V.
EDIT:
Even more than 87.2 V, because positive part of voltage is shorted by diodes.
When scope is connected to siggen directly, then leakage current is shorted by low
out impedance of siggen for probe connected to generator negative terminal.
 
Last edited:

Thread Starter

Elerion

Joined Sep 11, 2017
125
Because of current leakage from scope and siggen power supplies,
Thank you for pointing that out. Does it present a hazard?

When scope is connected to siggen directly, then leakage current is shorted by low
out impedance of siggen for probe connected to generator negative terminal.
Maybe that explains why when directly connected to negative terminal, the stability of the signal is not as good as it is when connected to the positive terminal? It is stable, but seems to fool the scope trigger at regular intervals.
 

Danko

Joined Nov 22, 2017
1,828
Maybe that explains why when directly connected to negative terminal, the stability of the signal is not as good as it is when connected to the positive terminal? It is stable, but seems to fool the scope trigger at regular intervals.
Exactly.
Thank you for pointing that out. Does it present a hazard?
It is hazardous for digital scopes with probes set in X1,
for sensitive parts of circuit under test,
for people, if leakage current is big enough.
To prevent this hazard, housings of scope, signal generator, power supply etc should be connected together and to ground wire of line.
For measuring voltage between two, both ungrounded points of circuit, may be used floating measurements.
 
Last edited:

Thread Starter

Elerion

Joined Sep 11, 2017
125
To prevent this hazard, housings of scope, signal generator, power supply etc should be connected together and to ground wire.
The scope is grounded, but the signal generator is not, because it is powered from a two prong cord (it seems to have an internal SMPS). It is all plastic, except for the BNC connectors.
The only way I know to fix this is to always connect the scope ground clips to the negative generator's outputs.
Do you mean that? Or is there any other way I could safely fix it?

Also, how can there be leakage even if I swap the polarity of the two prong cord to the siggen?
 
Top