Oscilloscope: input signal distorted by output

[raff5184]

Hi,
I am working on a 4 diode bridge rectifier, as discussed here: https://www.eevblog.com/forum/beginners/1mhz-circuit-diodes-which-ones/new/?topicseen#new

I want to do something very simple, visualize the voltage waveform "before" and "after" the rectifier.
Now, if just I measure the voltage of the input, that's ok (blue signal only).  But, why is the negative part of the input (blue signal) cut when I measure both the input and the output at the same time?
And, on the other hand, if I measure only the output (green), it is completely distorted on the oscilloscope?

Thank you

[ataradov]

Where do you connect the ground of the probes? Remember, grounds of all probes are tied together.

[raff5184]

ok, got it. I connected the probes to two different grounds one for the source and one for the output, as in the 8th figure here: http://www.allaboutcircuits.com/textbook/semiconductors/chpt-3/rectifier-circuits/

If I use the same gnd (the gnd of the source) the input is not cut, but the output is  a half wave rectified 

[ataradov]

It should be, since it is referenced against your AC signal. No matter what the input signal is, your individual AC input has one diode open towards DC output.

You need to have either two isolated scopes or just look at signals independently.

[raff5184]

right makes sense, I get it. I tried with two oscilloscopes but the input was still cut

[ataradov]

A this point you need to draw an actual diagram of what you are connecting where. In most scopes ground pins will also be connected to the the earth pin of the power socket. So two individual scopes will have grounds connected. You can test this using a multimeter.

The trick is to have one of the scopes floating - either battery powered or powered though an isolated transformer or isolated power supply. 

[johansen]

A this point you need to draw an actual diagram of what you are connecting where. In most scopes ground pins will also be connected to the the earth pin of the power socket. So two individual scopes will have grounds connected. You can test this using a multimeter.

The trick is to have one of the scopes floating - either battery powered or powered though an isolated transformer or isolated power supply.

the capacitance of the oscope is a problem at 1Mhz, even if it is floating.

the ground lead of mine is permanently broken, i have a neon lightbulb and resistor in series, so the light lights up when the scope is hot.

[raff5184]

everything is clear. Thank you very much

[Alex Eisenhut]

4 channel scope? Use CH1+(-CH2) for one measurement referenced to the same ground as CH3, then you can look at the two signals.

[alsetalokin4017]

Exactly ^.

Didn't we already cover this in that other thread? The scope probe ground reference leads are all connected together at the scope chassis. So you have to use a _differential_ measurement on the output of the diode bridge, where you do not connect the two output channel probe ground reference leads at all, because the input measurement probe already has its ground reference lead connected to one side of the input to the bridge and you don't want to short this to the output side through the probe reference leads.

So you use two scope channels + Math subtraction to obtain the output measurement. Evidently the OP has a good 4 channel scope so making this differential measurement should not be a problem, and in fact it's a good learning exercise.