Full-wave rectifier measured on one oscope channel

[Taelonious]

Howdy,

I'm trying to measure the output of my full-wave rectifying circuit. I would think that connecting my probe between the DC source and ground would give me the expected signal, but I only see half-wave rectification. I am only using one probe when do this, I was told to use two and invert one channel, but that seems wrong. Any help would be awesome!

I'm using a Siglent SDS1052DL

[alsetalokin4017]

 I think you might need a differential voltage probe to do it on one channel only.

But with two probes, one inversion, and math...




(ETA: sorry, no channel inversion, just math subtraction which has the same end effect)

[Taelonious]

Thank you for clearing up if using math is correct. Could you explain the probe difference?

[alsetalokin4017]

Please study the scopeshot carefully. There is a lot of information there.

If you are asking me to explain differential voltage probes, please google and read as much information as you like. You'll find much better explanations than I can give you.

Think about why simply connecting a probe+ground across the DC output of the bridge results in the half-wave rectification waveform.

[Taelonious]

My apologies. I'm tired and have been staring at this board for a while. I thought you typed "different" opposed to "differential". I also may have been unclear when I said the probe was referenced to ground. The probes grounding clip was connected to the opposite end of the load resistor.

I see how when both probes are referenced to ground, they will only see the voltage related to the side of the load they are on. As for why the scope only shows half the rectification when measuring across the load, my guess is something to do with the probe doesn't like being referenced there..?

When I was trying to figure this out earlier I stumbled on this video. It seems the guy makes the measurement using regular probes in the same way I am, only I have a load resistor between the DC terminals. https://youtu.be/7Q0W-IHu1UM

[Ian.M]

Obviously, I haven't seen your actual circuit setup on your bench so something else may be happening, but my comments and explanation is based on what beginners usually get wrong when trying to scope a bridge rectifier for the first time on a fully equipped lab bench.

Your problem is almost certainly because you are using a signal generator with a ground referenced output as a signal source.  Your scope adds a second ground connection, which shorts out one of the diodes in the bridge.   It also means that one of the other diodes leads directly to ground (via your scope probe ground clip), so its shorting your signal source on the half waves you are *NOT* seeing.
I've put together a LTspice sim (attached) to show you what I *think* is going on:

In order, the circuits and waveforms are:

• floating signal source, ground referenced probe
• ground referenced source, ground referenced probe
• ground referenced source, differential probe using two channels and subtraction

As you can see case #2 is no good.  For case #3, you must not connect the ground clips to any circuit node except the ground of the signal source.
 

The only reason you haven't damaged anything is because the signal source output impedance is limiting the current.   If you'd tried this with a mains supply or other low impedance ground referenced voltage source, you'd probably have blown up your scope.

Must watch:

Comment thread: https://www.eevblog.com/forum/blog/eevblog-279-how-not-to-blow-up-your-oscilloscope!/

[Taelonious]

Ian, that's exactly what I had done. Thanks for the reminder.

[alsetalokin4017]

Erratum: On my FWB scopeshot above, I may have given the impression that one channel was inverted. This is not the case for the specific display shown. The necessary inversion is handled by the math subtraction. (On analog scopes this is generally done by inverting one channel and using the "Add" vertical display mode, which does the same thing as not inverting, and then subtracting in Math, as is done in the DSO.)



On the Rigol, channel inversion is indicated by a "not" bar above the Channel Number on the screen display.