Full bridge rectifier with AC provided by one signal generator

Hi all
A member of my family is doing an experiment about full bridge rectifiers, and I think they are running into some trouble.

They have a signal generator making I believe 10v ptp AC being fed into the AC inputs of the full bridge rectifier (the negative half of the AC wave apparently being connected to signal generator ground), and the DC output being observed on an onscilliscope.

They were running into trouble where the negative half wave was not showing (googling and thinking about this, it's almost certainly due to shared grounds shorting the diodes that would deal with the negative part of the AC wave).

They can now see the "effect" of rectification using some built-in math operation on the scope.

This all seems weirdly wrong to me.

Surely:
1. There is no true AC current as the "bottom" diodes dealing with the negative wa e are only affixed to 0v ground
2. You can't have this circuit working without a transformer (or without 180 degree out of phase dual ac?)


A capacitor was placed (I presume in parallel to output of the bridge) and this appears to have skewed the 2 AC voltage traces on the scope in a strange "camel hump" way. Just seems the circuit I messed up, but I have no circuit diagram (and may not be able to get one).

Any thoughts please?

My only 2 times i have managed to get a full bridge working was via mains 240v AC and transformer (one of which I use daily in my lm3886 DIY build), so I'm a bit unsure if a bridge rectifier can be used without a transformer.
 
There can be no common ground between the AC and DC side of a bridge rectifier, and typically a signal generator and an oscilloscope are both grounded to the same safety ground. The concept of a common ground makes circuits easy to understand but floating voltages are very common, and you have to be aware of the implications. Because SPICE software bases everything on a common zero-volt ground, it is impossible to simulate floating voltages, which are undefined. Common switching power supplies use a bridge directly connected to the mains, save the RF filter, so the input circuit floats on the negative peaks of the power line.
 
  • Like
Reactions: 1 user
Nothing in SPICE stops you specifying a 10^12 ohm resistor or higher in series with a picoamp level current source of your choosing (perhaps 50 or 60Hz?) - that's basically a floating voltage with mains pickup! That could be connected to a MOSFET gate to simulate a floating gate quite well I suspect. You could add inter-pin capacitance and simulate cross-talk to floating inputs too.
 
If there are 2 scope channels that can be subtracted from each other ( or even just 2 channels), connect one to the bridge + and the other to the bridge-. Do not use the ground clips on the scope. This should allow you to see the FWB operation without shorting the diodes in the bridge through common grounds.