Recently I came across this and searched a lot online for an answer but couldn't find one...

I found it mentioned in basic electronics book under rectification...It was also mentioned that the cicuit will not function properly if one of the input terminals is grounded.

  • 1
    \$\begingroup\$ You'll need to provide the context for this statement. Hit the edit link below your question ... \$\endgroup\$
    – Transistor
    Commented Dec 1, 2019 at 15:51
  • \$\begingroup\$ Yes you must provide context, because this is not always so. In mains powered switch mode power supplies, the mains voltage is directly put via bridge rectifier to get unisolated DC supply for switching. \$\endgroup\$
    – Justme
    Commented Dec 1, 2019 at 16:05
  • \$\begingroup\$ Even if a transformer is used it doesn't have to be isolating. I remember the first time I worked on a TV set with 'hot' chassis - a literally shocking experience! (silly me thought the presence of a transformer meant it must be isolated - I hadn't considered that it might be an autotransformer). \$\endgroup\$ Commented Dec 1, 2019 at 20:37

4 Answers 4


A full wave bridge rectifier requires input to be derived from a transformer that provides dc isolation from the supply in situations where it is desirable to:

  1. connect the negative point of the supply to earth.

  2. protect the supply and connected circuitry from earth faults.

  3. have no point in the supply and connected circuitry to have a voltage with respect to earth that is higher than the supply voltage.

Some or all of the above may not be necessary and there are other means for satisfying the requirements such as providing the isolation at a latter stage such as demonstrated by the answer provided by @Oldfart. Therefore the requirement is not a firm requirement that applies in all circumstances.


Just for your sake I used a main web search engine and searched for:

"switched mode mains supply schematic".

The third image was a nice small one from here and shows this: enter image description here

As you can see a bridge rectifier direct connected to mains. So no a bridge rectifier does NOT need a transformer that provides dc isolation.

What then still remains is the question @Transitor asked: what was the context in which you found that statement.

  • \$\begingroup\$ In this context the transformer provides the ground isolation - the negative output of the bridge rectifier is "live" and cannot be connected to input ground or the output ground. \$\endgroup\$ Commented Dec 1, 2019 at 18:41


simulate this circuit – Schematic created using CircuitLab

The essence of the problem is that with both the input and output referenced to ground, there will be a short circuit between two of the leads.


Consider what happens if you don't.

Your bridge rectifies 120VAC into a V++ rail and a V-- rail. At any instant, the V-- rail is the most negative voltage available - it rides the zero line for the first half of the cycle, then the bottom of the sinewave for the second half ot the cycle. It dips as low as -170 volts compared to mains neutral and equipment safety earthing. (In a perfect world, neutral and safety earthing are quite near each other).

Your logic is, ok, you take this bouncing-ball DC and buck V++ it down to 5 volts or whatever, giving Vcc. Tie V-- to Vss. 5 volts is harmless enough! What could go wrong?

Except V--/Vss are still ranging as low as -170V compared to the outside world.

So Vcc is at -165V compared to the earthing system.

Now hook up an Ethernet cable to that. Person is plugging that cable into an (actually grounded) piece of equipment and what happens? Human touches equipment chassis while holding the ethernet plug, and they have 170 volts DC between them. See the rather shocking problem?

It gets worse.

You have a similar piece of equipment, also plugged in. Except this one is plugged in upside down (or there's a hot-neutral reversal in the wall or on the cheap Chinese power strip). That unit's sine wave is 180 degrees out of phase with the other's. So when unit 1 - is on the bottom of the sinewave (Vss=-170V), unit 2 - is on the top of the sinewave (Vss=0V). with Vss connected between the two units, this comprises a dead short betwen AC hot and AC neutral.

Ah, but you have internal fuses! I bet you decided to save 10 cents and not put a fuse on the neutral.

Switching supplies do it a little different

Note the schematic on the above switching power supply. At first glance it seems like mains DC comes around the transformer. Not so. IC2 and C7 actually provide an optical or capacitive gap.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.