Common mode ground and its connection to main ground

analogcommon-modeground

I am trying to understand concepts in common mode voltage and EMI due to it. Now, I understand that differential mode is the difference between the P and N wires/signals. The common mode is the – (P+N)/2.This is w.r.t ground. Now, the common mode currents are induced due to noise generated via long wires that are connected to PCBs(that carried signals or power).

Now, the common mode current is dumped into the ground and thus making my ground noisy. Now my doubts are the following –

  • Common mode noise is generated in power lines(Vcc and GND) or in differential signal lines(RS422,485) or in both ?
  • If common mode noise is dumped into my ground and if I have a common ground my whole ground will be corrupted,right?
  • How should the common mode ground be coupled to my main GND plane ?

Also, I have long wires bringingin in Vcc,GND,and RS422 signals. So, where all should I put common mode chokes to filter common mode noise ?

Best Answer

This is a thought experiment and something to think about rather than an answer to the direct questions. I believe that thinking about various scenarios like this is more helpful to the learning process than trying to focus on the specific questions raised.

So, if you had a small battery powered circuit that flashed an LED on and off a few times per second and then, with a wire, you connected the ground of that circuit to the live AC wire in your wall outlet what would you expect to see?

  • Would you expect to see the LED fry?
  • Would you expect to see the LED start flashing randomly?
  • Would you expect to see the LED carry on as if nothing had happened?

For the 2nd experiment, using the same battery and LED flasher circuit, if you routed your AC live wire through the 0V connections on your flasher and out to a 60 watt lightbulb that connected back to neutral would the flasher continue to operate without a hitch?

In both experiments I have tried to demonstrate a common-mode (or probably better stated as a longitudanal interference) problem and the first experiment was the ability of the LED flasher to deal with a common mode voltage and the 2nd experiment was dealing with common mode currents passing through the 0V in the flasher (as well as the common mode voltage being present).

You should think about this. What would be the bad practises that would make the LED flasher either die/fry or run intermittantly? This last question is a bit more searching but at the heart of it is the ability of the flasher circuit to cope with electric field gradients and magnetic field induction. You don't need to worry about this last bit yet, just concentrate on the two scenarios. Hope this helps.