Proper Grounding of a Three-Phase AC/DC Diode Rectifier Circuit

Thread Starter

EE_Dan

Joined Jan 18, 2016
3
Hello All!

This is a first post for me.

I was just analyzing a three-phase diode rectifier circuit - an AC/DC converter. Please see the attached. Three-Phase Rectifier.jpg The source is a wye configuration, and the rectifier involves the use of 6 diodes along with a smoothing capacitor across the load. It is a simple, but pretty sweet circuit. Whenever the phase voltage of any particular phase exceeds the voltage of the positive output terminal, a diode turns on for a fraction of a cycle, recharging the capacitor. Likewise, whenever the phase voltage of any particular phase falls below the voltage of the negative output terminal, a diode turns on for a fraction of a cycle, recharging the capacitor. As long as the capacitance of the capacitor is sufficiently high, the output voltage is essentially constant, and diodes only turn on for a tiny fraction of a cycle every cycle as soon as the output voltage just starts to diminish.

All of which is known, understood, and celebrated.

However, if we assume that the neutral of the source is grounded, then the DC output cannot share the same ground. In theory, this would cause diodes to burn up. In practice, I suspect, this would cause fuses to blow or circuit breakers to trip, etc.

So, my question is this: How do you properly ground the output of the AC/DC converter? i.e. How do you properly ground the load that you connect to the AC/DC rectifier? Or do you just live with a floating reference point? What are the best practices for proper grounding of this circuit?

Thank you.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
If the bridge is fed from a star grounded system then you cannot connect the DC -ve to earth as you suggest, it is either one or the other.
You can ground the DC if you leave the star connection not referenced to earth.
Max.
 

Thread Starter

EE_Dan

Joined Jan 18, 2016
3
If the bridge is fed from a star grounded system then you cannot connect the DC -ve to earth as you suggest, it is either one or the other.
You can ground the DC if you leave the star connection not referenced to earth.
Max.
Thank you. But then we have a situation in which either the source isn't grounded, or the load isn't grounded. Seems kind of dangerous. In typical three-phase power, the neutral is referenced to earth ground, right?
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
But there is still an earth ground safety aspect as the DC is still referenced to earth ground in the case of a fault, it is not as though it is floating above earth ground.
Max.
 

Thread Starter

EE_Dan

Joined Jan 18, 2016
3
But there is still an earth ground safety aspect as the DC is still referenced to earth ground in the case of a fault, it is not as though it is floating above earth ground.
Max.
You are saying (I think) that the potential of the negative terminal of the DC output will always be below that of the earth ground (in normal circumstances), or if there was a short between earth ground and the negative terminal of the DC output, then the potential of that negative terminal would be equal to that of ground. It is not as if the negative terminal of the output would float all over the place with respect to earth ground. It would usually be a negative constant voltage, or in the case of a fault, it would rise part way or all the way up to 0V. That all seems pretty reasonable to me.

Thank you.
 
Top