RCD fault - pos/neg sine wave

[Reg87]

I had a fault with a RCD the other day which no one could explain.

Just some background firstly:
In Australia we have RCD's which are rated to trip at maximum of 100ms at 30mA for the majority of purposes. Our power supply is 230V at 50Hz nominal.

Fault:
I test for nuisance tripping as well as actual fault condition trip times. I was using a RCD tester which could isolate the test to only one side of the AC sine wave. Ie positive or negative.
On the positive side whilst conducting the nuisance trip test it past. Ie: Didn't trip.
However when I conducted it on the negative side, it failed by tripping.

My question is, Why does it matter what side of the sine wave we test on(Its not a requirement by law)?

Although it didn't break any of Australia's Electrical Wiring Rules, I was thinking maybe this form of testing is carried out overseas.

Aplogies if this is in the wrong section

 

[zappedagain]

Sounds like you may have a ground loop or something that is biasing you to the positive side. Have you tried checking with a high voltage scope probe to see if Vpeak+ equals Vpeak-?

Are you sure your tester isn't half-broken?

Z

 

[Reg87]

Tester is good.

The fault is a easy fix really, just replace the RCD.

More concerned with why the tester comes with such a feature.

 

[ScottyUK]

The tester can simulate a pulsating DC earth fault (half-wave rectified AC), which is one of the requirements for a Type A RCD listed in IEC 61008 and 61009.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Reg87]

Thanks for indicating that Scotty. Wasnt aware of IEC standards. I think Australian standards (AS/NZS 3000:2009) still apply however.

So I understand how the tester works, I understand what codes/standards insruct me to perfrom the tests.

But why must I do the test? What is it proving?

 

[ScottyUK]

I think - and am willing to be corrected if anyone knows better - that the test proves that the RCD will operate correctly in the presence of a DC bias. In some RCD designs the presence of DC would cause saturation in the toroidal core and prevent detection of the fault.

The nature of the load will affect the choice of RCD - for example an RCD feeding a circuit supplying (say) an offline SMPS in a computer or a high-frequency lighting ballast will rectify the incoming mains and convert it to DC. The DC bus is still very much connected to the mains through the rectifier, and with this type of load there are credible failure modes which could result in an earth fault which would produce a pulsating DC current.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Reg87]

Cheers scotty. That makes sence with me.

I was also thinking that for the 6mA/130ms RCD's the tripping time may be slower if the fault accured during the positive side of the sine wave, and you were testing the negitive side. I realize 50Hz is quick but when your talking about 130ms for the RCD to trip, then the time taken for the fault condition to be realized may be to slow.

 

[waross]

Re DC bias. If there is corrosion of copper conductors in the ground path it may form a copper oxide rectifier and only pass one half of the sine wave to ground.
The worst example of this type of failure that I have seen was a DC motor in a printing press. As I remember it was about 15 HP. The connection between the brushes and the inter-poles became loose and overheated. The heat caused the formation of copper oxide and rectification. We were called in with the complaint that the motor would run forwards but not in reverse.
It is possible for a similar condition to develop in a grounding circuit. Hence, test but half waves to ensure that the RCD will operate under all possible conditions.

Bill
--------------------
"Why not the best?"
Jimmy Carter