Why RCDs/RCBOs sometimes go bang/smoke

[Clitheroe]

After reading at least a couple of topics elsewhere on RCDs/RCBOs etc going bang and smoking when pressing the test button, isn't the test button linked between the phase and neutral contacts internally via a high value resistor to create an imbalance?

If so, then my theory on this is the resistor probably losing it's value dramatically and going very low resistance hence when the button is pressed creating a direct short circuit - in a consumer unit RCD this would mean a short across the meter tails.........

 

[333rocky333]

I thought the test thing was wired between live and earth

 

[Spark123]

Nope, normally between phase and neutral, connected to one in andthe other out so it imballances the RCD. A lot of the good old type of RCDs don't have earth connections.
If it was wired as you say, between live and earth and if the earth was lost upstream for any reason then pressing the test button may not be very nice!!

 

[333rocky333]

Nope, normally between phase and neutral, connected to one in andthe other out so it imballances the RCD. A lot of the good old type of RCDs don't have earth connections.
If it was wired as you say, between live and earth and if the earth was lost upstream for any reason then pressing the test button may not be very nice!!

Only really fit M/Gerin ones and they have earth wires, so for some reason i thought that was what is was for.
Thinking again your right, the majority only have L and N

 

[NotHimAgain]

Square D had a product warning out on this 3 or 4 years ago. Since then there has been another warning for M & G C60, C60X and Square D KQE, SQOR, QOE - but this concerns the RCD not operating.

I have also had some MEM RCBOs fail in this way. I think that the 'explosion' is a result of the test resistor failing under load - but it is only my guess.

I visited one site where someone had cut all the 'white wire' earth connections off - probably to do on IR test . I rang MEM and they said that the RCBOs would still work on Line (phase) earth faults but would not respond to neutral to earth faults.

 

[Lectrician]

The ones with a white flywire are DC sence RCD's.

 

[JohnD]

and surely they could only overheat if they failed to trip? otherwise the current would only pass through the test resistor for 30mS or so

I suppose they would smoke best if they failed to trip, and the test button stuck closed

btw if you had a N/E fault, there would still be a current imbalance between the conductors, even if the white wire was cut/missing/not part of the design. so it should still trip. I don't understand the other idea. Lots of RCDs/RCBOs don't have a functional earth.

 

[NotHimAgain]

I simply tell you what MEM told me - ring them and pose the same question.

 

[Spark123]

iirc I have seen a pic on the front of one of the MEM ones which makes it look as though it is looking at the potential between neutral and earth as a secondary function. Not sure if this is true for all of the RCDs with a functional earth wire, the square D KQ ones don't have the same pic. They have the functional earth wire but only have the type AC logo on them, not type A (AC and DC pulse).

 

[Sailbadthesinner]

Some of the RCDs with a white earth do have a voltage trip if N-E exceeds 50v this protects against supply reversal or if the incoming N fails.

Sailbadthesinner