RCD tripping before rcbo

[jpadie]

I have a garden room with a cu fed from a time delayed RCD in the main consumer unit.

 

The garden room CU is populated with rcbos.

 

I was installing an isolator for a vmc this evening and when I came to turn the local rcbo back on (all circuits were turned off but the main isolator switch was not) I found that the RCD in the main CU had tripped.  

 

Any thoughts on why this happens?  

 

I've also had a recent experience where the RCD in the main unit tripped whilst none of the rcbos did. Related to a brief short to earth on a circuit in the garden room.  

 

All rcbos have tested good and earth is good, albeit the rod is much closer to the main house than the garden room.  

 

I'm swapping the main RCD for an mcb shortly so this is just trying to understand what's happening rather than looking for a solution

[andyscotland]

On 15/12/2024 at 22:45, jpadie said:

I have a garden room with a cu fed from a time delayed RCD in the main consumer unit.

 

The garden room CU is populated with rcbos.

 

I was installing an isolator for a vmc this evening and when I came to turn the local rcbo back on (all circuits were turned off but the main isolator switch was not) I found that the RCD in the main CU had tripped.  

 

Any thoughts on why this happens?  

 

When you isolate individual circuits, you are only isolating the phase (live) conductor. The neutral remains connected.

 

If other circuits in the installation are energised, there will be current flowing in the neutral. This, combined with the resistance of the neutral, means that the neutral will have a small voltage difference to earth/the neutral at the origin.

 

Normally that doesn't matter because the current can still only flow along the neutral and through the RCD.

 

However if you touch neutral and earth together (e.g. when cutting a cable) this creates a parallel path which will allow some current to flow through the circuit you have created. Since this bypasses the RCD, it causes an imbalance and causes it to trip.

 

On 15/12/2024 at 22:45, jpadie said:

 

I've also had a recent experience where the RCD in the main unit tripped whilst none of the rcbos did. Related to a brief short to earth on a circuit in the garden room.  

 

All rcbos have tested good and earth is good, albeit the rod is much closer to the main house than the garden room.  

 

Time delayed RCDs - like all RCDs - still display shorter tripping times as the current increases above the rated value. For example a 100mA S type RCD at 500mA is allowed to trip in 40ms - a lot less than the 130ms minimum with a 100mA current.

 

If you had a phase to earth dead short then the current imbalance was likely into the hundreds of amps, and therefore well outside the parameters the time delay was specified/designed/tested for, so all bets are off.

 

The time delay is really for situations where e.g. a human is in the loop, rather than cable shorts.

 

On 15/12/2024 at 22:45, jpadie said:

 

I'm swapping the main RCD for an mcb shortly so this is just trying to understand what's happening rather than looking for a solution