RCD nuisance tripping

RCDs (aka RCCB, ELCB) are devices that help detect earth leakage from a live circuit and prevent untoward incidents like fire and electric shock leading to injury or death. RCDs operate on the principle of Kirchhoff’s laws. These are very sensitive devices and trip instantly as soon as they detect any difference in current passing towards load via line, and that returning via neutral.

I moved to a new apartment a few months ago. And ever since have been facing nuisance tripping of the RCCB that protects the apartment. I am sharing my notes on how I investigated and mitigated this issue.

To begin with, this is how my apartment is powered:

Diagram 1. I will refer to the labels (in red) in my notes.

The device that used to trip is the lighting circuit RCCB (labeled LRCD in the Diagram 1). We started observation to identify any patterns when the RCCB trips. We soon realised that the device trips when the electric supply switches between the 2 sources: electricity board (EB) and diesel generator (DG). In coming days we were able to narrow it down to switchover specifically from DG to EB.

I consulted the apartments’ maintenance manager on this. He sent the apartment electrician to inspect, who mentioned 3 possibilities:

1. Some faulty appliance used at home that leaks current to earth. This is easy to validate by simple observation.
2. The RCCB may be faulty and tripped on any small surge of current at switchover (less that RCCB’s 100mA sensitivity), or the ACCL may be faulty and sent a large surge of current at switchover. This is easy to validate by replacing the devices.
3. There may be a wiring fault at home. He clarified that it is difficult to narrow down such a fault due to the concealed conduits.

We started investigating starting with first possibility. We started paying attention to the appliances, lights, other fixtures that were being used during the switchover. We did not see any pattern.

To narrow down this down further, I started isolating individual rooms. When the supply switched from EB to DG, I would manually turn off the MCBs for one of the room (labeled L1CB, L2CB, L3CB, L4CB, L5CB in the Diagram 1). Later when the supply switched from DG back to EB, I observed if the RCCB tripped with the particular MCB turned off. If it did not, then that MCB/room that was turned off would have the faulty appliance or fixture. The RCCB tripped in every case.

First possibility was at a dead end. There was no evidence of a faulty appliance or fixture at home that was leading to the RCCB tripping.

We decided to go to the second possibility. After a lot of coercion, the maintenance manager replaced the RCCB. But the tripping continued. He even replaced the ACCL in coming days, but the RCCB continued to trip.

He confirmed that the devices are tested and confirmed to be working correctly. This eliminated the second possiblity.

On to the third possibility. The maintenance team and electrician were able to reproduce the issue by simulating DG to EB switchover. But no luck in identifying the root cause. The electrician hypothesised that some device in my home may be pulling more than 100mA spike at the time of switchover. We have a 100mA sensitivity RCCB (LRCD in Diagram 1) and it will trip if there is a spike of more than 100mA. He recommended changing to 300mA sensitivity RCCB.

I was not convinced because they hadn’t discovered the actual root cause yet. Secondly, 30mA is considered critical threshold for electric shock protection. Sensitivity of 100mA was already too high, and I did not agree on the proposal to change it to even higher sensitivity of 300mA.

Diagram 2. RCCB connected as a dummy.

As a mitigation, the electrician changed the connections for RCCB to effectively operate it as a dummy (check Diagram 2). This prevented tripping. But in this configuration RCCB wouldn’t offer any protection because there can be no difference in line and neutral currents passing through the RCCB. This was not acceptable to me, and I asked the electrician to revert to original configuration.

Before I moved in, I had a few electrical sockets moved for ease of usage. The maintenance team then used a AC line fault detector to inspect all sockets that were moved. They detected missing earth connection on the H2 circuit in Diagram 1, and fixed it. He concluded that this must be causing the RCCB to trip. But why would this problem on the heating circuit cause lighting circuit RCCB to trip? I was asked to monitor and so I did. The RCCB continued to trip.

His new hypothesis was that some of the electrical conduits may have been drilled through when installing wall cabinets. This can cause the wires to get damaged and end up leaking current to the earth. His final solution was to change to 300mA sensitivity RCCB so that it ignores such leakage, assuming that the leakage is under 300mA. I asked him not to do that, I did not want to address the symptom but the root cause.

I decided to take up the RCA myself and started with verifying that all neutral wires for lighting circuits are indeed connected to the lighting circuit neutral busbar, and likewise for heating circuit. If this verification shows any cross-connection, it could explain the difference in current between line and neutral of RCCB.

I used a AC line fault detector for this test. I turned off the heating circuit RCBO and verified all sockets on lighting circuit using the fault detector. If there was any lighting circuit where neutral was connected to heating circuit busbar, this test should show a fault. I repeated the same for sockets on the heating circuit after turning off lighting circuit RCCB manually. The test passed successfully. There was no cross connection.

Next test was to confirm that neutral and earth do not connect with one another. I used a multimeter in the continuity test mode for this purpose. I turned off the heating circuit RCBO (HRCD) and the lighting circuit RCCB (LRDC) and started my tests. Meter readings between:

• Lighting circuit and heating circuit neutral busbars: No connection
• Earthing and heating circuit neutral busbars: No connection
• Lighting circuit neutral and earthing busbars: Some connection. The resistance fluctuates from 450Ω to 600Ω. This was the first indication of earth leakage from lighting circuit. Need to check further.
• External neutral and earthing wires (LN, HN and Earth in Diagram 1): Connected to each other. This is expected as neutral and earthing wires are meant to be bonded at the sub-station.

I had some data that proved earth leakage from the lighting circuit after these tests. I opened up the lighting circuit neutral busbar to inspect individual lighting circuits independently. I concluded that the connection between lighting circuit and earth is from L3 Neutral in Diagram 1. I reconnected all other neutral wires back to lighting circuit neutral busbar. I verified if L3 Line shows any such connection. None existed. So the culprit was L3 Neutral.

Next step was to identify the earthing wire that L3 Neutral is making some connection with. I opened up the earthing busbar and discovered that L3 Neutral isn’t showing any connection with wires used for earthing inside the house. Instead, L3 Neutral is making some connection with the external earthing wire that connects the earthing busbar with the pit (indicated as Earth in Diagram 1). L3 Neutral and Earth have independent conduits, and it is not possible for the two to have a physical connection.

The L3 circuit includes 3 parallel connections (refer Diagram 3).

Diagram 3. L3 circuit

I disconnected one of the parallel connection at a time, and retested the connections between L3 Neutral and Earth. The connection disappeared when I removed the under cabinet lights from the circuit. Culprit found.

The only conclusion is that the neutral wire for under cabinet lights is grounded somewhere. The electrician repeated his hypothesis that the conduits may have been drilled through when installing wall cabinets, resulting in unintended grounding of the neutral wire. Or the wire may be touching the cabinet (built in steel) itself. Either way the decorative lights have now been disconnected. This has resolved the RCCB tripping issue.

References:

1. https://www.linkedin.com/pulse/earth-leakage-nuisance-tripping-what-do-steve-kelly/
2. https://electricalinstallationwiringpicture.blogspot.com/2010/02/1-phase-elcb-connection-pictures.html
3. https://electrical-engineering-portal.com/what-is-the-difference-between-mcb-mccb-elcb-and-rccb
4. https://havellsindia.wordpress.com/2011/09/23/lets-start-with-an-a-for-accl/
5. https://circuitglobe.com/difference-between-grounding-and-earthing.html