Interesting fault? Lowish IR

[GBDamo]

Had a call out at a dental practice at 19:00 last night in the end I left the fault disconnect requiring further investigation.

This fault not the simply resolved call out fault but a "while you're there can you look at" Friday night special.

Very bizarre set up as the building looks to be a rather small corner terrace, two up two down, with a large single storey rear extention yet, just inside the front door is a three phase supply with one phase in use (supply 1 feeds DB1 & DB2) then in the rear room of the original build is a second three phase supply again with one phase in use (supply2 feeds DB3), both L3.

The "while you're there" fault I was looking at was an extractor fan that tripped when turned on.

The fan was fed from a B32 MCB, via a fused spur, on what looked to be a dedicated circuit, from the non-RCD protected side of a MEM split DB (DB2) in the same room. This DB was fed from an upfront RCD from supply 1(not DB1). It is this RCD that trips.

Brief (Friday@19:00) testing showed the up front RCD to be okay. IR @ 500V gave 999MOhms on L-N and L-E but 12.6 MOhms N-E, this was with the MEM board isolated with the boards main switch.

The fan was brand new but I still stuck a flex and plug on it and it works fine on another circuit, on the same RCD.

I know I've a lot to look at when I return but the low N-E IR, whilst not very, is baffling me. It could be a genuine fault but I wouldn't have thought it low enough to cause an RCD to trip. (?)

I could well be going off the deep end but I'm wondering if this is indicative of someone borrowing a neutral for a circuit fed from another board/supply.

If you made it this far and can make sense of this then we'll done. I've rewritten this so many time but it's still clear as mud.

 

[Pretty Mouth]

If it was borrowing a neutral from another circuit from another board, it probably would have given very low IR L+N to E. That's assuming the fan was left in circuit during the test, and that the other circuit hadn't had both L and N isolated from the supply.

Same assumptions: Borrowing a neutral from the same board, it would have given high IR L+N to E if the other circuit had been isolated from the supply, but very low if the other circuit was still energised.

Did you try energising the circuit with the fan disconnected, and if so, did it still trip the RCD?

Class I or II fan?

 

[GBDamo]

If it was borrowing a neutral from another circuit from another board, it probably would have given very low IR L+N to E. That's assuming the fan was left in circuit during the test, and that the other circuit hadn't had both L and N isolated from the supply.

Same assumptions: Borrowing a neutral from the same board, it would have given high IR L+N to E if the other circuit had been isolated from the supply, but very low if the other circuit was still energised.

Did you try energising the circuit with the fan disconnected, and if so, did it still trip the RCD?

Class I or II fan?

With the fan disconnected the circuit held.

The fan is class 2,

 

[timhoward]

Worth a ramp test of the up front RCD.
What else is on the non RCD side of the board. If they are not drawing current (no load) then a N-E fault on one of those could be having current returned through the fault when the fan is on. In theory this is ruled out by your global IR test but I’d double check.

 

[Pretty Mouth]

With the fan disconnected the circuit held.

The fan is class 2,

Did you test IR with the fan in circuit, or disconnected?

 

[GBDamo]

Did you test IR with the fan in circuit, or disconnected?

The fan was disconnected by removing the flex from the fan, meaning the flex and FCU were included in the IR test.

 

[HappyHippyDad]

I don't think the 12Mohm reading is any help. An RCD (30mA) trips around 0.01Mohms. 12 Mohms is fine.
I would test the RCD actually at the RCD to eliminate an unlikely faulty RCD.
There's no fault showing so it's almost impossible to find a fault. I'm not sure what to suggest!
Check connections in FCU.
Check connections in fan.
Change fan just in case it's intermittent.
Did it actually trip whilst you were there, or did the customer tell you it tripped and you didn't actually see it?

 

[Pretty Mouth]

The fan was disconnected by removing the flex from the fan, meaning the flex and FCU were included in the IR test.

Presumably IR tested at the CU? So a borrowed neutral, or possibly a discontinuous neutral with the side closest to the fan shorting to earth. Both of these would not have shown up under your IR test, I'd start by looking for these.

Good luck, let us know how you get on

 

[ImpededLoop]

My guess would be neutral/cpc reversal in the spur feeding the fan, so the fan is connected L/E rather than L/N.

 

[GBDamo]

I don't think the 12Mohm reading is any help. An RCD (30mA) trips around 0.01Mohms. 12 Mohms is fine.
I would test the RCD actually at the RCD to eliminate an unlikely faulty RCD.
There's no fault showing so it's almost impossible to find a fault. I'm not sure what to suggest!
Check connections in FCU.
Check connections in fan.
Change fan just in case it's intermittent.
Did it actually trip whilst you were there, or did the customer tell you it tripped and you didn't actually see it?

Exactly, there is no "fault" as such.

1, When wired to the FCU the fan trips the up front RCD every time it is turned on.

2, The fan is new and when run off a plug and plugged into a socket on the same RCD it doesn't trip.

The IRs were done at 500V.

This is a 6" fan that only runs at 10W.

As this was a "can you just" jobs at the end of a long day I've a nagging feeling I've missed something obvious, seems a simpler solution.

When they send me back, hopefully I'll have a clearer head and resolve this.

Promise I'll fess up and post the results.

 

[GBDamo]

My guess would be neutral/cpc reversal in the spur feeding the fan, so the fan is connected L/E rather than L/N.

Definitely not this, the circuit CPC and the flex earth were in the same brass terminal, remember it vividly.

 

[ImpededLoop]

Definitely not this, the circuit CPC and the flex earth were in the same brass terminal, remember it vividly.

Where was the flex neutral?