Another Intermittent Fault

[SJD]

I was asked to look at an intermittent fault in a rental property yesterday:

Apparently over the last month or so, the tenants have several times heard a loud bang, that they think may be coming from the kitchen area of the house (but they are not sure, being elsewhere in the building each time). Until Friday, nothing tripped, but on Friday when this occurred, the MCB for the main socket circuit tripped (first time this has occurred).

One pertinent fact is that approx 2 months ago the central heating was renewed - new boiler, new surface pipes to all the rads - due to the old pipework leaking under the concrete floor. There was a lot of disturbances in the kitchen - units out etc - to get the pipes behind them. The problems have only occurred since then.

The installation has a Memera 2000 CU, most circuits just protected by MCBs, one circuit (not this one) has an RCBO, otherwise there is no RCD protection.

The socket circuit is a RFC on a B32 MCB. I tested the circuit, with no real issues - dead tests at all the sockets I could find - a couple needed their connections tightening to get decent results. IR with everything unplugged & FCUs off measured 90M (L-N), 45M (L-E) and 55M (N-E), stable/repeatable. In summary, the circuit appears in good condition.

If there had been something flashing over, I'd expect carbon deposits etc to give much lower IR results? Unless perhaps a chaffed cable with exposed L copper is touching e.g. a CH pipe, and the cable gets thown away by the force of the flashover, only to creep back and touch again?

With loads connected (as presented), the L+N to E measured at 2.5M. I didn't have time to investigate which load(s) were dragging the IR down.

The kitchen appliances are all built-in, and I've not (so far) been able to remove and check any of them. But I figure if one of them was at fault, it would likely have blown the fuse in its' plug or FCU, and it would no longer work - which is not the case, all the appliances appeared to be working normally. I'm reluctant to start dismantling the kitchen when the tenants are not actually sure that is where the "bangs" are coming from.

Any recommendations on what my next step(s) should be?

 

[Octopus]

Have you taken off each socket and looked in the back noses / at the cables?

 

[davesparks]

This is actually a vary familiar sounding story.

I had one where mysterious bangs were heard under the kitchen floor. Some months previous a new water main had been blasted through the wall below floor level.

I ended up hacking a hole in the Lino and cutting up a bit of floorboard. Of course the moment I went to look down the whole BANG and a flash. I knocked main switch off and approached cautiously. There was a 2.5 which had been mullered when they blasted through for the new pipe. It was floating around and occasionally touching L-E

I set about summoning a chippy and plumber from the building company I was working for and we dislodged the kitchen, including removing the sink and appliances to get to it.

 

[SJD]

I looked inside the sockets in the kitchen that I could get at without too much difficulty (4 out of 7, if I recall).

 

[SJD]

This is actually a vary familiar sounding story.

I had one where mysterious bangs were heard under the kitchen floor. Some months previous a new water main had been blasted through the wall below floor level.

I ended up hacking a hole in the Lino and cutting up a bit of floorboard. Of course the moment I went to look down the whole BANG and a flash. I knocked main switch off and approached cautiously. There was a 2.5 which had been mullered when they blasted through for the new pipe. It was floating around and occasionally touching L-E

I set about summoning a chippy and plumber from the building company I was working for and we dislodged the kitchen, including removing the sink and appliances to get to it.

This is the sort of scenario that has been going through my mind, somewhere the plumbers have put a new CH pipe. But without a more precise location, I'm reluctant to start hacking and dismantling - the tenants are not yet even sure it is in the kitchen.

 

[marconi]

I was asked to look at an intermittent fault in a rental property yesterday:

Apparently over the last month or so, the tenants have several times heard a loud bang, that they think may be coming from the kitchen area of the house (but they are not sure, being elsewhere in the building each time). Until Friday, nothing tripped, but on Friday when this occurred, the MCB for the main socket circuit tripped (first time this has occurred).

One pertinent fact is that approx 2 months ago the central heating was renewed - new boiler, new surface pipes to all the rads - due to the old pipework leaking under the concrete floor. There was a lot of disturbances in the kitchen - units out etc - to get the pipes behind them. The problems have only occurred since then.

The installation has a Memera 2000 CU, most circuits just protected by MCBs, one circuit (not this one) has an RCBO, otherwise there is no RCD protection.

The socket circuit is a RFC on a B32 MCB. I tested the circuit, with no real issues - dead tests at all the sockets I could find - a couple needed their connections tightening to get decent results. IR with everything unplugged & FCUs off measured 90M (L-N), 45M (L-E) and 55M (N-E), stable/repeatable. In summary, the circuit appears in good condition.

If there had been something flashing over, I'd expect carbon deposits etc to give much lower IR results? Unless perhaps a chaffed cable with exposed L copper is touching e.g. a CH pipe, and the cable gets thown away by the force of the flashover, only to creep back and touch again?

With loads connected (as presented), the L+N to E measured at 2.5M. I didn't have time to investigate which load(s) were dragging the IR down.

The kitchen appliances are all built-in, and I've not (so far) been able to remove and check any of them. But I figure if one of them was at fault, it would likely have blown the fuse in its' plug or FCU, and it would no longer work - which is not the case, all the appliances appeared to be working normally. I'm reluctant to start dismantling the kitchen when the tenants are not actually sure that is where the "bangs" are coming from.

Any recommendations on what my next step(s) should be?

By Marconi Next, I suggest you do some IR measurements between the RFC and a) the MET and b) and all the other final circuits. Where is the boiler located? Suggest you also isolate the boiler at its FCU and do some IR checks on its wiring - but here I show my ignorance - I don't know if modern boilers use electronic switching of valves and motors rather than relays and switches so you may have to proceed with caution: that may mean disconnecting items. I am sure another will advise on this.

 

[davesparks]

What makes you think the boiler wiring will be at fault? It's far more likely to be damage to a cable on the rfc that's causing the mcb to operate

 

[Marvo]

IR with everything unplugged & FCUs off measured 90M (L-N), 45M (L-E) and 55M (N-E), stable/repeatable. In summary, the circuit appears in good condition.............


....................With loads connected (as presented), the L+N to E measured at 2.5M. I didn't have time to investigate which load(s) were dragging the IR down.

I'd chase the 2.5Meg IR fault to start with, it might be withing acceptable limits but it's substantial enough to warrant investigation given the intermittant flashing issue. Even the 45 & 55Meg readings on the naked circuit would be worth investigating, they're probably your best shot at localising the flashing. Do long IR tests (10 minute duration) using 10KV or the highest voltage test available as well using an instrument with a max hold facility.

Alternatively could you split the ring in the middle and convert it to 2x 20A radials, maybe just temporarily put them on RCBO'S and see which one trips. Once one of them does trip you can also move the point where the split is to narrow it down a bit further.

 

[marconi]

What makes you think the boiler wiring will be at fault? It's far more likely to be damage to a cable on the rfc that's causing the mcb to operate

By marconi A good question, as ever. My answer: It may be the boiler has a supply derived from the RFC as an FCU in the ring or as a spur off it. And if it is, there is switching taking place in the boiler electrical system as thermostats and timers operate. And, the new boiler/heating system is a major, recent change correlated to the onset of the intermittent fault. As ever - I may be wrong.

 

[SJD]

I'd chase the 2.5Meg IR fault to start with, it might be withing acceptable limits but it's substantial enough to warrant investigation given the intermittant flashing issue. Even the 45 & 55Meg readings on the naked circuit would be worth investigating, they're probably your best shot at localising the flashing. Do long IR tests (10 minute duration) using 10KV or the highest voltage test available as well using an instrument with a max hold facility.

Alternatively could you split the ring in the middle and convert it to 2x 20A radials, maybe just temporarily put them on RCBO'S and see which one trips. Once one of them does trip you can also move the point where the split is to narrow it down a bit further.

Thanks Marvo, since posting earlier, I was wondering perhaps to split the circuit into 20A radials, that would be fairly easy to do.

 

[SJD]

By marconi A good question, as ever. My answer: It may be the boiler has a supply derived from the RFC as an FCU in the ring or as a spur off it. And if it is, there is switching taking place in the boiler electrical system as thermostats and timers operate. And, the new boiler/heating system is a major, recent change correlated to the onset of the intermittent fault. As ever - I may be wrong.

The boiler FCU is on the RFC, which I turned off when conducting the IR tests on the circuit alone. But my reasoning is that if the fault took out the 32A MCB, then if on the boiler it ought to take out the FCU 3A fuse as well - and the boiler is still operational with an intact fuse.

 

[marconi]

The boiler FCU is on the RFC, which I turned off when conducting the IR tests on the circuit alone. But my reasoning is that if the fault took out the 32A MCB, then if on the boiler it ought to take out the FCU 3A fuse as well - and the boiler is still operational with an intact fuse.

By Marconi. More information for the brains trust to ponder. Thank you.

 

[marconi]

The boiler FCU is on the RFC, which I turned off when conducting the IR tests on the circuit alone. But my reasoning is that if the fault took out the 32A MCB, then if on the boiler it ought to take out the FCU 3A fuse as well - and the boiler is still operational with an intact fuse.

By Marconi First have you confirmed the size of fuse protecting the boiler circuit?

Second, whether the MCB operates before the cartridge or vice versa depends on their 'Isquaredt' (current x current x time) characteristics. Here is a graph for a 10Amp MCB and 5 Amp fuse I pinched off the internet:

https://www.google.co.uk/search?q=d...KbvUo7z-M:&usg=__l3H6IkxcmtJcsfbwCPTNuI4Z8w4=

When the line for the fuse is to the left of that for the MCB, the fuse ruptures before the MCB trips. Otherwise, the MCB operates first albeit with fault current even more to the right both might/will operate. The crossover of the fuse and McB characteristics takes place at about 100Amps. Remember too that the fuse relies on the thermal effect of the current but the MCB uses both thermal and magnetic effects.

I only know the theory. The practice of selecting overcurrent protection devices to achieve discrimnation was not something I have ever had to do. other commentators can/will no doubt amplify further.

 

[Octopus]

By Marconi First have you confirmed the size of fuse protecting the boiler circuit?

Second, whether the MCB operates before the cartridge or vice versa depends on their 'Isquaredt' (current x current x time) characteristics. Here is a graph for a 10Amp MCB and 5 Amp fuse I pinched off the internet:

https://www.google.co.uk/search?q=d...KbvUo7z-M:&usg=__l3H6IkxcmtJcsfbwCPTNuI4Z8w4=

When the line for the fuse is to the left of that for the MCB, the fuse ruptures before the MCB trips. Otherwise, the MCB operates first albeit with fault current even more to the right both might/will operate. The crossover of the fuse and McB characteristics takes place at about 100Amps. Remember too that the fuse relies on the thermal effect of the current but the MCB uses both thermal and magnetic effects.

I only know the theory. The practice of selecting overcurrent protection devices was not something I have ever had to do. other commentators can/will no doubt amplify further.

So whats your motivation matey - you claim to be a landlord, but you seem to be posting more and more - so are you really a sparky?

 

[marconi]

So whats your motivation matey - you claim to be a landlord, but you seem to be posting more and more - so are you really a sparky?

By Marconi My motivation - an abiding interest in electrical engineering. I spent time in the 1970s working as an electrician. I then joined the Royal Navy. I gained a first class honours degree in engineering specialisng in electronic, electrical, control and real time computing. I served at sea in two frigates, first as the deputy of one of te two engineering departments and later as head. In each case I had 30-40 odd highly trained and qualified technicians working for me. Alone at sea we only had ourselves and our wits to keep complex equipment operational so we had to good at diagnosis of faults, work together as a team and draw on each others knowledge and experience. I am a chartered engineer and chartered electrical engineer.

I will not bore by detailing my other experience save for my interest in electronics - recent projects being the electronics to control 4 sets of traffic lights and another an 8 floor lift.

You will note I only venture to post on certain topics - the wiring regulations and practice of wiring I stay clear of because it beyond my competency to comment. Fault finding is not.

I am fortunate to be retired. The flats are my wife's pensions (she is chronically ill and disabled) - so that in the unfortuante event of my ill-timed demise she has enough income to look after herself. So I am a landlord and do some volumtary work and act as a carer for my wife.

This forum provides some interesting mental stimulation. And I think I can help and I suppose provide some education once in a while.

Am I not welcome in the EF or by you in particular?