Totally stumped. Night Storage Heater tripping out MCB, but ok on daytime circuit.

[Steve Poulton]

Please help. I have a basic understanding of electirics, but 5 Electricians are stumped and I don't know where to go next. My mother in law heats her house by 4 Night Storage Heaters, all of which are wired to a Economy 7 supply/fuse board, with no RCD just MCB. The largest of the Night Storage Heaters began to fail and trip the MCD. To cut a long story short, she ended up having a new one fitted. Unfortunately this new one also now trips the MCB. The electricians have tried swapping the various supplies to the various heaters around in the E7 fuse board and the same heater is tripping the different MCB. It would be an obvious thing to consider that there is a fault on the circuit between the board and the night storage heater. However, they have now wired it into the day time board and it is not tripping. Has anyone got any ideas as we need to get it wired back into the E7 circuit??

 

[elsparko]

I can't see it influencing the MCB. The arc carries the same current as the circuit (incomer, in this case) and behaves like a (poor) conductor, producing an equal field to a copper wire carrying the same current. Because of the way the gas ionises, the effective resistance fluctuates in ways that generate wideband noise, including at RF. But the energy radiated from the circuit picked up by the radio is a matter of milliwatts, and would not couple into the low resistance MCB heater or trip coil to any extent. To trip an MCB with an external field derived from a strictly limited current (there's a resistor in series with the arcing circuit) would require putting a substantial coil around it or carefully aligned next to it.

Does anyone think the MCB tripped magnetically, or thermally?

what about magic? its that time of year

 

[Squid]

Does anyone think the MCB tripped magnetically, or thermally?

Its just like when the wife doesn't know how the large dent appeared in the car, its just one of those life mysteries and sh1te happens.

 

[westward10]

I very much doubt it was magnetically.

 

[Darkwood]

Looking at the switch the actual connection section does not seem to have carried the excessive heat as it is still shiny, I still favour thermal through the side wall tbh, the magnetic trip with arcing was just putting it out there, but like yourself I did express in a post yesterday that it was very unlikely on a resistive load to be the cause.

Thermal is my bet through the side wall as the MCB are already working close to rated load so a little extra heat would derate the breaker.

 

[elsparko]

I very much doubt it was magnetically.

so either magic or thermal?

 

[westward10]

We do not know the switch caused the problem unless said switch is replaced and the problem disappears.

 

[marconi]

LT21 - could you put me out of my misery and tell me:

a. what size mcb fed the defective nsh?
b. what size mcb feeds the new nsh?
c. was it a 16A fuse in the 24/7 CU which supplied the nsh?
d. Do the lights flicker/dim in the house when E7 turns on?
e. Could you describe the L-E fault in the cable to the old nsh? What had caused it? Where was the fault located?

For all - does anyone know

a. if a Main Switch in a domestic CU is designed so that the neutral is first make and last break?

b. the power factor of a 3.4kW NSH?

c. are modern NSH controlled by solid-state relays and if so of what type?

 

[Lucien Nunes]

I think that is what we've effecttvely been told - something about the switch or the low voltage caused 'the MCB to overheat'. Note that the thermal trip in an MCB is a heater, and the NSH is a heater, and they're in series. So if the MCB is correctly rated, if it overheats, the NSH element will too (maybe not enough to notice or cause damage).
I think I'm with DW on transferred heat so far, although all 4 MCBs would have had to be pretty darn close to the edge.

But I do like things to obey the laws of physics (even if the wife doesn't!). Troubleshooting electronics often takes one along many steps from cause to effect, e.g. bad backplane connection caused data errors, caused incorrect parameters loaded into drive, caused overspeed, caused limit overrun, caused output bridge overcurrent, caused transistor failure, caused heavier overcurrent, caused PCB ground track failure, caused driver bypass capacitor to explode. Tortuous, but each step is a logical consequence of the last.

Here, it feels like a non-sequitur. e.g. Bad connection caused smokes to get wet caused giraffes to get restless caused capacitor to explode.

E2A Marconi - DP main sw does not normally have leading neutral contact (only TP+N). PF of heater will be very close to unity, unless it has some fancy PWM drive or something spurious - was one of my earlier q's.

 

[ipf]

LT21 - could you put me out of my misery and tell me:

a. what size mcb fed the defective nsh?
b. what size mcb feeds the new nsh?
c. was it a 16A fuse in the 24/7 CU which supplied the nsh?
d. Do the lights flicker/dim in the house when E7 turns on?
e. Could you describe the L-E fault in the cable to the old nsh? What had caused it? Where was the fault located?

For all - does anyone know

a. if a Main Switch in a domestic CU is designed so that the neutral is first make and last break?

b. the power factor of a 3.4kW NSH?

c. are modern NSH controlled by solid-state relays and if so of what type?

Main switch has no lead/ lag, that I've ever heard of, at least.

Power factor should be unity.

Not heard of any solid state involvement with such NSH's.

 

[Dave OCD]

I'd like confirmation that it's all still working tomorrow or in a few days time, it's a simple circuit with a purely resistive load, so it had to be heat transfer from the main switch derating the MCB resulting in thermal operation, I'd replace that too.

 

[Dave OCD]

Main switch has no lead/ lag, that I've ever heard of, at least.

Power factor should be unity.

Not heard of any solid state involvement with NSH's.

Only the very modern units like Quantum which are dual supply and contain a fair amount of electronic control but that's not the case here.

 

[AJshep]

Well done finding the fault, however this doesn't make sense to be either and I'm quite glad !
If someone had hypothetically asked me what would happen if a main switch wasn't making a proper connection on a E7 DB (that has quite a large load) , I would tell them that They would expect to notice a fishy burning smell and see scorch marks on the tails, busbar and surrounding areas. Im surprised and glad the damage is quite localised. Running a 9KW shower with a lose connection for 15 mins is enough to really heat a cable up, this has been running all night ! Extremely lucky well done LT21.

 

[ipf]

I'd like confirmation that it's all still working tomorrow or in a few days time, it's a simple circuit with a purely resistive load, so it had to be heat transfer from the main switch derating the MCB resulting in thermal operation, I'd replace that too.

Agreed. Faults can often be found which are a result of......faults.
Should have put 'such' NSH's

 

[westward10]

I think it was on the verge of tripping, each night a spider goes for a stroll across the mcb toggles which broke the camels back.

 

[marconi]

Now I know there is no sequencing between the L and N contacts in a domestic CU Main Switch, I am pondering why it was only the L contacts to fail. In the narrative, associated with this failure is the L-E fault discovered by Deakon and LT21. Are they correlated or not?

Thought - Has someone opened the Main switch, closed the tripped MCB and then closed the Main switch in an attempt to re-energise the E7 Board in the presence of the L-E fault, thereby damaging the L contacts? The damage to the L contacts of the Main switch looks to be in the early stages of deterioration.