A strange fault

SUNRAY · 13 Jul 2020

So Sunray was woken at 4 this morning with a desperate plea for help from an employer.

A freezer building on a farm is blowing fuses 83 seconds after starting.

It's a fair distance from the power source and fed by a double pole Cast Iron fuse box with 30A fuse wire directly off the meter, well over 100m of armoured cable buried direct, MEM 30A switched fuse which feeds the freezer plant and a CU with type1 5A for lights and 10A for a 15A socket.

The freezer has internal D20 and B4 MCB's.

It was always the fuse in the 'L' source that blew.

I happened to have a couple of boxes of MCB's with me and fitted a C32 prior to the CI fuse box.

At 83 seconds a timer relay isolates a sensor with no sign of any current increase using a clamp in the MEM box at the I/C to the remote freezer room.

We meggered the whole of the outbuilding installation including the plant control panel and components and found nothing wrong.

Testing the armoured cable found ~600Ω L-E & 1.2MΩ N-E. Obviously wrong but nothing to blow a 30A fuse wire.

We swapped L & N to keep the freezer working and the farmer has promised to dig a trench in the next couple of days for a new cable.
He remembers watching the farm hands hand digging the original trench in the 50's and he has it in his mind it was second hand ex gov from Exchange and Mart

bernardgreen · 13 Jul 2020

SUNRAY said:
Testing the armoured cable found ~600Ω L-E & 1.2MΩ N-E. Obviously wrong but nothing to blow a 30A fuse wire.

Could the fault current that blew the fuse also have "blown out" the L-E fault in the SWA leaving charred insulation that provided the 600 Ω fault path.

SUNRAY · 13 Jul 2020

bernardgreen said:
Could the fault current that blew the fuse also have "blown out" the L-E fault in the SWA leaving charred insulation that provided the 600 Ω fault path.

Yes it could be and very likely something like that is the case.

Except it was still be blowing the fuse [now tripping the MCB] until we flipped it over. Viz:

RF Lighting · 14 Jul 2020

Did you manage to measure current somewhere at the supply end with the load offline or when the fuse was actually opening?

Did the fuse wire look like it was an overload or a short which blew it?

I’m wondering if the cable it’s self was drawing a load, and the addition of the freezer was enough to blow the fuse.

Jackrae · 14 Jul 2020

Sounds like you have a line to earth fault on the armoured cable. How did you measure the circa 600 ohms and at what voltage did you measure it at. Once it has been 'cooking' for a while with 240 AC on it, the resistance is probably dropping sufficient to blow the system fuse. Reversing L and N to place the neutral onto the defective wire merely masks the fault since N and E are near enough the same potential.

On a separate note, there shouldn't be a fuse in the neutral circuit within the CI box, it should be a solid link. If the neutral fuse blew the device on the end of the circuit would still be live.

SUNRAY · 14 Jul 2020

RF Lighting said:
Did you manage to measure current somewhere at the supply end with the load offline or when the fuse was actually opening?

Did the fuse wire look like it was an overload or a short which blew it?

I’m wondering if the cable it’s self was drawing a load, and the addition of the freezer was enough to blow the fuse.

No we haven't measured it during fault at source end but the fuse wire vapourised with a decent bang so it's a short.

It's only ever the source end that blows despite using identical fusewire at both ends. Having reversed the polarity of the cable, so the 600Ω is in the neutral, it hasn't blown since so we have blamed the cable which is some 70 years old. Additionally we included a B16 after the load end fuse during fault finding which never tripped.

The plant starts and runs ok until the sensor is disconnected and altering the timer setting to 2 mins alters the time the fuse blows so we know it's coincident. Nothing "switches" when the sensor is disconnected, no contactor, no relay, no motor just the operation of the timer relay [which we powered via a 0.5A MCB during fault finding]
It's been running like this for 30 hours:

We had to put oversize wire in 'L' fuse as 30A would also blow with MCB trip during fault finding.

So yes we believe the cable fault is the problem and seems to be... shall we say breaking down? under the tiny spike generated by the relay??? grasping at straws here.

SUNRAY · 14 Jul 2020

Jackrae said:
Sounds like you have a line to earth fault on the armoured cable. How did you measure the circa 600 ohms and at what voltage did you measure it at. Once it has been 'cooking' for a while with 240 AC on it, the resistance is probably dropping sufficient to blow the system fuse. Reversing L and N to place the neutral onto the defective wire merely masks the fault since N and E are near enough the same potential.

On a separate note, there shouldn't be a fuse in the neutral circuit within the CI box, it should be a solid link. If the neutral fuse blew the device on the end of the circuit would still be live.

Yes we have a L-E fault measured at 250V with a megger, opted not to raise the voltage in fear of inflicting more trouble, on a multimeter it was a little bit higher but under 1KΩ.
This reversal is only a stop gap to keep the freezer plant working until cable is replaced.
The switchgear will be replaced to bring up to modern standards but for now we left as neutral fuse as we don't really know what the fault is.

bernardgreen · 14 Jul 2020

Do you know what the timer and the sensor do ( or did ) in the system. ? Seems odd that they are there with no obvious function.

SUNRAY · 14 Jul 2020

bernardgreen said:
Do you know what the timer and the sensor do ( or did ) in the system. ? Seems odd that they are there with no obvious function.

Yes it's a door sensor which will make the freezer plant start in boost mode if the door is open.

bernardgreen · 14 Jul 2020

If changing from Boost mode to Normal ( or reverse ) involves a change of configuration of the motor then possibly the windings are not being switched around in the correct sequence ( a "break before make" function becoming a "make before break" function ). This could be due to worn mechanism(s) in contactors. This false sequence could result in transient peaks of current while the windings are incorrectly connected

SUNRAY · 14 Jul 2020

bernardgreen said:
If changing from Boost mode to Normal ( or reverse ) involves a change of configuration of the motor then possibly the windings are not being switched around in the correct sequence ( a "break before make" function becoming a "make before break" function ). This could be due to worn mechanism(s) in contactors. This false sequence could result in transient peaks of current while the windings are incorrectly connected

Agreed, it simply adds fan motors by the heat exchangers but unfortunately it's not relevant as it's not going into boost mode. Our initial testing concentrated on the control panel but the confusion was 2x30A fusewire and a D20/B4 in line and it was the one fuse which blew with quite a decent bang.
I've heard the trench should be dug tomorrow.

RF Lighting · 14 Jul 2020

It would be interesting to see the results of a 500V or 1000V test on the SWA once you can without risk of damaging the equipment. My money is firmly on a cable fault.

SUNRAY · 14 Jul 2020

RF Lighting said:
It would be interesting to see the results of a 500V or 1000V test on the SWA once you can without risk of damaging the equipment. My money is firmly on a cable fault.

Well yes we know there is [or at least seems to be] a cable fault but the test so far aren't conclusive.

If I get back on the job and I get the chance I'd like to find out more.

SUNRAY · 17 Jul 2020

Just got an update from the boss.

The plan was a good'n, meter installer booked for Tuesday.

2 of the farm hands were tasked with digging a trench as deep as possible and lay the replacement cable, 2" water pipe and 4" purple duct, the bonus for doing it quickly - they get to keep any of the old cable for scrap that they get out of the ground.

So the buckets on the JCB's are 1 metre wide, its reach is about 3m and the trench is about 180m long. So 600m³ of soil pulled out of the ground with the old cable showing half way up one side.

A 4 core 35 or 50mm² 4 core of old colours is coming up through the floor at each end and a pre used 4/12 way 3ph MEMshield board screwed on the wall at each end.

And we were going to get as much work pre-done so the new 3ph tails could go straight into the 'new board' with minimal down time.

So lunch time today the meter fitter arrived, had a look and very nicely ran new tails into the 'new board' and sealed the meter. Leaving no power to the whole farm.

My boss went out and cobbled the existing [mostly rubber/cotton] tails into the random selection of MCBs available to get it all back up and running.

We're still going there Tuesday to do the work, for now the freezer is still working on the old, faulty, cable without any functional problems.

RF Lighting · 17 Jul 2020

I much prefer letting farmers pull in their own armoureds. Last one I did(n’t) was 120m of 70mm² 4 core!

diynot

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.

Are you a trade or supplier? You can create your listing free at DIYnot Local

A strange fault

Similar threads