Open PEN faults

[Glan gluaisne]

There's been chat on various charge point option threads about open PEN fault protection, and why it is needed. Right now we have an open PEN fault, probably as a consequence of storm damage to an overhead nearby, as much of the village is supplied via ABC, and the DNO have told me that a tree has damaged a nearby cable.

Guessing from the symptoms I'm seeing at the moment, it seems probable that the PEN has gone open circuit, that the transformer is way out of balance and that consequently I'm seeing the voltage between neutral and "true" earth (i.e. my earth electrode for one of the charge points) floating between 50 VAC and 85 VAC. We're getting between 260 VAC and 285 VAC in the house (or were until I decided it was safer to isolate the incoming supply) between line and neutral.

The solar inverter seems to have packed up, not sure yet if it's permanently damaged or just shut itself down, and a phone charger's blown (big time, a bang and lots of smoke). I anticipate a claim going in to the DNO, as I took some photos of the screen on my multitester, to prove the supply had gone way outside the safe (and legal) limit before I shut it off. We're running on batteries for now, cut off from the grid until the fault gets fixed.

So, despite the best endeavours of the DNOs to try and make the PEN connections more robust in recent years, with their newer, high integrity jointing kits, it seems that open PEN faults can still occur, and can produce potentially (no pun intended. . . ) dangerous conditions at the consumer end. It's certainly reinforced my view that open PEN protection is essential, and made me glad that the charge point my car is plugged into at the moment still has an earth electrode, given the relatively high voltage that the PEN was floating up to.

 

[LukeUK]

Do you know any of the neighbours that might be on the other two phases? It would be interesting to confirm the imbalance.

Not that that is at the top of anyone's to-do list right now...

 

[Glan gluaisne]

Do you know any of the neighbours that might be on the other two phases? It would be interesting to confirm the imbalance.

Not that that is at the top of anyone's to-do list right now...

I had thought of asking if my neighbour's supply was OK, as I know that she's on one of the other phases, but she's also in her 80's, and in the current climate it would be too risky to go into her house and start measuring voltages.

Just measured our incoming supply again and it seems to have come down a lot, 245 VAC line to neutral, and barely half a volt between PEN and true local earth, so my guess is that SSEPD may have fixed the fault. I'll give it another half hour before I switch the incoming mains supply back on, though, just to be safe. I've not heard anything back from SSEPD, and they did promise to call me as soon as the supply was back to normal. Would have been nice to have confirmation of the details of the cable damage, although I'm near-100% certain that it was an open circuit PEN, just from the crazy voltage readings I was getting, they were floating around all over the place.

 

[LukeUK]

What are your thoughts on the failure mode?

I've worked on IT earth systems, where it is somewhat strange to observe a system with an initial phase-to-earth fault happily running as usual, but with the earth fault monitoring alarms ringing away.

I'd not really considered what would actually cause an open PEN fault in a TN system. A simple open circuit caused by mechanical failure? Cable damage? Caused by a fault current in the PEN at the transformer, or caused by (mal)operation of an overvoltage detection device or something similar?

 

[Glan gluaisne]

If I had to guess, knowing that pretty much all the overhead cable around here has now been changed to ABC (Aerial Bundled Cable) I would say that the tree that SSEPD say had brought down a cable, broke the PEN conductor in the ABC, leaving at least two of the phase conductors intact. That would explain both the high voltage between PEN and "true" local earth that I measured, and the crazy fluctuations between line and neutral. The latter was varying by 10 to 20 V faster than my meter display could keep up with. I know that we have an intermediate earth bonding the PEN on the underground run of Wavecon 95 that supplies our house, as I saw the jointing team run a long copper earth tape out of the pot joint, in the bottom of the trench I'd dug for them.

The fact that I was seeing a voltage of up to 80 VAC between that PEN and an earth electrode that's about 10m away suggests that there was a fault current flowing the "wrong" way down our supply cable and through the intermediate. The Ra of my electrode was low when I banged it in, around 22 ohms, and the DNOs intermediate will be a fair bit lower, as its deeper and running almost at the level of the local water table. If it had a resistance of, say, 5 ohms, then the ~80 VAC I measure suggests that it might have been conducting around 16 A of fault current, that should have been flowing through the PEN conductor. All a bit rough and ready, but probably in the right sort of ball park. The slightly scary thing is that if a couple of houses turned kettles on, that fault current could have risen by maybe 20 A, increasing the voltage at the PEN to around 180 VAC, enough with the wet ground conditions here to give someone a lethal shock if touching their supposedly "earthed" car.

 

[26ct2143]

Replying to this thread as probably more logical.
sounds like you've had quite an ordeal with your supply, quite scary.
Makes me think about some sort of overvoltage disconnect protection for the farm.

With regard to solar inverters, our (30kw) one and the CHP have a G59 relay, which quickly disconnects them from the grid if the voltage or the frequency is out of spec. So I suspect your invert will come back tomorrow when the sun comes up.
Might ask my electrician about EV protection due to grid issues, since our power supply can be a bit wayward sometimes.

 

[Glan gluaisne]

Yes, our PV inverter also has a G59 relay, so should come back online tomorrow. It did also trip the RCBO on the inverter supply, though, which makes me wonder if there might be some other fault, although the RCBO has reset, and may have tripped just because of all the odd things going on with the out of whack incoming supply.

The normal protection that's mandatory for any outlet intended to charge an EV should be OK for most fault conditions, and would have prevented any issues with my car, had it been plugged in at the time. The only thing I think I may do is replace our consumer unit with a new one that includes an SPD. These are pretty much mandatory now, and although our CU is only about 6 years old, there isn't room within it to add an SPD. I can probably fit a new metal CU, with an SPD, and, perhaps an AFD (although I'm far less convinced that AFDs are a significant benefit), for a couple of hundred pounds or so, and it should only be a couple of hours work, if that. Might be worth doing, just for a bit of peace of mind if we're likely to get any more voltage surges. Not sure quite what an SPD might have done in the conditions we saw today, though.

 

[26ct2143]

well good luck and I hope it all sorts itself out.
I have to admit, I tend to disconnect the car from the charger in extreme weather, because worry about some bad power damaging it. Which is why I generally prefer to charge when I'm about in the day, rather than during the night when power is cheaper. Not such an issue if it's my own electric I suppose.

 

[Glan gluaisne]

Just had SSEPD out, believe it or not. They've checked our incoming supply, agreed it's over the upper voltage limit, and that there was a PEN fault on the LV network earlier, that they fixed and they have now gone off to check the substation to try and bottom out why we're still over voltage. I have to say I'm pretty impressed with their response. They are going to call me as soon as they have word on the cause of the over voltage problem. I have to say I really didn't expect them to come out at this time of night, especially in the current circumstances.

 

[pgkevet]

It might be a strange rural benefit? We had them out late one sunday for a power outage affecting just us..turned out to be a squirrel had blown the fuse on the pole transformer from the 11K line - clue being fried squirrel on the ground under the pole.
Hope they get ya sorted pronto.

 

[Glan gluaisne]

Could well be. We're at the end of a long cable run, shared with one other house that's on a different phase to us, and the engineers have been busy around here all day, apparently, dealing with storm damage to overhead cables. I'd hoped that storm damage wouldn't be as common now, as a couple of years ago a storm took out a long run of old single conductor overhead cable and SSEPD replaced it all with ABC. ABC is generally very tough, poles tend to break before ABC breaks, so even if hit by falling trees the supply tends to stay on, with the cables just lying on the ground. The weird thing is why we've been left with a high voltage after the PEN fault was fixed.

 

[Glan gluaisne]

The SSEPD chap's called me, the voltage is too high coming from the transformer. It's pretty old, and may well be failing. He reckons that our supply probably wasn't directly affected by the tree damage, as that was on another nearby section of the network, but his thinking is that what we thought was an open PEN cable fault may have been a breakdown in the transformer that's since cleared. They're going to send a team out to replace the transformer either later tonight or tomorrow, and in the meantime have just asked me to keep an eye on both the voltage and to check Ze if the voltage does increase again, in case the PEN fault re-appears. I should have thought to do a loop impedance test earlier, rather than faffing around checking voltages, as if Ze rises, together with the high phase voltage, that's a pretty good indicator that there's a PEN fault.

 

[Glan gluaisne]

Looks like we may, finally, be on the way to getting this fixed. Getting a case of left hand not knowing what right hand is doing with SSEPN, though. I'd originally been told on the phone that they wouldn't alter the tap changer until after they had a week's worth of voltage measurements. I've repeatedly sent them photos of the screen of my MFT, proving that the voltage sits at around 254 VAc to 256 VAC most of the time, plus their chap that came out on the 27th December found much the same.

Last Thursday I had a call from a local SSEPN chap, to arrange to come around and install a voltage recorder first thing this morning. Last Friday I had a very belated email from SSEPN (only about two weeks after their 24 hour target) telling me that they were going to tap down the local PMT feeding our house. The chap that came this morning found our voltage sitting at 257 VAC, and checked the loop impedance again (still at its normal 0.3Ω). His view was the same as mine, there's not a lot of point in waiting a week for a load of voltage records, so he got on the phone to a colleague and requested an immediate tap change at the local PMT.

With luck we may, finally, get our supply back within the ESQCR voltage limits within a few days. Pity there seems to be the usual PITA management failings within the DNO though. As is often the case, the people actually doing the work are very efficient, with a lot of common sense, whereas their management often seem to struggle to organise even the simplest of things.

 

[Mr Miserable]

I won't pretend to have understood a word of this!
However, are you suggesting that in these circumstances a chargepoint with an earth spike would be safer/better than a chargepoint with inbuilt PEN protection?

 

[Glan gluaisne]

I won't pretend to have understood a word of this!
However, are you suggesting that in these circumstances a chargepoint with an earth spike would be safer/better than a chargepoint with inbuilt PEN protection?

Either should be fine, although technically an earth electrode may offer slightly better protection, if one can be fitted.

The problem is to do with the fact that an open PEN fault can cause the "earth" terminal, and hence anything "earthed" using it, to rise to a voltage high enough to cause an electric shock risk. It's to do with the fact that the low voltage supply network (the 230 VAC supply to each house) may not actually use a separate protective earth conductor, but, for reasons associated with economy as much as anything else, the supply may use the neutral conductor to provide both the neutral current return path and to provide the protective earth function.

The snag with this, pretty common, arrangement is that if the combined Protective Earth and Neutral (PEN) conductor goes open circuit anywhere on the incoming supply to the property, because of a cable fault, cable joint problem, etc, then the return current from equipment and appliances cannot flow back through it, and this then causes the voltage on the household "earth" connection to rise, potentially up to the same voltage as the incoming live supply.

Inside the house, this presents a pretty low electric shock risk, as all the exposed metalwork that's earthed (water and gas pipes, metal sinks and metal cased appliances) will now be sitting with their cases at the same high voltage, and the chances are that the floor of the house won't be conductive, so there's only a small electric shock risk. The risk arises outside, where perhaps damp ground will be close to true earth potential, and yet anything conductive connected to the house supply earth (such as a plugged in car) could be sitting at up to mains voltage. Clearly, just standing on the ground and touching a door handle could give a lethal shock under those conditions.

Exactly the same thing occurs with things like an electricity supply to a metal framed building, like a greenhouse, or to a hot tub installation or caravan hook up point, which is why there has always been a requirement to ensure installations like this have some form of open PEN protection. For years this has been done by wiring them as TT installations, with an earth electrode and RCD.

With the increasing prevalence of EV charge points being installed where installing an earth electrode might be difficult, perhaps impossible, due to the layout of the area around where the charge point is being installed, another solution was needed. The main issues with fitting an earth electrode are the proximity of things like outside taps, exposed conductive pipework, plus the distance from underground services that might fall within the influence zone of the earth electrode. The solution has been to develop open PEN protection devices, that try to determine if there is a PEN fault, and disconnect the supply, and, most importantly, the earth, from the charge point outlet.

The only question is how effective these open PEN devices may be. None are really as effective as a properly installed earth electrode and RCD, IMHO, but some come pretty close. The Viridian system, used in both their modular EVSE (used in a few brands of charge point now, and also used in a charge point connection box) is the best designed open PEN detection system I've seen, as it uses a combination of voltage measurement, plus current measurement in the earth conductor, to try and determine if there is a PEN fault. The matt-e O-PEN device seems to me to be less effective, as it seems to rely wholly on voltage sensing, and whilst seemingly compliant with the letter of the regs, there may be open PEN faults that it won't reliably detect.

My own view is that, if possible, wiring a charge point as a TT installation with an earth electrode, in the same way as any other outdoor equipment that may have earthed and exposed conductive parts, offers the highest degree of protection, as long as the installation is maintained in good working order. If it's not possible, or practical, to install an earth electrode, then an open PEN device is the next best thing.

 

[Glan gluaisne]

The saga continues. SSEPD initially agreed that there was an over-voltage problem, and agreed to tap down the local PMT. Another part of SSEPD seems to have disagreed with that decision, so an investigations chap came out last week, initially confirmed we had a supply voltage outwith the allowable tolerance but said he had to fit a voltage recorder, to be in place for a week, for confirmation. Meanwhile another bit of SSEPD emailed me to tell me that the PMT would be tapped down tomorrow anyway. The investigations chap came back this morning, knew nothing of the tap down instruction to a line team, confirmed it wasn't a line team from his depot, retrieved the voltage recorder and downloaded it to his laptop.

Our average supply voltage over the past week has been 252 VAC, regularly exceeding 260 VAC, and was in excess of the upper voltage limit for 47% of the time, so the high voltage problem has been confirmed (again). Whilst he had the main fuse out, and was downloading data from the voltage recorder (it connects via Bluetooth) I pulled the PE conductor out of the earth block and did another earth loop impedance test on the isolated incoming supply, which proved that if there had been an open PEN fault (I'm pretty sure there was one originally) then it has now cleared, as Ze was back to it's normal value of ~0.3Ω .

The investigations chap knew nothing about the tap down apparently scheduled for tomorrow, but reported the fault for immediate action, on the basis that the voltage was consistently higher than the allowable maximum under ESQCR. I have a suspicion that we may get two line crews turn up to do the same job now. I walked up and took a look at the PMT this morning, and it's one of the larger ones, with the tap changer on the top, so a bit more work to do a tap change, I think. We can see the PMT in the field on the opposite side of the valley from us, so may get early warning of the power going off when they isolated the 11 kV supply to do the work.

 

[ACarneiro]

Two separate crews will come to reduce the voltage.
By later tomorrow your voltage will be about 200VAC

 

[Glan gluaisne]

It seems that the problem has been sorted now, albeit with a lot more work that was first thought. It turned out that three local PMT (pole mounted transformers) were all putting out over 250 VAC, so all three have now been tapped down. We're now sitting at around 235 to 245 VAC. Talking to one of the (slightly grumpy) line team, his thought was that the real problem was on the 33 kV network, and that they were now going around tapping down the LV network because of that. It seems that, during the storm just after Christmas, there was some damage to the 33 kV feeder, with one phase being taken out.

Apparently, the fault damaged the 132 kV to 33 kV sub-station transformer, or its associated switchgear (my guess is that an explosive fuse probably operated to clear the line fault). The sub-station was repaired on the afternoon of 27th December, but it seems that during the process the output voltage from that increased. The view of the chaps doing the local work is that the old sub-station had probably been giving an output on the low side for years, and as a consequence all of the LV transformers fed from the 11 kV network it was feeding were tapped up to the maximum, in order to get the LV supply up to ~240 VAC.

When the repair work restored the 33 kV supply to the middle of its allowable range, that then pushed the 11 kV network voltage up, which then pushed the LV voltage up, and in the case of our supply, and that of those fed from two other PMTs nearby, we ended up with a supply voltage that was out of limits. Apparently the tap on our PMT was originally set to +5%, and has now been tapped down two positions to 0%, so that tallies with the 11 kV supply having originally been low at the time the PMT was installed, around 40 years ago.

 

[pdk42]

Were the line men too grumpy to be singing the song?

 

[Glan gluaisne]

I don't know what his problem was, TBH, but I suspect it was coming out to do a simple 30 minute tap down job on one PMT, then finding that he had to chase around tapping down two other PMTs, and, as a consequence, have the 11 kV supply to the area turned off for a couple of hours, resulting in complaints about the power being off.