Speak EV - Electric Car Forums banner

1 - 12 of 12 Posts

·
Registered
Joined
·
2 Posts
Discussion Starter · #1 ·
I'm new to the forum. My wife has a Zoe and had a Rolec Wallpod (untethered HomeSmart 7.2 kW EVHS2020) installed 10 months ago (still under warranty). It's been fine until now but it's just started playing up the last few days.

It randomly shuts off several times a day when the Zoe is NOT plugged in. We hear a clunk from where the 'black box' (Matt:e Single Phase EV Voltage Monitoring and Protection Unit with Type A RCBO - I think) is fitted next to our fuse box indoors, and the light goes off on the Rolec charging unit outside. N.B. it is not tripping (the switch in the Matt-e unit doesn't move and neither do any switches in the fuse box); we just hear a clunk from the Matt-e unit and then the normal hum/buzz from the black box stops. Usually about 10 minutes later we hear another clunk and the power to the charger comes back. This happens several times a day (not sure how often, but we hear it every few hours or so). The rest of our electrics in the house appear to be fine.

A few days before we had a power cut followed by low-voltage to the house for half a day, and then full power was restored. It's possible this problem preceded that (possible we just didn't hear it), but it's a hell of a coincidence.

As I say, the Zoe has not been plugged in when this has been happening. The charging cable is permanently plugged into the charger outside with the other end hanging up (not touching the ground, with its cover on).

Rolec are stumped so far. They suggested turning the Matt-e unit on and off again, but that hasn't made any difference. I guess the next step is to get the original installer out. We may have to also speak to our electricity supplier to see if anything odd is happening with our supply (but if that were the case why is only the Rolec being affected?).

Any ideas what the problem might be? And do you think it's safe to try to charge the Zoe in the meantime?

Thanks.
 

·
Registered
Joined
·
268 Posts
Maybe your voltage is going out of range. You can check it with a multimeter or most energy monitoring plugs can show voltage. The unit is designed to switch off the supply if the voltage goes under 207v or over 253v because that indicates a potential PEN fault (although there are other causes). If it’s happening regularly and that is the reason, report it to the DNO. If it happens more when the sun is out then it’s possibly the solar panels around you putting the voltage up.

Alternatively the DNO might check it out if you simply report that your voltage monitor is tripping.

I don’t know exactly how the Matt:e works but the other way of detecting PEN faults is if the voltage between the earth conductor and true earth is over 70v. That’s harder to verify but if that is the reason your monitor is tripping then there is almost certainly a fault at the substation.

Rolec are stumped so far.
Not a cause for surprise.
 

·
Registered
Joined
·
1,884 Posts
Sounds like you may have a grid voltage fluctuation. There are lots of reasons for this, and one major flaw with the matt-e is that is uses the voltage between N and L (not the voltage to PE) to guess when there might be an open PEN fault. There can be other events, perhaps transient, that can cause the supply voltage to go out of limits and if this happens the matt-e will shut down, wait for a time, then power back on.

The matt-e is an approved device for open PEN fault detection, but in my opinion, and I am sorry if this seems a bit harsh, it is probably the most crap way to reliably, and more importantly selectively, detect such a fault. My own view is that the matt-e does not comply with Section 722 in BS7671:2018, even though the manufacturer has hoodwinked the IET into pretending it does.

There are other, very much better, open PEN fault detection methods and devices available, but in all honestly it is very difficult to beat the foolproof solution of just using an earth electrode and suitable RCD. Admitedly, there are some use cases where it isn't practical to use an earth electrode, but these really aren't very common, and it is worth remembering that 50 or 60 years ago earth electrodes were the normal way of providing protection to supplies, and still are in many countries around the world. The main reason we switched to using PME systems here in the UK is because it's cheap, not because it's safer.

Edited to add:

For completeness, it's worth noting that a loose connection can make the matt-e trip out, too. In this case I don't think that's very likely, because, unlike many other EVs, the Zoe does an earth loop impedance test when it starts charging, and is a bit fussy if that's too high. It is still a possibility though, so getting all the terminations checked to ensure that are torqued up to the right value would be a useful precaution.
 

·
Registered
Joined
·
2 Posts
Discussion Starter · #4 ·
Thanks very much for the detailed replies. Our local electrician thought the same. He opened up the Matt-e (didn't touch anything inside, though he is EV-trained but not on Matt-e/Rolec) and there's an LED that flashes green to indicate, I believe, that the voltage is within range. He spoke to Matt-e, who said that if the unit shuts off and the LED is flashing once this indicates voltage below range and if it's flashing twice it indicates voltage above range (actually might be the other way round!). The Matt-e shut off every day for a week after the power cut (usually once or twice late evening, but occasionally several times during the day). Since we've been monitoring it this week and looking to see what the LED indicates it has not happened! I suppose this is good news (perhaps an issue at the substation has been fixed?) but it would have been interesting to see what the LED told us. I'll update if we learn any more.
 

·
Registered
Joined
·
1,462 Posts
Agree that it's voltage fluctuation on the supply- the previous power cut is the smoking gun here - could be that they hooked up a temporary feed pending repairs which would explain problems for a while and then OK later
 

·
Registered
Joined
·
268 Posts
it is probably the most crap way to reliably, and more importantly selectively, detect such a fault.
I agree, you are going to get more false positive detections than real ones.
My own view is that the matt-e does not comply with Section 722 in BS7671:2018, even though the manufacturer has hoodwinked the IET into pretending it does.
I’d be interested to know why. At face value it appears to comply with option 4 of this (a device which electrically disconnects line, neutral and protective earth connections, when the supply voltage is greater than 253 V or less than 207 V rms) but I haven’t looked at it in detail.
it is very difficult to beat the foolproof solution of just using an earth electrode and suitable RCD
I agree. Where you can achieve the separation, I’d much rather go TT for the charger because I’m not a fan of connecting the electrode to the MET (para ii) because you end up providing the earth for not just your own installation but also a return path for fault current in the event of a fault on the supply cable coincident with a loss of substation earth. There are also questions about maintenance and routine testing for earth electrodes that make them less suitable for domestic installations.

My charging point is from 2016 and still connected to the PME earth. Wishing I had got the guys who built the extension to bury an earth mat in some Marconite whilst they were digging.
 

·
Registered
Joined
·
1,884 Posts
I’d be interested to know why. At face value it appears to comply with option 4 (a device which electrically disconnects line, neutral and protective earth connections, when the supply voltage is greater than 253 V or less than 207 V rms) but I haven’t looked at it in detail.
It complies with the letter of the changes in Amendment 1, but it seems clear that those changes came about because of lobbying, probably by charge point/open PEN fault device companies that wanted their solution to be allowed.

The main issue is that there can still be a PEN fault with the voltage between phase and neutral remaining within the 207 to 253 VAC limit, plus there can be other faults unrelated to an open PEN that can cause the voltage to go outside those limits. That makes the device pretty non-selective and not really an open PEN fault device at all, it's just a voltage sensing relay (and I am aware of at least one other off-the-shelf voltage sensing relay that has been re-badged as an open PEN fault device now).

To work reliably, a single phase supply open PEN fault device really needs to ensure that the PE connected to the exposed conductive parts (the car in this case) cannot ever rise above 70 V rms with reference to the local earth potential. Obviously to do that really means having a local earth reference, which means having an earth electrode close to the point of use. Just sensing between L and N on the supply doesn't give a true indication of the potential on the local PE at all, as there could still be a fault that pushes the PE potential up but that doesn't take the potential between L and N out of range (it depends on the current flows on the fault side of the LV network).

If an earth electrode is really not an option, then the next best thing is to sense both the voltage between L and N but also sense the current flowing between the connected device and the incoming PE. This is the approach taken by at least one other stand-alone open PEN fault detection device, and seems inherently safer, as the leakage current flowing in the CPC is a pretty good analogue for the local PE potential, in terms of the real world electric shock risk. Coupling that with the phase to neutral voltage measurement and the same disconnection arrangement for both live conductors and the CPC, seems to be the best possible way of dealing with the risks from an open PEN fault to me (other than fitting an earth electrode and suitable RCD).

I agree, although I’m not a fan of connecting it to the MET (option 1) because you end up providing the earth for not just your own installation but also a return path for fault current in the event of a fault on the supply cable coincident with a loss of substation earth. There are also questions about maintenance and routine testing for earth electrodes that make them less suitable for domestic installations.

My charging point is from 2016 and still connected to the PME earth. Wishing I had got the guys who built the extension to bury an earth mat in some Marconite whilst they were digging.
I don't know if you've watched "droning John Ward's" YouTube channel recently, but he's been testing conductive concrete and a large conductive puck type earth electrode system. The results look promising and it could be a good solution for areas where it's too risky to bang a couple of rods in in order to get Ra down low enough. We're lucky here, as we have a stream running alongside the house, so a single rod pretty much reaches the water table and Ra is always very low. I put a replacement rod in for the garage and my wife's charge point a few weeks ago and Ra was just 24Ω and that was after several dry days. We also have no underground services, other than a run of 95mm² wavecon that comes in under our meter cabinet (which is some way from the house), so there are no issues with putting earth electrodes in, they are a dead cheap and effective way to tackle this issue.[/QUOTE]
 

·
Registered
Joined
·
268 Posts
I agree the voltage monitors will detect various faults including some open PEN. On that basis they are probably better than nothing, but far from discerning.

An open PEN fault where the imbalance doesn’t take the L-N voltage over or under limit will also not take the PE to earth voltage over 70v so is a lower risk.

Either way not a solution I’ll be using any time soon.
 

·
Registered
Joined
·
1,884 Posts
You can easily get a voltage greater than 70 V from the CPC to true earth, with the apparent voltage between line and neutral within the limit, with a broken PEN, depending on the phase balance in the local LV grid. It depends which phase the supply is on, and the IET have assumed that the risk of the charge point being connected to one of the phases that doesn't go out of range is small. I am completely unconvinced by this argument, as it seems quite probable that there will be a phase imbalance in the local grid, and even more probable that there may be a phase imbalance at the time there is a PEN fault (just because the PEN fault implies damage to the cable somewhere).

I'm not alone in this view, as the chap from Zappi felt so strongly about it that he built a test rig to prove the problem is real, and probable, and persuaded the Efixx people to make a video about it, to raise awareness of the issue. Now clearly Zappi have a vested interest in promoting their solution, but nevertheless it is pretty damning evidence that the matt-e approach doesn't work under all, or even perhaps most, occasions when a PEN fault occurs. Given that there are several hundred PEN faults reported on the network every year, then I think there is a need for reliable detection of them, with measures to reduce the risk of shock.
 

·
Registered
Joined
·
268 Posts
Thinking about it again the principle does depend on being on the right phase to detect the imbalance through measuring phase voltage.

The HSE don’t report the cause of PEN faults in the data I’ve seen but I’d think a significant proportion of them are to do with copper theft at substations rather than cable faults.
 

·
Registered
Joined
·
2,323 Posts
The Zappi 2 disconnects when it detects an over voltage situation. That said, with a smart meter it is possible to view the actual Grid voltage as is the case with some PV Solar Inverters. Having had this problem with my Zappi 2, I was able to get advice from MyEnergi on how to recalibrate their EVSE. I am not sure if this is possible with a Rolec. I should add that recalibrating a Zappi 2 requires nothing more than a few button presses - no screwdrivers involved. My Zappi 2 menu now shows the same Grid voltage as my smart meter.
 

·
Registered
Joined
·
1,884 Posts
The Zappi also has a far better open PEN fault protection system than the matt-e, or the other systems that are essentially just voltage sensing relays.
 
1 - 12 of 12 Posts
Top