TESLA Charger Install - No earth

[Airbag80]

Hi Guys,

I ordered my Model 3 LR a few weeks ago.... been on here for weeks, reading info, looking at the shipping news etc. Got my hidden VIN, so fingers crossed in a few weeks ill be on the road.

Anyways I thought this might be of help, I had a very mixed bag when having discussions with getting my TESLA charging point installed. Lots of electricians even the TESLA recommended guys made a bit of a meal out of the process of quoting. One electrician even suggested it would take him 3hrs to look at my house before he could quote., many guys couldn't even communicate, how people like this earn a living ill never know. Anyways.

So I thought I'd post on here what I learned. My install was fairly easy, consumer unit/main fuse is in my garage, and charger was going on outside wall of the garage.... In the end I got a EV specialist installer, install cost was £550 inc VAT.

What he used is the following unit... Made in the UK. BS:7671 18th Edition compliant
SP-EVCP-R from matt:e is a simple single connection unit, which allows for the connection of 1 x 32amp single phase charge point to the existing PME earthing facility.
http://matt-e.co.uk/wp-content/uploads/2019/02/SP-EVCP-R.pdf
Matt:e Single Phase EV Voltage Monitoring and Protection Unit with Type A RCBO (SP-EVCP-R) | CEF
Its £150 inc VAT from CEF.

Great bit of kit... removes the need for an earthing Rod....
Its got a Relay or Contactor as is known in the electrician trade... which disconnects the earth/live/neutral in the event of a fault.

So thats it for post number 1... hopefully this is of use..

Cheers,
A>

 

[Airbag80]

sorry... dont know how to edit posts... but to be clear install cost £550 inc VAT... included the cost of all materials (inc the matt:e unit)...

 

[Glan gluaisne]

The matt-e has been mentioned before as an apparent way to avoid the need for either an earth electrode or another form of Section 722 compliant solution. It's definitely not true to say that the unit is installed without an earth, though, it must be properly earthed to be safe to use.

The other discussion on the matt-e is here, and might be worth a read: Home Charging Stations (there are other posts in that thread discussing the matt-e and whether it really does comply with BS7671:2018 Section 722). My view is that at the moment it cannot be compliant with the regs as they are currently written, but it may become compliant if that section is amended, as it may be before long.

 

[Airbag80]

Thanks for the reply Jeremy ... sorry I should have been clearer. Of course it’s connected to the earth, and safe... or that’s what the Cert says anyways. I’ll do my reading on the other posts very helpful... if I’ve any concerns I’ll be getting the bloke back to sort it.

 

[MrBadger]

Just curious, where does that gets its 6mA DC protection? And I'm not sure what this offers over and above an earth road that seems to cost a few £ providing the ground is suitable.

 

[pdk42]

I thought the Tesla charger needed a type E RCBO (residual DC current protection)?

 

[MrBadger]

Its needs either Type A + 6mA DC protection (manufacturers marketing of said device calls this Type EV) or a Type B - now available at much more reasonable price than they were even 4 months ago.

So a straight Type A does not meet the regs imho unless this box also does something else that they are not shouting about, which of course you would if your device sorted that side of things too.

 

[Glan gluaisne]

Just curious, where does that gets its 6mA DC protection? And I'm not sure what this offers over and above an earth road that seems to cost a few £ providing the ground is suitable.

It doesn't. The matt-e is quite clearly (from the wiring diagram) only monitoring voltage. It has no current sensing at all, and seems to be assuming that an upstream PEN fault will create a voltage disparity that can be reliably detected.

The wording of the relevant bit of Section 722 of BS7671:2018 is, at the moment:

722.411.4.1 A PME earthing facility shall not be used as the means of earthing for the protective conductor contact of a charging point located outdoors or that might reasonably be expected to be used to charge a vehicle located outdoors unless one of the following methods is used :

(i) The charging point forms part of a three-phase installation that also supplies loads other than for electric vehicle charging and, because of the characteristics of the load of the installation, the maximum voltage between the main earthing terminal of the installation and Earth in the event of an open-circuit fault in the PEN conductor of the low voltage network supplying the installation does not exceed 70 V rms.

NOTE 1: Annex 722, item A722.2 gives some information relating to (i).

NOTE 2 : See also Regulation 64 1.5 when undertaking alterations and additions.

(ii) The main earthing terminal of the installation is connected to an installation earth electrode by a protective conductor complying with Regulation 544.1.1. The resistance of the earth electrode to Earth shall be such that the maximum voltage between the main earthing terminal of the installation and Earth in the event of an open-circuit fault in the PEN conductor of the low voltage network supplying the installation does not exceed 70 V rms.

NOTE: Annex 722, item A 722.3 gives guidance on determining the maximum resistance required for the earth electrode in (ii)


( iii) Protection against electric shock is provided by a device which disconnects the charging point from the live conductors of the supply and from protective earth in accordance with Regulation 543.3.3.10 l(ii) within 5s in the event of the voltage between the circuit protective conductor and Earth exceeding 70 V rms. The device shall not operate if the voltage exceeds 70 V rms for less than 4s. The device shall provide isolation. Closing or resetting of the device shall be by manual means only. Equivalent functionality could be included within the charging equipment

Where buried in the ground, a protective conductor connecting to an earth electrode for the purposes of (ii) or (iii) shall have a cross-sectional area not less than that stated in Table 54.1.

I believe that the matt-e device is attempting to comply with sub-section (iii) of the above (the bit I've underlined in the quote). The problem I have (and I'm not alone, I think) is that in order to comply with that sub-section a "true" earth reference is needed, i.e. the actual potential of the ground adjacent to the vehicle. Without an earth electrode this unit can only make a "guess" as to whether the touch voltage may have exceeded 70 V, by seeing an anomaly between line and neutral (as that's clearly all it's monitoring, the monitoring bit is the board to the left in the diagram below, which only has line, neutral and contactor switching connections):

 

[MrBadger]

It doesn't.

So for the Tesla wall connector, it sounds like its a non compliant install as no 6mA DC protection - ie it needs an additional >6mA DC protection mechanism, ie the Type A-EV or Type B RCD or a stand alone protection device?

 

[Glan gluaisne]

So for the Tesla wall connector, it sounds like its a non compliant install as no 6mA DC protection - ie it needs an additional >6mA DC protection mechanism, ie the Type A-EV or Type B RCD or a stand alone protection device?

The question is whether the matt-e box is a compliant stand alone protection device. I believe that @arg has mentioned that perhaps this part of Section 722 may be amended before long, so as to clarify how voltage sensing devices (like the matt-e) may be used to provide an equivalent level of protection. Using voltage sensing isn't new, as before we had reliable RCDs houses with TT earthing systems used a voltage sensing earth fault protection device. Those did rely on having a good earth electrode connection to local earth though, something the matt-e device doesn't use.

The bottom line is that the requirement is to prevent any metalwork that can be touched whilst standing on the ground outside (like the car body) from reaching a dangerous touch voltage (over 70 V). This can be done by monitoring the AC and DC leakage current, or by monitoring voltage. Until recently, monitoring AC and DC leakage current has been the preferred way, but it seems that the makers of the matt-e unit believe that their voltage monitoring system can provide an equivalent level of protection.

 

[bangtidy]

I've also just had one of these units fitted along with a Tesla charger that I was lucky enough to be gifted by my borther in law.

Unfortunately though, I'm experiencing some issues with the matt:e box randomly cutting the power to the charger. Presumably that means that it's either detecting a fault with the supply voltage or it's faulty itself. Sometimes it seems to work though which is puzzling, but if it's not working reliably I can't use it at all.

When the power gets cut I have to turn the matt:e unit off and then back on again to get power back to the charger.

The matt:e is suppled from a 32A fuse in a consumer unit about 4m away. The consumer unit itself is supplied by an underground cable to the garage from the house that itself is connected to a 100A fuse that goes straight back to the supply at the meter. Never had problems before with the power in the garage so I 'm a bit stumped.

Hopefully the electrician that installed it will be able to figure out what's wrong but if anyone has any ideas in the meantime I'd love to hear them!

 

[bangtidy]

Incidentally, not sure if this is relevant at all, but the one that I've been supplied with seems to be the one that incorporates a type A RCBO ie. their SP-SCVP-R model.

 

[rincewind]

That £550 install cost - isn't it covered by the government OLEV grant in the UK? My Rolec unit was a no cost install (install charge was £480)

 

[Glan gluaisne]

Incidentally, not sure if this is relevant at all, but the one that I've been supplied with seems to be the one that incorporates a type A RCBO ie. their SP-SCVP-R model.

If the RCBO (shown in the wiring diagram for the model you have below) isn't tripping, but the voltage sensing circuit is triggering (which seems to be the case) then that sounds like either you have a fault that is causing a big enough voltage anomaly to trip the matt-e, or the voltage sensing part of the matt-e may not be functioning as it should.

It should be possible to test the installation fairly quickly to see if there is a fault somewhere upstream of the unit, in particular I'd get the electrician to check the loop impedance, as that may show up a possible cause for the trip. If it were me, the first thing I've do would be a physical inspection to make sure everything looks to be in good order, followed by a check on the supply voltage at the unit, then do some loop impedance testing.

 

[MrBadger]

I'm well past the earthing issue, I'm questioning the 6mA DC protection from Tesla Wall Connector - Type B / Type A-EV RCD with Tesla stating that their units must be fitted either with Type A-EV (one of a kind and hence very expensive) or Type B (not quite so expensive) RCD protection and @arg agreeing that a basic Type A was not compliant - also backed by other EV charge point suppliers that I have communicated with.

 

[Glan gluaisne]

I'm well past the earthing issue, I'm questioning the 6mA DC protection from Tesla Wall Connector - Type B / Type A-EV RCD with Tesla stating that their units must be fitted either with Type A-EV (one of a kind and hence very expensive) or Type B (not quite so expensive) RCD protection and @arg agreeing that a basic Type A was not compliant - also backed by other EV charge point suppliers that I have communicated with.

The point is that you don't have to have an expensive type B/Type EV RCD/RCBO with a DC capability, you can opt to detect an unsafe touch voltage as an alternative.

It's this specific sub-section that allows for voltage sensing as a suitable means of protection when the earth provided by the incoming supply is used, and it would work every bit as well with a Tesla charge point as it would with any other make that doesn't have built-in protection (as many now seem to have):

( iii) Protection against electric shock is provided by a device which disconnects the charging point from the live conductors of the supply and from protective earth in accordance with Regulation 543.3.3.10 l(ii) within 5s in the event of the voltage between the circuit protective conductor and Earth exceeding 70 V rms. The device shall not operate if the voltage exceeds 70 V rms for less than 4s. The device shall provide isolation. Closing or resetting of the device shall be by manual means only. Equivalent functionality could be included within the charging equipment

 

[bangtidy]

Thanks for that advice Jeremy - much appreciated. I'll post back here if and when I manage to get this resolved.

 

[bangtidy]

If the RCBO (shown in the wiring diagram for the model you have below) isn't tripping, but the voltage sensing circuit is triggering (which seems to be the case) then that sounds like either you have a fault that is causing a big enough voltage anomaly to trip the matt-e, or the voltage sensing part of the matt-e may not be functioning as it should.

...and yep, this is exactly what's happening.

 

[Airbag80]

Hi Jeremy, I’ve been reading the other posts. And the ones above. So it seems like the Matt-e is compliant to the regs if interpreted a certain way? And the Voltage sensing fiction is a reasonable approach to safety. (I think so)

I was wondering should I get the bloke back to rip it out and ask for different install? ... but to be fair seems like the unit gives a lot more protection than if I got an EV Charing point installed a few years ago without any sort of contactor.

 

[arg]

The point is that you don't have to have an expensive type B/Type EV RCD/RCBO with a DC capability, you can opt to detect an unsafe touch voltage as an alternative.

This isn't right - they are separate requirements.

Earthing is in clause 722.4 ("protection for safety"), and in particular the prohibition of PME earthing in 722.411.4.1. It is the prohibition of PME earthing that can be waived by use of the unsafe touch voltage detection (722.411.4.1(iii)).

The requirement for RCDs is in 722.5 ("Selection and erection of equipment"), in particular 722.531.2.101. This requires an RCD of "at least Type A" for all types of EV charging (except where "electrical separation" is used, ie. an isolating transformer), and then for any chargepoint with EN62196 plugs or sockets (ie. Type1 or Type2) it imposes an additional requirement of Type B or DC protection.

As to whether or not this is reasonable:

• It is reasonable that the two requirements are separate. The concern about the PME earth relates to open neutrals, and no type of RCD will protect you from that problem. Conversely, parts of the charging system (cables etc.) are not protected by earthing so they need an RCD (potentially this could have been avoided if cables with earthed armour/screen were adopted, but I've never seen that advocated and it's reasonable that the regulations ignore this unlikely possibility).
• The requirements about earthing are backed by data showing how often open neutrals occur and reasonable analysis of the likely rate of incidents with a larger EV population. As with all risk assessments, you can argue over the acceptable level of risk.
• The requirements for RCDs are more complex. I think there's no difficulty arguing for the need for a basic RCD. Likewise, raising the bar to Type A is easy to justify as clearly there are many plausible faults in an EV charger that could result in pulsed DC leakage currents and Type A aren't that expensive anyhow; the only questionable issue is that the argument for needing Type A applies equally to a huge range of modern appliances, not just EV charging. The step to Type B is much harder to justify, and in particular the fact it is only mandated for a chargepoint with Type1/Type2 connectors but not, for example, a commando socket.
• There are conceivable fault scenarios in an EV charger that could give rise to pure DC leakage, but they are probably rare, are probably equally likely in a range of other domestic appliances, and certainly equally likely regardless of whether the EV has a commando or Type2. This doesn't in fact seem to be the justification (see below).
  
• One of the concerns with small amounts of DC leakage is that the DC leakage itself might be harmless (and so not noticed) but have the effect of saturating the coils of standard RCDs, preventing those RCDs (anywhere in the installation) from reacting to ordinary AC faults that might arise. To meet the standard of Type A, an RCD is required to continue working (detecting AC faults) in the presence of up to 6mA of DC. So if this is indeed a concern, it makes sense to require either a Type B (which will react to all faults AC or DC), or a Type A plus separate arrangements to handle the case of >6mA DC.
• The one rational explanation I have heard for the distinction between Type1/Type2 connectors vs commando sockets etc. is that the EN62196 charging standard has the pilot signal fed with 12V DC relative to earth, with a 1K source impedance. This means that if you short out the pilot signal, you will get a 12mA DC current. If you short it to neutral rather than earth, you will get that 12mA flowing through the earth/neutral bond and so it will be seen as a 12mA DC leakage by any RCD in the circuit. So the failure scenario is a cable on your chargepoint where the insulation has been heavily damaged: the pilot shorts to neutral, disabling the RCD and preventing charging but not actually tripping anything; the user then goes to see what's going on and grabs hold of the live - and gets electrocuted because the RCD isn't working to save him.
• Personally, I find this argument a bit unsatisfactory. On the one hand, if you believe this really is a problem then it's still going to be a problem with commando sockets, because the way commando sockets are going to be used in practice is with a portable EVSE which then brings exactly the same problem. It would maybe make sense if EN62196 mandated DC leakage detection in portable EVSE, but currently (and historically) it does not. On the other hand, the level of risk seems quite small in either case - it's true that 12mA of the maximum possible DC induced by a fault on the pilot signal is greater than the 6mA minimum tolerated by a Type A RCD, but not my much: real-life faults will give slightly less than 12mA, and real-life TypeA's will tolerate rather more than 6mA, so in many cases the worst-case fault will still end up safe.

So in summary, yes the regulations do require both the superior RCD and a solution to the earthing issue, and yes there are reasons for both, but the level of risk being avoided is relatively low.