Consultation on Electric vehicle (EV) chargers for residential applications (DZ 6011:2020) closes 22/01/2021

Forums › Transport (cars, bikes and boats) › Consultation on Electric vehicle (EV) chargers for residential applications (DZ 6011:2020) closes 22/01/2021

1 | 2 | 3

cthombor

81 posts

Master Geek

Lifetime subscriber

#2633070 10-Jan-2021 13:15

SomeoneSomewhere:

PE and neutral must never be connected downstream; they are connected in one location only: the MEN link in your MSB.

RCD is unrelated to this. The only reason I mentioned it (I shouldn't have) is that it's easy to accidentally trip the RCD when testing the upstream earth-neutral link so your test circuit needs to be appropriately designed.

RCDs only trip on current imbalance. Nothing else at all. They have no connection to earth and do not care about earth, for about the fifth time.

Hmm, I now see that I was confused by the 2.74k resistor from CP to PE in the J17772 wiring diagram (from the not-entirely reliable Wikipedia article, copied below).

SAE J1772 Wiring Diagram from Wikipedia

You're right, this circuit isn't showing any connections to earth ground so -- once again -- I'm back to square-one in my (mis)understanding of how this could be protecting against a (true) ground fault, or for that matter against faults in the earth and neutral inputs to the EVSE (except for the faults which are covered by the upstream RCD-A).

Nevermind, there's no need to reply, I really don't need to understand how this circuit operates! In my planned submission on this NZ PAS, I'll restrict myself to noting that I see no definition of the sort of "effectiveness" that must be monitored continuously for "the earthing connection to the vehicle" -- in particular I'm wondering if the intent is to impose a requirement for continuous monitoring of any touch-voltage hazards on the vehicle being charged, or if the intent is instead to require that the charging cease in a failsafe manner if the CP-PE circuit between the EVSE and the vehicle becomes inoperative.

SomeoneSomewhere

1785 posts

Uber Geek

Lifetime subscriber

#2633076 10-Jan-2021 13:28

You're misusing terms and it's confusing you.

A ground fault is a fault to ground. RCDs protect against that well, because it causes a current imbalance as some current is passing through ground, not returning via neutral. It needs to be alternating current or pulsating DC to trip a Type A RCD; high frequency or smooth DC may or may not operate it.

The 'earthing connection to the vehicle' is quite clearly simply the earthing connection between the EVSE and the vehicle. This is monitored by monitoring that 2.74k resistor. They make no mention of touch voltages nor of independent earths; they simply want all power to be disconnected if that earth conductor goes open circuit.

cthombor

81 posts

Master Geek

Lifetime subscriber

#2633078 10-Jan-2021 13:30

SomeoneSomewhere:

RCDs only trip on current imbalance. Nothing else at all. They have no connection to earth and do not care about earth, for about the fifth time.

Hmm, you may be interested to read this

https://www.standards.org.au/StandardAU/Media/SA-Archive/Documents/Wiring_Rules_FAQs_March_2010.pdf:

FAQ 007/2009: AS/NZS 3000:2007—CLAUSE 3.8 & TABLE 3.4 & CLAUSE 5.2.2

3.8 IDENTIFICATION

TABLE 3.4 CONDUCTOR COLOURS FOR INSTALLATION WIRING

5.2.2 EARTHING FUNCTIONS—Functional earthing (FE)

Question 007/2009. If an RCD is provided with a functional earthing conductor (typically coloured white or pink), may this conductor be connected to an earthing terminal/connection or bar?

Answer. Yes. The purpose of the conductor is to allow a return path for the electronic circuit, if the neutral is disconnected, to ensure the RCD still operates correctly. The conductor does not perform a protective earth function.

I'm happy to accept that such "VD with FE RCDs" are rarely -- if ever -- found in NZ. I'm also happy to accept that their use in a residential NZ EVSE circuit would not mitigate many of the risks of a poorly-earthed residential supply to an EVSE.

SomeoneSomewhere

1785 posts

Uber Geek

Lifetime subscriber

#2633081 10-Jan-2021 13:39

Yes, I avoided mentioning those. AFAIK the only RCDs that actually have those connections are currently illegal for other reasons (does not switch both earth and neutral, although many are still installed), and it still does not mean that they actually use it for much; it's just to supply the electronics in the event of a lost neutral.

elpenguino

3389 posts

Uber Geek

#2633303 10-Jan-2021 23:36

cthombor:

Hmm, you may be interested to read this

https://www.standards.org.au/StandardAU/Media/SA-Archive/Documents/Wiring_Rules_FAQs_March_2010.pdf:

FAQ 007/2009: AS/NZS 3000:2007—CLAUSE 3.8 & TABLE 3.4 & CLAUSE 5.2.2

3.8 IDENTIFICATION

TABLE 3.4 CONDUCTOR COLOURS FOR INSTALLATION WIRING

5.2.2 EARTHING FUNCTIONS—Functional earthing (FE)

Question 007/2009. If an RCD is provided with a functional earthing conductor (typically coloured white or pink), may this conductor be connected to an earthing terminal/connection or bar?

Answer. Yes. The purpose of the conductor is to allow a return path for the electronic circuit, if the neutral is disconnected, to ensure the RCD still operates correctly. The conductor does not perform a protective earth function.

Protective and functional earths are two different animals.

Read this brief explanation of the difference.

Most of the posters in this thread are just like chimpanzees on MDMA, full of feelings of bonhomie, joy, and optimism. Fred99 8/4/21

cthombor

81 posts

Master Geek

Lifetime subscriber

#2633489 11-Jan-2021 11:49

elpenguino:

Protective and functional earths are two different animals.

Read this brief explanation of the difference.

Thanks for the reference. But... I have learned (the hard way!) not to rely on Wikipedia -- except as a starting-point. In this particular case: when I get-round-to-it I might take the time required to amend the section on "Equipment Grounding" -- to make it less US-centric, and so that it'll have a reference to something more authoritative than some no-longer-available course-notes from a U Wisc extension course. The section on Functional Earthing is a real mess IMHO, as it references only a very stale version of BS7671 -- so has limited relevance here in godzone and would be downright confusing for anyone outside the UK's orbit. If you take a look at the Wikipedia article on Isolated Ground you'll get an idea of the complications facing anyone (not me!) who would attempt to amend the section on Functional Earthing in the Wikipedia article you reference.

The section on Earthing and Bonding in the Wikipedia article on electrical wiring in the UK is perhaps somewhat more relevant and reliable than the source you cite... but ... of course here in godzone we don't follow the UK wiring code, and we have developed somewhat different terminology over the years!

But I do totally agree with you: there are very important distinctions to be made when speaking about "earth". PE has a definite meaning in NZ. FE has a definite meaning in the context of a particular electrical part, such as a particular type of RCD, which has this label on one of its terminals. A reputable vendor (NEXANS) offers the following definitions.

Protective Earthing concerns safety i.e. low frequency currents with a high level of magnitude which present a safety hazard to personnel.

Functional Earthing concerns other effects such as EMC i.e. high frequency currents with low levels of magnitude, which may not constitute a hazard to personnel but can degrade data signalling. Functional Earthing also relates to the provision of reference signalling voltages or return paths.

But... back to the subject of this thread! I started this thread because I was genuinely confused over the intended meaning of requirement 1.7.b on IC-CPDs, that they must "incorporate or provide a system that continuously monitors earth continuity and automatically disconnects the supply in the event that the earthing connection becomes ineffective".

I had (incorrectly!) guessed that the requirement was to require the IC-CPD to ensure (somehow!) that its (upstream) earthing terminal is effectively connected to earth in the vicinity of the IC-CPD.

It has now become abundantly clear to me that the intended meaning of this requirement is that the IC-CPD must have means to ensure that its (downstream) PE terminal has a low-impedance connection to its (downstream) CP output -- and of course this would allow the PE to be "effective" at providing protection (if we're happy to believe that the IC-CPD's upstream earthing terminal is actually connected to earth somewhere nearby).

So... I'm left with two concerns. One is definitional -- and would be easily addressed by a revision to the NZ PAS if its authors accept that other readers might be confused. The other is safety. What risk do I run whenever I plug an IC-CPD into an NZ outlet at a friend's house, or at a motel, if (for whatever reason!) it is not actually earthed? I'm half-tempted to see what happens if I plug one of my IC-CPDs into a socket that doesn't have an effective connection to earth -- that test gear is easily rigged and what could possibly go wrong? Well in best-case my IC-CPD will flash a warning light, or an RCD might trip... but in worst case? Yeah... well ... as Dr Magic says, "Don't try this at home"!

However I just might get-around-to measuring the DC voltage between my Leaf's chassis (while it's charging) and the earth terminal on the IC-CPD supply. I really doubt it'll be in the tens of volts but I also really doubt it'll be below 1V. Any bets? ;-)

gregmcc

2145 posts

Uber Geek

ID Verified

Trusted

Lifetime subscriber

#2633734 11-Jan-2021 16:42

cthombor:

However I just might get-around-to measuring the DC voltage between my Leaf's chassis (while it's charging) and the earth terminal on the IC-CPD supply. I really doubt it'll be in the tens of volts but I also really doubt it'll be below 1V. Any bets? ;-)

this may not be easily done, how are you going to access the earth terminal in the IC-CPD with something plugged in to it?

Got to remember the whole point of the earthing system is to (try) to keep and exposed conductive items that are in contact with the ground at the same potential as the supply earth, so that in the even of and earth fault (or leakage to earth) the voltage rise doesn't exceed 50V, this is achieved by putting in an earth pin, bonding (earthing) metal water pipes that go in to the ground. In the 50's, 60's and 70's there were lots of parallel earth returns, water pipes etc were all metal. The modern house construction has very little in the way of conductive connections to the ground, this is why the earth pin is so important, and this is why there has been recently been a bit of talk about using the TCNS system (as opposed to the MEN system) in order to ensure a good earth return.

Why is there a proposed requirement to monitor the earth continuity? Think about the construction of the electronics in EV's, the inverters, chargers and batteries are liquid cooled, it's not uncommon for seals to fail and when that happens this can liven the car body. Plug a charger in and although the car is isolated from earth (the tyres) there would exist a potential difference between the system earth and the body of the vehicle. So this is the reason why there should be an earth continuity check going on before the power is put on to the car, so if there is an earth fault in the EV the RCD will see the imbalance and trip, rather than waiting for an unsuspecting person to touch the car and create the imbalance.

Quic Broadband

Move to New Zealand's best fibre broadband service (affiliate link). Note that to use Quic Broadband you must be comfortable with configuring your own router.

SomeoneSomewhere

1785 posts

Uber Geek

Lifetime subscriber

#2633783 11-Jan-2021 17:25

I had (incorrectly!) guessed that the requirement was to require the IC-CPD to ensure (somehow!) that its (upstream) earthing terminal is effectively connected to earth in the vicinity of the IC-CPD.

It has now become abundantly clear to me that the intended meaning of this requirement is that the IC-CPD must have means to ensure that its (downstream) PE terminal has a low-impedance connection to its (downstream) CP output -- and of course this would allow the PE to be "effective" at providing protection (if we're happy to believe that the IC-CPD's upstream earthing terminal is actually connected to earth somewhere nearby).

What? No, we've told you about 3x that's not the case. It's checking that the upstream earth is connected as far as the main earth bar and MEN link.

or an RCD might trip... but in worst case?

As noted, RCDs can't trip on earth loss and aren't connected to earth except for a very small number and not for the reasons being discussed here.

cthombor

81 posts

Master Geek

Lifetime subscriber

#2634298 12-Jan-2021 13:38

gregmcc:

Think about the construction of the electronics in EV's, the inverters, chargers and batteries are liquid cooled, it's not uncommon for seals to fail and when that happens this can liven the car body. Plug a charger in and although the car is isolated from earth (the tyres) there would exist a potential difference between the system earth and the body of the vehicle. So this is the reason why there should be an earth continuity check going on before the power is put on to the car, so if there is an earth fault in the EV the RCD will see the imbalance and trip, rather than waiting for an unsuspecting person to touch the car and create the imbalance.

Thanks, your explanation has helped me clarify some of the risks I'm wondering about.

Consider the following cases, all of which are based on the following scenario.

Scenario: I have driven my 24 kWh Leaf to a friend's house that's about 50 km away from my house. Because my battery has only about 15 kWh of usable capacity (SOH 70% / 9 bars), I'd have severe range-anxiety if I were to try to drive home without a recharge -- even if it's a warm summer day, and even if there's only a 200m elevation difference between our two houses. Unfortunately, there's no public-charge facility within a 40 km radius of my friend's house. My friend offers me the use of a wall socket in the garage. My visual inspection tells me that it's a standard 10 A socket. I have an IC-CPD with a display showing its charging current and its supply voltage. When this IC-CPD is in "8 A" mode, its display informs me that it's charging at 7.2 A.

Cases:

1. Charging an EV on a socket which feels loose, and which has no dedicated RCD-A. Yikes! My reaction if faced with an option to charge my Leaf in this case would be, in the first instance, to attempt to enter case-2. I'd also explain to my friend why I think it'd be a good idea to avoid using this socket until it has been inspected by a registered electrician. If I can't perform a case-2 charge at my friend's place, or at any friend-of-friend's place, then... I'd (partially!) mitigate the risk of a case-1 charge by a) charging my car (if possible) inside my friend's garage on a dry floor; b) using my portable RCD-A (if I remembered to bring it, and if my friend's wall-socket doesn't seem to have a local RCD-A); c) confirming that my IC-CPD doesn't show a fault-indicator when I plug it in; d) confirming that the AC supply voltage is currently between 220 and 240 (fortunately my IC-CPD displays this on its screen); e) looking for fault-indications (audible arcing in the socket, on the display of my IC-CPD, and on the displays of my Leaf) when I attempt to commence the charge; f) confirming that the charge is at 7.2 A, as displayed on my IC-CPD; g) monitoring the temperature of the plug (by touch) after 5 min, 15 min, 30 min, 1 hour, then hourly thereafter. Rather a difficult procedure to follow but hey, safety-first. And I really do want to visit my friends, even if I'm not confident of the electrical supply at their place!

2. Charging an EV on a socket which "feels" ok (and doesn't audibly arc), which is out of range of my IC-CPD cable but which is close enough to my Leaf that I can use a 3m extension lead with no (immediately-obvious ;-) safety defects. But: there's no dedicated RCD-A on this socket. Yikes! I'd try to enter case 3 below, but if I can't... I'd discuss the fire and electrocution risks with my friend. We'd probably agree it'd be better to charge my Leaf in this (rather unsafe) way, than to go to the (considerable!) expense and delay of arranging a tow to the nearest public charge-point or all the way back to my home. (If we were in Wellie, we'd have a third option of getting a charge from a mobile EV charger -- which is rather a spendy beast, but perhaps Auckland will someday have enough EV owners to make these economically feasible here?)

3. Charging an EV on a socket which has no apparent defects and which is equipped with a (possibly portable) RCD-A. The charging risks in this case are considerably lessened over case-2 charging... but I'd still be at least partly exposed to two not-quite-insignificant hazards: a) my friend's socket may not have a low-impedance path to earth (through the MEN-tie in their residential supply), b) there's no RCD-B on the supply to my IC-CPD, and c) there may be a major fault in the insulation of the 480 VDC circuit on my Leaf which (for some reason!) hasn't caused it to enter a failsafe state with DTC code P0AA6 HYBRID BATT VOLT SYS ISOLATION, "VCM detects a insulation resistance calculated based on IR sensor signal sent from Li-ion battery is 380 kW or less." Well... the risk of a) would be at least partly mitigated if I could get an RCD-B supply at my friend's place (or at some nearby friend-of-friend who has a mode-3 charger that had been installed to Worksafe May 2019 requirements)... and "Someone somewhere" in this forum has (repeatedly!) asserted that I need not worry about the "earth continuity from the EVSE to the supply source" because my IC-CPD (perhaps in conjunction with an RCD-B) is adequately performing this verification. This person may well be correct in their assertion but I still am unconvinced due to a lack of any explanation that I can understand, and due to my observation that the Worksafe 2019 guidelines have three separate safety steps: verifying the "earth continuity from the EVSE to the supply source", "verifying the RCD type and residual current rating for all RCDs", and verifying "the presence of an earth continuity monitoring system in the EVSE". I'm aware that a ground-loop impedance (as might be measured between the ground near my car and the earth supply on the AC socket that's supplying my IC-CPD) *might* reveal a dangerously-high impedance -- at which point the ground-fault currents through my body *might* reach dangerous levels without triggering any RCD or by any safety-circuitry in an IC-CPD (because in the case of a high-impedance earth connection in the supply then the MEN-tie in this system could effectively be through me if I'm touching the chassis. In other words: my Leaf cannot be equipotentially bonded to earth through an AC socket which provides only a high-impedance path to earth.

4. Charging an EV on an (apparently recently and well-installed) socket which has its own RCD-A, using an IC-CPD which isn't obviously unsafe (not one of those banned units which may have an unpersuasively pasted-on label claiming "230 VAC 50 Hz", but which is clearly Nissan-standard gear that was designed to run only on their domestic 100 / 200 VAC, 50 / 60 cycle supplies).

5. Charging an EV on an (apparently recently and well-installed) socket which has its own RCD-A, using a (spendy, rarely seen) IC-CPD with a built-in RCD-B or RDC-DD.

6. Charging an EV in full compliance with DZ PAS 6011.

Here's a quiz on your personal risk-assessments: what is the *lowest* numbered case in which you'd be willing to charge your EV?

cthombor

81 posts

Master Geek

Lifetime subscriber

#2634311 12-Jan-2021 14:07

A friend of mine recently had a very bad experience with a case-1 charge (as defined in my previous post). Here's my best understanding of his experience:

Overnight (unattended) charging of a Leaf using an IC-CPD that's plugged into a socket that was wobbly.

* This was at a motel somewhere in NZ, and I won't reveal any possibly-identifying details here!

* I don't know whether any or all of the following were checked before going to bed: the IC-CPD's display (charging current, charging voltage, fault-indicator light), the Leaf display (three large blue LEDs visible through the windscreen) or the distinctive "thunk" noise a Leaf makes at the start and end of every charging session.

* In the morning, the Leaf was unresponsive -- wouldn't start. My friend called a tow truck. The helpful towie found no fault in the 12V supply, so plugged his OBD2-laptop testgear into my friend's Leaf... and hey presto the Leaf would once more take a charge!

* I'm *guessing* that the towie's testgear had cleared some DTC in the Leaf.

My friend's Case-1 charging misadventure has persuaded him, and his partner, to sell their Leaf and to purchase a petrol-fuelled auto -- perhaps a hybrid.

Being a tech-geek who loves a challenge... I have learned how to hard-reboot my Leaf. Will treat it the way I treat a Windows PC -- if there's no obvious fault (e.g. in the 12 V system, whose voltage I do monitor!) then reboot and hey maybe it'll start working again. Well... if you're using a faulty (e.g. root-kitted) PC you're "only" risking your finances and personal-privacy, whereas if your hard-boot clears the DTCs on a Leaf which had entered a failsafe immobilised mode because of a dangerous fault in your Leaf, then you're at mortal risk. The underlying fault in your Leaf is probably not even diagnosable until it rears its ugly head again and resets the DTC you had so foolishly cleared... so my intention is to go to the (major) expense of engaging a towtruck unless I can figure out how to read and record the DTCs before rebooting. I have an OBD2 that I trust only enough to use once a month, while my Leaf is parked -- so that I can contribute fully to FlipTheFleet. So I guess I'll now upgrade to LeafSpy Pro so that I'm able to read DTCs on my (not yet campervan-converted) e-NV200 on long trips within NZ, that'll mean many charge-sessions in cases that aren't fully compliant with recommendations in DZ PAS 6011, and a dodgy-charge session that immobilises my EV could happen even on a mode-4 charger if the voltage goes out of its design range... and I don't know of any EVSE that was designed to operate over the full 200 - 250 VAC range allowed for AC power supply in NZ!

1 | 2 | 3