2012 Leaf killing EVSEs

I have a 2012 Leaf which has now blown up 3 EVSEs. The car is an ex-Japan grey-market import and I'm using it in NZ (a 230V 50Hz country). Two of the EVSEs are the stock JDM Nissan plug-in 12A one, and the third is a locally sourced 6A one. The first died a few months ago, the second a couple of weeks later, and the third the first time it was used. In all cases the car was charging on timer in the middle of the night, and gained some extra range before the EVSE died, so it wasn't at turn-on.

I've pulled apart one of the dead EVSEs, and the failure is in the primary coil of the transformer which powers the low-voltage signal circuit. The primary's open circuit, and the transformer smells burnt. When powered directly with 18VAC the low voltage system still seems to work (at least the LEDs come on - haven't plugged it into the car), and the current draw is low - 70 mA or so.

Since it's killed multiple EVSEs of more than one brand, I assume the problem's not the EVSE, it's either the wall power or the car. I can't measure anything wrong with wall power, and the circuit is a new install by a registered electrician. It seems unlikely that surges would be killing a transformer, and in any case the EVSE has surge protection built in, and nothing else in the house is dying. More likely is that the car is drawing too much current through the signal circuit and cooking the transformer.

Anyone recognise the symptoms, or have any ideas what could be causing this and how to check it?

Thanks in advance
Dave

Distraxi said:

I have a 2012 Leaf which has now blown up 3 EVSEs. The car is an ex-Japan grey-market import and I'm using it in NZ (a 230V 50Hz country). Two of the EVSEs are the stock JDM Nissan plug-in 12A one, and the third is a locally sourced 6A one. The first died a few months ago, the second a couple of weeks later, and the third the first time it was used. In all cases the car was charging on timer in the middle of the night, and gained some extra range before the EVSE died, so it wasn't at turn-on.

Click to expand...

Uhh... is the "stock JDM Nissan plug-in 12A one" only a 100 volt EVSE? IIRC, typical Japanese power is 100 volts at 50 or 60 hz, depending on the region (I've visited Japan 4x but only stayed in hotels). Feeding in 200+ volts will definitely fry it relatively quickly, just like it will fry a US 120 volt stock L1 EVSE (60 hz in the US): http://evseupgrade.com/?main_page=faq_info&fcPath=4&faqs_id=24" onclick="window.open(this.href);return false;.

I know nothing about the "locally sourced 6A one". Make and model? Have some pics of the labels of these?

cwerdna said:

is the "stock JDM Nissan plug-in 12A one" only a 100 volt EVSE?

Click to expand...

No, it's rated 200V, 15A. Japan runs both 100V and 200V domestic supply.

It's a little over rated voltage when used at 230V, but using the 200V JDM charger and just changing out the plug for a NZ one is the standard approach here, and it doesn't seem to have caused a general problem. Almost all Leafs in NZ are grey market JDM, as official dealer pricing has been crazy high till recently.

In the specific case of the signal voltage system, the transformer is almost certainly just the same as a 230V-spec one, just with a few less turns on the primary, so the extra volts shouldn't be hurting it. Its output voltage will be a little higher than spec, so the 12V regulator will be working a bit harder, but the regulator's not the bit that's died.

There have been several reported cases of the 2011 / 2012 on board charger sensor voltage detection failing. In these failures the on board charger was reading higher voltage than the EVSE was providing.
See http://www.mynissanleaf.com/viewtopic.php?f=30&t=15372&hilit=Chattanooga+dealer&start=60#p401360" onclick="window.open(this.href);return false;.

Seems unlikely that specific failure would cause the EVSE failures you have experienced.
But would seem likely that some type of failure in the LEAF on board charging system is causing the EVSE failures.

You should also note that there was an error in the 2011 / 2012 software related to implementation of L2 charging. But that error resulted in failure of the LEAF on board charger when used with some GE EVSE.
The P3227 software change includes correction of that.
Seems a bit unlikely that would be causing your EVSE failures though.

Distraxi said:

It's a little over rated voltage when used at 230V, but using the 200V JDM charger and just changing out the plug for a NZ one is the standard approach here, and it doesn't seem to have caused a general problem.

Click to expand...

I don't buy the "standard approach" at all.

It looks to me as though Nissan (actually Panasonic who makes the EVSE that comes with the car) used a different transformer for the Japanese market. It's obviously failing at the higher voltage. The "works for a while then dies" failure mode is the same for an unmodified EVSE that's had it's plug changed to 240V here. Ask Ingineer (surprised he hasn't weighed in yet, he probably will) who modifies these things for a living!

The diode failure of the charger won't kill the EVSE. The car won't charge from an EVSE that checks. The 2012 EVSE does not do the diode check so this doesn't apply here.

The voltage sensor failure causes excessive current to be drawn on the power lines if the voltage is 120-208V. That's not the case here. I believe if you had this failure mode you wouldn't be able to charge as ~230V should be enough to trigger an erroneous "high mains voltage" DTC. You can check for incorrect voltage sense with LeafSpy and look for DTC's with LeafSpy pro. Further, even if the charger pulled excessive current from the EVSE (18 amps maximum), you'd probably see it as a contactor failure.

Would love to see a picture of the nameplate of the 6A "local" EVSE.

My name is not Ingineer, but I am an electrical engineer, familiar with power electronics and magnetics.

Based on your problem description, I think it is highly likely that your LEAF is totally innocent, and that the real root cause is in fact the 230V 50Hz vs. 200V EVSE input rating. Connecting a 230V nominal 50Hz mains up to a 200V rated transformer is likely to cause problems, especially if kept plugged in for many hours at a time (or 24/7).

Transformers are normally designed so they operate just below (or mildly into) magnetic "saturation". The core can only withstand a certain volt-second product, and if you subject it to too much (ex: by applying a higher voltage than it is rated for, or applying the correct voltage, but at too low of a frequency), the core will saturate and the primary side magnetization current will increase exponentially.

Your 230V 50Hz mains has a tolerance (ex: +/-5% or something) associated with it, and the actual supplied voltage will drift throughout the day. In the middle of the night, the voltage will often be highest, since electrical loads on the system will be less, causing less resistive voltage loss through the distribution system. Therefore, your 230V may actually be up to ~241V or higher in the middle of the night, and this will heavily bias the transformer into saturation. This can easily cause 2x-3x the magnetization current from what it was designed for at 200V. However, since resistive losses in the primary coil go as the square of the current (P = I^2 * R), the primary winding copper losses will increase 4-9 times normal. Under normal operation, the copper temperature rise might by a manageable 30C, but at 7x dissipation, you get 210C rise above ambient, which will melt the enamel coating on the magnet wire (on the primary winding). Once the insulation melts, adjacent turns (or layer to layer) shorts will occur, causing massively higher currents, causing more melting/arc flash/vaporization of the copper wire, which eventually results in an "open circuit" measurement on the primary winding. The secondary winding doesn't see any of the magnetization current (and heating), so it can remain unharmed. This matches your described symptoms perfectly.

Chances are you can fix your problems by repairing one of your broken EVSEs by replacing the broken transformer with a proper 230V 50Hz transformer (with same or higher output current and same output voltage rating as the original), and everything will likely work fine.

There is a chance that replacing the transformer might not 100% fix everything, as there may be something else inside the EVSE that may also be running directly from the 230V AC (perhaps a relay coil). If there is a relay coil running from the AC directly (instead of the low voltage DC), you can add a small resistor (but relatively high wattage, say 1W) in series with the relay coil, to dissipate the extra ~30V AC.

how about the 50 hz vs 60 HZ? isn't japan 60 HZ? Running any transformer or anything with a coil usually overloads things if you run it lower than its rated frequency. Might not be the whole problem, but it certainly exacerbates it.

@JeremyW, @ExpensiveLettuce,

I agree with your comments, and yes, running over rated voltage is likely hard on the transformer.

My biggest problem with it is that my car's blown multiple EVSE's in a fairly short period of time, and there seems to be no local knowledge of any of the other 100 or so JDM Leaf's running round NZ having blown any. If it was a line voltage problem, that would require my house to be running higher peak voltages than anyone else. Which is certainly possible, but not a default theory till I've done some more diagnostics. Certainly during the day I'm seeing the 232-235V that's typical for urban NZ.

I don't have the NZ spec charger it blew up, that's gone back to the supplier for analysis. But I certainly plan to follow up with them as to what they're doing to generate their low voltage supply.

@Johnrhansen, AFAIK, Japan runs both 50 & 60Hz, so I'd expect local-spec gear to be 50/60 rated. The EVSE label doesn't mention frequency rating.

Dave

johnrhansen said:

how about the 50 hz vs 60 HZ? isn't japan 60 HZ?

Click to expand...

Japan is 50 or 60 hz depending on the region. This became a huge problem after the 3/11 tsunami: http://www.japantimes.co.jp/news/2011/07/19/reference/japans-incompatible-power-grids" onclick="window.open(this.href);return false;.

OP, you seem reasonably technical and have been using EVSE (yay!) but sometimes mixing it up with "charger" (boo, confusing) . For L1 and L2 AC charging, the cord set, charging dock, "trickle charge" cable, charging station, etc. are all EVSEs. They on-board charger is in the back, under the hump (http://www.mynissanleaf.com/viewtopic.php?p=69224" onclick="window.open(this.href);return false. See http://www.mynissanleaf.com/viewtopic.php?f=6&t=14728&p=332668#p332668" onclick="window.open(this.href);return false; and links there.

For DC fast charging, the charger is external to the car...

I have a 15 amp variac at home. You can get them on ebay for like 50 bucks. then you could just adjust the voltage to where the EVSE is happy.

Update:
I replaced the LV transformer with an oversized, 230V rated one (dangling outside the EVSE box, yay safety!), and measured the current on the 18VAC output from the transformer.

When not connected to the car, the transformer is drawing 70mA RMS. When connected to the car but not charging, that rises to 110mA. When the car starts charging, it rises to 310mA.

310mAx18V=5.6W RMS. That seems like a lot for a transformer the physical size of this one - checking an online components supplier, transformers of roughly similar size seem to be around 5-6 VA rated, and that's open frame, without a bunch of potting to keep the heat in. Unfortunately the transformer doesn't have a rating on it and the part number doesn't give any hits in a Google search.

Plus when I run it for a few minutes there's warm air and hot smells coming out of the gap in the EVSE casing where I haven't closed it up completely. And that's with the transformer on the outside, so not contributing internal heat. I suspect the voltage regulator is also working pretty hard.

However 310-70=240mA seems like too much for the J1772 protocol, according to the Wikipedia article on that standard. If I read it correctly, the EVSE has an internal 1k resistor in the pilot circuit, which is the only circuit which changes between charging and idle. That should only be allowing 12mA of extra current to be flowing even if the car is presenting no resistance at all. Even adding a bunch of extra losses for the 12V regulator in the EVSE, I can't see 240mA. So now I'm confused - where am I going wrong here?

At any rate, current theory is that the car's somehow drawing too much current from the EVSE pilot line, overloading the LVSE power supply. Because the transformer is only 200V rated, that's the weak link and is failing.

Next step, find another Leaf and do the same measurement for comparison.

Dave

No idea, but one would imagine the EVSE is supposed to emit a pilot duty cycle (https://code.google.com/p/open-evse/wiki/J1772Basics" onclick="window.open(this.href);return false that is commensurate with 100 volts or 200 volts at whatever amperage max it can handle. If you're feeding in 230 volts and the pilot duty cycle is unchanged, that might be causing the car's OBC to draw more than is safe for certain components in the EVSE.

Just a wild guess. I'm no EE.