Dead LEAFs and GE Chargers

Alex L. Dykes
by Alex L. Dykes

The GE Wattstation killed my Leaf! That’s the story being reported by the New York Times as well as PlugInCars.com. As the tale goes, 11 Leaf owners have had their chargers “damaged” while charging with GE’s Wattstation home charging station. The relative significance of only 11 failures aside, the Nissan Dealer in San Pablo, CA confirmed to PlugInCars.com that Nissan North America has notified dealers of a potential problem with the Leaf and the GE home charging station. TTAC contacted Hilltop Nissan and they have yet to return our calls. Rather than just parroting back the usual news reports we dug deep. We contacted GE and Nissan, consulted some professional electrical engineers and read though hundred of pages of boring SAE documents. Click past the jump to learn more about EV charging than you ever wanted to know.

Before we dissect the dead Leaf issue, we must first understand how EV charging stations work. Because there were a wide variety of charging connectors prior to the Nissan Leaf and Chevy Volt coming on the scene, we’re going to focus on just the SAE J1772 standard. Other than Tesla, who has decided to off on a tangent with their “prettier” charging connector, all EVs and PHEVs on sale in America use this connector. The short list includes the Leaf, Volt, Karma, Coda, Prius Plug-in, i MiEV, Fit EV, RAV4 EV, Focus Electric, Smart EV, and ActiveE.

The first thing you need to understand is that a “charging station” or “EV home charger” is a confusing term that seems to imply that the home unit is “doing all the work.” In reality, the J1772 connector and station you plug your car into is simply a “smart” extension cord for your car. All the circuitry required to charge the battery, monitor the rate of charge, and keep tabs of the heath of the battery are driving around with you all the time.

Wait! Why is my charger in my car, isn’t that inefficient? While the extra weight of hauling around your charger sounds crazy at first, it is the easiest way to make the charging infrastructure both universal and cost-effective. Because each EV’s battery differs in cell count, voltage, chemistry, cooling characteristics and capacity, it is easier to supply AC power directly to the car and allow the car’s electronics to charge the battery the way it sees fit. The charging connector is simply responsible for communicating to the vehicle what kind of power is available and providing protection to the electrical circuit on which it is installed.

The J1772 connector has 5 pins: AC1, AC2, ground, “control pilot” (aka data) and “proximity detection.” AC1, AC2 and ground are fairly self-explanatory. When using 120VAC AC1 is power and AC2 is neutral. When connected to a 240V circuit AC1 is power and AC2 becomes power as well.

OK, why not just use an extension cord? What’s going on in my charger? Excellent question. Inside the charger we essentially get some relays that turn on and off the power to the pins when the car requests it. In addition we have electronics that communicate the power type (120VAC or 240VAC) and maximum charge current available from the charging station.

So how does it work? When the station is not connected to a car, the AC1 and AC2 pins are not active. This is a safety measure to keep you safe if you should decide to go probing with a paperclip. When you start to connect the vehicle, the first thing that connects is the ground pin because it is longer than the others. The ground in theory helps prevent (among other things) static discharge that could harm the electrical components. The next pins that connect are the power, data and proximity detection pins. Now that everything is connected, the car sees the proximity detect line, establishes a data connection with the charging station, and tells the car what voltage and current options are available. Part of this process involves checking for the presence of a small silicone diode SAE refers to as D1. According to SAE, the purpose of D1 is that it “insures that an EV is actually connected and can be discriminated from other potential low impedance loads.” In other words, D1 isn’t involved in actually charging, just in the verification that there’s actually a car connected to the plug. Next, the car tells the charger what kind of charging it will be doing and requests power. The station energizes AC1 and AC2, the car begins drawing power and charging the battery.

During the charge the station and the car are both monitoring the connection and either device can end the charge at any time. When you unplug the car, the first connections broken are the proximity and control pilot connections, which cause the station to stop power to AC1 and AC2 within milliseconds.

Back to that whole GE thing

The core of the issue seems to be the circuitry that communicates with the charging station. According to SAE, “the minimum control circuit necessary on the EV to use in conjunction with the inlet uses one diode, one capacitor, and one resistor“. Based on the way these components are connected between the “control pilot” pin and the ground connector, the station knows a car is connected and can tell basic charge status.

With me so far? Let’s dig deeper. Suppose that for some reason you had a bad building ground that caused some sort of transient voltage, or ground fault in your home’s electrical system, or a massive power surge from your utility. If this was the case, it is possible to damage the diode which, in this case, is the most likely component to be damaged from a reverse voltage situation. Again, this is possible because the station just connects the car’s on-board charger to the mains. This is likely enough that the J1772 spec outlines that D1 should be rated for at least 100V because “this diode is exposed directly to cable transients.” If D1 fails, charging stations that adhere strictly to J1772 won’t energize AC1 and AC2 because they will think that there is no vehicle connected or there is a fault in the connection.

Actual failures

From my forum research is appears that the Leaf failures fall into two broad categories: D1 failure and a failure involving more than just D1 on the control side of the charger.

The failures can be identified from one another by using the EVSE (Electric Vehicle Supply Equipment) that came with your car. Users on MyNissanLeaf.com report that if D1 has failed, then the car will still charge with the EVSE as it does not check for the existence of D1 in the first place. If the car fails to charge period, then there’s more wrong.

Assuming there is no design fault inherent in the Wattstation’s “control pilot” design (and we might assume this logically because the issues are limited to Nissan Leaf vehicles only), the most likely possibility is a problem with the an underrated or faulty D1 diode in the Leaf’s charger that makes the control pilot circuit more susceptible to transient current and failure. While it does seem fishy that the problems are only reported with the Wattstation and not the popular Leviton and Nissan branded chargers, the issue likely comes down to surge suppression and bad luck. It is likely that Nissan uses a D1 diode with a lower rating (and therefore affording less protection) than the Volt and Prius plug-in. With so few EVs on the road, and little public information on the specifications of electrical components in the chargers it is hard to say for sure.

With US Leaf sales at approximately 12,841 through June 2012, Volt sales at 16,814, i MiEV at 413 and some 2,000 Plug-in Prius sales to the same date, there are some 32,000 EV/PHEVs on the road (not counting the smaller volume vehicles). GE won’t release specific sales numbers simply citing sales in the “thousands.” Even if we assume this means only 2,000, then the number of actual problem units is just over half a percent. If you assume that half the units went to Volt owners and half went to Leaf owners, then the problem percentage raises only to about 1% of all the units being used with the Leaf. Maybe.

Confirming our hypothesis which is that the root cause of the failure is a factor external to the Leaf we got a statement from Nissan at the 11th hour.

We are aware of several isolated instances of Nissan LEAFs sustaining damage due to voltage current spikes from the power grid. These isolated instances, while resulting in component damage to the on-board charger, did not result in any injuries or fires. Some of these reported occurred while LEAFs were charging at GE WattStations. Nissan and GE are working to investigate every issue and determine root cause of the charging issues. While this issue represents a handful of incidents out of millions of charging events involving the Nissan LEAF, we are doing everything we can to get to the bottom of the issue.-Katherine Zachary, Nissan North America

While a power surge/spike is the most likely cause, it seems to highlight a possible shortcoming in the Leaf’s charging circuitry that may make it more susceptible to this type of damage. But it’s probably not all D1’s fault, your home might be killing your leaf. If you live in a home built before 1960, your home was likely built without grounded outlets, and possibly without the neutral line being connected to ground properly. If the neutral is “floating,” there is the possibility of having some very strange voltage potentials at the charging connector to your car. We have no real way of knowing whether the Leaf or the Volt is more likely to fail from this sort of event, but we can assume that it may manifest itself in the Leaf first as Volts don’t have to be plugged in to operate. There are a few steps you should take regardless.

1. Get a surge suppressor. You put one on your computer, your TV, your stereo and even your fridge. Why wouldn’t you put one on the most expensive appliance you’ll ever buy? Regardless of the outcome of the GE/Nissan investigation, the few hundred you’ll spend on a surge suppressor is insurance well spent, especially if you live in a lightning prone area. According to GE, the Wattstation has an internal 6kV surge protection per UL2231-2 & IEC 1000-4-5 which is the same standard that Leviton and other competitors meet. Buy a whole house surge suppressor anyway.

2. Have an electrician checkout your electrical system before you have a station installed. This may seem like a no brainer, but if you’re just asking your electrician to install an outlet to plug the station into, or hard wire the station, they may not check your main panel to see what’s going on. Be sure they check your main and all sub-panels (at the least) to see if everything is kosher.

3. If you live in a lighting prone area, have lightning rods professionally installed on your home.

In the end this is a textbook example of the power of the internet. The fact that a very small percentage of problems can make a New York Times article is amusing to say the least. But it also tells us something else: As EVs gain market share and our cars become essentially expensive electrical appliances with expensive computers inside, we need to re-think how we view the quality of the power in our homes.

According to the New York Times: A spokeswoman for Nissan North America, Katherine Zachary, said in an e-mail, “There’s no official Nissan policy instructing customers not to use G.E. WattStations.”

We contacted GE ourselves and got the following response:

Since its launch in 2011, GE’s WattStation Wall Mount has performed as designed, thousands of units have been shipped, and it has received positive reviews from EV drivers. Regarding the charging issue raised by 11 Nissan Leaf owners who had GE WattStations, GE’s current analysis does not indicate that the WattStation is the cause of the reported failures. GE has been actively working with Nissan to help determine the source of this issue. The GE WattStation has not encountered a similar issue with other brands of electric vehicles. GE’s WattStation is also designed and tested to the SAE J1772 and appropriate UL standards and these tests have been validated by an independent third party. And there have been no design changes to WattStation since its 2011 launch.

The GE WattStation has surge protection per UL2231-2 & IEC 1000-4-5 which will protect the internal circuitry of the charger in the event of a surge up to 6kV. This is consistent with what is seen with our competitors. –Sean Gannon, GE Energy Spokesperson

Alex L. Dykes
Alex L. Dykes

More by Alex L. Dykes

Comments
Join the conversation
3 of 23 comments
  • Steve65 Steve65 on Jul 17, 2012

    "the most likely possibility is a problem with the an underrated or faulty D1 diode in the Leaf’s charger that makes the control pilot circuit more susceptible to transient current and failure." "Confirming our hypothesis which is that the root cause of the failure is a factor external to the Leaf..." These statements seem to directly contradict each other. What am I missing?

    • Alex L. Dykes Alex L. Dykes on Jul 18, 2012

      It's convoluted, but the cause was external to the Leaf and the GE WattStation but it still may have been prevented, IMHO, by using a more robust D1.

  • Evnut Evnut on Aug 07, 2017

    Maybe I'm the 12th person? My 2011 Leaf has this issue (after using a WattStation at work). Nissan wants to replace the charger, but there's nothing wrong with the charger, only the diode. Has anybody seen any other news on this in the past five years?

  • Tassos Jong-iL Communist America Rises!
  • Merc190 A CB7 Accord with the 5 cylinder
  • MRF 95 T-Bird Daihatsu Copen- A fun Kei sized roadster. Equipped with a 660cc three, a five speed manual and a retractable roof it’s all you need. Subaru Levorg wagon-because not everyone needs a lifted Outback.
  • Merc190 I test drive one of these back in the day with an automatic, just to drive an Alfa, with a Busso no less. Didn't care for the dash design, would be a fun adventure to find some scrapped Lancia Themas or Saab 900's and do some swapping to make car even sweeter. But definitely lose the ground effects.
  • MRF 95 T-Bird These 164s, as documented by its owner have to be constantly sorted, as they say. They are nice drivers. I’d rather find a, under the 25 year rule nice and easier to deal with Type 916 Alfa Romeo GTV/Spyder.
Next