LEAF Software Caused OBC To Fail, Not WattStation-fix coming

[DaveinOlyWA]

this is a joke. i wonder what would have played out if it was Blink's EVSE having the issue and not GE; one of the largest and most powerful company in the world?

brownouts?? oh really?? so how does Phil's EVSE work just fine at 208 volts??

what a bunch of BS.

now we have Nissan has to adjust their software to fit a 3rd party vendor?? its like telling Microsoft to fix Windows because that Adobe's software is not running correctly.

does any of this make sense??

 

[jackal]

brownouts?? oh really?? so how does Phil's EVSE work just fine at 208 volts??
does any of this make sense??

Why not? Specification frameworks are known to be never 100% complete. One can follow the framework exactly and yet introduce an unprecedented scenario.

 

[DaveinOlyWA]

brownouts?? oh really?? so how does Phil's EVSE work just fine at 208 volts??
does any of this make sense??

Why not? Specification frameworks are known to be never 100% complete. One can follow the framework exactly and yet introduce an unprecedented scenario.

you are missing the point. brownouts is when line voltage drops below its spec. the charger on the LEAF will work at 208 volts so the brownout excuse is exactly that; an excuse.

 

[Nubo]

brownouts?? oh really?? so how does Phil's EVSE work just fine at 208 volts??
does any of this make sense??

Why not? Specification frameworks are known to be never 100% complete. One can follow the framework exactly and yet introduce an unprecedented scenario.

you are missing the point. brownouts is when line voltage drops below its spec. the charger on the LEAF will work at 208 volts so the brownout excuse is exactly that; an excuse.

It depends
-When does wattstation cut off? Lower than 208V?
-what is the specification?
- what is the tolerance of the LEAF?

It's conceivable that wattstation is within specs, but LEAF needs a narrower range than allowed by the specifications.

 

[smkettner]

you are missing the point. brownouts is when line voltage drops below its spec. the charger on the LEAF will work at 208 volts so the brownout excuse is exactly that; an excuse.

What about when your 208 supply has a brown out? That would be 187 or less.

 

[TomT]

Good point!

GE is much more proactive about addressing these issues than Nissan... If Nissan reacted after 11 LEAFs lost a bar they'd have tested the affected LEAFs months ago.

 

[Ingineer]

The brownout excuse is silly. Keep in mind the Leaf's OBC (On-Board Charger) module was originally designed for the Japanese market where 100v on Level 1 and 200v on Level 2 is the norm. If it is indeed the "diode" how can a brownout in the OBC affect this?

All switch-mode power supply systems are prone to high-stress under brownout conditions if they don't incorporate some sort of LVCO circuitry. It's hard to imagine that the Leaf's OBC doesn't when many low-cost consumer "wall wart" switchers do.

As an engineer extremely experienced with EVSE's, something doesn't jive.

Though one interesting thing to note is the 2011/2012 Leaf's OBC is made by Nichicon, and they apparently did not get the contract for the 2013 Leaf.

-Phil

 

[Nubo]

The brownout excuse is silly. Keep in mind the Leaf's OBC (On-Board Charger) module was originally designed for the Japanese market where 100v on Level 1 and 200v on Level 2 is the norm. If it is indeed the "diode" how can a brownout in the OBC affect this?

All switch-mode power supply systems are prone to high-stress under brownout conditions if they don't incorporate some sort of LVCO circuitry. It's hard to imagine that the Leaf's OBC doesn't when many low-cost consumer "wall wart" switchers do.

As an engineer extremely experienced with EVSE's, something doesn't jive.

Though one interesting thing to note is the 2011/2012 Leaf's OBC is made by Nichicon, and they apparently did not get the contract for the 2013 Leaf.

-Phil

Hey Phil, do you have the J1772 specs? I'm curious if there is any specification for LVCO for the EVSE itself which might explain why the GE unit was involved but not others -- perhaps they have a more conservative cutoff?

 

[davewill]

The brownout excuse is silly. Keep in mind the Leaf's OBC (On-Board Charger) module was originally designed for the Japanese market where 100v on Level 1 and 200v on Level 2 is the norm. If it is indeed the "diode" how can a brownout in the OBC affect this? ...

I was thinking that it might be the EVSE's +/- 12v supply that was the problem in the brownout condition. The wrong voltage could burn out the diode, but that would make it the EVSE's fault to my way of thinking.

 

[padamson1]

...big snip...
As an engineer extremely experienced with EVSE's, something doesn't jive.

Though one interesting thing to note is the 2011/2012 Leaf's OBC is made by Nichicon, and they apparently did not get the contract for the 2013 Leaf.

-Phil

I think this is more of GE winning "on a technicality" as lawyers would say. GE probably followed the J1772 spec exactly in low voltage conditions where Nissan only followed the spec only as closely as all the other EVSE's on the market. Since GE didn't deviate from the spec, they get to say that Nissan was at fault, even though every other EVSE on the market works fine with the LEAF.

 

[garygid]

How can the LEAF's diode blow out (open) if the EVSE's +/-12v supply
for its Control Pilot signal does not exceed the +/- 12v specs?

Does not make sense with the information I see here.

 

[DaveinOlyWA]

How can the LEAF's diode blow out (open) if the EVSE's +/-12v supply
for its Control Pilot signal does not exceed the +/- 12v specs?

Does not make sense with the information I see here.

amen to that. how often have you seen a device that works with a multitude of accessories but fails to work with "only" one and its the device's fault??

what are the other accessories doing right? or is it wrong? or is it what i suggested before in that GE is powerful enough to "part the waters?"

 

[DarkStar]

How can the LEAF's diode blow out (open) if the EVSE's +/-12v supply
for its Control Pilot signal does not exceed the +/- 12v specs?

Does not make sense with the information I see here.

Exactly.

 

[Ingineer]

Here's what SAE J1772 has to say about the diode:

The diode shown on the vehicle side is intended to be a common small signal silicon diode. Reverse voltage ratings
of at least 100V are readily available and are recommended since this diode is exposed directly to cable transients.

The cable capacitance from the Pilot wire to the Ground wire will probably be around 25 pF per foot, and many cables
are 15 to 20 feet long. If the EVSE’s contactor closes when the line voltage is near a positive or negative peak, then
the voltage on the contactor output can rise from 0 to 170 V in just a few nanoseconds. This fast, high-voltage
transition can easily be coupled through the capacitance of the cable. In addition, with the contactor closed during
charging, any transients such as might be generated by nearby industrial equipment or lightning strikes can be
coupled through. It is highly recommended that transient protection be installed on both the EVSE output and the
vehicle input.

-Phil

 

[Ingineer]

Here's the J1772 and UL stuff related to power quality:

7.5 EVSE Electrical Fast Transient Immunity

The EVSE shall be tested in accordance with and shall meet the requirements for electrical fast transient immunity specified in UL 2231-2.

7.6 EVSE Voltage Dips, Short Interruptions and Voltage Variations Immunity
The EVSE shall be tested in accordance with and shall meet the requirements for voltage dips, short interruptions and voltage variations immunity specified in UL 2231-2.

7.8 EVSE Capacitor Switching Transient Test
The EVSE shall be tested in accordance with and shall meet the capacitor switching transient test requirements specified in UL 2231-2.

7.9 EVSE Voltage Surge Test
The EVSE shall be tested in accordance with and shall meet the requirements for voltage surge specified in UL 2231-2.

Here's the relevant UL sections:

22.5 Electrical fast transient immunity

22.5.1 CENELEC EN 50082-2, electromagnetic compatibility general immunity standard part 2: industrial environment (1995), is to be used as the basis of requirements. IEC 1000-4-4, electromagnetic compatibility (EMC) part 4: testing and measurement techniques - section 4: electrical fast transient-burst immunity test - basic EMC publication (1995), is to be the standard for testing methods and to specify multiple levels of limits based on installation environment Level 3 (2kV transients on power lines) is to be the generic test limit

22.6 Voltage dips, short interruptions and voltage variations immunity

22.6.1 CENELEC EN 55104, electromagnetic compatibility immunity requirements for household appliances, tools and similar apparatus product family standard (1995), is to be used as the basis for requirements. IEC 1000-4-11, electromagnetic compatibility (EMC) part 4: testing and measurement techniques – section 11: voltage dips, short interruptions and voltage variations immunity tests (1994), is to be the standard for testing methods. The protective aspects of the device are not to be comprised under the following power line conditions:

a) 100 percent voltage dip for 10 ms,
b) 60 percent voltage dip for 200 ms, or
c) 30 percent voltage dip for 1 s.

The protective device is permitted to turn OFF during these disturbances as long as:
d) This removes the power to the protected unit, and
e) Operation is automatically restored when input power is restored to at least 85 percent of rated voltage.

22.8 Capacitor switching transient test

22.8.1 A device shall not trip when exposed to a variable ringing wave. The device shall be connected to a rated load and the ringing wave is to be injected onto the input power line. The amplitude of the ringing wave is variable but shall not exceed 40 percent of the amplitude of the fundamental (60 Hz) and is to decay exponentially with a time constant of 1.6 ms. The ringing wave test frequencies are 1, 2, 3, 4, and 5 kHz. The ringing wave shall be applied to start at 90 degrees on the fundamental wave-form and end at 270 degrees.

22.9 Voltage surge test

22.9.1 The representative device shall be capable of withstanding voltage impulses as specified in 22.9.2, using a full-wave 1.2 x 50 microsecond impulse in accordance with Standard Techniques for High-Voltage Testing, ANSI-IEEE 4-1978, and having a crest value in accordance with 22.9.2.

22.9.2 The representative device shall be tested such that each terminal intended to be connected to an ungrounded circuit conductor is subjected to the following surge voltage impulses in the order given:

a) Three applications of a positive impulse voltage with a 6kV crest and three of a negative impulse voltage with a 6kV crest. The follow current shall be a minimum of 3kA available. When flash-over is noted on any of a group of three, an additional nine impulses of the type which had flash-over are to be applied. Any flash-over during the additional impulses is unacceptable.

b) Three applications of a positive impulse voltage with a 3kV crest and three of a negative impulse voltage with a 3kV crest. Any tripping of a device as a result of the impulses is unacceptable.

-Phil

 

[Ingineer]

Note there is no specific voltage range I can find in SAE J1772, or anything pertaining to Brown-out operation.

-Phil

 

[GeekEV]

Here's what SAE J1772 has to say about the diode:

The diode shown on the vehicle side is intended to be a common small signal silicon diode. Reverse voltage ratings
of at least 100V are readily available and are recommended since this diode is exposed directly to cable transients.

The cable capacitance from the Pilot wire to the Ground wire will probably be around 25 pF per foot, and many cables
are 15 to 20 feet long. If the EVSE’s contactor closes when the line voltage is near a positive or negative peak, then
the voltage on the contactor output can rise from 0 to 170 V in just a few nanoseconds. This fast, high-voltage
transition can easily be coupled through the capacitance of the cable. In addition, with the contactor closed during
charging, any transients such as might be generated by nearby industrial equipment or lightning strikes can be
coupled through. It is highly recommended that transient protection be installed on both the EVSE output and the
vehicle input.

-Phil

I may be way off base, since I'm not an EE, but if I'm reading what you quoted correctly I'm left wondering if perhaps the LEAF's diode isn't adequately rated for the higher voltages the spec suggests (since it's only SUPPOSED to be +/-12v). Combine that with (perhaps) inferior shielding on the cable and a higher current draw due to a brown-out. I suppose that could potentially result in higher voltages on the pilot line (as the spec says can happen) and blow out the LEAF's diode. Right?

Though I don't see how a software update could fix that!

 

[Nubo]

22.6 Voltage dips, short interruptions and voltage variations immunity

22.6.1 CENELEC EN 55104, electromagnetic compatibility immunity requirements for household appliances, tools and similar apparatus product family standard (1995), is to be used as the basis for requirements. IEC 1000-4-11, electromagnetic compatibility (EMC) part 4: testing and measurement techniques – section 11: voltage dips, short interruptions and voltage variations immunity tests (1994), is to be the standard for testing methods. The protective aspects of the device are not to be comprised under the following power line conditions:

a) 100 percent voltage dip for 10 ms,
b) 60 percent voltage dip for 200 ms, or
c) 30 percent voltage dip for 1 s.

The protective device is permitted to turn OFF during these disturbances as long as:
d) This removes the power to the protected unit, and
e) Operation is automatically restored when input power is restored to at least 85 percent of rated voltage.

So, the EVSE is "permitted to turn off" during brownouts, which seems to imply that it is not *required* to do so. Sounds to me like the EVSE is allowed to pass low voltage to the vehicle as long as it maintains electrical safety.

Now, what if the LEAF engineers had tested all along with EVSE that chose to turn off during brownouts at a voltage above that which turns out to be damaging to the LEAF charger? And what if GE did not take that apparently optional approach with their EVSE?