Blink / Rav4 Blows Out a Contactor Pin (with gory pics)

Ingineer: Care to comment on advisability/effectiveness of AFCI detection especially if you are correct that worn-out plugs/inlets are inevitable? Or do you think the best approach is that worn-out/damaged plugs/inlets are able to be routinely repaired/replaced? I am concerned about the possibility that a worn-out/damaged public charger plug could spread damage to car inlets (and vice versa)...

Tony: If you are conceding that the Blink may not have been at fault, to be fair and not to mislead people who aren't able to read all this threads' posts: shouldn't you change the title of your post?

MikeD said:

Ingineer: Care to comment on advisability/effectiveness of AFCI detection especially if you are correct that worn-out plugs/inlets are inevitable? Or do you think the best approach is that worn-out/damaged plugs/inlets are able to be routinely repaired/replaced? I am concerned about the possibility that a worn-out/damaged public charger plug could spread damage to car inlets (and vice versa)...

Click to expand...

I have not tested AFCI's for this application. On a standard Leaf charging at only 16A, I just don't see this happening, but when we start getting up to around double that, the chances of problems go up markedly.

The right thing to to is for the software in the on-board charger (OBC) to be modified to detect things like this and issue a warning. On the CHAdeMO port, they have a thermistor to detect problems like this, but this is because there can be well over 100 amps of high-voltage DC flowing (Arcing would be disastrous!). Obviously the addition of a thermistor to the J1772 port would also go a long way toward detecting these kinds of failures.

I would design the algorithm to reduce charging power and alert the owner in the event of a high resistance (sagging voltage) or arcing event.

Keep in mind, if a public charger gets a failed contact like Tony's, I seriously doubt it will "spread". It will kill one hungry EV, then be damaged too badly to be used again.

It's obviously prudent to at least take a glance at the condition of J1772 handle connectors before blindly inserting them into your EV. Report any apparent damage to the station owner so they can fix it.

It's too bad the J1772 pins were designed so thinly. They are about the same cross-section as a #12 AWG wire (3.6mm), which nobody would not consider sending 70 Amps though, let alone 30, but someone decided it was ok.

-Phil

Are there different standards for the connector based on the current delivered? For example, these three Leviton J1772 cables: would the difference be only in the conductors, or also the plug? If the latter, would that be standards-based or at the discretion of the manufacturer?

http://store.leviton.com/b/5742800011" onclick="window.open(this.href);return false;

hmmm... actually looking at the details, they all report to have 10 gauge wires. :?

Nubo said:

Are there different standards for the connector based on the current delivered? For example, these three Leviton J1772 cables: would the difference be only in the conductors, or also the plug? If the latter, would that be standards-based or at the discretion of the manufacturer?

http://store.leviton.com/b/5742800011" onclick="window.open(this.href);return false;

hmmm... actually looking at the details, they all report to have 10 gauge wires. :?

Click to expand...

There is no compliance testing for J1772 (That I am aware of), so interpretation of the specifications is up to the manufacturer.

To answer you question, yes, on the higher current capacity handles they always use different contacts, though the contact pin diameter is always 3.6mm.

I guarantee the Levition (ITT) cables are not all 10AWG. That must be a typo.

-Phil

J1772 was originally designed with 32A in mind (J1772-2001), but was later updated to support a maximum of 80A (J1772-2009).

It does seem that some sort of thermistor would be prudent - along with the onboard charging monitoring voltage drop as it ramps up, though it would have a hard time distinguishing voltage drop caused by a single hot-spot compared to a long service run.

I wonder if a some sort of dielectric grease or switch grease on the pins might help or make things worse?

J1772 power pins are 3.6mm in diameter or about the same as 7 AWG - it does appear that the pins should provide plenty of surface area which should result in any hot spots being focused on the base of the pin provided that good contact is being made.

Looking at Tony's pics, though, make it clear that the heat was focused on the inside half of the pin.

The pins on the inlet appear to be some sort of bi-metal construction - maybe a combo stainless/copper pin perhaps to reduce the electrical resistance of the pin as stainless has a much higher electrical resistance than copper.

It would be interesting to do a proper failure analysis on Tony's connector and inlet.

Nubo said:

Are there different standards for the connector based on the current delivered? For example, these three Leviton J1772 cables: would the difference be only in the conductors, or also the plug? If the latter, would that be standards-based or at the discretion of the manufacturer?

http://store.leviton.com/b/5742800011" onclick="window.open(this.href);return false;

hmmm... actually looking at the details, they all report to have 10 gauge wires. :?

Click to expand...

That is strange. I always assumed that the pins were all rated at 75 amps, but that the inlets and plugs were just de-rated based on the wiring that was connected to them. Now I see that they all have the 10ga wiring, so that is confusing. 10 seems a little light for anything over 30 amps. Not sure how they can do this for 70 amps.

drees said:

I wonder if a some sort of dielectric grease or switch grease on the pins might help or make things worse?

Click to expand...

You would want conductive grease, not dielectric! I would advise against any grease though, as it will attract dirt which will make it worse. The J1772 contact design is self-cleaning.

-Phil

Personally, I doubt that arcing was involved, at least at the beginning of the problem. I would be much more suspect of just poor contact on the pin. All of the current went through a small part of the surface of the pin, causing the heating.

As other have stated, using voltage sag to detect a problem could be problematic: a long service run could cause the same sag which, distributed over a must larger area, would be harmless. A thermistor to detect heating would be much better.

We'll have to wait to see how many other RAV4/Tesla/other high charging current EVs see similar problems.

I rather doubt that the Blink had anything directly to do with this per se since this is obviously a connector issue and many EVSEs use pretty much the same connector. Thus, it could also have happened to any of a number of 30 amp EVSEs.... For all we know at this point, it might have been the car's fault, in fact, and had nothing to do with the EVSE.

solardude said:

Simply Inferior.
Plug In. Get Stranded.
Blink

Click to expand...

It's possible that this is a deficiency of the Rema connector design used on the Blink. I compared it to the reference standard (Yazaki), and it appears that there are 3 deviations at work here that could be synergizing to create this failure:

1. The female contacts are captured very tightly in the housing, such that they are not as free to "self-align" to the male pins in the receptacle. The Yazaki has more freedom of movement.

2. The connector body is more loose in the receptacle housing.

3. The handle shape is more flat, so the cable exit is closer to straight out rather than angled down. This is causing the heavy cable to impart a lot more downforce on the body.

All these factors working together mean there is a good chance the self-aligning contact design of J1772 is being defeated in this implementation. Without sawing one in half, it's hard to know for sure, but based on what I can see, it sure looks bad.

-Phil

While not a substitute for connectors that can handle the load, it would be relatively simple to add a subminiature thermostat in parallel with the proximity button switch on the J1772 handle. This would drop the charge current based on a connector temperature of one's liking (70-90C?) and restore charging when the temperature recedes, effectively imposing a thermally controlled charging duty cycle to protect the connector and perhaps alert when there is a heating issue. There's not a lot of space to internally fit the thermostat close to the pins with the LEAF inlet (without drilling into the plastic?), but the image of the RAV4 inlet shows voids between the pin sockets that should easily fit a small (~8mm) thermostat. I suppose the thermostat could be mounted on the EV inlet side rather than the EVSE plug/handle by dropping the resistance on the inlet side instead of paralleling the button on the handle. It just depends which side you want to protect all the time, and which you are willing to modify.

Howdy

Ingineer: "It's too bad the J1772 pins were designed so thinly. They are about the same cross-section as a #12 AWG wire (3.6mm), which nobody would not consider sending 70 Amps though, let alone 30, but someone decided it was ok.".

I agree with your J1772 pin diameter, but not about your comparing that pin's diameter to that of a 12AWG conductor and therefore your conclusions about J1772 current carrying capacity.

Perhaps you misread a table, as according to NEC 2008/Table 8 (Conductor Properties) 12AWG (single strand) has a diameter of 2.05 mm, 10AWG (single strand) has a diameter of 2.588 mm, 8AWG (single strand) has a diameter of 3.264 mm. Using Wikipedia 7AWG (single strand) has a diameter of 3.665 mm and 6AWG (single strand) has a diameter of 4.115 mm.

I am not an expert on the current carrying capabilities of copper wire, but as you know the maximum current allowed for a given size wire is affected greatly by the thermal capacity of that wire's insulation. For example according to NEC 2008/Table 310.15 the 8AWG copper wire allowed ampacity for 140 degrees F insulation is 40, for 167 degrees F insulation is 50, and for 194 degrees F is 55. The corresponding numbers for 6AWG is 55, 65, and 75. Ampacities for 7AWG copper wire is not given, but since 7AWG's diameter is about half way between 6AWG and 8AWG, it is reasonable to estimate its corresponding ampacity numbers as 47, 57, and 65. There are other factors that can reduce the maximum current that can be handled safely, but 65a seems to me a ballpark figure for those pins.

It is important to note the J1772 inlet power pins are NOT insulated like conventional wires -- but they lead eventually to conductors that are and may be 6AWG or larger (anybody know J1772 inlet wire gauge for Tesla Model S?). So heat dispersal from those pins lead to those conductors and so the heat rating of that insulation must be taken into account. I don't know about the possibility of heat sinks between the power pins and the insulated conductors, but it doesn't seem to be a stretch to think that 3.6mm power pins might not be a problem at higher amps just looking at its current carrying capacity.

Now the reliability of the J1772 connection may be another matter entirely!

I know the Honda Fit EV has a thermistor of some sort on the j1772 inlet. Some have reported problems with heating on the Facebook page.

JeremyW said:

I know the Honda Fit EV has a thermistor of some sort on the j1772 inlet. Some have reported problems with heating on the Facebook page.

Click to expand...

Hmmmm and the Fit EV has a 6.6 kW charger. I wonder if we'll start hearing more of these as other vehicles w/6.6+ kW chargers are deployed. It seems like it's just about everyone else other than the Volt, 2011-2012 Leaf, PiP and i-Miev.

cwerdna said:

JeremyW said:

I know the Honda Fit EV has a thermistor of some sort on the j1772 inlet. Some have reported problems with heating on the Facebook page.

Click to expand...

Hmmmm and the Fit EV has a 6.6 kW charger. I wonder if we'll start hearing more of these as other vehicles w/6.6+ kW chargers are deployed. It seems like it's just about everyone else other than the Volt, 2011-2012 Leaf, PiP and i-Miev.

Click to expand...

I have not heard of a similar incident before. For what it's worth, I've been charging the ActiveE at about twice the rate of what the Leaf used to pull for the past eight months. There was a host of issues and problems with the car, but an overheated J1772 inlet was not one of them. That said, I only used a Blink station once, at Turbo3's place.

surfingslovak said:

cwerdna said:

JeremyW said:

I know the Honda Fit EV has a thermistor of some sort on the j1772 inlet. Some have reported problems with heating on the Facebook page.

Click to expand...

Hmmmm and the Fit EV has a 6.6 kW charger. I wonder if we'll start hearing more of these as other vehicles w/6.6+ kW chargers are deployed. It seems like it's just about everyone else other than the Volt, 2011-2012 Leaf, PiP and i-Miev.

Click to expand...

I have not heard of a similar incident before. For what it's worth, I've been charging the ActiveE at about twice the rate of what the Leaf used to pull for the past eight months. There was a host of issues and problems with the car, but an overheated J1772 inlet was not one of them. That said, I only used a Blink station once, at Turbo3's place.

Click to expand...

Yeah, since I was there too, it wasn't being charged on his Blink EVSE for very long.

cwerdna said:

Yeah, since I was there too, it wasn't being charged on his Blink EVSE for very long.

Click to expand...

Yes, indeed. I'm fairly confident that the inlet on the ActiveE is from a different supplier. Several factors to consider here, and I would not pin it solely on the higher power draw. I've been enjoying the 6.6 kW charging for many months, and did not observe overheating.

surfingslovak said:

cwerdna said:

Yeah, since I was there too, it wasn't being charged on his Blink EVSE for very long.

Click to expand...

Yes, indeed. I'm fairly confident that the inlet on the ActiveE is from a different supplier. Several factors to consider here, and I would not pin it solely on the higher power draw. I've been enjoying the 6.6 kW charging for many months, and did not observe overheating.

Click to expand...

Here's something to consider. Tesla and Toyota drove the test mules and prototypes around with Tesla HPC connectors, and didn't add the J1772 until the car was closer to production.

TonyWilliams said:

surfingslovak said:

cwerdna said:

Yeah, since I was there too, it wasn't being charged on his Blink EVSE for very long.

Click to expand...

Yes, indeed. I'm fairly confident that the inlet on the ActiveE is from a different supplier. Several factors to consider here, and I would not pin it solely on the higher power draw. I've been enjoying the 6.6 kW charging for many months, and did not observe overheating.

Click to expand...

Here's something to consider. Tesla and Toyota drove the test mules and prototypes around with Tesla HPC connectors, and didn't add the J1772 until the car was closer to production.

Click to expand...

Interesting...

And... Toyota's only other US plug-in, the PiP has a small battery and doesn't have a 6.6 kW charger. In fact, I think it barely goes much over 2 kW at 240 volts (this is OTOH... I don't recall what people have stated on Priuschat) and most of them probably just use the low amperage 120V Toyota provided EVSE (brick) since it's less than 3 hours at 120 volts to fully charge.