Bad news using my retrofitted EVSE with the LEAF :(

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Spies

Well-known member
Joined
Jul 31, 2010
Messages
838
Location
Foster City, CA
The good news is that my retrofitted EVI MCS 100-3 EVSE with a brand new Yazaki J1772 plug and cable from the fine folks at Clipper Creek will in fact charge the LEAF but the bad news is none of the LEAF timers work.

This of course creates a number of issues for me including not being able to join the 80% club and also having to continue to rely on the pool timer hooked to the EVSE so the LEAF will charge during off-peak time. I suspect I will not have the ability to pre-heat or pre-cool the Leaf cabin with grid power either.

This situation seems to be triggered by the fact that the LEAF does some sort of "testing" of the EVSE when first plugged in when using a timer. This "testing" will cause the relays in my EVSE to close but then the relays will open up again after just a few seconds. I suspect my EVSE does this as a safety measure because it senses no power being drawn. Once this cycle happens the timers will come and go without the relays closing again.

Now if I don't set a timer or I override the timer the moment I pug in the EVSE the relays will close and the LEAF will charge.

Next to going out a buying a different EVSE does anyone have suggestions on how I might be able to get this combination to work? How are the LEAF timers supposed to function with the EVSE? I will also be contacting Clipper Creek to see what they think. Thanks in advance for your suggestions.
 
well that is a bummer i guess. so i have no real idea what you did, so what did you do? how much it cost and are u in a TOU area?
 
Interesting problem, Spies. I wouldn't have expected that much 'smarts' to be between the Leaf and EVSE. Your hypothesis sounds reasonable.. the EVSE doesn't like something the Leaf does during connection, or the Leaf does something differently when the timer hits.

Definitely give CC a call and walk them through the scenario. It's likely a software fix from them, and they'll want to get it right to service the biggest EV rollout of 2011.
 
Spies: I would be VERY interested to find out what you learn from ClipperCreek.

This "relays-not-closing" issue could also become a problem for the battery balancing or top-off phase we've been learning about from other threads.

Also ... be aware that your MCS is not aware of the "proximity" signal in J1772-2010. That means you should cut power on the upstream side (at the MCS or "higher") before removing the J-plug from the car. (You probably already knew this ... but just so everyone else is aware.)

(Side comment: the MCS is SO OLD ... software upgrade(s) from ClipperCreek are ... NOT going to happen.)
 
I just want to be clear that my EVSE manufactured by MCS is not a Clipper Creek product and was purchased when I got my Honda EV Plus way back in 1998. I don't want to suggest that any Clipper Creek product does not work properly with the LEAF. Clipper Creek was kind enough to sell me a UL approved Yazaki J1772 plug and cable for $350 plus tax and shipping so I can do the retrofit and suggested that my EVSE should work properly with the LEAF but of course there were no guarantee since no had a LEAF to test with yet.

I have been trying to come up with a workaround this morning to get the pool timer and the LEAF timers to somehow work together but to no avail. If the LEAF allowed one to set the charge to 80% without using the built in timers would be one way I could pull it off. Another way would have been if the LEAF timers had a "window" of charging but alas it seems they only have start time and if that time passes and the LEAF is not plugged in nothing will happen until that timer comes around again the next day.

Just a note, the pool timer was a low tech way to get around the fact that neither the Honda EV Plus or the EVSE had timer capability. Worked great for the most part exact one time the clock mechanism in the timer got stuck. Luckily I noticed in time to get enough charge in the car before I needed to get to work since I worked in the afternoon in those days. Just cost me more money to charge that day. I have been on the PG&E E9a rate since I got the Honda EV Plus.
 
Does the proximity pin (not the PWM pin) go all the way back to the EVSE? If so and the EVSE is turning that off, that is your problem. Pin 5 (proximity) should be looped back to pin 3 with a 150 ohm and 330 ohm resistors inline. This tells the vehicle that the EVSE is connected.

Sounds like the EVSE might be cutting off that signal, making the car think it's not connected anymore.
 
LEAFer said:
Also ... be aware that your MCS is not aware of the "proximity" signal in J1772-2010. That means you should cut power on the upstream side (at the MCS or "higher") before removing the J-plug from the car. (You probably already knew this ... but just so everyone else is aware.)
LEAFer are you sure about the needing to cut power part? Certainly the MCS is not J1772-2010 but it is J1772 and produces the J1772 pilot signal. The MCS behaves with the LEAF just as it did before with other EVs except when the timers are involved. Also does this "proximity" signal come from a different line on the on the Yazaki connector? The reason I ask is the only lines in the Yazaki connector were the same ones that were in the Avcon connector I replaced and are the following:
Green wire = Ground
Black Wire = Line 1
White wire = Line 2
Blue Wire = Pilot (communication)
 
DarkStar said:
Does the proximity pin (not the PWM pin) go all the way back to the EVSE? If so and the EVSE is turning that off, that is your problem. Pin 5 (proximity) should be looped back to pin 3 with a 150 ohm and 330 ohm resistors inline. This tells the vehicle that the EVSE is connected.

Sounds like the EVSE might be cutting off that signal, making the car think it's not connected anymore.
You might be onto something here but it still does not feel quite right to me based on the behavior. I suppose I need to figure out what is opening the relays back up in the EVSE after plugging in with timers set on the LEAF. Do other EVSE keep the relays closed when a timer is set? I don't know at this time if it is something the EVSE is doing to open the relays back up or something that the LEAF is doing when a timer is set to open the relays back up.
 
Spies said:
LEAFer said:
Also ... be aware that your MCS is not aware of the "proximity" signal in J1772-2010. That means you should cut power on the upstream side (at the MCS or "higher") before removing the J-plug from the car. (You probably already knew this ... but just so everyone else is aware.)
LEAFer are you sure about the needing to cut power part? Certainly the MCS is not J1772-2010 but it is J1772 and produces the J1772 pilot signal. The MCS behaves with the LEAF just as it did before with other EVs except when the timers are involved. Also does this "proximity" signal come from a different line on the on the Yazaki connector? The reason I ask is the only lines in the Yazaki connector were the same ones that were in the Avcon connector I replaced and are the following:
Green wire = Ground
Black Wire = Line 1
White wire = Line 2
Blue Wire = Pilot (communication)
The "cut power ... before removing J1772 plug" is a precaution that applies mostly when the car is still charging with power running thru the cable. The disconnect motion when you remove the J-plug does indeed "cut" the pilot signal and in essence tell the MCS to cut power automatically. But unlike the "strain relief" (at the EVSE end) safety feature, where the pilot signal gets pulled first if the cable gets yanked, there is no equivalent function, EXCEPT thru the proximity signal, on the J-plug side. So ... small as the possibility may be ... depending on timing and plug motion ... if you have power in the J-plug, and the proximity signal is NOT functioning as designed, then you could end up with a situation of arcing across the high-voltage pins while pulling the J-plug.

Ask Clipper Creek when you talk to them ... that's how I understand the J1772-2010 function of the proximity signal.

EDIT: I too need to eat a little bit of crow today ... see my post on the next page made at 6:02pm. The above statement
But unlike the "strain relief" (at the EVSE end) safety feature, where the pilot signal gets pulled first if the cable gets yanked, there is no equivalent function, EXCEPT thru the proximity signal, on the J-plug side.
is incorrect or at least misleading. Also see Gary's info in the next post below.
 
The Proximity pin is connected through a resistor to ground in the nozzle, and is not (normally) connected back to the EVSE at all.
The car uses the Proximity "load" to tell that a "nozzle" is plugged in (to the car).

Your old EVSE is probably not reacting "properly" to the car's "loading" of the Control Pilot (PWM) line. I do not have the old J1772 specs to see what the old EVSE was expected to do.

Normally, loading the +12v CP line (is output from the EVSE through a 1k resistor) down to 6v tells the EVSE to close the relay, and going back to 9v (CP line less loaded) tells the EVSE that the car is still there, but to open the relay (because the car does not want power right now).

Maybe your EVSE expects one cycle to Power ON (6v), then Power Off (9v), but then refuses to close the relay again until the No-Car condition is detected (12v = no load)?

With two resistors and a diode this is not difficult to test the EVSE (when it is not connected to the car).
 
LEAFer said:
The "cut power ... before removing J1772 plug" is a precaution that applies mostly when the car is still charging with power running thru the cable.
Thanks LEAFer, its always better to be safe than sorry.

I did go out this afternoon and try to figure out what exactly is the behavior of the LEAF charge timers. My assumptions about the charge timers not being a charge window may not be correct after all. If this is the case I will be able to use the old pool timer in conjunction with the LEAF charge timer so I can charge to only 80%. I will just need to be sure the LEAF charge timer triggers before the pool timer powers on the EVSE. Why Nissan did not make the charge now function settable to 80% is beyond me. So it seems if a timer event passes and the EVSE is plugged into the LEAF without power and then power is applied to the EVSE charging appears to start. I will set this up tonight and report back. Perhaps the charge window after the timer triggers is not very big? If having to still use the pool timer with the EVSE to give me access to the 80% charging function I can live with that for now.
 
garygid said:
Maybe your EVSE expects one cycle to Power ON (6v), then Power Off (9v), but then refuses to close the relay again until the No-Car condition is detected (12v = no load)?

With two resistors and a diode this is not difficult to test the EVSE (when it is not connected to the car).
Gary! That totally sounds like what is happening! If you could walk me through the test I would be extremely grateful!

I seem to remember there was at least one other member on this board that had the same EVSE so this should benefit that person as well.
 
At the nozzle, with the EVSE On:

WARNING: DO NOT attempt this if you are not competent to do it. DEADLY voltages will be present.

1. Use a voltmeter to check that you have +12v DC on the Control Pilot pin (to the ground pin).

2. Connect a small diode diode (approx. 10 ma (or more) rating) in series with a 2.7K resistor, and connect them from the nozzle's Control Pilot pin to the ground pin. This should tell the EVSE that an EV is connected. The "arrow" of the diode should be "pointing" towards the ground pin.

3. Connect a 1.3K resistor in parallel with (across the) 2.7K resistor to tell the EVSE to close its Power Relay. You should hear the relay click closed. Caution: 240v AC should be present on the POWER lines.

4. Remove the 1.3K ohm resistor to tell the EVSE that the car no longer wants power. The relay should click OFF, and there SHOULD no longer be 240v AC on the Power pins.

5. Connect the 1.3K resistor again. The Power relay SHOULD close again. If it does not, the EVSE is not doing "modern" sequencing.

6. Disconnect the 1.3K, then the 2.7K resistors to "tell" the EVSE that the car has been unplugged.

8. If necessary, do the test again.

PLEASE, be careful.
NOTE: I think these instructions are correct, but I take NO responsibility for any use of them.

If in doubt, consult the relevant J1772 standards, and all appropriate safety procedures.
 
Spies said:
LEAFer said:
The "cut power ... before removing J1772 plug" is a precaution that applies mostly when the car is still charging with power running thru the cable.
Thanks LEAFer, its always better to be safe than sorry.
(I am chewing on a little bit of crow. I made a correction to my 12:07pm post above.)

I had a chance to look at a customer Volt this afternoon. In particular ... it was my first close-up and careful inspection of the J1772-2010 receptacle (inlet). ( I have not even driven a LEAF yet, if you can believe that. ) Although the J-plug seems to have all pins positioned at about the same distance, except for the ground pin ... IOW, GND connects first, then the two hots ... it appears that (on the Volt ... check your LEAF; they should be identical) in the receptacle the two signaling pins (Pilot and Proximity) are set back a little ways. In other words they connect last upon insertion, and disconnect first upon removal. There should, of course, not be any power present in the J-plug upon insertion, so that part is benign. And, based on what I found empirically (a theoretical inspection of the new "paper" standard could have the same result), during removal the "cutting" (interruption) of both signaling lines appears assured to happen before the power pins are disconnected. If the MCS reacts quickly enough (based on the old standard) to the "lost" pilot signal, you're in good shape; it should cut power (to prevent arcing) in due time, and that's assuring.

But yes .. better safe than sorry ... when mixing the old & new standard hardware. So I still advise (and would do so myself) ... cut the power up stream before disconnecting an energized J-plug.

Back to the behaviour with timers: I would guess that the LEAF temporarily energizes the EVSE in order to check voltage and maximum available current (pilot signal) in order to perform the calculation about necessary charge duration. It then decides start time based on the power available, current SOC, and desired final SOC.
 
1. A check by the car to see if the EVSE is responsive, and to see what voltage and current it offers, is a very likely initial "check-test".

2. Will EVProject log it as a very-short charge, saying that (with 2 Eq-Spikes) you "charge" on L2 four times a day/night, thus L2 stations are "very necessary"?

3. Before removing the nozzle, one has to press a "button" (lever) to unlatch the plug from the socket. That signals the car that Proximity is lost, and the car should (very quickly) tell the EVSE to cut off the AC Power, allowing the current to stop flowing BEFORE you actually begin to slide the nozzle (plug) out of the socket.

So, a Suggestion:
To better preserve the socket and plug's high-current contacts, press the "release" button, and then pause for about one second, before moving the nozzle (pulling it out of the car's socket).
 
garygid said:
3. Before removing the nozzle, one has to press a "button" (lever) to unlatch the plug from the socket. That signals the car that Proximity is lost, and the car should (very quickly) tell the EVSE to cut off the AC Power, allowing the current to stop flowing BEFORE you actually begin to slide the nozzle (plug) out of the socket.

So, a Suggestion:
To better preserve the socket and plug's high-current contacts, press the "release" button, and then pause for about one second, before moving the nozzle (pulling it out of the car's socket).

It sounds to my old ears like there is a micro-switch inside the silver 'button' on the EVSE plug. So what you're saying makes perfect sense...although if I ever have to unplug while the EVSE is actually charging, I'll hit the OFF button on it before grabbing the plug just to be safe. :?
 
This special process, just to be safe, sounds like over-analysis gone wild.. I have no intention of walking to the EVSE to stop it before disconnecting, or pausing unnaturally in disconnecting.

I trust that the J1772 standard, and the electronic speed of light are sufficient to conduct an orderly disconnect and zero the current before I could possibly get the main contacts to disconnect.

After all, if it requires any non-intuitive or special handling, none of the public L2 stations will last more than a few uses.
These are mass-market cars, meant to be in service for a decade or more. It has to be fool-proof.
 
Jimmydreams said:
It sounds to my old ears like there is a micro-switch inside the silver 'button' on the EVSE plug. So what you're saying makes perfect sense...although if I ever have to unplug while the EVSE is actually charging, I'll hit the OFF button on it before grabbing the plug just to be safe. :?
I agree that there seems to be a micro switch inside the Yazaki plug that is triggered by pushing the release button/lever on the plug. No idea what it does though.

Gary, thanks so much for the instructions to test the EVSE! Seems simple enough and I can even do the test without the Yazaki plug and cord even attached to the EVSE since it is from a time that they just used a NEMA 14-50R plug. Should be safer to test that way. I will give it a try and report back soon.

You know I went back and pulled out my paperwork on this EVSE and I realize that they never sent me the documentation on it. I do however have this great letter dated February 7, 1998 that reads and I quote "To help you get to know your charger better, we've enclosed the following materials: MCS-100-3 Electric Vehicle Connecting Device (EVCD) Owner's Manual**. Be sure to refer to this manual for all of your technical charger questions. **Note: The MCS-100-3 (EVCD) Owner's Manual is not available at this time, but will be mailed to you on a later date." I think I better give up on receiving that manual now ;) I did mange to find this website that describes the unit however http://www.bowzerbird.com/eviweb/products2.html#mcs100
 
Spies said:
Jimmydreams said:
It sounds to my old ears like there is a micro-switch inside the silver 'button' on the EVSE plug. So what you're saying makes perfect sense...although if I ever have to unplug while the EVSE is actually charging, I'll hit the OFF button on it before grabbing the plug just to be safe. :?
I agree that there seems to be a micro switch inside the Yazaki plug that is triggered by pushing the release button/lever on the plug. No idea what it does though.
I thought at first there's a micro switch in that button too, but after seeing the Volt today, I am much less sure. The button looks more like a purely mechanical release. A micro-switch seems to make perfect sense, but my noticing the shorter signaling pins has me believe that the safe disconnect function does not rely on such a switch.

GroundLoop said:
This special process, just to be safe, sounds like over-analysis gone wild.. I have no intention of walking to the EVSE to stop it before disconnecting, or pausing unnaturally in disconnecting.

I trust that the J1772 standard, and the electronic speed of light are sufficient to conduct an orderly disconnect and zero the current before I could possibly get the main contacts to disconnect.

After all, if it requires any non-intuitive or special handling, none of the public L2 stations will last more than a few uses.
These are mass-market cars, meant to be in service for a decade or more. It has to be fool-proof.
Groundloop, you are generally correct. But you are forgetting that this particular thread is about MIXING the OLD technology J1772-1996 (MCS unit) for use with the NEW standard (J1772-2010). The mixture, IMHO, requires more caution. This kind of “special procedure” caution is indeed unwarranted for your standard LEAF hooked to your new standard EVSE.
 
Back
Top