iMiEV charging, J1772 testing

[bladecar]

Moderator note: this topic started in Any iMievs for Sale? Moved here due to topic drift.

130km on a charge, holy moly, that's the best I've seen yet (and I bought mine brand new)

Yes, my 2012 has charged to 116 or so (do you call that surface charge?). That's with 73% battery condition?? I reckon you could get to 120 or more by leaving your location and descending to a much lower altitude with regen (but should you ever do that??). Anyway, with 116 showing as a possible range on the guessometer, any driving away for a short period quickly drops that back a little.
This subject car must have a newish battery.

[antiscab]

Getting 130km to show on the guessometer, and actually getting 130km are two very different things. A battery with all 16kwh remaining can do it.

[ Edited Coulomb: Changed topic subject following topic split; changed on many subsequent posts as well. ]

[mark_hetho]

I reckon you could get to 120 or more by leaving your location and descending to a much lower altitude with regen (but should you ever do that??).

We do this on our work commute, the regen slowly tapers off as the battery reaches max state of charge. I'm not sure it is great for the battery, and I wish the car could be set charge to 95% to allow us to get the top 5% on regen at the start of our commute.
Incidentally as our commute involves climbing a hill shortly before getting home, our guess o meter often reads into the 60s when fully charged, but by the time we reach town after 30kms it usually hasn't dropped at all.

LauraiMiEV's car sounds like a really good deal if is indeed getting well over 100km on a charge. Cheaper than what I'm looking at spending to replace my imiev's batteries.

[nuggetgalore]

Getting 130km to show on the guessometer, and actually getting 130km are two very different things. A battery with all 16kwh remaining can do it.

Yesterday my fully charged toy showed 98 km on the guessometer, after I drove 27.5 km it *guessed*
46 km!
A while ago I took this EvBatMon screen shot : Battery has 32.4 Ah capacity and that is 67.5 % of new.
https://imgur.com/a/0EtRVyu

[nuggetgalore]

We do this on our work commute, the regen slowly tapers off as the battery reaches max state of charge. I'm not sure it is great for the battery, and I wish the car could be set charge to 95% to allow us to get the top 5% on regen at the start of our commute.

I did this until the other week before the charger called it a day. The idea was that NOT fully charging the battery may make it live longer (I now doubt that). It is possible that killing the charging without letting the machine know is bad ( pressing the trigger on the gun,changing the resistance between earth and PP pins from 150 to 480 Ohms is supposedly ending charging more gently,though checking it with a clamp meter the drop in amperes is immediate ) .

[ Edited Coulomb: fixed quoting. ]

[mark_hetho]

I wouldn't be able to set a timer accurately enough to hit 95%, given the length of time needed also varies. I did briefly play with a raspberry pi connected to the OBD port to monitor charging and somehow trigger a charging stop but I could never reliably get it to read the messages and track charging status. Then the 12V battery died and I wanted to avoid any extra loads and abandoned it. Definitely a feature I'd like in my next EV though.

[coulomb]

It is possible that killing the charging without letting the machine know is bad ( pressing the trigger on the gun,changing the resistance between earth and PP pins from 150 to 480 Ohms is supposedly ending charging more gently,though checking it with a clamp meter the drop in amperes is immediate ) .

Yes, I have this theory that interrupting a charger with a PFC stage (that would be essentially all EV chargers) is hard on the electronics. It might even have been the cause of your charger's retirement.

Using the trigger button may seem immediate, but ramping to zero in say a hundred milliseconds would look the same as instantaneous disconnection to a multimeter. But I believe that the electronic ramp down will be far less harsh on the electronics.

It would be nice to know how fast the ramp-down actually happens, and how hard an instantaneous disconnect actually is on the electronics. I don't have the equipment to find out.

[nuggetgalore]

I wouldn't be able to set a timer accurately enough to hit 95%, given the length of time needed also varies. I did briefly play with a raspberry pi connected to the OBD port to monitor charging and somehow trigger a charging stop but I could never reliably get it to read the messages and track charging status. Then the 12V battery died and I wanted to avoid any extra loads and abandoned it. Definitely a feature I'd like in my next EV though.

My timer is actually a PLC. I monitor lots of things just for the heck of it (standalone solar etc).
When the charging starts I record the current drawn from the grid every 100 ms for about 40 seconds
( the ramp-up is just shy of this time) .Then it logs the KWh and current and a few more things every minute. I could if I wanted monitor the KWh and/or the Kw as it slows the charging towards the end and terminate the charger 10 or 5 % short of full. But before I would do that (I doubt I will) I would build a relay into (onto) the gun to simulate pressing the trigger.

[coulomb]

But before I would do that (I doubt I will) I would build a relay into (onto) the gun to simulate pressing the trigger.

A robo EVSE plug!

Maybe you should go the whole hog and adapt a Tesla snake "charger" (actually it's just the cord part of an EVSE). Just don't bend over when that thing is active

[nuggetgalore]

I have questions for the brains-trust here:
1)How and where is the command given to the charger of how much current is drawn from the grid?
I know my model 10 charges with approx 12.5 A whether I use a common cable or one with the "brick".
The "brick"reduces the charge to about 8 + A on the model 12s.
I also know that when charging to full, for about half an hour or so the rate gradually reduces to about 1A and then shuts off ( unverified data indicates ramp down and shut off may be quicker if charging the battery when it was not down to the last few bars).
2) Activating the trigger on the gun increases the resistance between earth and PP from 150 to 480 Ohms and this signal shuts off the charging, what would happen if an other resistor ( a pot?) would increase it to something in between?
Could that be used to ramp down the charger or would it blow it up?
Things one thinks off while locked up because of the virus.......

[brendon_m]

https://openev.freshdesk.com/support/so ... -sae-j1772

This may answer some of the questions


Maybe you could use something like this to ramp down the charger
https://shop.gwl.eu/Electric-Cars/EVSE- ... cable.html
Just modify the resistance of the pot to lower the charge rate

[coulomb]

2) Activating the trigger on the gun increases the resistance between earth and PP from 150 to 480 Ohms and this signal shuts off the charging, what would happen if an other resistor ( a pot?) would increase it to something in between?
Could that be used to ramp down the charger or would it blow it up?

It won't blow up the charger, but it won't throttle the charge current/power either. The circuit that the trigger connects to (the proximity signal) is a present/not present thing; there is no in-between.

The power control is all in the PWM ratio of the pilot signal, as Brendom_m's link indicates.

[nuggetgalore]

https://openev.freshdesk.com/support/so ... -sae-j1772

This may answer some of the questions


Maybe you could use something like this to ramp down the charger
https://shop.gwl.eu/Electric-Cars/EVSE- ... cable.html
Just modify the resistance of the pot to lower the charge rate

Thanks to both of you for the info.
That GWL Pilot DIY kit looks interesting... but I believe it will not work on my vehicle because there is probably some thing missing ( could it be the wire from the *pilot pin* to the charger?)
A while ago we went to check out an adjustable EVSE (.evolutionaustralia.com.au) but my car ignored the 6A or 10 A settings and charged with 12.5 A (tripping the AMPfibian).
I heard that apparently it needs some modification, I suppose that is what I am after to be able to charge on a 10 A power-point.

[jonescg]

Andy - yes there's a missing connection in the car's wiring between the charge port and the on board charger. I recall Gelco in SA was providing replacement kits.

[rhills]

Would that be Product no GS1011 on this page? It's about half-way down a long page, so use your browser's search function to look for the Product code.

[mark_hetho]

That the brick reduces charge current on MY12+ but not 2010s tells me that the 2010 charger ignores the signal from the EVSE about how much current it should draw.

As I understand it, all the 2010 mods are just enough to fool an EVSE to turning the juice on. They don't make the charger smarter, but provide the right pilot signal to the EVSE by bridging a resistor across the right pins. Without the mod the pin is physically missing from the socket - a good look in good light should confirm.

Almost no EVSE checks the EV is drawing the right amount of current, and cuts the power if the current drawn is too high - generally drawing 12-13A from an EVSE is not going to let the smoke out, but it's not ideal for a 10A socket.

[nuggetgalore]

Without the mod the pin is physically missing from the socket - a good look in good light should confirm.

The pin itself is there, question is if there is a wire attached and where does it go?
I may have to trace continuity from that pin to wherever , possibly just the other side of the inlet socket(actually it is the plug as far as the contacts are concerned).
GELCOservices may have the harness required but want one in return I think, don't know how that works, I sent them a message.

[coulomb]

I may have to trace continuity from that pin to wherever

The easy way would be with a multimeter on diode range from pin 4 to pin 3, because of the following path:

Diode check.png (33.38 KiB) Viewed 1513 times

The pins are arranged as follows; you want pilot (4) and ground (3):

J1772 socket pins.jpg (25.66 KiB) Viewed 1513 times

You should see about 0.5 - 1.0 V with the pilot connected to multimeter positive, and infinite (open circuit, overload) in the other direction. You might also check that pin 3 really does connect to the metal chassis of the vehicle.

Edit: thinking about it some more, it's not certain that this would work because of the relatively large resistance (2.74 kΩ). At a test current of 1 mA, this might appear as open circiut. Better to test on ohms range; you should see some conductivity in the forward direction (I'd guess 100 kΩ or less), and nothing (infinite resistance, overload) or at least several mega ohms in the other.

Edit 2: First image adapted from the Wikipedia J1772 article.

[nuggetgalore]

I may have to trace continuity from that pin to wherever

Edit: thinking about it some more, it's not certain that this would work because of the relatively large resistance (2.74 kΩ). At a test current of 1 mA, this might appear as open circiut. Better to test on ohms range; you should see some conductivity in the forward direction (I'd guess 100 kΩ or less), and nothing (infinite resistance, overload) or at least several mega ohms in the other.

Edit 2: Image adapted from the Wikipedia J1772 article.

Thanks coulomb.
I checked what you suggested (car switched off):
Ground pin to chassis is connected.
Ground to pilot pin 1646 Ω (on the 2000 Ω range) and 1743 (I assume Ωs) on the diode check setting.
In reverse it is open both methods.
If I measure this correctly , that is no where near the 2740 Ω as per diagram.
What am I missing?
On another note, this thread is getting way off topic, do you think it is worth it to make a new thread?

[coulomb]

In reverse it is open both methods.

That's the most important thing.
[ Edit: and I believe it means that your car was at one stage updated with the Gelcoservices GS1011 harness. Otherwise, I believe that you would not be seeing diode-like behaviour. Or possibly, it had the harness replaced with one from a 2012+ model. ]

If I measure this correctly , that is no where near the 2740 Ω as per diagram.

I cropped off other parts of the circuit, and the circuit is powered down, so that will affect the measured resistance.

Your diode seems fine from that test.
[ Edit: But the question is, is that the diode in the charger, or from some modification to just get the EVSE connected. ]

this thread is getting way off topic, do you think it is worth it to make a new thread?

I was thinking that myself recently. I'll see what my fingers can come up with

Edit: done. Please pardon me only changing the subject lines for the first two posts, and this one. [ Edit: I relented and changed them all. It was just too distracting/confusing. ]

[nuggetgalore]

In reverse it is open both methods.

That's the most important thing.

If I measure this correctly , that is no where near the 2740 Ω as per diagram.

I cropped off other parts of the circuit, and the circuit is powered down, so that will affect the measured resistance.

Your diode seems fine from that test.

The diode and resistor in my mates 2012 model reads pretty much the same as on mine, yet charges with a reduced rate.As I understand, the EVSE sends a signal via the pilot pin to the charger but on my vehicle there is something else missing to take note of said signal. So where is this signal going to, and could that be added or activated in my charger?

[coulomb]

So where is this signal going to, and could that be added or activated in my charger?

It looks like it goes to the on-board charger. But also it may get processed (buffered? amplified? filtered?) and sent on to the EV-ECU:

Pilot signal to charger.png (77.6 KiB) Viewed 1433 times

Complete schematic that the above was taken is in the first post here: http://myimiev.com/forum/viewtopic.php?t=4079

could that be added or activated in my charger?

Good question. It might be worth running a wire from the front of the car to the charger at the back, to pin 9 of the connector in the charger, or if you are really lucky, find the grey [ edit: was green ] wire in the harness and connect to that without having to open the charger. If that works, neaten it up and make it permanent. [ Edit: there is no point, see next two posts. ]

Edit: the pilot processing hardware may be missing from the 2011 chargers, but then how can that mob offer a harness that fixes the problem?

Edit 2: you may also have to add the beige / pink / violet wire or parts thereof to the EV-ECU,

[ Edit: oops, wrong photo. Added second question. ]

[mark_hetho]

Edit: the pilot processing hardware may be missing from the 2011 chargers, but then how can that mob offer a harness that fixes the problem?

I think with the modified harness, it's just enough to convince the EVSE a car is there and to close a contactor. I don't think the car then reads the control signals to know what speed to charge at - it just pulls 13A. I think it's working on the assumption that you are using a public charger, and public chargers are going to be 15A+.

In my car the Control pilot pin is entirely missing, so the EVSE never knows there is a car there and the charging plug remains dead. The Gelco harness could just involve having the pin present and a fixed resistor?

[reecho]

I have questions for the brains-trust here:
1)How and where is the command given to the charger of how much current is drawn from the grid?
I know my model 10 charges with approx 12.5 A whether I use a common cable or one with the "brick".
The "brick"reduces the charge to about 8 + A on the model 12s.
I also know that when charging to full, for about half an hour or so the rate gradually reduces to about 1A and then shuts off ( unverified data indicates ramp down and shut off may be quicker if charging the battery when it was not down to the last few bars).
2) Activating the trigger on the gun increases the resistance between earth and PP from 150 to 480 Ohms and this signal shuts off the charging, what would happen if an other resistor ( a pot?) would increase it to something in between?
Could that be used to ramp down the charger or would it blow it up?
Things one thinks off while locked up because of the virus.......

No PWM charge current control on 2010 models.

[reecho]

So where is this signal going to, and could that be added or activated in my charger?

It looks like it goes to the on-board charger. But also it may get processed (buffered? amplified? filtered?) and sent on to the EV-ECU:


Pilot signal to charger.png

Complete schematic that the above was taken is in the first post here: http://myimiev.com/forum/viewtopic.php?t=4079

could that be added or activated in my charger?

Good question. It might be worth running a wire from the front of the car to the charger at the back, to pin 9 of the connector in the charger, or if you are really lucky, find the grey [ edit: was green ] wire in the harness and connect to that without having to open the charger. If that works, neaten it up and make it permanent.

Edit: the pilot processing hardware may be missing from the 2011 chargers, but then how can that mob offer a harness that fixes the problem?

Edit 2: you may also have to add the beige / pink / violet wire or parts thereof to the EV-ECU,

[ Edit: oops, wrong photo. Added second question. ]

2011 onwards I-Miev have a separate control box mounted on top of the AC charger. I suspect that does the PWM charge current control and in turn sends commands to the AC charger below it. I'm unaware of the protocol between that box and AC charger. It may well be CAN BUS.