BMW i3 CCS/charge port controller

Development and discussion of fast charging systems eg Chademo , CCS etc
bitterandreal
Posts: 55
Joined: Fri Jan 15, 2021 2:30 pm
Location: Netherlands
Has thanked: 2 times

Re: BMW i3 CCS/charge port controller

Post by bitterandreal »

PP of the vehicle charging port is not connected to the charging station.
The PP - PE resistor in the charging connector indicates the AC current capacity of the cable.
For CCS Type 2 fast charging it is usually 1K5.
That resistance would be measured higher by the LIM if the parallel resistor in the charging port is missing. (1K5 instead of 1K14)
But since the LIM works on most stations I suspect it is not the ABB station problem.
For AC charging higher resistance means lower cable current allowed which is probably not so much of a problem.
bitterandreal
Posts: 55
Joined: Fri Jan 15, 2021 2:30 pm
Location: Netherlands
Has thanked: 2 times

Re: BMW i3 CCS/charge port controller

Post by bitterandreal »

Jack Bauer wrote: Sun Oct 24, 2021 10:28 am Right, resistance testing update.
2k7 between pp and pe : No change to ac or dc charge behavior. ABB stations still say "charging will begin once power is available" and then time out. All other dc stations work as before. Only negative is the cp door flag from the lim stays in error. Presumably because it sees the resistance as a plug inserted.
Might it be possible that we are missing a signal or a whole CAN message which has information the LIM uses in the communication to the charging station and only ABB stations care about it?

Possible missing messages:
0x0FF Battery status/errors
0x2F5 Max battery voltage
0x1B5 Alive and CRC checksum
0x828 seconds counter
0x382 Hospitality
User avatar
CCSknowitall
Posts: 105
Joined: Fri Jun 04, 2021 1:47 pm
Has thanked: 1 time
Been thanked: 28 times

Re: BMW i3 CCS/charge port controller

Post by CCSknowitall »

bitterandreal wrote: Tue Oct 26, 2021 8:12 pm PP of the vehicle charging port is not connected to the charging station
That is not true for DC. PP/Proximity is used by DC chargers in both CCS1 and CCS2. Meaning, the charger actually monitors/measures proximity. It's primarily used to detect plug out in case of a broken lock or latch.
User avatar
CCSknowitall
Posts: 105
Joined: Fri Jun 04, 2021 1:47 pm
Has thanked: 1 time
Been thanked: 28 times

Re: BMW i3 CCS/charge port controller

Post by CCSknowitall »

bitterandreal wrote: Tue Oct 26, 2021 8:56 pm
Might it be possible that we are missing a signal or a whole CAN message which has information the LIM uses in the communication to the charging station and only ABB stations care about it?

Possible missing messages:
0x0FF Battery status/errors
0x2F5 Max battery voltage
0x1B5 Alive and CRC checksum
0x828 seconds counter
0x382 Hospitality
From what I could tell in the previous PCAPs, there were no missing data once PLC was established and the high level communication was running. The only issue was the EV maximum battery voltage which needed to change during charge parameter discovery. I remember specifically this caused the tritium veefill 50kW to only give ~8A, as that charger stored that value from charge parameter discovery and used it for the whole charge as the maximum voltage it would allow (essentially went into voltage mode if you requested higher current).

This particular issue seems to be something related to the initial condition at the ABB station that is preventing it from giving the 5% duty cycle, thus preventing the LIM from sending PLC packets to set up SLAC and start high level coms.

I highly suspect something's up in the proximity circuit of Damien's set up. Of course it could be some CAN message that is messing with proximity in some way. I don't know if that's likely though. Other stations should be monitoring PP too. But I suspect ABB is particular about the resistance being very accurate to the spec.
bitterandreal
Posts: 55
Joined: Fri Jan 15, 2021 2:30 pm
Location: Netherlands
Has thanked: 2 times

Re: BMW i3 CCS/charge port controller

Post by bitterandreal »

CCSknowitall wrote: Wed Oct 27, 2021 1:26 am I highly suspect something's up in the proximity circuit of Damien's set up. Of course it could be some CAN message that is messing with proximity in some way. I don't know if that's likely though. Other stations should be monitoring PP too. But I suspect ABB is particular about the resistance being very accurate to the spec.
Thanks for your explanation.
I very much hope this is the problem!
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: BMW i3 CCS/charge port controller

Post by Jack Bauer »

Will test asap. Only thing left is the resistor value. Seems 4k7 might be a good place to start.
I'm going to need a hacksaw
bitterandreal
Posts: 55
Joined: Fri Jan 15, 2021 2:30 pm
Location: Netherlands
Has thanked: 2 times

Re: BMW i3 CCS/charge port controller

Post by bitterandreal »

Jack Bauer wrote: Wed Oct 27, 2021 10:53 am Will test asap. Only thing left is the resistor value. Seems 4k7 might be a good place to start.
Yes 4k7 is most likely what you need.
bitterandreal
Posts: 55
Joined: Fri Jan 15, 2021 2:30 pm
Location: Netherlands
Has thanked: 2 times

Re: BMW i3 CCS/charge port controller

Post by bitterandreal »

I try to understand the differences in the signaling circuits.
I modified the J1772 version to match the IEC61851 / CCS2 circuit.
Please correct me...
IEC61851_signaling_circuit-01.png
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: BMW i3 CCS/charge port controller

Post by Jack Bauer »

Well, bad news. 4k7 resistor had no effect on the Ionity ABB station. Same "charging will commence once power is available" nonsense. While charging on a Efacaec station I repeatedly connected and disconnected the 4k7 resistor. No effect. Speaking of efacaec, Day before yesterday I charged no problem at a station I use frequently. Plug in, start up all good. 24 hours later I go back. 20 attempts no charge. LIM will not close contactors. A model 3 pulled in and tried to start charge. Elon was not happy and flashed his big red T in anger several times. The Tesla owner was on 5% and stuck badly. Then a shiny new VW ID4 pulls in. The station wont even recognise its plugged into a car. Now at this stage we have all been playing musical chairs as there is only one parking spot at the station that will reach with the ccs cable. I'm so used to the LIM doing its contactor dance when I start the car after a session that I hardly notice anymore. So I decide to give it one last go before calling a tow truck. Ping! contactors close and away we go. I notice on that station display thats all new and sexy so clearly a software update had been applied to get rid of all that pesky useful information like power delivered etc. So anyway the goose gasses up just fine and I pull out. In goes Elon and Ping! a happy green T. I left at that stage but I suspect the ID4 also gassed up fine afterwards.

Comments on a postcard please.
I'm going to need a hacksaw
MattsAwesomeStuff
Posts: 871
Joined: Fri Apr 26, 2019 5:40 pm
Has thanked: 278 times
Been thanked: 165 times

Re: BMW i3 CCS/charge port controller

Post by MattsAwesomeStuff »

*slow clap*

Was it perhaps doing a software update literally as you were waiting there?
User avatar
CCSknowitall
Posts: 105
Joined: Fri Jun 04, 2021 1:47 pm
Has thanked: 1 time
Been thanked: 28 times

Re: BMW i3 CCS/charge port controller

Post by CCSknowitall »

Jack Bauer wrote: Mon Nov 01, 2021 8:39 am Well, bad news. 4k7 resistor had no effect on the Ionity ABB station. Same "charging will commence once power is available" nonsense.
Sigh. I thought that would work. :x I'm really at a loss here.

Efacec stations suck. Blows my mind that it didn't at least go to an unavailable status while updating.
User avatar
CCSknowitall
Posts: 105
Joined: Fri Jun 04, 2021 1:47 pm
Has thanked: 1 time
Been thanked: 28 times

Re: BMW i3 CCS/charge port controller

Post by CCSknowitall »

Hm. Besides the resistance, what about that 5V source? Do you think that's enabled somehow over CAN?

The second proximity line on the LIM that's only for US - could that actually be the 5v output used to drive the voltage divider?

I don't doubt some stations are not actually monitoring proximity. It's important to do so in CCS1 markets as drivers can often overpower the lock and push the button on the handle down enough to trigger the switch. For CCS2 there's no such mechanism and the locking is completely controlled via the car, so maybe there is more trust on the station side.

ABB is known for following the standards pretty tightly. They are very likely monitoring proximity.

Efacec is known for being garbage. They are likely NOT monitoring proximity. :P

Also something to look into - monitoring pilot PWM on a scope to see if/when the 5% duty cycle is turned on.

Anyone want to probe prox to gnd with a DMM set to DC volts? Should see a bit less than 5V.
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: BMW i3 CCS/charge port controller

Post by Jack Bauer »

With a 4k7 fitted I see 4.69v between PP and PE. 5v with no resistor fitted.
I'm going to need a hacksaw
User avatar
marxx
Posts: 6
Joined: Wed Nov 03, 2021 11:46 am
Location: Attenkirchen / Germany
Contact:

Re: BMW i3 CCS/charge port controller

Post by marxx »

Dear all,

a good friend off mine sent me a link to some very boring clips on Youtube :D From there I found my way to this exciting thread.

First a little introduction ...

I live 50km north of Munich/Germany and I'm into e-mobility since 2014. I think it is fascinating that I could witness two technical revolutions in my lifetime. First the digital revolution (I started with a Sinclair ZX81) and now the e-mobility revolution. As I'm fascinated by both, I like to play in both worlds and try to combine them. You can read about some of my "adventures" on my website http://www.arachnon.de/wb/pages/en/welcome.php

My latest "project" was hacking the comfort can of the VW e-UP to write a module for OVMS for the e-UP. I started that module with the "T26" part directly connected to the comfort can. That project developed over time in collaboration with fellow nerds and we now have a module that is connected to the comfort can and to the OBD gateway. It was a steep learning curve and loads of fun. Here a link to the OVMS copy in my Github repository: https://github.com/devmarxx/Open-Vehicl ... /index.rst

But enough about me :)

As the e-UP OVMS Project is now slowing down, I'm looking for some new fun stuff and the BMW i3 LIM seems to be a great candidate. As I'm not a hardware guy, but willing to learn, this seems to be a good starting point to learn a lot 8-)

We also own a Renault Twizy with an owned battery. So my initial thought was, could it be possible to use the BMW i3 LIM to enable CCS charging for the Twizy? The Twizy has a 6 kWh battery and has a charging voltage maximum input of 58.8 V (as far as I understood). So my first question is, could it be possible to ask CCS chargers via the BMW i3 LIM for 58.8 V DC charging voltage? And how much ampere could we ask for (if possible at all)?

Second question is about these LIM's I found on Ebay: https://www.ebay.de/itm/154528022067

They don't have a MAC address on the sticker and have Chademo Pins, but here https://www.lllparts.co.uk/en/product/c ... 1355A01E54 is written that they are "COMBO" LIM's? Any ideas on those?

Well, enough for now. I'm looking forward to see where the BMW i3 LIM can take us :D

Greetinx

marxx
Charging station with Raspberry Pi, own Open Source Software
OVMS Leaf/e-Up, hitch Leaf, GE Android App, SteVe OCPP Raspi Image
http://www.arachnon.de/wb

Renault Twizy Life (owned battery) 06/2016,
VW e-Up! 07/2020.
Tesla Model 3 09/2019, LR AWD,
User avatar
CCSknowitall
Posts: 105
Joined: Fri Jun 04, 2021 1:47 pm
Has thanked: 1 time
Been thanked: 28 times

Re: BMW i3 CCS/charge port controller

Post by CCSknowitall »

Unfortunately CCS spec is 200V minimum. Now, in theory you could ask for a lower voltage but it’s likely you’ll fault the station or you won’t even get past charge parameter discovery (when you state your max pack voltage of ~59V). Now, CHAdeMO is spec’d down to 50V, so with a station that also has a CHAdeMO cable the power conversion hardware *should* support as low as 50v. However, in practice, not all stations can go that low, as evidenced by the trouble Zero motorcycle owners had with CHAdeMO and their ~100V pack voltage. You’ll also be under 50V most of the battery’s charge curve anyway.

The other problem is that pack is very small, you could maybe ask for 10-20A from the station and get 6-12 kW. This is L2 speeds. I seriously doubt the pack is going to support a 2C+ charge, unfortunately. You’ll just be tying up a 50kW+ station for a L2 charge which could cost more than L2, and tie up a resource that could be better utilized by a larger battery EV
User avatar
CCSknowitall
Posts: 105
Joined: Fri Jun 04, 2021 1:47 pm
Has thanked: 1 time
Been thanked: 28 times

Re: BMW i3 CCS/charge port controller

Post by CCSknowitall »

Jack Bauer wrote: Wed Nov 03, 2021 8:00 am With a 4k7 fitted I see 4.69v between PP and PE. 5v with no resistor fitted.
Ok the 4k7 resistor seems correct, that’s the right drop. I guess proximity is good. Might be good to monitor pilot and prox with a scope when connected just to be sure nothing funny is going on.

This is very bizarre though, if i3’s couldn’t charge at ionity, that would certainly make the EV news and be fixed by ABB and/or BMW.
mikeselectricstuff
Posts: 120
Joined: Sun Nov 08, 2020 11:33 am
Been thanked: 2 times

Re: BMW i3 CCS/charge port controller

Post by mikeselectricstuff »

I wonder if the I3 LIm has ever had any firmware updates to fix compatibility issues.
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: BMW i3 CCS/charge port controller

Post by Jack Bauer »

So the weird thing here is that my first ever actual successful CCS session was on an ABB/ Ionity station :


I also visited this station with a friend in his 2019 i3 and it charged just fine. Logs here :
https://github.com/damienmaguire/BMW-i3 ... rtup1a.csv

A pcap from an attempt earlier in the year shows we don't even get to SLAC not to mind any parameter exchanges :
https://github.com/damienmaguire/BMW-i3 ... fail1.pcap

So hopefully tomorrow I'm going to head up again and run some scope traces of CP and voltage readings on PP in the hope it may shed some light.
I'm going to need a hacksaw
User avatar
marxx
Posts: 6
Joined: Wed Nov 03, 2021 11:46 am
Location: Attenkirchen / Germany
Contact:

Re: BMW i3 CCS/charge port controller

Post by marxx »

@CCSknowitall: Thanx for your response. It was worth a thought.

marxx
Charging station with Raspberry Pi, own Open Source Software
OVMS Leaf/e-Up, hitch Leaf, GE Android App, SteVe OCPP Raspi Image
http://www.arachnon.de/wb

Renault Twizy Life (owned battery) 06/2016,
VW e-Up! 07/2020.
Tesla Model 3 09/2019, LR AWD,
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: BMW i3 CCS/charge port controller

Post by Jack Bauer »

I'm going to need a hacksaw
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: BMW i3 CCS/charge port controller

Post by Jack Bauer »

I'm going to need a hacksaw
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: BMW i3 CCS/charge port controller

Post by Jack Bauer »

So this is where we are. Went to Ionity. We get PP voltage and 5% duty from the station. Not only do we see it on the scope but the lim recognises it as 5% not ready. The problem is the LIM just sits there and the station does its authentication dance, times out and drops pilot to static and LIM shunts down. Drive over the the Efacec and works first time. Just to refresh some memories I also had a fancy new Delta UFC (https://evcharger.delta-emea.com/ufc200/en/) behave in a similar way a few months ago. Something changed between approx June 5th and June17th this year as I was able to charge on Ionity on the 5th but this nonsense started on the 17th.

The only physical difference between the non working and the working stations is the plug in and auth dance. Efacec and the working stations do auth first then plug in. Inoity and Delta want the plug in first. Does that mean anything? Don't know myself but I'm pouring over CAN logs today and some sort of a picture is emerging from the mist. Just not sure what it is yet:)
I'm going to need a hacksaw
User avatar
CCSknowitall
Posts: 105
Joined: Fri Jun 04, 2021 1:47 pm
Has thanked: 1 time
Been thanked: 28 times

Re: BMW i3 CCS/charge port controller

Post by CCSknowitall »

Ok we’re getting somewhere. Slowly.

In DIN spec, you can set up communications and sit in the stage of “contractAuthentication” for some time (minute or two, depending on vehicle). A number of stations/networks encourage or require plugging in first, which will result in setting up SLAC, TLS (if the car wants), picking DIN/ISO 15118, a few other odds and ends, then contract authentication where you’ll sit until getting the ok from the backend (EVSE Processing: finished), then onto charge parameter discovery, cable check…

It appears that the LIM is somehow not ready (no real surprise there) and is ignoring the 5% duty cycle. ABB has a strict timeout (probably around 10 seconds) and so times out before the LIM can boot the Qualcomm chip.

For the ones requiring authentication first, they’re likely staying in state B1, 9V with no PWM, for some time that allows the LIM to fully boot before then the LIM is reacting correctly to B2 5% PWM.

Sooo… look at your setup compared to the BMW i3 can logs. Perhaps it’s doing some setup/wake up that you’re doing slightly later. It wouldn’t surprise me that there’s a bug in the LIM that causes a race condition where it gets hung up if you give it 5% PWM before it’s ready. Something that the rest of the modules need to code around, booting the LIM once the port flap is open or something.
muehlpower wrote: Wed May 26, 2021 2:38 pm 0x3B4 D4 low nible: status pilot
0=no pilot
1=10-96%PWM not charge ready
2=10-96%PWM charge ready
3=error
4=5% not charge ready
5=5% charge ready
6=pilot static
What does “not charge ready” mean in this context? DIN spec says you are supposed to go through SLAC within 8s of receiving 5% duty cycle from the EVSE.
bitterandreal
Posts: 55
Joined: Fri Jan 15, 2021 2:30 pm
Location: Netherlands
Has thanked: 2 times

Re: BMW i3 CCS/charge port controller

Post by bitterandreal »

Good question what is pilot status 4 (5% not charge ready) meaning?
- Is the Qualcomm chip not ready? (It seems the Qualcomm chip is ready very quickly on the Efacec station)
- Is the station sending the 5% PWM but no PLC (charge parameter discovery) which hangs or times out the LIM?
- Is a charge ready signal on the CAN bus missing?
- Or do we send one signal to early? Something we should not send before pilot status 5 (5% charge ready)?
...
Post Reply