Renault Zoe BMS

joel
Posts: 6
Joined: Tue Apr 12, 2022 3:17 pm
Location: Germany
Been thanked: 1 time

Re: Renault Zoe BMS

Post by joel »

Hi wizo, here you go:
Screenshot from 2022-07-15 15-26-18.png
Screenshot from 2022-07-15 15-26-55.png
The connector is made by Yazaki, part numbers are 7282-8854-30 for male and 7283-8854-30 for female connector sides, respectively.

You can get them on a lot of online shops, cheapest is probably AliExpress, e.g. here:
https://de.aliexpress.com/item/4000174903780.html

I hope these infos are something you can work with!

Do you also have a working Zoe or only a battery? I am sort of stuck on my process of reverse-engineering the handshake and could use somebody who is able to use a filter (e.g. laptop or microcontroller with two Can Interfaces) in order to let only some messages through. That way we could identify which can frames are needed for happy-flow operation and which can frames are unnecessary.

Best Regards,
Joel
User avatar
Sophiefxx
Posts: 5
Joined: Fri Dec 31, 2021 3:55 am
Location: Goole, East Yorkshire, UK
Been thanked: 6 times

Re: Renault Zoe BMS

Post by Sophiefxx »

I've got a working Zoe 40kWh Joel, from 2018. I'm about to add Chademo charging, and will be splicing into the CAN wiring from the battery.

PM me if you like, and I am happy to experiment or get logs if it helps you.

Sophie
:!: Unfinished projects clutter lives,
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Zoe BMS

Post by TheMiller »

pemtek wrote: Sat Apr 23, 2022 5:16 pm OK so solved that somewhat. Thanks to paaa I used one of the can logs and found that it wakes up and starts sending data when I send 0x423 and then stops again after the log is turned off. Now to work out what everything means..

Phil
Did you connect to the bms-CAN directly ? Which protocol and baudrate did you use ? I've got a 2011 pack from a Kangoo with the "white BMS" and trying to connect to it but yet without luck. Because I always receive a "can-error" it seems to be a very basic problem. When powering up the BMS it only changes CAN levels to high for about half a second and then drops them to low again. No message - nothing. So I have to wakre up this sleepy box by any means. Unfortunately there almost no information to be found in the internet regarding the Kangoo battery. Though the BMS looks like the "white Nissan BMS" there are differences to be found in the wiring. I'm stuck.

Gerald
woleg
Posts: 43
Joined: Thu May 05, 2022 6:09 pm
Has thanked: 10 times
Been thanked: 9 times

Re: Renault Zoe BMS

Post by woleg »

TheMiller wrote: Tue Oct 11, 2022 7:23 am I always receive a "can-error" it seems to be a very basic problem.
0x423 needs to be sent every 100ms like clockwork.
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Zoe BMS

Post by TheMiller »

woleg wrote: Tue Oct 11, 2022 1:14 pm 0x423 needs to be sent every 100ms like clockwork.
Thanks Woleg - sure this is important. Unfortunately it seems that I haven't found the correct protocol yet. 11Bit ID/500kbaud (ISO15765-4) does not work. Leads to "can-error". So sending the 0x423 is still out of sight for me... I'd like to have ONE little chat with the BMS before I get an Arduino to do the job. Just to proof that its living.

sigh
Gerald
woleg
Posts: 43
Joined: Thu May 05, 2022 6:09 pm
Has thanked: 10 times
Been thanked: 9 times

Re: Renault Zoe BMS

Post by woleg »

I'd check the simple stuff. Powerdown the BMS, disconnect your can interface, check the Kangoo can resistance between CanL & CanH (120ohms). If that's fine then confirm CanL & CanH on whatever PC interface you are using.

The frame you should be sending is id: 0x423 payload: [0x33, 0x00, 0xFF, 0xFF, 0x00, 0xE0, 0x00, 0x00] @ 500kbit/s every 100ms

Still SOL? Use SavvyCan (all platforms) or Busmaster (Windows only), Savvy with a cheap ESP32 + transceiver works fine for me.
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Zoe BMS

Post by TheMiller »

Yeah, thanks for the hint. Indeed I did already check this but I did it again (you never know...) and it's OK. Because I am connecting to the BMS's CAN, maybe I'd need another protocol ? Unfortunately nobody ever noted anything about the BMS-CAN. It looks like the BMS is simply dead. But for me it's hard to believe that BOTH boxes are dead. I even opened the boxes to look (and sniff) for signs of burnt items - nothing. As far as I know this battery was in service I bought it. They told me that besides some mechanical problems the Kangoo won't charge anymore and there is a problem with the charger (bla bla bla). But it was a good deal anyway. The battery only cost me $ 1.000,-.

Just to let you know what my project is all about: Re-using an APC Smart UPS VT with 40kW(!) 3 phase as a PV storage. Replaced the original (dead) lead batterys with the two stacks of the Kangoo battery. The only thing missing is yet the BMS. Sure I could go with Batrium, Rec or Emus, but my intention is to reuse as much as I can. Man this thing humms ! 40kW on 3 phases, 17kWh (the Kangoo battery is a little worn). The UPS cost me another $ 1.500,-. I will put a documentation online as soon as I got this problem solved.

cheers
Gerald
woleg
Posts: 43
Joined: Thu May 05, 2022 6:09 pm
Has thanked: 10 times
Been thanked: 9 times

Re: Renault Zoe BMS

Post by woleg »

The Kangoo BMS will not send any frames without a correct message on 0x423, it will appear dead. The fact you are getting a rise in current draw suggests that it is receiving frames just not interpreting them properly.

I would build a SAVVYCAN setup with a $6 ESP32 + ESP32RET https://github.com/collin80/ESP32RET
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Kangoo BMS

Post by TheMiller »

Took me some time to get an Arduino with a CAN-shield up and running and receiving data from the BMS. Your hint with the 100ms "clock" was THE key-information! Now this looks promising but the BMS stops responding to the 0x423 payload: [0x33, 0x00, 0xFF, 0xFF, 0x00, 0xE0, 0x00, 0x00] after about 15 to 20 cycles - this is different after every restart.
Usualy I receive 10 responses with id 0x155 payload: [0xFF, 0x87, 0xD2, 0x94, 0xFF, 0xF8, 0x03, 0xFF] where the second and third byte are changing sometimes and after those 10 responses there's one with id 0x445 payload: [0x80, 0x55, 0xF8, 0xFF, 0x2A, 0x8C, 0x00] - only 7 bytes.
So only two questions left: Did anybody already decode this garbage and where can I find this valuable information (many hours of searching the net did not reveal anything for me) AND why does the BMS stop responding - maybe it's waiting for some "Well done, BMS. Go on!" or something like this ?
Sure I could find out by myself but re-inventing the wheel does not make it "rounder" (outch!)

cheers
Gerald
User avatar
dimonlipko
Posts: 20
Joined: Thu Apr 02, 2020 9:28 pm
Location: Ukraine, Kiev
Been thanked: 5 times
Contact:

Re: Renault Zoe BMS

Post by dimonlipko »

You can connect elm 327 to battery can bus, and use app CanZE to read battery data.
woleg
Posts: 43
Joined: Thu May 05, 2022 6:09 pm
Has thanked: 10 times
Been thanked: 9 times

Re: Renault Zoe BMS

Post by woleg »

Which is fine but not much use when you have an embedded device connected which can interpret can frames.

If readers go back through the history of this thread they can find CanZe information. Important data is held in ISO TP frames. https://en.wikipedia.org/wiki/ISO_15765-2
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Zoe BMS

Post by TheMiller »

Well, because connecting an ELM327 directly to the BMS-CAN and using CANZE did not work, I decided to go with the Arduino. The BMS-CAN seems to use different IDs. The packets CANZE sends do only work when connected to the OBD-CAN.
So Woleg's hint with the 0x423... was essential. Finaly I could find what Woleg ment with "If readers go back..." and dug deep into the CSV-tables. They look a lot like mine, but with a lot more stuff, that I haven't seen on mine yet.
I must have overlooked these loggings in the first place - thank you again Woleg!
And you are absolutely right - the Arduino connected to the BMS-CAN gives me much more opportunities...
At the moment I'm trying to find out the meaning of the packet IDs 0x425 and 0x445 (although I don't even know what the payload of the 0x423 packets is made of, yet...).
And yes, ISO-TP docs hold a lot of valuable information to help in this matter - 'll have to dive deeper into the protocol ;-/
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Zoe BMS

Post by TheMiller »

Woleg: Your hint with the 0x423 was key to receiving back data from the BMS. Now I've been studying CANlogs for about a week but with no result. All the findings of other people on the web are about the Nissan Leaf BMS, but as it seems this is VERY different from the Kangoo's. Using the well documented 0x79x sequence does not wake up the Kangoo's BMS. The CAN packets the BMS sends in answer to the 0x423 sequence have IDs 0x155, 0x424, 0x425 and the payload seems to have nothing in common with the Leaf packages.
It's hard for me to believe that I'm the first one trying to decode this garbage (?)
My findings so far:
* Kangoo BMS seems to have NOTHING in common (except the hardware) with the Nissan Leaf's
* Nissan CAN message don't work on the Kangoo
* until now the messages I received from the BMS did not reveal any schematic to me.

As you are the ONLY person who ever mentioned the 0x423 sequence, I'm sure that you know much more about this... 8-)
Any little hint ?
I'd like to post some beer here for you, but I fear It won't work out as expected.. ;-)
woleg
Posts: 43
Joined: Thu May 05, 2022 6:09 pm
Has thanked: 10 times
Been thanked: 9 times

Re: Renault Zoe BMS

Post by woleg »

I’m on a few forums and lose track of what is already posted where. This spreadsheet is fairly accurate and in the public domain (not my copy). Tabs you are looking for are Battery (far right) for rapid data, and LBC for cell level data. To be fair, most of the module data in Battery covers most of your needs. If you fail to extract a certain value or it returns garbage let me know, there’s a few tricks you have to do for the odd one, plus the sheet isn’t perfect for this application. Happy hunting!

https://docs.google.com/spreadsheets/u/ ... M/htmlview
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Zoe BMS

Post by TheMiller »

WOW that must have been a LOT of work to build this table!
Funny that the 0x155 messages in the battery tab are completely unknown. These are the messages I'm receiving from the BMS in masses. I tried to decode the few 0x445 messages with the help of this table but with no luck. The easyest test would have been to look for highest and lowest cell voltage but this did not work out as expected.
So what I will do in the next days is to fiddle with the cells and look for changing values in the 0x155 messages. It realy looks like the cell voltages must be in there although I have not find any evidence of some kind of index yet.
When I succeed with this I'll complete the sheet with my findings.
Again it seems to me that the Kangoo BMS is different to any other BMS from Renault and Nissan so far.
Many thanks for this valuable link - this makes a difference in "hunting" for me.
And yes, I'm also reserching in so many forums that I lost track completely. Good to hear that I'm not alone with this problem :)
woleg
Posts: 43
Joined: Thu May 05, 2022 6:09 pm
Has thanked: 10 times
Been thanked: 9 times

Re: Renault Zoe BMS

Post by woleg »

I have a few values decoded from 0x155, not at my computer at the moment so will edit this comment later.

Edit: I think these notes for 0x155 are accurate..or close anyway. https://i.imgur.com/beWYytk.png
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Zoe BMS

Post by TheMiller »

Still my records don't correlate with anything I've seen so far.
When I read that the 0x445 records hold information about highest and lowest cell voltage, I thougt that having only those two values would be efficient for my task: Monitoring the pack not to damage cells.
So I tried decoding these records first and leave the 0x155 for some time later. The result was stunning:
Looking at only the 0x445 records (CAN-filter):
80, 57, aa, 73, ff, ff, 80
80, 55, f8, 73, ff, ff, 80
80, 55, f8, 73, ff, ff, 80
80, 55, f8, 73, ff, ff, 80
80, 55, f8, 73, ff, ff, 80
80, 55, f8, 73, 29, 8b, 80
80, 55, f8, 73, 29, 8b, 80
80, 55, f8, 73, 29, 8b, 80
80, 55, f8, 73, 29, 8b, 80
80, 55, f8, 73, 29, 8b, 80
80, 55, f8, 73, 29, 8b, 80
..... until the end of the world...
So IMHO this does not look like it would hold information about highest and lowest cell voltage. Why should it change from first to second record and then again from fifth to sixth record and then remain the same forever ?
Also this sequence looks the same EVERY time I start my logging (changing after 1st and 5th record).
IF the 0x445s should realy hold HV and LV then in my understanding - without knowing the details - every record should look the same, until one cell leaves the reported range ? Right ?
My pack is lying around doing nothing. There is no load and no charger connected. You see what my point is...

If you don't have the current sensor connected, the 0x445 records show ff in B4 and B5.
Again, if B4 and B5 are holding HV and LV why would disconnecting this current sensor make a difference ? HV and LV have nothing to do with the current flowing in or out of the whole pack. Verrrrry strrrrrange.

This evening I'll take the time an measure ALL the 96 cells and see if I can find a correlation with the records - stay tuned.
woleg
Posts: 43
Joined: Thu May 05, 2022 6:09 pm
Has thanked: 10 times
Been thanked: 9 times

Re: Renault Zoe BMS

Post by woleg »

That's not an ID I use and haven't previously decoded it.

Looks like

Code: Select all

Temp 17, highest cell volts 3970mV, lowest cell volts 3790mV
although I'm not sure on the low cell volts accuracy, need comparative data from 0x79b to check.

https://play.rust-lang.org/?version=sta ... 0025371250
TheMiller
Posts: 9
Joined: Tue Oct 11, 2022 7:09 am

Re: Renault Zoe BMS

Post by TheMiller »

Wow! Found a typo in my code when comparing it to yours. Now I got the right readings!
Got to finish my PV storage project now to stop my wife from complaining......before I sit down and try to decode the rest ;)

So finaly: Many, many thanks to you. Your hints led me in the right direction. Would have taken me for ever without them.

Documentation is to follow in the next weeks.

Cheers
Gerald
woleg
Posts: 43
Joined: Thu May 05, 2022 6:09 pm
Has thanked: 10 times
Been thanked: 9 times

Re: Renault Zoe BMS

Post by woleg »

Noticed that there is still no documentation on this protocol so I've put together something.

https://github.com/rand12345/toucan_con ... n/ze40_bms

Edit, updated link.
wochuku123
Posts: 1
Joined: Tue Aug 01, 2023 10:48 am

Re: Renault Zoe BMS

Post by wochuku123 »

Hi,

I tried to capture logs from OBD2 port but there is no data. My car is Renault Zoe 2021. OBD port is near the brake pedal. Could you help me ?
User avatar
dimonlipko
Posts: 20
Joined: Thu Apr 02, 2020 9:28 pm
Location: Ukraine, Kiev
Been thanked: 5 times
Contact:

Re: Renault Zoe BMS

Post by dimonlipko »

In Zoe 2021 OBD port connected to gateway, and if you want listen CAN BUS you need connected after gateway to powertrain can.
geamanug
Posts: 6
Joined: Wed Oct 18, 2023 4:10 pm
Has thanked: 1 time

Re: Renault Zoe BMS

Post by geamanug »

Hello there. Reading the whole thread I got to the conclusion that I should follow the same path to get can logs out of a Dacia Spring 27 kw battery ( same Renault group). Any suggestion greatly appreciated.Great work by the way...
geamanug
Posts: 6
Joined: Wed Oct 18, 2023 4:10 pm
Has thanked: 1 time

Re: Renault Zoe BMS

Post by geamanug »

I would really appreciate some assistance...I already made a communication extension cable ( original connectors) so I can sniff data out of the battery. Using a p-can device on busmaster or savvycan I am pretty sure I will be successful on getting a log out of the battery. Not sure what message should I send but I have a hunch protocol might be similar to renault zoe...Thanks in advance.
Post Reply