Volvo V60 battery pack

[Bart]

Hej all,

I recently bought a Volvo V60 hybrid battery pack, and since I did not find any info online, I decided to share some pictures and basic findings:

- The Li-ion battery pack is advertised as a 12 kWh pack, and estimating from my first charge, it seems to hold about 10-11 kWh (pack is from 2013).
- Each (LG Chem) battery block is built in a 10s configuration, and there are 10 of those, making it 370 V nominal voltage.
- The approximate cell block size is: 120x310x185 mm.
- There is a BMS (with thermistor) attached to each block, and I have no idea how to communicate with it. Maybe someone does, or sees similarities with other (LG) systems sold?
- The pack comes with two contactors and a current sensor.
- The battery box is built to be stored inside a car, it is built of rather thin sheet metal, and seems are sealed with rubber putty.
- The volumetric energy density seems quite high, compared to other hybrid batteries.

All in all, a very nicely built pack.

EDIT:
- I have weighed the battery modules, and these blocks weigh 11kg per block. So, I guess the entire pack might weigh about 120-130 kg.
- The total pack size is about 115x65x21 cm (largest dimensions).

Cheers,
Bart

[PacEmaker]

Very good work Bart - do you have approximate dimensions of the full pack?

It's times like these I wish I lived in Europe - as usual the interesting models never appear in Australia until 10 years later ... or never at all.

[MattsAwesomeStuff]

370v in such a small pack is a great beginner pack for an EV conversion.

[Isaac96]

370v in such a small pack is a great beginner pack for an EV conversion.

It's very interesting that Volvo is using 100S instead of the more common 96S. Maybe they wanted more modules while still having them all identical. In any case, it could possibly pose issues since most OEM inverters are designed for 96S and might have overvoltage lockouts.
The density does look great though. Maybe we can get a similar deal to that for the BMW hybrid packs and make these another standard conversion pack.

-Isaac

[johu]

I don't think the extra ~16V are going to bother any inverter.
Can you find out the total weight or module weight?

[tom91]

Any diagrams of the wiring internal of the pack? As it looks canbus controlled so would be good to grab some captures.

[arber333]

I don't think the extra ~16V are going to bother any inverter.
Can you find out the total weight or module weight?

Well i trieed with my 102S pack and at 420Vdc inverter shuts down. So i use my cells up to 4.01V per cell.

[Bart]

Thanks for all the replies! I will come back with answers after I can have a look at the battery this weekend.

EDIT:
- I have weighed the battery modules, and these blocks weigh 11kg per block. So, I guess the entire pack might weigh about 120-130 kg.
- The total pack size is about 115x65x21 cm (largest dimensions).
- I have no wiring schematics yet, and it would be very good to try to collect some CAN captures. Unfortunately, I am not very well versed in CANbus yet, so it will take some time for me to get more information about. Could it be similar for LG Chem modules built in different cars, perhaps?

/Bart

[zilion]

Thanks for all the replies! I will come back with answers after I can have a look at the battery this weekend.

Any luck with this battery hacking, Bart?

[Bart]

Hey Zilion,

Thanks for your interest. I haven't done much hacking on it yet. I did realize that splitting the pack in two (2x50s instead of 1x100s) would be best for my setup, and I don't know how the current electronics in it will deal with that. I am anticipating building replacement boards for each 10-pack, possibly with an openinverter BMS system, if it exists, and communicating with them. I know Johannes is working on something like that, and since I am quite slow, it might already exist when I start needing it .

But anyhow, it might be possible to communicate to each 10-pack via CANbus, but as said currently there is more CANNOT than CAN to it for me. Maybe I should invest time in it and see if I can get to talk to any of the boards in this battery via CANbus.

I did find out that this pack fits very elegantly as-is behind the front seats of a Peugeot 106 electrique, I guess we can call that luck, as well. My main focus now is on the inverter.

Cheers,
Bart

[FJ3422]

...But anyhow, it might be possible to communicate to each 10-pack via CANbus...

That's possible. I am going to use these packs for my build, and reverse-engineered the BMS-slave communication. Alle slaves are uniquely addressed, and reply the cellvoltages & temperatures after a CAN-request message. From my scratchpad:
- 0X17 (1byte, 100ms rate) requests all slaves to report cell voltages
- 0x1B (1byte, 100ms rate) requests all slaves to report temperatures (2 in the pack, one on the PCB (around the bleeder resistors)
- 0x1F (2bytes, 100ms rate) request some other status messages.

- Balance resistors are activated with telegram 0x1C (3bytes, 100ms rate), but there are some restrictions on the maximimum number of resistors you can simultanously switch on (I guess to limit dissipation).

I will document the messages it one of these days and put it online, or put my BMS-software online when it's fully working.

[zilion]

...But anyhow, it might be possible to communicate to each 10-pack via CANbus...

I will document the messages it one of these days and put it online, or put my BMS-software online when it's fully working.

I have opportunity to buy 2 of these packs. Your comments bring me closer to this decision.

FJ3422, I'm eager to see this CAN messages. I think, that there could be some CAN-dumps for V60 D5 Hybrid on Swedish Volvo forums?

[FJ3422]

FJ3422, I'm eager to see this CAN messages. I think, that there could be some CAN-dumps for V60 D5 Hybrid on Swedish Volvo forums?

Remember that just a CAN-log of the HS-CAN or SENSOR/HV-CAN would not help, the internal CAN-bus of the pack is the one you need to use. There is a BMS-master in the pack (large right-up standing module) that I didn't use, I just simulated its function in an arduino (like the SimpBMS does).
If you would like to experiment, just supply the pack with 12V and you can log the communication between the BMS-master & 10 slaves.

[Clanarn]

Nice work. Is there any similarities to the chevy volt bms? It seems like it also uses the L9763. And do you have the pinout for 9pin connector on the slave board?

[FJ3422]

Yes there are some similarities, but scaling & adressing is completely different.
From my scratchpad:
Pin 1 = +12V ('rough vehicle 12V', switched by the BMS master, not regulated)
Pin 2 = CAN-L
Pin 3 = CAN-H
Pin 4 = Digital input (internally 50K pulled up)
Pin 5 = GND (only first slave has this pin connected to the master, guess for some kind of 'first slave present' detection)
Pin 6 = Digital output (open collector with 1K in series)
Pin 7 = GND
Pin 8 = CAN-L (parallel to pin-2)
Pin 9 = CAN-H (parallel to pin-3)

Pin 4 & 6 are daisy-chained through the slaves. I suspect for slave addressing. I might have triggered it because the modules on my bench have different addresses to what I initially marked them with. Will test that within a few weeks when having a full pack wired up.

[Peter]

Hi FJ3422. I am using the Texas Instruments software to read the BMS modules (using the bq76PL455A) but would like to use Arduino to communicate. Would it be possible to see the programme you wrote for the Arduino please? Thanks in advance. Peter

[FJ3422]

Hi FJ3422. I am using the Texas Instruments software to read the BMS modules (using the bq76PL455A) but would like to use Arduino to communicate. Would it be possible to see the programme you wrote for the Arduino please? Thanks in advance. Peter

Hi Peter,

Sure. See attached an unfinished version that requests & reads the 10 cell voltages & 3 temperatues from every slaves. Balancing untested / presumably not yet working, did only test it manually by sending the right CAN-messages. Uses parts of the SimpBMS code, but converted in several ways to work with my own vcu-hobbyproject (dualCAN Teensy 3.6 thing, I can send you the schematics if interested).

VCU_LG-BMS_V1_2021-08-03.rar

(18.77 KiB) Downloaded 669 times

[Peter]

Thanks FJ3422. I will try and extract the BMS part only. Schematics would be a big help and very kind of you. Peter

[victor_norberg]

I have a pack like this for sale in Sweden.

Good condition.

Open for offers, shipping within EU

[manny]

Hi I have two of these volvo packs that i like to use for home energy storage.

I was reverse engineering the can communications before I found this forum. The cell voltages and temp sensors I am able to read. the balancing is not yet working. Hopefully someone can help me out?

Thanks manny

[Peter]

Hi Manny. Look on the Battery Management section under New bms for bq76...., catphish has a board that can do all. Peter

[manny]

hi Peter,

I think the battery packs and bms slave boards are different between the V90 and the V60.

The V60 has 10S modules and uses can communication

The V90 has 16S modules and uses "proprietary daisy chain interface" (catphish github).

Or a am missing something.

Manny

[mjc506]

Likely the same bms IC, just configured differently? Catphish's daisy chain interface should work for both (might need to tell it only 10 cells etc) as should my UART option (https://github.com/mjc-506/bmsSerial)

[FJ3422]

The V60 has 10S modules and uses can communication
The V90 has 16S modules and uses "proprietary daisy chain interface" (catphish github).

Correct ! Completely different design. V60 BMS has full insulation between LV / HV sides and onboard dc-dc converters for supplying the BMS-chip. So no hassly with putting the BMS-chip into sleep mode preventing it draining the cells.

[manny]

Hi FJ3422

Did you found out how the ballancing works?
If so can you tell me how it works.

Thanks Manny