[WIP] Mercedes 310E (310D campervan) - GS450H/GS300H

[Jacobsmess]

And some pics of the module dimensions...

[Jacobsmess]

A few more details to confirm.
Big modules are 9S3P (27 cells) and weigh 27.5kg 8 per pack
Small modules are 8S3P (24 cells) and weigh 24.7kg 3 per pack

Cells are Panasonic prismatic and are 51ah each according to spec.
So large module is around 5kWh and small module 4.5kWh.

[Jacobsmess]

Battery module connectors
Large modules seem to have different wiring colours but same pins used in connectors.
Small modules are the same wiring colours and same pinouts by the looks of things (yet to confirm on all modules)

[Jacobsmess]

Contactors, eight of them. One board with two small precharge contactors (G3841-48030) and two larger ones (AEVT760122)
Then a larger board with four large contactors doing positive and negative connections. They are all AEVH900122M07.

[Jacobsmess]

Battery computer assembly (BMS?) And Battery voltage sensor... (Other BMS?)

[Jacobsmess]

Innards if BMS Battery computer thingies

[tom91]

The one with the 5Ics looks to be a "dumb slave" board. 5x the same BMS front end chip with no higher level controller.

The one with the big IC is the "battery controller" I would say or the brains of the operation.

What are the part numbers on the 5 ICs on the "dumb slave" board?

[Jacobsmess]

What are the part numbers on the 5 ICs on the "dumb slave" board?

This one?

[tom91]

This one?

yes that one, the one with the repeating circuit layout.

[Jacobsmess]

yes that one, the one with the repeating circuit layout.

SF784
11D2416

Google doesn't turn up anything for me....

Tried scanning the QR but can't get my camera to focus on it well enough

[Jacobsmess]

HC4051AS 050 on the other side x5 much smaller so assume you meant the larger previous ones

[Jacobsmess]

9176J F61J6 14 of these on the side with the big ICs

[tom91]

Shit is fuly custom spun Denso chip that is why the "D" logo at the bottom.

https://www.denso.com/global/en/news/ne ... 01013-g01/

[Jacobsmess]

Shit is fuly custom spun Denso chip that is why the "D" logo at the bottom.

https://www.denso.com/global/en/news/ne ... 01013-g01/

Yes I read that, the image they have has a similar number but not the same. So not usable at all? Would it be possible to swap them out for something not custom or is there a way of reusing the BMS in anyway?

[tom91]

For you the best bet and more cost and time effective is replacing the BMS.

More electronically and programming inclined could attempt to reverse engineer parts of the BMS or try talking to the whole thing over CAN.

[tom91]

Good info on the pack in the dismantler info https://www.lexus-tech.eu/HYBRID/HVDM/EN/hvdm.pdf

[Jacobsmess]

For you the best bet and more cost and time effective is replacing the BMS.

More electronically and programming inclined could attempt to reverse engineer parts of the BMS or try talking to the whole thing over CAN.

Hmm, I'm not sure on cost, it's a 288 cell pack so even the OI BMS monitoring all cells would be around £1600 plus connectors....
I'm tempted to see if I can get anywhere with this, would there be any obvious place to start? Given it's split into two modules, is it possible the dumb module could be controlled? Seems I might as well test this to destruction and potentially learn some things along the way. I'm in no massive rush to get the BMS sorted so time is on my side there. Unfortunately it's unlikely I'd find another Lexus UX300E owner willing to provide can logs, but that article mentions the use in Yaris BMS's so perhaps that is a useful source of information....

[tom91]

You need to look at getting as much info on the RX300e as possible and learn about electronics and CAN. These kinds of skills take time to learn and master before you can even hope to get started on reverse engineering the BMS and getting it to work.

Relying on others to offer to help you or hope that info will be available somewhere is not going to get this resolved.

If you do not value your time then of course getting things you have working is cost effective. However when things need to get done money is the only way to solve an issues involving time.

[Jacobsmess]

I appreciate that and I'm aware I'm often in need of help but I'm also willing to learn.
What I need mostly is pointing in the right direction. I also have friends who are much more proficient with software and electronics than I am (although not necessarily automotive) whom I can call on for help when needed. I also suspect if this IC is used in the lexus UX300E and at least the Yaris, it's likely used in a lot of other Toyota kit and is likely that anything that can be reversed engineered with this BMS would likely have greater implications for people refusing other Toyota BMS kit

[Jacobsmess]

I had some more time to look at these modules today and I'm a little confused.
I've been trying to rematch the wiring harness with the module (silky me for not taking photos whilst dismantling!)
Anyway,
The modules seem to have 17 wires to their terminals, however, the connectors on the BMS harness only has 16, so there is one unused cell tap.
Each module has 3 temp sensors, with two wires each, then 9 (on the 8S3P module) or 10 (on the 9S3P module) cell taps on the positive and 1 on the negative. Is this normal BMS architecture?

[tom91]

Start by mapping out by looking at all the module connectors and physical layout. Identify if the modules and their pinouts are the same this will give you clues.

Due to the layout of the cell voltage monitoring slave the question arises is where the reference between the various modules is done. There will be 5 cell groups to match the number of ICs.

As you have 96 cells in series this does not nicely divide into 5 so there are some tricks going on in the harness. So it will be a lot of fun for you to label all connectors and buzz them through one by one and document your findings.

This is why planning for what you are doing is the way to go.

[Jacobsmess]

Thanks, I'll try do that this morning. I've gone through all the connectors between the BMS and batteries. Some have obvious matches where the BMS harness connector only fits the battery. Others fit may but also have wiring that does not match to any of them. There are always unused pins on these ones.
Difficult to explain in words but I'll try get some pics and draw it all out this morning.

[Jacobsmess]

Some scribbled diagrams of the wiring. Most seem to have a repeating pattern, I've still got to check the others.
C = cell tap
T = thermistor
Pink highlight = positive terminal

The connectors are drawn from the wires from the battery side.
As can be seen, some have 17 wires, some 16. It seems on the BMS side, only 16 wires, with one of the 17 cell taps remaining unconnected.

Any insight is appreciated, I'll type this all up in a more readable format later on...

[Aragorn]

It doesnt really matter wether the tap is on the positive or negative, they're joined together. The tap sits between each parallel block. They surely must all reach the BMS??

You will need to map out the exact order of the cell taps vs the connector. Colors alone are not enough as there appear to be duplicates.

You appear to have an extra cell tap on certain modules on the bottom two drawings but not the first?

[Jacobsmess]

You appear to have an extra cell tap on certain modules on the bottom two drawings but not the first?

Also, I plan to map them out later on also but the thermistor wires seem consistent with their colouring (paired white, yellow and pink)
Yes that's the confusing thing. Some 9s3p modules have 11 taps some have 10.
And yes it seems some duplicated taps (where a group of 3 have two taps) don't make it to the BMS