Electronics > Power/Renewable Energy/EV's

BMW iX Xdrive40 75kWh battery teardown

(1/2) > >>

I posted this over at AEVA, but I thought it'd be interesting here too.

I was fortunate enough to get hold of a dismantled BMW iX Xdrive40 battery.

It's 330-350V in ten modules connected in series.
75kWh total.

Each module has a connector with 18+ wires, exposing all the individual cell voltages and temperature sensing.

They are all bussed together in a loom to a CSC module:

the CSC might be on a CANBus, but we couldn’t see any useful output directly. It turns out it’s transformer isolated, but didn’t investigate whether it can talk yet.

Inside the CSC is a bunch of microcontrollers and some isolation:

And some surface mount power resistors. It seems it does some balancing, but obviously the balance current is pretty low - small wire gauge and small resistors:

I’ve done some basic capacity testing of the batteries. At 75kWh and 330V quoted capacity they should be around 220Ah.

The top cover is just plastic and comes off easily. I didn’t need to cut the main power cables, I made some threaded connectors that mate to the orange screw terminals pretty well.

- 18 cells in a series/parallel arrangement  in each module. 9S2P arrangement.
- one module weight about 40kg.
- Just bare cells with a connector that exposes each cell voltage and three temperature sensors
- no BMS or balancing circuit - because it’s in the CSC module
- initial voltages as I received it all 3.936V, all within about 1mV of each other - very good matching.
- charged them all to 4.05V (max nominal might be 4.1 or 4.2, depending on chemistry)
- discharged with an approx 60A load. Discharged to approx 3.05V, which is a common endpoint for lithium cells.
- main current carrying terminals are a TE connectivity thing with a conductive ring and an internal M4 thread.

Curve below:

Shows we got just about 200Ah between 4.05 and 3.05V at a ~4 hour discharge. That’s 27V to 36V for a module.

I think this is pretty typical, and we’d expect about 10% more capacity if we pushed it to the limits, so it seems bang on spec.

Internal resistance measured at about 3.5milliOhms(?) for the pack. Seems high, maybe - should check again.

For what it’s worth I bought one of these:


Which happily runs off the pack and feeds a good 1200W into the grid.

I can supply more photos or measurements on request if anyone is interested.  There's other parts that I didn't get good photos of; the parts that interface to the rest of the car and the charging subsystem and the safety cutout and so on. This is what I had, for what it's worth.

cheers, hope someone finds it entertaining,

How are these cooled?

It is nice to see this is a serviceable battery pack (you can replace modules) which should give the car a better resale value / lower depreciation.

--- Quote from: hojnikb on November 14, 2023, 01:06:27 pm ---How are these cooled?

--- End quote ---
The top picture shows two black tubes and tubing between the batteries so I guess it is liquid cooled. AFAIK no sensible car manufacturer uses passively cooled batteries in a BEV nowadays.


--- Quote ---How are these cooled?

--- End quote ---

there's an aluminium plate "sandwich" on the bottom with coolant channels stamped into it.


I'm surprised how much volume is not used by battery cells - I guess they use the same casing for the 105kWh models.

330V is quite a low nominal voltage for an EV.  Normally it is closer to 400V.  But if they want to maintain a similar architecture, that means they stay within 450V by adding 4x extra modules.  I would guess rapid charging and performance is more limited for 330V model.

Looks like overall a good design of pack, as nctnico points out looks very serviceable, it'll all ultimately come down to whether these are good quality cells or not.

I would guess these are CATL or Samsung prismatic cells.


[0] Message Index

[#] Next page

There was an error while thanking
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod