BMW i3 CCS/charge port controller

[muehlpower]

The VIN is transferred to the LIM via CAN. Damien found that it works without a VIN. Only if the wrong one is transmitted will the LIM be blocked. The VIN is not passed to the EVSE. If necessary, the MAC address that is printed on the LIM and cannot be changed is used for identification and is transmitted to the EVSE.

[JaniK]

Thanks for explaining it, now I see my error in chain of thought.

So it is MAC address identification that is used for plug and charge without using a tag or app?

[EV_Builder]

Thanks for explaining it, now I see my error in chain of thought.

So it is MAC address identification that is used for plug and charge without using a tag or app?

Yeah that might be.

[tom91]

Yeah that might be.

Nope not that easy, plug and charge is a whole other handshaking back end connecting process. It requires the car to be connected to the cloud and the charger. So getting that sorted in a "DIY" setup is not happening anytime soon, as who wants to deal with backend compatibility checks for some random persons own backend.

[JaniK]

ok so lots of tags for the win then

[muehlpower]

Fastned has a system that only needs the MAC address.
here is a description of how it works
https://generationstrom.wordpress.com/2 ... devorgang/
"
A unique identification code (MAC address) is transmitted via the CCS connection, which enables Fastned to identify a vehicle precisely. Registered users and the vehicle can be linked and clearly assigned via the connection to the app activation.
"

should also work with our handicraft solution

[tom91]

Fastned has a system that only needs the MAC address.
should also work with our handicraft solution

Yes this is Fastned using the Fastned backend and payment system quite cool, someone should try this.

[EV_Builder]

So yes that might be, and that's because MAC adress is unique.
Nobody talked about easy anyway...
Chargers are iot'ed anyway to handle payment I guess.

[mikeselectricstuff]

Gridserve are also introducing what they call autocharge,using the MAC.
Unfortunately VAG cars use randomized MACs for privacy reasons, apparently

[EV_Builder]

Unfortunately VAG cars use randomized MACs for privacy reasons, apparently

Randomized Macs? A MAC is always unique. The first part is fixed and is coupled to VAG or BMW etc. The last part is de serial.
On a network all MACs need tobe unique.

[chrskly]

Randomized Macs? A MAC is always unique. The first part is fixed and is coupled to VAG or BMW etc. The last part is de serial.
On a network all MACs need tobe unique.

Yeah, Android has had this feature for a while. Not surprising to see it popping up elsewhere.

Using the mac (and only the mac) to authenticate payment for a charge seems like asking for trouble if you ask me. Especially since some of the docs I've read seem to imply that all of the chargers (and by extension, cars) at a particular charging location are on the same network. So, potentially, all I would need to do is have my car passively capture the mac addresses of the other cars on the network and I can get free charging. Assuming this is true, we'd see skimming activity like this soon enough.

[asavage]

Randomized Macs? A MAC is always unique.

Sadly, this hasn't been true for a number of years.

[tom91]

I have been involved with more testing of the BMW I3 CCS controller. Main focus was does the code I modified to listen to more limits given by the SimpBMS work? Also have a Chinese isolation monitoring device that is being tested, which turns out might be causing some issues even before the contactors close.

Two things to note came to light:

• After Cable test the charger did not return to 0V but hovered around 55V, so tweaked the code so it no longer waits till 0V to count down to proceed. The charger did not mind

• Has anyone been able to exceed the currently hard coded power limit of 44Kw (set in code can be changed) or seen the following behaviour? the charger seems to overshoot by listening to FC current request and then start backing down to Power Request without indicating any limits on the charger side

[Jack Bauer]

Have seen the cable test voltage behavior a few times. Not sure if you are looking at the charger limit flags in code but I can whack the current request through the roof and the charger will roll on until it triggers a flag (either max power or current) and then back down. Seems here in Ireland at least they use the flags when throttling chargers as they will happily report say 125A avail but trip a current limit flag at much less.

[mikeselectricstuff]

Randomized Macs? A MAC is always unique. The first part is fixed and is coupled to VAG or BMW etc. The last part is de serial.
On a network all MACs need tobe unique.

Unique is not the same thing as consistent in each session. Nowadays the MAC doesn't travel far through a network so global uniqueness isn't particularly important

[mikeselectricstuff]

Yeah, Android has had this feature for a while. Not surprising to see it popping up elsewhere.

Using the mac (and only the mac) to authenticate payment for a charge seems like asking for trouble if you ask me. Especially since some of the docs I've read seem to imply that all of the chargers (and by extension, cars) at a particular charging location are on the same network. So, potentially, all I would need to do is have my car passively capture the mac addresses of the other cars on the network and I can get free charging. Assuming this is true, we'd see skimming activity like this soon enough.

Possible, but rather difficult in practice. Obviously the charger should b e filtering so only essential traffic appears on the PLC interface, but even it it doesn't, how are you going to access the network through the CCS controller?
And even if successful, the potential reward is low, and risk of getting detected very high - wouldn't be hard to correlate use of spoofed MAC to CCTV
I wouldn't be at all surprised to see a talk on it at Black Hat/CCC/Defcon at some point....

[EV_Builder]

Unique is not the same thing as consistent in each session. Nowadays the MAC doesn't travel far through a network so global uniqueness isn't particularly important

That's true. I was more theoretically speaking but yes. MAC can be changed etc.

[evMacGyver]

Does LIM control CCS cable lock motor automatically as I don't see lock control on DBC, there is only lock status? Is it possible to control lock motor somehow from sw? Maybe unknown 764 lock button?

[tom91]

Does LIM control CCS cable lock motor automatically as I don't see lock control on DBC, there is only lock status? Is it possible to control lock motor somehow from sw? Maybe unknown 764 lock button?

Be good to know if there is a message with vehicle locked status. So it inhibits unlocking while vehicle is locked, which the BMW I3 does.

[Jackk]

Hi all,

Does the lim control pre charge ? From what I can see from the pinout there are only outputs for the main and CCS contactors.

What is the best way to handle pre charge?

Before lim module was fitted my gs450h V2 board dealt with it and obviously main contactors.

[EV_Builder]

Pre-charge is done at the charger side in the essence that if I'm corre t it matches the vehicle voltage before contacts are closed. So then there are no arks.

[Jackk]

Thanks but I was actually referring to the inverter pre charge?

I am right in thinking the lim controls the main contactors?

Still would like to know why this is actually necessary? Is it just an extra safety measure?

J

[tom91]

Thanks but I was actually referring to the inverter pre charge?

I am right in thinking the lim controls the main contactors?

Still would like to know why this is actually necessary? Is it just an extra safety measure?

J

You are quite lost it seems. The CCS has its own contactors controller by the LIM, these are connected to the HV bus so you need to activate the HV bus before being able to charge. So you need an external system like the inverter to control the contactors and do precharge.

If you are asking this level of basic questions please have a good look at the wikipedia pages.

[SciroccoEV]

Thanks but I was actually referring to the inverter pre charge?

I am right in thinking the lim controls the main contactors?

Still would like to know why this is actually necessary? Is it just an extra safety measure?

Inverter (and charger/DC-DC) precharge is usually handled by the VCU (Vehicle Control Unit).

The LIM is a charging controller, so it only handles the rapid charge (CCS) contactors.

If you connected the Inverter straignt to the battery without precharge, the DC-Link capacitors would draw huge current, like in the 1000s of Amps range. This will degrade the capacitors and weld or burn out the contactor contacts.

[asavage]

(aside from the missing charge port cover sensor) Is this diagram (from the BMW i3 LIM Wiki) still a current representation of a viable high-level system wiring configuration?




Also, would anyone care to enumerate the downside(s) of having the LIM's contactors connect to the HV battery on the HV side of the main contactors (the LIM's contactors would be HOT on one side at all times)? Other than not being able to run the DC-DC Converter? I don't think I'd want to do that, but I'm not immediately seeing why it couldn't be done.