BMW i3 CCS/charge port controller

[asavage]

TIL something about a variation of Plug-n-Charge for non-PnC-capable vehicles: EVgo (and Fastned in EU has/is developing similar) has "AutoCharge+", which not only uses the MAC address (obtained in the SLAC conversation),if it can be determined that it's a unique MAC address, one can enroll in AutoCharge+ and EVgo will deliver PnC-like capability to older non-PnC vehicles.

AutoCharge+ is already in place and has lots of subscribers right now.

OK, that sound nice and all . . . but the EVgo rep said they are going a lot further than that, interfacing with some vehicles' telematics (via vehicle mfgrs' public APIs) to see if the GPS location of the vehicle matches the MAC being presented. Or, noting if multiple vehicles with the same MAC are trying to charge at different locations.

The EVgo rep notes that they are aware of which mfgrs use non-unique MACs in their CCS implementations, and those vehicles are likely exempt from qualification to the AutoCharge+ program.

I mention this as things to think about when cloning LIMs, for example.

Link to the specific segment of a 53-minute interview:

(Thanks to Simotronic for recommending I watch the whole vid.)

Also, the rep mentions they currently "fingerprint" manufacturer's models by the quirks their CCS implementations exhibit: they can easily discern a Volvo XC40 Re from an F150, for example.

[Simotronic]

I was thinking some more about this auto connect feature. I think the security surrounding it needs to be hammered out because I see a potential for misuse. We also have a number of people who aren't too keen on geo-location and all the telematics. Well unfortunately that's the world we live in so it behooves us to at least put some stringent requirements on the data that's being passed around.
Plugging your car into your own charging station, AKA the EVSE at home doesn't generally call the mothership. OK, the Tesla one probably does because you can identify the TWC as your home charger...
Charging up on a road trip is generally advertising your location which is a safety point right there. So I believe it's critical that any data being shunted around needs to be secured in an acceptable manner that prevents unauthorized access.
Right off the bat I can see a market in charging adapters that are actually charging anomolizers that will perhaps have a revolving set of credentials to permit a charging session to occur with valid payment, but the real vehicle data is shielded. Or maybe they can be used for older EVs that don't speak the native language of the charging station.

[muehlpower]

I mention this as things to think about when cloning LIMs, for example.

The MAC address of the LIM cannot be changed with E-Sys, and I don't know if that's even possible!

[christoph]

Hi

I have a problem charging at a HyperCharger station. It gets inte state 5 but then waits for the connectors to close. But apparently they dont. I havent really be able to figure out why. Any hints? See attached log. Ignore the CHAdeMO frames in it

Thanks
Christoph

This problem is still there, tried with another station of the same kind. I've seen that tom91 had exactly that problem where the precharge voltage was there but contactors wouldn't close. I think Jack Bauer pointed him that he did set the maximum A in 0x2F1 to 0 which apparently was the cause. This doesn't seem to be my problem, though...

[uhi22]

If you could run the BMW LIM against a simulated charger using pyPlc and homeplug modem (see discussion in parallel chat), you could record a pcap trace with wireshark, and maybe this would reveal what it is the problem.

[christoph]

If you could run the BMW LIM against a simulated charger using pyPlc and homeplug modem (see discussion in parallel chat), you could record a pcap trace with wireshark, and maybe this would reveal what it is the problem.

This sounds like a cool thing to do, did just read a bit...

[ivankovff]

Good day! I'm from Ukraine. I find your topic very interesting. It's a pity that I only now found this forum. I have a 2013 Ford Focus Electric. I made him a new 55 kWh battery. I want to join and also try to make a ccs port on my car. I have already ordered all the elements. I apologize for my English because I use Google Translate.

[ivankovff]

Bytes 0x00 and 1x00 in message 0x112 show battery current and in all your logs normally it is mentioned 0xf9 and 1x1f, are these indicators stable (fixed) or these bytes have to show real values?

[christoph]

Bytes 0x00 and 1x00 in message 0x112 show battery current and in all your logs normally it is mentioned 0xf9 and 1x1f, are these indicators stable (fixed) or these bytes have to show real values?

Byte 0 and 1 are battery current, right. Note that it has a factor of 0.1 and an offset of 819. So 0xf9 0x1f, being 8185 translates to -0.5A.

[ivankovff]

Byte 0 and 1 are battery current, right. Note that it has a factor of 0.1 and an offset of 819. So 0xf9 0x1f, being 8185 translates to -0.5A.

Thank you for the answer, I understand it. My question was how theses values change while charging?

[christoph]

Thank you for the answer, I understand it. My question was how theses values change while charging?

Charging is a positive amps value. So if you e.g. charge with 100A it will be (100 + 819) * 10 = 9190 -> 0xe6 0x23

[ivankovff]

Charging is a positive amps value. So if you e.g. charge with 100A it will be (100 + 819) * 10 = 9190 -> 0xe6 0x23

thanks for the answer
I understand that I have to make a request for the Amp needed in the message 0х112 d0 and d1 and this request must necessarily correlate with the request in Watts from the message 0х3Е9 d0 and d1

[uhi22]

The meaning of that current value is not clear, I guess. It could be the actually measured battery current (just for information or plausibilization in the LIM, to be compared to the measured current with is reported by the charger via CCS PLC to LIM). Or it could be the requested current, which is transferred from the LIM to the charger via CCS PLC. Both could be possible, I have no answer, only see different understandings and want to point this out. Would be great to have a clear specification in the wiki.

[christoph]

thanks for the answer
I understand that I have to make a request for the Amp needed in the message 0х112 d0 and d1 and this request must necessarily correlate with the request in Watts from the message 0х3Е9 d0 and d1

Message 0x112 comes from the BMS and should the be the actual amperage. Not the wanted charging amperage.

[muehlpower]

I'm not sure that byte 0 + 1 is evaluated by the LIM. In my documentation for the I3 from 2013-2017 it is shown in black and that means:

google translated:
"Positions in the frame occupied by signals included in the current configuration context but not in the current selection"

[ivankovff]

I'm not sure that byte 0 + 1 is evaluated by the LIM. In my documentation for the I3 from 2013-2017 it is shown in black and that means:

google translated:
"Positions in the frame occupied by signals included in the current configuration context but not in the current selection"

I studied messages: 0х112 d0 - d1 and 0х3Е9 d0- d1 from the log in the link below.

It turns out that the values in the messages 0х112 d0 -d1 and 0х3Е9 d0- d1 are only for LIM from the BMS, and the control of the charge power from the car is carried out by the message 0х3Е9 d3 - d4

https://github.com/damienmaguire/BMW-i3 ... y_mida.csv

[muehlpower]

I do not believe that. The amps are specified with 3E9 byte 5, and that controls the charging process. 7D = 125A.

[ivankovff]

I do not believe that. The amps are specified with 3E9 byte 5, and that controls the charging process. 7D = 125A.

If this log is from an original BMW I3, then here in two messages we have the following indicators:
112 E0,24,0C,0E,AB,71,75,5A
Soc 112 (D5 D4) 42.7%
Amps 112 (D1 D0) 94.4 A
Volts 112(D3 D2) 359.6 V
Volts 112 (D7) 360 V

3E9 5C,8F,21,A1,7C,7D,20,4B
Ampere request 3E9(D5 125 A
Watts request 3E9 (D4 D3) 48825 Watts
Watts 3E9 (D1 D0) 36700 Watts

Why does the car choose 36700 watts and 94.4 amps, maybe through 360 volts?
Maybe 125 amperes is 50,000 watts at 400 volts, but at 360 volts there should still be 94 amperes and 36,700 watts, respectively?
I could be wrong, but most likely this algorithm is needed when displaying in messages?

[muehlpower]

24E0 = 9440, 9440-8192= 1248 -> 124,8A (Not 94,4A),
7CA = 1994, 1994*25= 49850 -> 49850W forecast_power_charging
36700 are Wh!!, the battery capacity.

[ivankovff]

24E0 = 9440, 9440-8192= 1248 -> 124,8A (Not 94,4A),
7CA = 1994, 1994*25= 49850 -> 49850W forecast_power_charging
36700 are Wh!!, the battery capacity.

thank you
now I realized that I did not apply the offset

36700 are Wh!!, the battery capacity -this is the total capacity of the battery?

[christoph]

36700 are Wh!!, the battery capacity -this is the total capacity of the battery?

Yes!

[nathaniel]

I have been helped a number of times on the forum in the journey to my own conversion project. and now I have arrived at the last part, CCS.
my setup is: Zombieverter VCU(only for the LIM), BMW LIM 61 35 6 828 052, 96 cell 24kw BMS hybride pack.
with the AC charging the lock works and i can do contctor test successfully. the DC inlet voltage sensor also works properly, when I apply the various voltages I read them back correctly (on a few volts, I don't know if that's good enough) via 0x3B4.
I went to a shell recharge station to test, I've come a long way but the chance that everything would go well in 1x is of course quite small. the charging station starts and I see a screen with the SOC at 50% (comes from the VCU), I hear pumps and fans come on in the station. but then it stops. no error messages on screen or VCU. in the VCU the last CCS_State is 4. at that moment I also hear the CCS slot open, it seems as if the session is just being stopped. when I go through the CANlog I see the charger in Pre_charge state trying to match the inlet voltage and then it seems to stop. I think this is the time for the relays in the car to switch on? i couldn't hear any relays switching because there was too much noise in the area. so I'm not sure if they work, what I also noticed is that I don't hear them switching during the weld test. I actually assumed that this would happen, but the weld test is apparently completed successfully. the relays work just fine when I put 12 volts on them.
the relays come from the BMW s box and have a coil resistance of 30 ohms so I put a 30 ohm resistor in parallel with the coils to get the 15 ohms.
would someone who has been working with the LIM module for a while take a look at my log files?
I am curious what the problem could be, if more information is needed please let me know.

ccs poging 1.csv

(2.72 MiB) Downloaded 69 times

ccs poging 3.csv

(6.73 MiB) Downloaded 61 times

[nathaniel]

I think I have already found a hint, when I look in the code of the VCU it is checked in state 4 whether the inlet voltage is < 20 volts of the charger. when I then look at the log the value I get back appears to be very unstable, so I have to make that stable first before continuing to the next step.

I will give an update when I have solved this, if any of you have another insight, I would like to hear it.

[nathaniel]

one more question, in anticipation of the next step, my current sensor reads a positive value when I use current from the battery when driving, and a negative current when I charge. so when i go to charge it will report negative current to the lim, is this correct or not?

[nathaniel]

Some progress, fixt the unstable init port measurement and now I come to stage 6 but fails after that. I have a log but I want to look at this first. But I don't hear the relays closing and they should be at that stage, so I think there is something there.
During the weld test I also don't hear them closing, assuming this would be the case. Can anybody confirm that?
Screenshot from the signal before and after, the problem was clear