BMW i3 CCS/charge port controller

[Shockazulu1]

The log indeed looks like the charger does not want to wait at the requested voltage. It ramps up to it then immediately hits it and stops providing any power to keep it there and thus is just lets its output caps discharge. Very strange compared to any other chargers I have seen.

Question woud be if reducing the software requirement to see the voltage stable above a 20V delta to the requested voltage, will make this charger work.

As a precautionary measure.
I'm gonna drive my actual BMW i3 over to this charger with the issues.
I wanna verify that it will charge the stock vehicle. It's been a while since I charged it using this particular charger.

[Shockazulu1]

As a precautionary measure.
I'm gonna drive my actual BMW i3 over to this charger with the issues.
I wanna verify that it will charge the stock vehicle. It's been a while since I charged it using this particular charger.

I got the chance to make it back to the charger with the issue.
It worked with my stock BMW i3. Went back with my new CCS1 setup on my Toyota RAV4, and same issue.
Voltage almost matches and than it slowly falls down to zero and fails.
I think I have worked out how to change the code if I knew were to look in the program.
A am already runing program on the USA CCS1 variant that's not out in circulation yet.
I'd like to change the time it waits while within 20 volts. Instead of 2 seconds I wanna try 1 second or half a second if that doesn't work.

[Bigpie]

Will PM you a new test binary.
The type 1 fix has been merged in to master so will be in the next release.

[Shockazulu1]

Will PM you a new test binary.
The type 1 fix has been merged in to master so will be in the next release.

Thank you for the edited file.
That got it to advance past state 4.
It went to state 6 for 1 second than failed.
I ended up with a failed weld test.
I cleared weld test fault and tried again several times with the same result.
The only way I have to clear the weld fault is by changing the web interface setting, Interface i3 to interface unused.
It runs the contactor test when I close the charger door.
If I don't change from i3 to unused it will not run the contactor test to clear the fault.
Somehow I believe my problem it that I need to have unused selected to run a weld test.

log file attached

[tom91]

Somehow I believe my problem it that I need to have unused selected to run a weld test.

Does the BMW I3 you tested with do this? It would be helpful if you get a log from it while it charges at this exact charger.

I highly doubt that the problem is due to doing a weld test, as it is not possible to do so after precharging. Due to the dumb fact the voltages across the contactors should be the same.

[Shockazulu1]

Does the BMW I3 you tested with do this? It would be helpful if you get a log from it while it charges at this exact charger.

I highly doubt that the problem is due to doing a weld test, as it is not possible to do so after precharging. Due to the dumb fact the voltages across the contactors should be the same.

Where would be the easiest location to tap into the canbus for the LIM on my BMW i3?

[uhi22]

If you see the chance to open the LIM and connect five wires to it, it would be great to record some SPI logs of successful charging sessions, to better understand the communication between the car and the charger. More details here: viewtopic.php?p=57541#p57541

[Shockazulu1]

If you see the chance to open the LIM and connect five wires to it, it would be great to record some SPI logs of successful charging sessions, to better understand the communication between the car and the charger. More details here: viewtopic.php?p=57541#p57541

Sure, can work on that later. Will be easy to connect to my i3 LIM in my conversion vehicle.
I'd like to get logs for Tom after work today.
Logs from the LIM conversion are easy to access.
But he wants logs from my stock BMW i3, from the one charger that will not work on my conversion.
I've been going back through the thread to see where the best location it for me to tap into the CCS1 canbus in my car. I can only find where not to find the right wires. Would like to not have to remove the bumper and rear fender.

[tom91]

I've been going back through the thread to see where the best location it for me to tap into the CCS1 canbus in my car. I can only find where not to find the right wires. Would like to not have to remove the bumper and rear fender.

I have sent you a PM with request for your VIN. I have found some info for a BMW I3 US spec, you need to find the PT CAN. Easiest is the Body controller in the footwell. Info from ISTA in pdf attachment.

[tom91]

I have spent a few days testing many different CCS1 chargers in my area.
Tested 6 different versions of chargers, works on 5 of them and 1 faults out in state 4.

Please provide logs of working sessions for comparison.

[Shockazulu1]

Please provide logs of working sessions for comparison.

Will do. I will get the logs of the working sessions from my "stock BMW i3" and also logs of the failed sessions with the "LIM conversion project". Both from the same charger with the issue. I'm gonna go look for the correct canbus wires now on my BMW i3 now.

[tom91]

Will do. I will get the logs of the working sessions from my "stock BMW i3" and also logs of the failed sessions with the "LIM conversion project". Both from the same charger with the issue. I'm gonna go look for the correct canbus wires now on my BMW i3 now.

The working logs I meant in addition to the BMW I3 working log, on the charger where the LIM "standalone" fails, is a log of a working session of the LIM "Standalone".

I would hope you saved some as you were testing chargers, having a working log to compare against is essential, even more so if it is your own setup.

[Shockazulu1]

The working logs I meant in addition to the BMW I3 working log, on the charger where the LIM "standalone" fails, is a log of a working session of the LIM "Standalone".

I would hope you saved some as you were testing chargers, having a working log to compare against is essential, even more so if it is your own setup.

Long night testing.
I accessed the canbus wires under the foot well, but couldn't get it to work.
Will take more research.
I did go and record a successful charge and we can compare to the one that failed at state 6.
But I did get the problem charger to work.
I changed charge mode to EXT_DIGI on a hunch.
Than drove across town to the problem charger and it worked.
Attached.
Failed at state 6
Successful charge
Photo of settings that worked
.

works good.csv

(3.62 MiB) Downloaded 41 times

greenlots failed state 6.csv

(1.33 MiB) Downloaded 40 times

[tom91]

I did go and record a successful charge and we can compare to the one that failed at state 6.
But I did get the problem charger to work.
I changed charge mode to EXT_DIGI on a hunch.
Than drove across town to the problem charger and it worked.
works good.csv
greenlots failed state 6.csv

did you log the one where you got the "problem" charger to work? Also did you try starting charging multiple times after it worked?

what is the works good log?

[Shockazulu1]

did you log the one where you got the "problem" charger to work? Also did you try starting charging multiple times after it worked?

what is the works good log?

I did get it to work multiple times at the previously "failed charger"

After hours unsuccessfully attempting to connect to the canbus on my BMW i3.
I headed out to a charger that already has worked (with the project) to get the logs of a working charge session.
That is the log name "works good"
The log name "greenlots failed state 6" is from earlier made at the "problem charger"
Unfortunately, my LINUX laptop battery didn't have enough battery left to record the successful charge from the previously "problem charger". I have ordered a new battery for laptop. For now I'll bring a power inverter so I can keep the laptop charged up.

My next step I'll be more organized and I'll get several logs from various conditions.
I plan to see if going back to the original 2 second wait file will still cause the failed charge session.
Then I will know if the change from 2 seconds to 1 seconds is needed or not.

[Shockazulu1]

More logs attached.
I went to the "problem charger" I'll name this one 54059.
I started the canbus recording, started the charger, and got a failed charge.
I had planned to record the successful charge.
I tried lots of things to get the charger to work, than had a idea, maybe it's electrical interference from the climate control turning on and off constantly.
My contactor box sits directly on top of my heater module.
I turned off the heat/climate control, and it charged successfully.
I was able to duplicate the successful charge a few times with the heater turned off.
I also got a capture of the successful canbus logs.
I had planned to test both files, 1 second vs 2 second wait, for contractors.
But ran out of time. Hope to have more time to do this after work today.

BTC Power54059 working heat turned off.csv

(1.4 MiB) Downloaded 35 times

BTC Power54059 Failed.csv

(3.8 MiB) Downloaded 35 times

[Shockazulu1]

More logs attached.
I went to the "problem charger" I'll name this one 54059.
I started the canbus recording, started the charger, and got a failed charge.
I had planned to record the successful charge.
I tried lots of things to get the charger to work, than had a idea, maybe it's electrical interference from the climate control turning on and off constantly.
My contactor box sits directly on top of my heater module.
I turned off the heat/climate control, and it charged successfully.
I was able to duplicate the successful charge a few times with the heater turned off.
I also got a capture of the successful canbus logs.
I had planned to test both files, 1 second vs 2 second wait, for contactors.
But ran out of time. Hope to have more time to do this after work today.


BTC Power54059 working heat turned off.csv
BTC Power54059 Failed.csv

Replying to my own previous post.
I was wrong about the heater causing the issue.
The cause of the problem was with the settings in the web interface.
At any other charger it didn't matter if I had Interface set to i3LIM.
But at this charger it needed to be set as unused.
All the other chargers also work with interface set to unused, so I won't need to change it back and forth.

I also tested this "problem charger" with both the 2 second wait and the 1 second wait versions
It only works with the 1 second wait version.
Also only works with interface set to unused, cannot have interface set to i3LIM.
Other chargers were ok with interface set to i3LIM.

As a bonus, having the interface set to unused, makes clearing a weld fault as simple as closing the charger port door.

[Shockazulu1]

did you log the one where you got the "problem" charger to work? Also did you try starting charging multiple times after it worked?

what is the works good log?

This is log of charger that was originally a problem, and using the revised 1 second delay file fixed it.
At the time I believed having the heater turned off was necessary.

BTC Power54059 working heat turned off.csv

(1.4 MiB) Downloaded 39 times

The saga continues.
The problem i thought I had solved, isn't completely solved.
All the chargers in my area worked, accept one. It appeared like the charger was waiting to long for the both voltages to be within 20 volts of each other. With a revised file Version 2.03a it solved the problem. The wait time was changed from 2 seconds to 1 second.
It worked. That charger now works perfectly.
Until today, someone was on the one side of that charger, so I used the other side. It's actually 2 separate chargers.
Guess what? The other charger is behaving the same way. (even though I am running the 1 second delay file, it acts like even 1 second wait is too long)

Btw at the moment the charger goes into fail,I hear the contactors clicking, and the logs say "failed to complete contactor test.
It's at this point I need to close the charge port door, and I'll here the contactors clicking and the weld test fault clears.

I wonder if I now need to try revising the 1 second wait to .5 second wait and try that?
I attached log file here.

54061 failed t15stat on.csv

(2.32 MiB) Downloaded 35 times

[tom91]

I wonder if I now need to try revising the 1 second wait to .5 second wait and try that?
I attached log file here.

54061 failed t15stat on.csv

The contactors actually close, the request arrives and the LIM responds the contactors are shut.
There is an issue, unsure why the log reads a current request of 255 which does not seem right.

And yes the contactor info changes to weld test cannot be carried out, this is true as we are charging thus cannot influence the voltage at charge port for checking. This may not actually be what is causing the faults.

From your testing and comments it sounds like there might be some wiring issues.

[Shockazulu1]

The contactors actually close, the request arrives and the LIM responds the contactors are shut.
There is an issue, unsure why the log reads a current request of 255 which does not seem right.

And yes the contactor info changes to weld test cannot be carried out, this is true as we are charging thus cannot influence the voltage at charge port for checking. This may not actually be what is causing the faults.

From your testing and comments it sounds like there might be some wiring issues.

image.png

I think I understand a little better.
Before the issue was waiting too long and it never tried to close the contactors, when it had a close enough voltage match.
This was fixed by changing it to not wait as long.
This issue is different because it is actually closing the contactors soon enough but than it believes something is wrong and wants to do a weld check.
I have tested the original issue charger with and without the faster contactor close and that charger only works with the faster close setting.
I'm gonna have to go to this 2nd issue charger and see if it works with the original slower close timing setting.(the 2 second wait file)
I'll also try a few other options with the web interface.

[asavage]

Is there a relationship between the BMS (whatever BMS is in use) and the LIM/ZV? If so, how are the existing users implementing this relationship?

The Wiki currently mentions this only in simple terms:

WIki entry for BMW i3 LIM: BMS

[EV_Builder]

BMS communicates limits...these limits are forwarded/applied to charger.

[asavage]

Please expand on this.

The BMS communicates limits . . . via CAN? To the ZV -> LIM (via CAN) -> DCFC EVSE (via PLC)?

If the BMS is communicating limits to the ZV, does the ZV have preset or user-defined communication profiles for various BMSs?

[tom91]

The BMS communicates limits . . . via CAN? To the ZV -> LIM (via CAN) -> DCFC EVSE (via PLC)?

Nope, the Zombie currently follows the limits from the Web Interface for the DC fastcharging. It does not change the current limit or voltage limits on its own.

[asavage]

Tom, that's what I was remembering, but EV_Builder seems to have a different setup, or knows someone how does.

At a minimum, I would want to restrict the charge rate at very low or very high pack temperatures, and this -- as well as other limiting factors -- is best communicated from the BMS, I would think. But I have a very simplistic view of what the BMS does and the factors that limit charge rate, and I'm hoping that others who are already using the LIM/ZV for DCFC can describe what they're doing.