Mitsubishi i-MiEV BMS

Topics concerning OEM and open source BMSes
Post Reply
User avatar
FJ3422
Posts: 113
Joined: Fri Jul 10, 2020 9:55 am
Location: Netherlands
Been thanked: 1 time

Mitsubishi i-MiEV BMS

Post by FJ3422 »

Hi,

I have a complete i-MiEV batterypack around for testing purposes, and was trying to get the SimpBMS running on it. Unfortunately currently getting some weird behaviour. I have been trying different things, but as an Arduino-beginner might be in need for some tips.

After startup, the BMS starts 'finding' the modules. Most modules are correctly monitored, except for module 6 & 12 (the only 4-cell modules). After about 10 seconds when the SOC-calculation is done (see "/// SOC SET ///"), all cell voltages turn negative.

Below the console output (search for "///" to find the SOC calculation point), and a screenshot of the present CAN-bus data. CAN-bus data stays steady, no lost messages. Using a MCP2551 CAN transceiver, oscilloscope displays clean signal at pin 4 of the Teensy. MCP2551 gives 5V at the RXD line, but imho the Teensy 3.2 should be 5V tolerant.

Tried with the most recent .HEX of Tom's Github as well as a recompiled version with the updated ADC library (modified the adc->adc0 calls), same behaviour.

What am I doing wrong ?

IMIEV_BMSmonitor_2020-08-16.txt
(62.9 KiB) Downloaded 142 times
IMIEV-Batt_CAN-screenshot.PNG
tom91
Posts: 1305
Joined: Fri Mar 01, 2019 9:15 pm
Location: Bristol
Has thanked: 102 times
Been thanked: 216 times

Re: Mitsubishi i-MiEV BMS

Post by tom91 »

the can data look unstable/incorrect to me.
Founder Volt Influx https://www.voltinflux.com/
Webstore: https://citini.com/
User avatar
FJ3422
Posts: 113
Joined: Fri Jul 10, 2020 9:55 am
Location: Netherlands
Been thanked: 1 time

Re: Mitsubishi i-MiEV BMS

Post by FJ3422 »

Just figured some things out:

- After the first SOC calculation, variable 'SOCset' is set. This triggers the 'bms.setSensors' which sets through 'setOffset' the temperature offset ('settings.TempOff', -52) to all the CellOffset of all modules.

- Voltage scaling of the i-MiEV CMU's is different; looks like it is (read value * 5mV + 2,1V). Will test with an external voltage source and confirm this.

- My CMU ID 06 is defective. Displays temperatures but no cell voltages.
tom91
Posts: 1305
Joined: Fri Mar 01, 2019 9:15 pm
Location: Bristol
Has thanked: 102 times
Been thanked: 216 times

Re: Mitsubishi i-MiEV BMS

Post by tom91 »

Any progress? If you could share this I can have a look at updating the main SimpBMS code
Founder Volt Influx https://www.voltinflux.com/
Webstore: https://citini.com/
User avatar
FJ3422
Posts: 113
Joined: Fri Jul 10, 2020 9:55 am
Location: Netherlands
Been thanked: 1 time

Re: Mitsubishi i-MiEV BMS

Post by FJ3422 »

I just sent you per e-mail the version I used for testing.
It was my very first Arduino-experience at that time, so I am pretty sure I screwed up a lot (bashing the "compile" button until it compiled , and took out a lot of code that I didn't need / understand). So watch out if copying parts. :-)

If I remember correctly; in the function "BMSModule::decodecan(int Id, CAN_message_t &msg)" I did just modifiy the CellVolt calculation, and wrote "2.1" there into the "CellOffset" as a quick workaround for both issues.

I sold this pack because plans changed a bit, so did not use it in a vehicle.
tom91
Posts: 1305
Joined: Fri Mar 01, 2019 9:15 pm
Location: Bristol
Has thanked: 102 times
Been thanked: 216 times

Re: Mitsubishi i-MiEV BMS

Post by tom91 »

Thank you for sharing I will look at what you sent and put it back into the SimpBMS code.
Founder Volt Influx https://www.voltinflux.com/
Webstore: https://citini.com/
Post Reply