RIVIAN battery 141kW shows incorrect data

[tehnik-ua]

Hello dalathegreat, I use rivian battery 141kW for inverter deye 50kw 3ph. I use a power supply with a voltage of 13.8V 3A. BE use LilyGo T-CAN485 with addon can 2515. BE shows incorrect data.
google say:

The part number PT00057223-F belongs to the 1st Generation (Gen 1) Rivian electric vehicles. Specifically, this part number is the 141 kWh (Large pack) High-Voltage Lithium-Ion Battery Assembly used in the original generation of vehicles:

• Rivian R1T (Model years 2022–2024)
• Rivian R1S (Model years 2022–2024).
  Rivian completely overhauled its battery architecture and vehicle platforms starting with the 2025 model year (Gen 2), meaning this particular part number is strictly tied to the first-generation platform.

battery sticker Can bus logs:

canlog_00-21-42.txt
canlog_00-03-27.txt
canlog_00-39-18.txt

I also drew the pinout of the battery connector according to your description, add it to the wiki

[tehnik-ua]

Hello, thanks for your answer. Once again I double-checked the connection of the LV connector, everything is as in the wiki description. There is a 120 ohm resistor on the can platform (pin 4,8) and the liligo also has one. That is, the can bus is terminated (60ohm)
I forgot to write that when the BE is turned on, the contactors close and the DCFC output has a voltage of 398V.
I installed HVIL jumpers on all unused connectors.
BE connects normally to the deye inverter and transmits to it all the data that BE sees, but it is not correct. I tried raising the supply voltage of the BE and HV batteries from 13.8 to 14.2V, but this did not give any result.
I tried changing the channel buses in liligo (internal and external) for communication with the battery - nothing changes, the data is absolutely incorrect. I tried disabling deye in BE settings, also without results.
In events, the connection to the battery shows 1 time, and the low battery voltage error is constantly growing.

[tehnik-ua]

Hi, I also double-checked the speed of the canbus with an oscilloscope. The pulse duration is 2 µs, i.e. 500kbit, as it should be.

So I would stop changing wiring or supply voltage: your new measurements and logs have already isolated the problem to telemetry decoding.

I initially thought that the problem was in decoding. Before writing here I checked everything several times.
I saved the logs using my canhacker.

(Platform CAN 500 kbit/s. These messages are with the battery connected and BE. The contactor is closed and there is voltage at the DCFC.)

ID       DLC Data                    Period Count  Comment
00A      8   AD 0A 00 00 00 00 00 00 20     4473   
06E      8   F3 07 00 00 00 00 20 00 10     8950   
078      8   08 01 00 00 00 00 00 00 100    896    
096      8   00 00 00 00 00 00 00 00 500    179    
0A0      7   00 00 00 00 00 00 00    1001   90     
100      6   B4 C6 DB F2 59 06       10     8949   
120      8   DD 83 BB DC 3D B6 07 00 50     1790   
130      8   9A 06 00 1C 1B CA 26 04 10     8948   
145      6   B7 09 00 04 00 00       249    358    
152      8   00 00 00 00 00 00 00 00 500    179    
153      8   00 00 7D 7D 7E 78 7D 7D 100    896    
154      8   37 13 0A A2 21 01 12 01 50     1790   
156      6   B2 53 61 13 24 00       50     1790   
160      8   B1 FA CF 07 D0 F7 CF 07 20     4474   
162      2   8C 16                   100    895    
164      1   00                      100    895    
1E3      3   04 04 12                50     1790   
219      8   22 09 23 04 23 04 22 09 200    447    
243      5   12 B1 19 00 00          100    894    
260      8   0F F7 00 0A 06 0A 00 00 100    894    
262      1   FC                      50     1790   
265      8   00 07 C0 14 21 3C DD 80 3002   30     
267      5   00 00 00 00 00          100    894    
269      8   12 01 ED 12 01 15 00 00 32861  18     
273      7   73 0B 00 00 00 00 00    20     4476   
274      8   E3 07 00 00 00 00 00 00 9      8947   
275      8   00 00 00 00 00 00 00 00 100    896    
276      8   00 00 00 00 00 00 00 00 10     8949   
278      8   BA 61 03 11 8E 0E B8 86 100    894    
279      8   D3 01 00 00 00 00 00 00 100    894    
287      7   7B BC 60 96 C8 00 00    50     1788   
288      6   7F FF 3F FF 80 00       50     1790   
299      7   F4 01 13 65 A0 00 01    100    895    
314      8   BE 71 44 44 44 0B 2D 91 100    895    
315      1   00                      1000   90     
31B      6   D1 DC 2F D6 2F 06       500    179    
329      2   01 58                   100    895    
340      8   74 C4 64 D8 5A EC 14 46 1000   90     
353      8   BA 21 E4 8C DC 8E 13 72 100    895    
405      2   03 00                   100    894    
42D      3   08 91 02                100    894    
553      2   51 01                   1000   90     
558      8   0A 40 00 00 00 26 C0 00 1000   90     
55B      6   0F B0 0F 0E 0F 78       200    447    
592      8   00 00 00 00 00 00 00 00 500    179    
593      8   00 00 36 80 7B 17 74 0B 1000   90     
594      8   00 00 00 00 00 00 00 00 500    179    
6DD      8   00 00 00 88 10 8E 03 00 1000   90     
6DE      8   CC 39 00 00 00 00 00 00 100    895    
6DF      5   00 00 28 00 00          100    895    
6E1      6   00 00 27 A6 02 1B       1000   89     
7EC      2   01 00                   0      1      

I recorded several logs with different conditions. the conditions are specified in the file name.
can_BE_batt_rivian_log.zip

I hope this will help display the battery parameters correctly.

[tehnik-ua]

Hi Mohammad Rezwan Khan, I have changed the voltage and current handling code. BE began to launch without errors and warnings, the voltage in the emulator began to be displayed more correctly, and so did the current.

in RIVIAN-BATTERY.cpp

// ============ я добавил этот код для обработки напряжения и тока батареи =================
// Функция сборки 8 байт массива в 64-битное число (Intel / Little-Endian)
uint64_t bytes_to_uint64(const uint8_t *bytes) {
uint64_t data = 0;
for (int i = 0; i < 8; i++) {
data |= ((uint64_t)bytes[i] << (i * 8));
}
return data;
}

case 0x120: //Voltages [Platform CAN]+
datalayer.battery.status.CAN_battery_still_alive = CAN_STILL_ALIVE;
full_data_voltage = bytes_to_uint64(rx_frame.data.u8);
raw_v1 = (full_data_voltage >> 12) & 0x1FFF; // Распаковка четырех 13-битных полей напряжения
raw_v2 = (full_data_voltage >> 25) & 0x1FFF;
raw_v3 = (full_data_voltage >> 38) & 0x1FFF;
raw_v4 = (full_data_voltage >> 51) & 0x1FFF;
DCFC_contactor_voltage = raw_v2 * 0.5f-0.9f;
voltage_reference = raw_v3 * 0.05f;
main_contactor_voltage = raw_v2 * 0.5f;
pre_contactor_voltage = raw_v2 * 0.5f;
break;

case 0x160: //Current [Platform CAN]+
datalayer.battery.status.CAN_battery_still_alive = CAN_STILL_ALIVE;
full_data_current = bytes_to_uint64(rx_frame.data.u8);
raw_c1 = (full_data_current >> 12) & 0xFFFF; // Распаковка четырех 13-битных полей напряжения 0x1FFF
raw_c2 = (full_data_current >> 25) & 0x1FFF;
raw_c3 = (full_data_current >> 38) & 0x1FFF;
raw_c4 = (full_data_current >> 51) & 0x1FFF;
battery_current = raw_c1;
break;

in RIVIAN-BATTERY.h

uint64_t full_data_voltage; // склеенные байты напряжения в одно число для обработки
uint16_t raw_v1; //Распаковка четырех 13-битных полей напряжения
uint16_t raw_v2;
uint16_t raw_v3;
uint16_t raw_v4;
uint64_t full_data_current; // склеенные байты тока в одно число для обработки
uint16_t raw_c1; //Распаковка четырех 13-битных полей напряжения, 12-битное поле тока
uint16_t raw_c2;
uint16_t raw_c3;
uint16_t raw_c4;

Please tell me what values should be in these variables:
voltage_reference
main_contactor_voltage
pre_contactor_voltage ?

difference between measured and actual dcfc voltage

After the emulator has been running for about a minute, the battery contactor opens and the events contain messages about battery undervoltage, the battery is empty. Why does this happen?

[tehnik-ua]

Hi Dala, I started the battery. She worked for three days. On the fourth day, when there was no load, the battery was charged to 99.96%, the contactor turned off and the battery stopped responding to BE commands. Rebooting and resetting events does not give any result. The contactor does not turn on. I took the logs, maybe this will help in finding the problem.
I don’t understand why BE shows a voltage of 2.6v?
How can I see the voltage inside the battery before the contactor?

log_00-03-22.txt
canlog_00-08-14.txt

[dalathegreat]

Sorry I have no clue, there are no RX (incoming messages), so that means the BMS is powered OFF

[tehnik-ua]

Hi Dala, I took a log from the moment the battery was turned on using сan hacker. Before this, the battery most likely fell asleep and I could not read the log properly.
Maybe you have a diagnostic can id to reset bms errors? I think that the contactor is blocked by the BMS itself due to overcharging. I connected the battery to the inverter in voltage mode. And I set it to 450v, but the inverter when charging the battery in the application showed 457v. It was a day off and there was no load on the inverter,
except for charging the battery from solar panels. After this situation, the contactor in the battery opened. For three working days it worked normally in this mode.
rivian_log_contactor_no_close_start.zip