txubelaxu commented 1 month ago

I have tested the esphome-jk-bms last version with my config and it goes OK. But when I launch esphome-jk-bms-can last version (development branch), there is a not normal behaviour comparing to previous version (previous yambms). I have tried to tune the yamls to be similar to those that I have working, but I cannot get it working.

I am testing more to know why it is happening to my system.

Sleeper85 commented 1 month ago

What problem are you facing exactly ?

txubelaxu commented 1 month ago

I am running multi-bms_JK-PB_RS485_mode2.yaml

It starts receiving some info, but after some seconds, it stops:

[12:56:11][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0D [0010000000000000] [12:56:12][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0E [0010000000000000] [12:56:13][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0F [0010000000000000] [12:56:14][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x01 [0010000000000000] [12:56:15][I][jk_rs485_sniffer:346]: SILENCE: 1157.000000 ms [12:56:15][D][jk_rs485_sniffer:365]: .......................................... [12:56:15][D][jk_rs485_sniffer:369]: Buffer before number 1: 55.AA.EB.90.02.05.F2.0C.F1.0C.F1.0C.F1.0C.F2.0C.F1.0C.F1.0C.F1.0C.F2.0C.F2.0C.F2.0C.F2.0C.F2.0C.F2.0C.F2.0C.F0.0C.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.FF.FF.00.00.F2.0C.01.00.00.01.53.00.50.00.4F.00.4A.00.4D.00.4B.00.4D.00.4D.00.53.00.4E.00.4D.00.4A.00.4C.00.4D.00.52.00.51.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0A.01.00.00.00.00.1C.CF.00.00.00.00.00.00.00.00.00.00.F0.00.F6.00.00.00.00.00.00.00.00.61.08.7E.04.00.80.A3.04.00.00.00.00.00.23.8F.00.00.64.00.00.00.68.8F.1E.01.01.01.00.00.00.00.00.00.00.00.00.00.00.00.00.00.FF.00.01.00.00.00.9A.03.00.00.00.00.60.54.3F.40.00.00.00.00.B6.14.00.00.00.01.01.01.00.06.00.00.38.B0.ED.09.00.00.00.00.0A.01.EF.00.F9.00.9A.03.FF.DD.A4.08.9C.00.00.00.80.51.01.00.00.00.00.00.00.00.00.00.00.00.00.00.00.FE.FF.7F.DD.2F.01.01.B0.07.00.00.00.9D.01.10.16.20.00.01.04.4B (308) [12:56:15][D][jk_rs485_sniffer:640]: ###############################Sequence found SIZE: 308 [12:56:15][I][jk_rs485_sniffer:512]: NODES AVAILABLE CHANGED: address 0x01 (0->1) [0010000000000000] --> [0110000000000000] [12:56:15][D][jk_rs485_sniffer:699]: Frame received from SLAVE (type: 0x02, 309 bytes) 01 address [12:56:16][D][jk_rs485_sniffer:074]: Received Event from BMS.. [address:0x01] @ 2 --> WORKING ! #####

last_cell_info_request_received_OK [12:56:16][D][jk_rs485_sniffer:371]: Response: 12: [12:56:16][D][jk_rs485_sniffer:381]: Buffer empty [12:56:16][W][component:237]: Component jk_rs485_sniffer took a long time for an operation (334 ms).

[12:56:16][C][jk_rs485_sniffer:723]: RX timeout: 500 ms [12:56:17][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x03 [0110000000000000] [12:56:18][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x04 [0110000000000000] [12:56:19][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x05 [0110000000000000] [12:56:20][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x06 [0110000000000000] [12:56:21][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x07 [0110000000000000] [12:56:22][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x08 [0110000000000000] [12:56:23][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x09 [0110000000000000] [12:56:24][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0A [0110000000000000] [12:56:25][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0B [0110000000000000] [12:56:26][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0C [0110000000000000] [12:56:27][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0D [0110000000000000] [12:56:28][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0E [0110000000000000] [12:56:36][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x08 [0110000000000000] [12:56:37][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x09 [0110000000000000] [12:56:38][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0A [0110000000000000] [12:56:39][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0B [0110000000000000] [12:56:40][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0C [0110000000000000]

I have checked that ESP32 is sending to network from another RS485 client.

El lun, 5 ago 2024 a las 12:23, Sleeper85 @.***>) escribió:

What problem are you facing exactly ?

— Reply to this email directly, view it on GitHub https://github.com/Sleeper85/esphome-jk-bms-can/issues/61#issuecomment-2268728003, or unsubscribe https://github.com/notifications/unsubscribe-auth/BFHIJMEGJUDMBOHXCIQTSBDZP5HDZAVCNFSM6AAAAABL77RWYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENRYG4ZDQMBQGM . You are receiving this because you authored the thread.Message ID: @.***>

txubelaxu commented 1 month ago

It is very very incredible: when I touch some free pins of the esp32 with my hands, it starts working!

Astonished!

lol

El lun, 5 ago 2024 a las 12:59, Txu Belaxu @.***>) escribió:

I am running multi-bms_JK-PB_RS485_mode2.yaml

It starts receiving some info, but after some seconds, it stops:

[12:56:11][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0D [0010000000000000] [12:56:12][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0E [0010000000000000] [12:56:13][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0F [0010000000000000] [12:56:14][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x01 [0010000000000000] [12:56:15][I][jk_rs485_sniffer:346]: SILENCE: 1157.000000 ms [12:56:15][D][jk_rs485_sniffer:365]: .......................................... [12:56:15][D][jk_rs485_sniffer:369]: Buffer before number 1: 55.AA.EB.90.02.05.F2.0C.F1.0C.F1.0C.F1.0C.F2.0C.F1.0C.F1.0C.F1.0C.F2.0C.F2.0C.F2.0C.F2.0C.F2.0C.F2.0C.F2.0C.F0.0C.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.FF.FF.00.00.F2.0C.01.00.00.01.53.00.50.00.4F.00.4A.00.4D.00.4B.00.4D.00.4D.00.53.00.4E.00.4D.00.4A.00.4C.00.4D.00.52.00.51.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0A.01.00.00.00.00.1C.CF.00.00.00.00.00.00.00.00.00.00.F0.00.F6.00.00.00.00.00.00.00.00.61.08.7E.04.00.80.A3.04.00.00.00.00.00.23.8F.00.00.64.00.00.00.68.8F.1E.01.01.01.00.00.00.00.00.00.00.00.00.00.00.00.00.00.FF.00.01.00.00.00.9A.03.00.00.00.00.60.54.3F.40.00.00.00.00.B6.14.00.00.00.01.01.01.00.06.00.00.38.B0.ED.09.00.00.00.00.0A.01.EF.00.F9.00.9A.03.FF.DD.A4.08.9C.00.00.00.80.51.01.00.00.00.00.00.00.00.00.00.00.00.00.00.00.FE.FF.7F.DD.2F.01.01.B0.07.00.00.00.9D.01.10.16.20.00.01.04.4B (308) [12:56:15][D][jk_rs485_sniffer:640]: ###############################Sequence found SIZE: 308 [12:56:15][I][jk_rs485_sniffer:512]: NODES AVAILABLE CHANGED: address 0x01 (0->1) [0010000000000000] --> [0110000000000000] [12:56:15][D][jk_rs485_sniffer:699]: Frame received from SLAVE (type: 0x02, 309 bytes) 01 address [12:56:16][D][jk_rs485_sniffer:074]: Received Event from BMS.. [address:0x01] @ 2 --> WORKING ! #####

last_cell_info_request_received_OK [12:56:16][D][jk_rs485_sniffer:371]: Response: 12: [12:56:16][D][jk_rs485_sniffer:381]: Buffer empty [12:56:16][W][component:237]: Component jk_rs485_sniffer took a long time for an operation (334 ms).

[12:56:16][C][jk_rs485_sniffer:723]: RX timeout: 500 ms [12:56:17][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x03 [0110000000000000] [12:56:18][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x04 [0110000000000000] [12:56:19][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x05 [0110000000000000] [12:56:20][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x06 [0110000000000000] [12:56:21][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x07 [0110000000000000] [12:56:22][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x08 [0110000000000000] [12:56:23][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x09 [0110000000000000] [12:56:24][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0A [0110000000000000] [12:56:25][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0B [0110000000000000] [12:56:26][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0C [0110000000000000] [12:56:27][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0D [0110000000000000] [12:56:28][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0E [0110000000000000] [12:56:36][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x08 [0110000000000000] [12:56:37][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x09 [0110000000000000] [12:56:38][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0A [0110000000000000] [12:56:39][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0B [0110000000000000] [12:56:40][D][jk_rs485_sniffer:417]: SCANNING TO DISCOVER...0x0C [0110000000000000]

I have checked that ESP32 is sending to network from another RS485 client.

El lun, 5 ago 2024 a las 12:23, Sleeper85 @.***>) escribió:

What problem are you facing exactly ?

— Reply to this email directly, view it on GitHub https://github.com/Sleeper85/esphome-jk-bms-can/issues/61#issuecomment-2268728003, or unsubscribe https://github.com/notifications/unsubscribe-auth/BFHIJMEGJUDMBOHXCIQTSBDZP5HDZAVCNFSM6AAAAABL77RWYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENRYG4ZDQMBQGM . You are receiving this because you authored the thread.Message ID: @.***>

Sleeper85 commented 1 month ago

It is very very incredible: when I touch some free pins of the esp32 with my hands, it starts working!

Amazing! are all GND pins connected together? Including to the RS485 bus ?

txubelaxu commented 1 month ago

Yes, of course. I thought about a grounding problem, as well. But it works great with esphome-jk-bms...

It reverse to the same issue after some seconds, and If I touch some pins it recovers. I think it is related more to the can-pins. I am not using UART to CAN converter. Could I somehow disable this "can-layer"?

El lun, 5 ago 2024 a las 13:15, Sleeper85 @.***>) escribió:

It is very very incredible: when I touch some free pins of the esp32 with my hands, it starts working!

Amazing! are all GND pins connected together? Including to the RS485 bus ?

— Reply to this email directly, view it on GitHub https://github.com/Sleeper85/esphome-jk-bms-can/issues/61#issuecomment-2268822736, or unsubscribe https://github.com/notifications/unsubscribe-auth/BFHIJMHFA77LNO6AUMIRSYDZP5NE7AVCNFSM6AAAAABL77RWYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENRYHAZDENZTGY . You are receiving this because you authored the thread.Message ID: @.***>

txubelaxu commented 1 month ago

Anyway, I have just tested again the previous esphome-jk-bms-can version, and it works great with the same hardware configuration (I have changed nothing), and I don't need to touch the pins to recover...

El lun, 5 ago 2024 a las 13:49, Txu Belaxu @.***>) escribió:

Yes, of course. I thought about a grounding problem, as well. But it works great with esphome-jk-bms...

It reverse to the same issue after some seconds, and If I touch some pins it recovers. I think it is related more to the can-pins. I am not using UART to CAN converter. Could I somehow disable this "can-layer"?

El lun, 5 ago 2024 a las 13:15, Sleeper85 @.***>) escribió:

It is very very incredible: when I touch some free pins of the esp32 with my hands, it starts working!

Amazing! are all GND pins connected together? Including to the RS485 bus ?

— Reply to this email directly, view it on GitHub https://github.com/Sleeper85/esphome-jk-bms-can/issues/61#issuecomment-2268822736, or unsubscribe https://github.com/notifications/unsubscribe-auth/BFHIJMHFA77LNO6AUMIRSYDZP5NE7AVCNFSM6AAAAABL77RWYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENRYHAZDENZTGY . You are receiving this because you authored the thread.Message ID: @.***>

Sleeper85 commented 1 month ago

Anyway, I have just tested again the previous esphome-jk-bms-can version, and it works great with the same hardware configuration (I have changed nothing), and I don't need to touch the pins to recover...

Which version? Old 1.17.5 or YamBMS 1.4.1?

Sleeper85 commented 1 month ago

Last change since the version without TALK PIN.

Sleeper85 commented 1 month ago

There have been major changes in the yambms_canbus.yaml file since YamBMS 1.4.2.

See the changelog :

CANBUS 2.3.1 : Improved the procedure for sending canbus frames with reduced loop time, rewritten of the canbus link validation code and added Inverter Heartbeat Monitoring function
CANBUS 2.3.2 : Added LuxPower protocol with updated can_id 0x355, 0x356, 0x359 and 0x35C

You can remove the import of the canbus1 package to test.

  canbus1: !include
    file: packages/yambms/yambms_canbus.yaml
    vars:
      # YamBMS ID
      yambms_id: 'yambms1'
      # CANBUS vars
      canbus_id: 'canbus1'
      canbus_name: 'CANBUS 1'
      canbus_node_id: 'canbus_node1'
      canbus_light_id: 'esp_light'
      # The CANBUS link will be considered down if no response from the inverter (ID 0x305) for 5s
      # The Deye inverter sends the ACK (can_id 0x305) only when it receives the can_id 0x356
      canbus_link_timer: '5'

Sleeper85 commented 1 month ago

@virus100b just shared this with us on the DIY solar forum.

No issues since i have replace can board. I forget to mention that i have replace the talk pin RS485 board with the one without talk pin(The one with LEDs on TX, RX and voltage supressor on A, B ) with no issues. Today i notice the @Txubelaxu modified the code to work without talk pin but i did not update yet. For now is working fine. Uptime for over 24h.

He is now using a RS485 board without TALK PIN. I asked him to come here to talk about it.

txubelaxu commented 1 month ago

Anyway, I have just tested again the previous esphome-jk-bms-can version, and it works great with the same hardware configuration (I have changed nothing), and I don't need to touch the pins to recover...

Which version? Old 1.17.5 or YamBMS 1.4.1?

Old 1.17.5, I think, because is not any "yambms" param or value

txubelaxu commented 1 month ago

There have been major changes in the yambms_canbus.yaml file since YamBMS 1.4.2.

See the changelog :

CANBUS 2.3.1 : Improved the procedure for sending canbus frames with reduced loop time, rewritten of the canbus link validation code and added Inverter Heartbeat Monitoring function

CANBUS 2.3.2 : Added LuxPower protocol with updated can_id 0x355, 0x356, 0x359 and 0x35C

You can remove the import of the canbus1 package to test.
  canbus1: !include
    file: packages/yambms/yambms_canbus.yaml
    vars:
      # YamBMS ID
      yambms_id: 'yambms1'
      # CANBUS vars
      canbus_id: 'canbus1'
      canbus_name: 'CANBUS 1'
      canbus_node_id: 'canbus_node1'
      canbus_light_id: 'esp_light'
      # The CANBUS link will be considered down if no response from the inverter (ID 0x305) for 5s
      # The Deye inverter sends the ACK (can_id 0x305) only when it receives the can_id 0x356
      canbus_link_timer: '5'

Yes, that was the cause of the issue: I have commented and it works well now.

Many thanks!!!

Sleeper85 commented 1 month ago

@txubelaxu

Not using canbus with your inverter ?

txubelaxu commented 1 month ago

I have not one yet. :-)

I am creating my system from bottom to top. lol

El lun, 5 ago 2024 a las 14:44, Sleeper85 @.***>) escribió:

@txubelaxu https://github.com/txubelaxu

Not using canbus with your inverter ?

— Reply to this email directly, view it on GitHub https://github.com/Sleeper85/esphome-jk-bms-can/issues/61#issuecomment-2268988041, or unsubscribe https://github.com/notifications/unsubscribe-auth/BFHIJMA3JLE3FL4ZCTSBBXLZP5XT3AVCNFSM6AAAAABL77RWYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENRYHE4DQMBUGE . You are receiving this because you were mentioned.Message ID: @.***>

Sleeper85 commented 1 month ago

Maybe the problem is that you don't have a canbus board ?

More tests should be done now to confirm that the latest YamBMS 1.4.2 and CANBUS 2.3.2 code still works correctly with your RS485 component.

There are currently 2 guys using your RS485 component with my project @widget4145 and @virus100b

A 3rd guy will soon be using your RS485 component with my project and the isolated RS485 board without TALK pin.

@widget4145 has problems and can't monitor more than 5 BMS with his Atom S3 and this also happens when he uses your code only without YamBMS. Did this happen after upgrading your component to version 1.3 or 1.3.1 ? Or is it related to the version of esphome or HA? He recompiled again yesterday with your latest changes but it still doesn't work with his 7 BMS.

txubelaxu commented 1 month ago

Yes, it was the problem. Removing can component from yaml, it works (YamBMS 1.4.2)

I have analyzed @widget4145 user's issue about why is not working with 7 BMSs. I cannot recreate that config here. From 1.3 to 1.3.1 there is a bug correction on the code:

That is why I unpupblished v1.3.

Sleeper85 commented 1 month ago

@widget4145 has been using your component for a long time and in early May 2024 he was the first to use my multi-bms version with your component in version 1.2. Everything worked well for 2 months with 7x BMS and an Atom S3.

Then you fixed bugs and released version 1.3.1.

I suggested to @widget4145 to recompile the same multi-bms code that I had provided to him to benefit from your bug fixes and since then he has never managed to monitor more than 5 BMS. At that time he also updated esphome and HA.

Do you have a backup of version 1.2 of your components ?

We could thus check if the problem comes from the latest changes or not ?

I still have the YAML that I sent him in May and June when your component was in version 1.2.

txubelaxu commented 1 month ago

I will review the changes: https://github.com/txubelaxu/esphome-jk-bms/compare/v1.2...v1.3.1

El lun, 5 ago 2024 a las 17:22, Sleeper85 @.***>) escribió:

@widget4145 https://github.com/widget4145 has been using your component for a long time and in early May 2024 he was the first to use my multi-bms version with your component in version 1.2. Everything worked well for 2 months with 7x BMS and an Atom S3.

Then you fixed bugs and released version 1.3.1.

I suggested to @widget4145 https://github.com/widget4145 to recompile the same multi-bms code that I had provided to him to benefit from your bug fixes and since then he has never managed to monitor more than 5 BMS. At that time he also updated esphome and HA.

Do you have a backup of version 1.2 of your components ?

We could thus check if the problem comes from the latest changes or not ?

I still have the YAML that I sent him in May and June when your component was in version 1.2.

— Reply to this email directly, view it on GitHub https://github.com/Sleeper85/esphome-jk-bms-can/issues/61#issuecomment-2269333346, or unsubscribe https://github.com/notifications/unsubscribe-auth/BFHIJMBCVVSU7UI6HXFKUD3ZP6KCHAVCNFSM6AAAAABL77RWYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENRZGMZTGMZUGY . You are receiving this because you were mentioned.Message ID: @.***>

txubelaxu commented 1 month ago

Yes, it was the problem. Removing can component from yaml, it works (YamBMS 1.4.2)

I have analyzed @widget4145 user's issue about why is not working with 7 BMSs. I cannot recreate that config here. From 1.3 to 1.3.1 there is a bug correction on the code:

That is why I unpupblished v1.3.

Well... it works for some time. After a few minutes it restarts. :-(

txubelaxu commented 1 month ago

I have checked esphome-jk-bms and has worked for all the night with no reboot:

But something happens to esphome-jk-bms-can

virus100b commented 1 month ago

For me no issues so far. Using yambms 1.4.2 version with RS485 board without talk pin. i am using previous txubelaxu component. I did not update to last verdion.

Screenshot_20240806_103612_Home Assistant

txubelaxu commented 1 month ago

Maybe it is because I deactivated the can layer. But... without can layer, should it work as stable as esphome-jk-bms?

El mar, 6 ago 2024 a las 9:40, virus100b @.***>) escribió:

For me no issues so far. Using yambms 1.4.2 version with RS485 board without talk pin. i am using previous txubelaxu component. I did not update to last verdion.

Screenshot_20240806_103612_Home.Assistant.jpg (view on web) https://github.com/user-attachments/assets/20e07fc6-4d5a-4318-9b7f-d16f82ad4cfe

— Reply to this email directly, view it on GitHub https://github.com/Sleeper85/esphome-jk-bms-can/issues/61#issuecomment-2270594309, or unsubscribe https://github.com/notifications/unsubscribe-auth/BFHIJME3AQH4I2V7RJV36QTZQB4VDAVCNFSM6AAAAABL77RWYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENZQGU4TIMZQHE . You are receiving this because you were mentioned.Message ID: @.***>

Sleeper85 commented 1 month ago

Maybe it is because I deactivated the can layer. But... without can layer, should it work as stable as esphome-jk-bms?

Normally yes. The CAN layer only retrieves information from YamBMS and sends it over the CAN bus. The canbus is declared in board.yaml but I don't think it should cause reboot problems even if you don't use it.

Which ESP32 board do you use? How many BMS?

Uptime of my ESP32 DevKit V1 with YamBMS 1.4.2 and 2x BMS (UART + BLE). The BLE stack takes a lot of memory, I'm at the limit but it's stable.

Capture d’écran du 2024-08-05 12-28-36

Same ESP32 with only 1x BMS UART, noticed the difference in available memory without the BLE stack.

Did you compile the latest version from my git or from your fork (it's not up to date) ?

txubelaxu commented 1 month ago

I have 2xJK-PB connected using RS485. Addresses: 0x01 and 0x02. So, I am using mode2.

I got the "development" branch from your git yesterday. I have syncrhonized again now and I have compiled again with the this last version. After this same issue appears.

I think the esp32 is very powerful for this code:

Successfully created esp32s3 image. Linking .pioenvs/esp32-s3/firmware.elf RAM: [= ] 6.6% (used 34624 bytes from 524288 bytes) Flash: [= ] 10.9% (used 881893 bytes from 8126464 bytes) Building .pioenvs/esp32-s3/firmware.bin Creating esp32s3 image... Successfully created esp32s3 image. esp32_create_combined_bin([".pioenvs/esp32-s3/firmware.bin"], [".pioenvs/esp32-s3/firmware.elf"])

The issue: after some seconds/minutes esp32 restarts

El mar, 6 ago 2024 a las 12:15, Sleeper85 @.***>) escribió:

Maybe it is because I deactivated the can layer. But... without can layer, should it work as stable as esphome-jk-bms?

Normally yes. The CAN layer only retrieves information from YamBMS and sends it over the CAN bus. The canbus is declared in board.yaml but I don't think it should cause reboot problems even if you don't use it.

Which ESP32 board do you use? How many BMS?

Uptime of my ESP32 DevKit V1 with YamBMS 1.4.2 and 2x BMS (UART + BLE). The BLE stack takes a lot of memory, I'm at the limit but it's stable.

Capture.d.ecran.du.2024-08-05.12-28-36.png (view on web) https://github.com/user-attachments/assets/6cdea03f-72a6-407a-b03d-2a50ec7aed5c

Same ESP32 with only 1x BMS UART, noticed the difference in available memory without the BLE stack.

image.png (view on web) https://github.com/user-attachments/assets/c4f0834c-1427-4f36-aef7-ee3ba8a2a4da

Did you compile the latest version from my git or from your fork (it's not up to date) ?

— Reply to this email directly, view it on GitHub https://github.com/Sleeper85/esphome-jk-bms-can/issues/61#issuecomment-2270916717, or unsubscribe https://github.com/notifications/unsubscribe-auth/BFHIJMHOK3YXN4KTRRTMSSTZQCO57AVCNFSM6AAAAABL77RWYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENZQHEYTMNZRG4 . You are receiving this because you were mentioned.Message ID: @.***>

txubelaxu commented 1 month ago

Hi @Sleeper85,

I have just made a change on my git. Could @widget4145 try with this change?

txubelaxu commented 1 month ago

I have opened an independent issue for "7 slaves issue".

I have detected that

esphome-jk-bms:

esphome logs esp32-S3-example-jkpb-rs485_0master_2slaves.yaml | grep "nullptr\|for object with address"
INFO ESPHome 2024.7.3
INFO Reading configuration esp32-S3-example-jkpb-rs485_0master_2slaves.yaml...
INFO Starting log output from esp-rs485-gw.local using esphome API
INFO Successfully connected to esp-rs485-gw @ 172.26.1.217 in 0.089s
INFO Successful handshake with esp-rs485-gw @ 172.26.1.217 in 0.523s
[20:36:43][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca30d0 [184.000000]
[20:36:45][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca2ad0 [184.000000]
[20:36:45][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca2b28 [184.000000]
[20:36:45][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca2c04 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2c5c [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2cb4 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2684 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2e10 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca2e68 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2ec0 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2f18 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2f70 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2fc8 [184.000000]
[20:36:46][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca3020 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca2524 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca257c [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca25d4 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca262c [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2708 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2760 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca27b8 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca2810 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2868 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca28c0 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2918 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca2970 [184.000000]
[20:36:48][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca29c8 [184.000000]

but, esphome-jk-bms-can: esphome logs multi-bms_JK-PB_RS485_mode2.yaml | grep "nullptr|for object with address" INFO ESPHome 2024.7.3 INFO Reading configuration multi-bms_JK-PB_RS485_mode2.yaml... INFO Detected timezone 'Europe/Madrid' INFO Starting log output from esp32-s3.local using esphome API INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 0.488s INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.022s

[20:25:45][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8770 [8064.000000]
[20:25:46][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca87c8 [8064.000000]
[20:25:46][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca8820 [8064.000000]
[20:25:46][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca8878 [8064.000000]
[20:25:46][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca88d0 [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8928 [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8980 [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca89d8 [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca8a30 [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8484 [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8b8c [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8be4 [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8c3c [8064.000000]
[20:25:47][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca8c94 [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca5200 [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca8274 [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca82cc [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8324 [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca837c [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca83d4 [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca842c [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca8508 [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8560 [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca85b8 [8064.000000]
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8610 [8064.000000]
[20:25:49][E][jk_rs485_bms:1297]:   Object changes its nullptr !!!--------- NOT PUBLISHED 0x4DB2
[20:25:49][V][jk_rs485_bms:1292]: Publishing state 1 for object with address 0x3fca86c0 [8064.000000]
[20:25:50][V][jk_rs485_bms:1292]: Publishing state 0 for object with address 0x3fca8d44 [8064.000000]

This is an issue that I corrected, about a corruption on certain zone of memory caused because when creating instances without arguments, it is necessary a "default constructor" (that is why I added line 17):

So, the problem reapears again here, but only with "esphome-jk-bms-can" and not with "esphome-jk-bms" alone. Umm... I have think a lot about this. lololololol

Any ideas?

Sleeper85 commented 1 month ago

@txubelaxu

ESP32-S3 is powerful enough for sure.

I just compiled (6/08 22:05) the code on an ESP32-S3 without importing the yambms_canbus.yaml package with supervision of 3x JK-B* BMS. We'll see if I encounter the same problem as you or not.

Wait and see

Sleeper85 commented 1 month ago

@txubelaxu

Here is the uptime of my ESP32-S3 test board without the yambms_canbus.yaml package, so no reboot at home.

The only difference is that I'm not using your RS485 component because I do not have a JK-PB BMS.

But @virus100b who uses your RS485 component confirms that it is stable for him too.

If you think it comes from the canbus pins you could try to modify the board_esp32-s3-devkitc-1.yaml file and comment out the block below.

canbus:
  # CANBUS NODE 1
  - platform: esp32_can
    id: canbus_node1
    tx_pin: 38
    rx_pin: 39
    can_id: 1
    bit_rate: 500kbps

If you don't have an inverter, you can easily simulate an inverter with two CAN transceiver boards I can explain how to do it if you are interested.

SN65HVD230 3V3 does not require a 4.7K resistor.

virus100b commented 1 month ago

I have not updated to the last @txubelaxu component. Below is my uptime using clasic esp32.

Screenshot_20240807_103517_WhatsApp

Can module TJA1050 is hiden under esp32 board. I have bypassed 4.7k resitor in my desperate troubleshooting when CAN was not working. Now i have direct connection on RX Can to ESP. I will remove bypass and use 4.7k resistor once i will be home.

Screenshot_20240807_102941_Home Assistant

txubelaxu commented 1 month ago

@txubelaxu

Here is the uptime of my ESP32-S3 test board without the yambms_canbus.yaml package, so no reboot at home.

The only difference is that I'm not using your RS485 component because I do not have a JK-PB BMS.

But @virus100b who uses your RS485 component confirms that it is stable for him too.

If you think it comes from the canbus pins you could try to modify the board_esp32-s3-devkitc-1.yaml file and comment out the block below.
canbus:
  # CANBUS NODE 1
  - platform: esp32_can
    id: canbus_node1
    tx_pin: 38
    rx_pin: 39
    can_id: 1
    bit_rate: 500kbps
If you don't have an inverter, you can easily simulate an inverter with two CAN transceiver boards I can explain how to do it if you are interested.

SN65HVD230 3V3 does not require a 4.7K resistor.

Yes, I have that code block commented out all the time. After last changes I made last night, the "pointer bug" has gone away, but not the restarts. Please, do not ask why these changes have corrected that bug because I do not know. lol In the similar previous bug, I was googling and openaing and encountered that suggestion.

These are the rebootings I have detected during this night:

WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API
INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 1.036s
INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.022s
WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API
INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 1.491s
INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.018s
WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API
INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 0.574s
INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.021s
WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API
INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 0.842s
INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.018s
WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API
INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 2.348s
INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.024s
WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API
INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 0.135s
INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.115s
WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API
INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 0.408s
INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.021s
WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API
INFO Successfully connected to esp32-s3 @ 172.26.1.217 in 0.867s
INFO Successful handshake with esp32-s3 @ 172.26.1.217 in 0.020s
WARNING esp32-s3 @ 172.26.1.217: Connection error occurred: [Errno 104] Connection reset by peer
INFO Processing unexpected disconnect from ESPHome API for esp32-s3 @ 172.26.1.217
WARNING Disconnected from API

I will continue investigating this issue.

Many thanks!!!

Sleeper85 commented 1 month ago

@txubelaxu

WiFi problem ?

txubelaxu commented 1 month ago

It could be, not sure 100%. It is not restarting when using "esphome-jk-bms". Only with "esphome-jk-bms-can". I have to test much more with settings.

txubelaxu commented 1 month ago

I think I can sat that the problem has gone. 😃

I think it was because Cell info and Settings were 1 time / 5s. I set the period of second one ("settings") to 10000ms and problem disapeared.

virus100b commented 1 month ago

To answer @Sleeper85 question form diysolarforum "Which version of YamBMS are you using?" I am using :

Updated : 2024.08.02

Version : 1.4.2

yambms_canbus:

Updated : 2024.07.29

Version : 2.3.1

My uptime:

Sleeper85 / esphome-jk-bms-can

JK-PB RS485 : compiled OK but rebooting problem (YamBMS) #61

Updated : 2024.08.02

Version : 1.4.2

Updated : 2024.07.29

Version : 2.3.1