Using on JK-B5A24S Balancer Only unit, mostly works but some issues

[JSladen]

Hi,

I am trying to use this on a Balancer Only unit (JK-B5A24S), and while most of it works there are a couple of bits which don't, and I was wondering if they could be investigated.

I've disabled all the sensors/switches which I don't think apply to the Balancer, which has left the following sections visible, but empty and non functional.
[Configuration] Balance Trigger voltage
[Configuration] Cell count
[Configuration] Max Balance Current
[Configuration] Power Off Voltage
[Configuration] Voltage Calibration
[Sensors] Temperature Sensor 1

I can see most of these values are correctly extracted from the settings frame, but are not displayed in HA, and attempting to send them isn't allowed. I can set them absolutely fine over bluetooth.

[08:51:03][I][jk_bms_ble:1116]: Device info frame (300 bytes):
[08:51:03][I][jk_bms_ble:1168]:   Vendor ID: JK-B5A24S
[08:51:03][I][jk_bms_ble:1169]:   Hardware version: 8.X
[08:51:03][I][jk_bms_ble:1170]:   Software version: 8.0.3M

[08:51:04][I][jk_bms_ble:1025]: Settings frame (300 bytes):
[08:51:04][D][jk_bms_ble:1052]:   Unknown6: 00 00 80 3F (1.000000)
[08:51:04][I][jk_bms_ble:1061]:   Cell count: 16
[08:51:04][I][jk_bms_ble:1064]:   Power off voltage: 3.200000 V
[08:51:04][D][jk_bms_ble:1075]:   Unknown74: 3.033129
[08:51:04][I][jk_bms_ble:1083]:   Start balance voltage: 4.250000 V
[08:51:04][D][jk_bms_ble:1086]:   Unknown102: 3.400000
[08:51:04][I][jk_bms_ble:1089]:   Trigger Delta Voltage: 0.008000 V
[08:51:04][I][jk_bms_ble:1092]:   Max. balance current: 4.500000 A
[08:51:04][D][switch:037]: 'esphome-jkbalancer balancer': Sending state OFF
[08:51:04][I][jk_bms_ble:1096]:   ADC Vref: unknown V

[09:01:29][D][number:054]: 'esphome-jkbalancer power off voltage' - Setting number value
[09:01:29][D][number:113]:   New number value: 3.100000
[09:01:29][E][jk_bms_ble.number:019]: This number entity isn't supported by the selected protocol version

The following is a list of the unknown parts when decoding the data, if any of that is of use?

[08:58:51][D][jk_bms_ble:788]: Unknown210: 00 00 00 00 (always 0x00 0x00 0x00 0x00)
[08:58:51][D][jk_bms_ble:791]: Unknown214: FF FF 00 00 (0xFF 0xFF: 24 cells?)
[08:58:51][D][jk_bms_ble:794]: Unknown218: 02
[08:58:51][D][jk_bms_ble:797]: Unknown219: 0C
[08:58:51][D][jk_bms_ble:805]: Unknown221: 01
[08:58:51][D][jk_bms_ble:812]: Unknown226: 00 00 00 00 00 00 00 (always 0x00...0x00?)
[08:58:51][D][jk_bms_ble:816]: Unknown233: 64 0D 9D 4A (5146290.000000)
[08:58:51][D][jk_bms_ble:820]: Unknown237: 40 00 00 00 (always 0x40 0x00 0x00 0x00?)
[08:58:51][D][jk_bms_ble:834]: Unknown290: 00 E0 82 3F (always 0x00 0x00 0x00 0x00?)
[08:58:51][D][jk_bms_ble:837]: Unknown294: 00 90 D5 3F
[08:58:51][D][jk_bms_ble:840]: Unknown298: 00

If there is anything else I can provide then please let me know.
Thank you for any assistance you can provide.

[syssi]

You are right. The JK04 support isn't complete yet. Could you increase the log level to VERY_VERBOSE? I'm interested in some raw data (notification frames) of your device:

https://github.com/syssi/esphome-jk-bms/blob/main/components/jk_bms_ble/jk_bms_ble.cpp#L219-L220

[JSladen]

Is this what you are after?

[10:03:34][VV][jk_bms_ble:220]: Notification received: 55.AA.EB.90.02.AC.FF.A9.54.40.9D.95.54.40.D9.6C.54.40.FF.A9.54.40.3B.81.54.40.9D.95.54.40.D9.6C.54.40.FF.A9.54.40.3B.81.54.40.D9.6C.54.40.3B.81.54.40.FF.A9.54.40.3B.81.54.40.9D.95.54.40.3B.81.54.40.3B.81.54.40.00.00.00.00.00.00.00.00.00.00 (80)
[10:03:34][VV][jk_bms_ble:220]: Notification received: 00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.F7.32.99.3D.E9.3E.A1.3D.E4.7C.9D.3D.8A.09.9A.3D.70.90.9A.3D.02.B2.92.3D.5B.82.94.3D.04.2D.93.3D.FA.3C.92.3D.66.EB.93.3D.88.3A.93.3D.52.24.97.3D.E6.53.98.3D.03.9A.95.3D.D9.65 (80)
[10:03:34][VV][jk_bms_ble:220]: Notification received: 92.3D.11.F4.9A.3D.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.CB.91.54.40.00.36.60.3C.00.00.00.00.FF.FF.00.00.0D.01.01.01.3C.F8.8C.40.00.00.00.00.00.00.00.54.39.92.4A.40.00.00 (80)
[10:03:34][VV][jk_bms_ble:220]: Notification received: 00.00.56.3C.5A.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.D0.02.0D.02.00.00.00.00.00.00.00.00.00.00.03.00.00.00.00.00.00.00.00.00.08.06.09.00.00.58.24.40.00.A8.D5.3F.00.C6 (60)

I Think I remember reading somewhere else that it uses a 320 byte frame rather than a 300, so I don't know if there is any information missing from the end of it?

[syssi]

This frame looks good & complete (80+80+80+60 = 300, Preamble 55.AA.EB, CRC at the EOF). You have provided the raw data of a cell info frame:

55.AA.EB.90.02
            ^^ cell info frame

Could you capture a settings frame and device info frame?

55.AA.EB.90.01 # frame_type: settings
55.AA.EB.90.03 # frame_type: device info

The frames are requested right after the established BLE connection. You could use the BLE switch to disconnect + resume the BLE connection:

switch:
  - platform: ble_client
    ble_client_id: client0
    name: "${name} enable bluetooth connection"

[JSladen]

Certainly, below should be all 3 frames from a single pass. let me know if there is anything else I can grab for you.

[10:03:02][I][jk_bms_ble:1025]: Settings frame (300 bytes):
[10:03:02][VV][jk_bms_ble:1026]:   55.AA.EB.90.01.72.00.00.80.3F.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.10.00.00.00.CD.CC.4C.40.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.C9.1E.42.40.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.88.40.9A.99.59.40.6F.12.03.3C.00.00.90.40.01.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00
[10:03:03][VV][jk_bms_ble:1027]:   00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.6F (140)

[10:03:00][I][jk_bms_ble:1116]: Device info frame (300 bytes):
[10:03:00][VV][jk_bms_ble:1117]:   55.AA.EB.90.03.72.4A.4B.2D.42.35.41.32.34.53.00.00.00.00.00.00.00.38.2E.58.00.00.00.00.00.38.2E.30.2E.33.4D.00.00.CC.B3.15.01.0A.00.00.00.4A.4B.2D.42.35.41.32.34.53.00.00.00.00.00.00.00.31.32.33.34.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00
[10:03:00][VV][jk_bms_ble:1118]:   00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.C0 (140)

[10:03:03][VV][jk_bms_ble:679]: Cell info frame (300 bytes):
[10:03:03][VV][jk_bms_ble:680]:   55.AA.EB.90.02.72.15.44.54.40.87.FB.54.40.3B.81.54.40.87.FB.54.40.D9.6C.54.40.C3.D2.54.40.77.58.54.40.4B.24.55.40.3B.81.54.40.61.BE.54.40.9D.95.54.40.61.BE.54.40.3B.81.54.40.E9.0F.55.40.9D.95.54.40.D9.6C.54.40.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.F7.32.99.3D.E9.3E.A1.3D.E4.7C.9D.3D.8A.09.9A.3D.70.90.9A.3D.02.B2.92.3D.5B.82.94.3D.04.2D.93.3D.FA.3C.92.3D.66.EB.93.3D.88.3A.93.3D.52.24.97.3D (150)
[10:03:03][VV][jk_bms_ble:681]:   E6.53.98.3D.03.9A.95.3D.D9.65.92.3D.11.F4.9A.3D.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.6C.8B.54.40.00.72.37.3C.00.00.00.00.FF.FF.00.00.07.02.02.01.3C.5C.26.C0.00.00.00.00.00.00.00.80.6D.91.4A.40.00.00.00.00.25.3B.5A.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.D0.02.D0.02.00.00.00.00.00.00.00.00.00.00.03.00.00.00.00.00.00.00.00.00.EA.05.09.00.00.98.91.3F.00.78.D5.3F.00.1E (150)

[syssi]

As next step I would be happy about a BLE traffic capture (btsnoop). Don't worry, it's quite simple as long as you have an Android device. The btsnoop capture is part of the bug report bundle (if enabled beforehand):

https://medium.com/@charlie.d.anderson/how-to-get-the-bluetooth-host-controller-interface-logs-from-a-modern-android-phone-d23bde00b9fa

I try to summarize the goal for a brief overview:

1. Turn on the developer mode of your android device + enable the btsnoop capturing
2. Turn off bluetooth, turn on bluetooth again
3. Open the Jikong Balancer App
4. Go to the settings and note down the current values and set some new values.
5. Use adb to retrieve the bug report bundle ZIP.
6. Extract the bundle und provide the btsnoop_hci.log

I will extract the write commands / instructions from the capture so we can guess the proper payload of the JK04 protocol. Without having this capture there is a risk to write garbage to your device.

[JSladen]

bt-snoop-JK-B5A24S.log

Hopefully this includes what you are after. I've removed the before and after as it was quite a large file, but I don't think anything should be missing.

I set each of the following values and then reverted them in this order
Initial Changed Return
Cells Count 16 17 16
Trigger Delta Vol.(V) 0.008 0.108 0.008
Max Balance Cur.(A) 4.500 3.500 4.500
Power Of Vol(V) 3.200 2.200 3.200
ADC Vref 52.xx 51.xx 52.xx

The ADC Ref figures I'm not 100% certain on, as they change, and it came up with the wrong figure next time I opened it.

Please let me know if anything else can be of use.

[syssi]

Thanks for preparing such a well structured and documented capture. I will try to go the next steps later the week and will report back if something is missing.

[syssi]

Extracted write commands:

aa5590eb 97 00 b44cb1791a3fa86121c638de6f09 # Request device info
aa5590eb 96 00 3dd6b2ad5d1a244642b3f70b136d # Request cell info
aa5590eb 96 00 4a93c058a9fe5358b092ca4add8a # Request cell info

aa5590eb 1c 02 1100 04d2e36fb8c5d4912ec07e1f # Set cell count 17
aa5590eb 1c 02 1000 378e33f7ada8dbf65ea26926 # Set cell count 16

aa5590eb 64 04 1b2fdd3d 22b35cce3623a75a4eed
aa5590eb 64 04 6f12033c ba4c253e8976b0f6ef9f

aa5590eb 68 04 00006040 8790eaa88382efbc3413
aa5590eb 68 04 00009040 9da3c98226fbf6539a45

aa5590eb 20 04 cdcc0c40 8aac6f3768b50a286b19
aa5590eb 20 04 cdcc4c40 2c0eaedfa413b82a4366

aa5590eb 44 04 717a3e40 6e7d4a145009cd1a30e4
aa5590eb 44 04 178e4540 ede90757c043ca4d2862

aa5590eb 96 00 a141ed784905039bca92f3db47b4 # Request cell info

[syssi]

See also #94

[iovcharyk]

@syssi thanks a lot for your work done, just to confirm it's working fine via BLE and JK-B2A20S20P HW V11.XW, SW V11.25H

[syssi]

@iovcharyk Thanks for your feedback! Could you do me a favor and create a new issue + report your device as supported? I don't want to reference this issue because your report is off-topic here.

[JSladen]

Hi,

Please can I check if there was any more progress on fully supporting JK04/balancer units?

I am happy to try and provide further scans or test things out if that's of any assistance?

Many Thanks