AEK-POW-BMS63EN + AEK-POW-BMSHOLD: VBAT ≈ 50 V but VREF/VCOM = 0 V, LEDs off, SPI reads 0/NaN

kaleed · ‎2025-10-01

Hello ST team,

I’m evaluating AEK-POW-BMS63EN with the AEK-POW-BMSHOLD. I’m seeing a bring-up problem where the pack voltage is correct but the board appears “dead” (no refs, no LEDs, SPI values = 0/NaN).

Setup

Boards: AEK-POW-BMS63EN (2 units available; this post is about the connection/reading behavior).
Battery holder: AEK-POW-BMSHOLD.
Cells: 18650 “INR 18650 MJ1”.
Population: progressed from 6S to 14S (final VBAT ≈ 50 V).
MCU/comms: AEK-MCU-C4MLIT1 + AEK-COM-ISOSPI1; UART 115200 to PC.
Firmware: AutoDevKit example reading cell/temperature; balancing disabled.
Power-up: mate the full holder connector once, with MCU unpowered; then power MCU and start SPI (no pin-by-pin hot-plugging).

Measurements (board powered by pack)

VBAT at the holder/connector: ≈ 50 V (matches cells).
VREF, VCOM and other reference rails on the BMS board: 0 V (some readings slightly negative).
LED indication: none (docs suggest at least one LED when VBAT > ~9 V).
SPI: V1…V14 = 0.000 V, T1…T14 = NaN (make sense since the temp sensors not used, B1…B14 = 0. (Terminal screenshots attached.)

What I’ve checked

Per-cell on the holder side with DMM: each adjacent tap is ~3.6–4.2 V; total ≈ 50 V.

Questions

On AEK-POW-BMS63EN + AEK-POW-BMSHOLD, can you confirm the exact mapping/order of C0…C14 between holder and board connector for 6–14 cells?
For fewer than 14 cells, how should unused cell inputs be handled—firmware disable only, or any termination/hardware requirement?
Is there any jumper (e.g., V_Ext, pre-charge/hot-plug network) that must be set for refs/LEDs to come up with pack power?
Any bring-up checklist you recommend before connecting a populated holder (e.g., resistance checks from each Cn to C(n-1), expected VREF/VCOM levels, etc.)?

Max VIZZINI · ‎2025-10-02

Hi,

You cannot leave cells not mounted in the 14 cells connector or the IC will not work.

There are some cells that have to be present always (first two and last two) or the IC reading will not work.

Please refer to the datasheet https://www.st.com/resource/en/datasheet/l9963e.pdf

Best Regards,

AutoDevKit Team

kaleed · ‎2025-10-02

Thanks for your reply, ok, then for the full cells setting, I still faced the same issue with the Vref, Vcom, based on the documentation, it should read 3.3-5V, in my case where I add all the 14 cells, still the V references are zeros or milli volts! I think the problem related with the l9963e since no blinging was found in the AEK-POW-BMS and the voltage reading is still zeros from both the MCU or by using Putty to read directly the SPI signals even though the Vbatt approx = 50 volts! What could be the reasons taking into accounts all wired is connected based on autodev or SPCStudio! I think the issue related with AEK-Pow specifically l9963e but since I am not expert, I wanted to know what should I look or measure to know which device making issue

PS: We tried to replace the AEK-POW_BMS and connect each AEK-POW board with the other battery holder and when we added 6 cells the two l96EE burn out same pin, same location, we are thinking also to buy a new one but we are afraid it will burn again! Is there specific settings, should we check the wiring of the battery holders (maybe wrong connections lead to that because when we start from 1 cell to 4, the l96ee was not blinking but no burn found) any kind of guide will be appreciated

Thanks in advance,
Khaled.

SRomeo · ‎2025-10-02

Hi there,
Based on the scheme you shared, you are using the Wrong ISO port. (so communication with BMS won't take place)

Please Plug the AEK-COM-ISOSPI1 board into ISOL.

BR,
Simone
AEK_Team