FAQs
cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 

AEK-POW-BMS63EN: second board burned near L9963E after swapping in

kaleed
Associate

‎2025-10-01 8:21 AM - last edited on ‎2025-10-01 8:34 AM by

Hello ST team,

Following on from my “connection/reading” thread, I have a separate issue about physical damage.

Timeline

  • With 4 cells both of my AEK-POW-BMS63EN boards read correctly.

  • When we moved to 6 cells on Board #1, we began seeing the “dead” behavior (refs/LEDs off, SPI zeros/NaNs).

  • We then swapped in Board #2, keeping the same holder and harness. Shortly after connection, Board #2 burned in the same area near the L9963E as Board #1 (photos attached).

  • Later we populated all the way to 14 cells (VBAT ≈ 50 V) to re-check; still no refs/LEDs and SPI zeros/NaNs on the remaining board.

Current status

  • Two boards visibly damaged in the same L9963E area.

  • We avoided pin-by-pin live plugging; balancing was off.

  • Per-cell voltages at the holder side look normal and monotonic with a DMM.

Asks

  1. Could you help with failure analysis based on the photos and symptoms? Do these marks suggest tap mis-order over-stress or a hot-plug/inrush event?

  2. What is the correct RMA/replacement path for two AEK-POW-BMS63EN units (Italy, academic lab)?

  3. Before we purchase another board, please share the approved connection/power-up sequence and any firmware settings (cell-enable mask, unused-tap handling) to avoid a repeat.

Thanks in advance
Khaled.

Labels:
Preview file
1577 KB
Preview file
2518 KB
Preview file
2734 KB
0 Kudos
1 REPLY 1
Max VIZZINI
ST Employee

‎2025-10-02 1:39 AM

Hi,

These are developments tools for evaluation purpose only so no RMA / failure analysis is available.

From the burn mark we can foresee a possible EOS (electrical current overstress), but we cannot judge if it was an inrush, a wrong connection or something else.

The 'proper' bring up and sequence procedure is based only on our tools and our SW, as specified in the user manual. If you change MCU, HW, SW, etc. we are unable to identify the damage source. In that case you would have to engage a consulting company, because we do not provide this kind of service. We can provide via private email some contacts for that in the Turin area.

I do NOT suggest keeping buying additional boards until you do not get around the root cause otherwise the same effect of board burn-out could happen again.

The best case is to start from a working case based on our Manual and then try to modify it.

The 4/6 cells board holder is NOT equivalent to the 14 cells board holder because pin cannot be left floating in the IC (sensing and balancing pins) otherwise you will experience a wrong IC behavior. This is also explained in the manual.

https://www.st.com/resource/en/user_manual/um3185-getting-started-with-autodevkit-battery-management-system-bms-evaluation-boards-stmicroelectronics.pdf

 

Best Regards,

AutoDevKit Team

 

0 Kudos
Top