Flipsky Dual FSESC6.6 - Idiot Damaged

[caswal]

I am making a little E-Bike, as a pit bike for motorsport events. And using this project as a learning experience to move from Arduino and designing my own PCB piggyback boards. To making my own PCB with an integrated Micro. I am making a controller to drive a 128x64 GLCD, and be able to select different drive modes, show pack voltage, current draw, speed etc. Talking to the Flipsky controller via UART. I would put my skills firmly in the meddling hobbyist level.

While messing around, I was the idiot. Shorted an M3 Bolt across some tantalum bypass caps (I thought I had the board with the M3 fully taped up), shorting across the main 32v battery rail and blew out a track:




This is a dual BLDC Motor Controller. So the left and right sides are duplicates of each other. With a selectable canbus interface between them (enabled/disabled by the Dual-CAN switch in the middle).

Amusingly the side (Side 2) with the blown track and set of caps still works. Side 1 went non-responsive.

Probing round Side 1 with a voltmeter and a scope. 3.3v and 5v rails are still good and each side have their own separate rails. As I was checking and comparing SWCLK and SWDIO of each side, I noticed that the MCU on Side 1 was very hot. Checking with my thermal camera, it gets to 45C+ in just a few seconds. The bench powersupply was doing a few hundred milliamps, at 31v.

This board is heavy copper. I don't have a hot plate and the hotair station had no luck. So I cut the pins on the MCU with a scalpel and removed it. Cleaned up the pads with an iron. Powered the controller back up, and it draws 45mA.

So I assume during the short, some stray current has found its way into the MCU on Side 1 and damaged it, gaining an internal short.

The MCU is an STM32F405RGT6

Turning my lemons into lemonade. This is quite an expensive controller, so I would like to save it and turn this into a learning experience.

Side 2 looks to be easily repairable. With some new bypass caps for phase A2.

What do you think my chances are of replacing the MCU for Side 1. I don't have the exact schematic for this controller, but it is based on an open design: https://vesc-project.com/node/311

Should I check all the pads for anything fishy, and if it checks out try my luck with a replacement MCU?

How is the STM32F405 flashed? Are they shipped with the USB DFU Bootloader. So solder it on the board, plug it in and flash it with the STM Software?

Checking Element14/Mouser this all looks a bit academic, as the STM32F405RGT6 is on backorder to 2022. STM32F405RGT6V is available sooner, maybe I am blind but can't find anything on the datasheet as to what the difference is, and on a naming conventions breakdown, the V is off the end.

[caswal]

So probing the pads of the removed STM, and the 2 connected to the canbus transceiver were at 5.15V and 4.4V. So I removed it, but then I found out the STM is 5V tolerant most of its pins. So maybe that was premature.

Some aliexpress sellers appear to have some cut reel STM32F405RGT6, so I'll roll the dice and get one. Appears the BOOT0 pad is not connected.