You are here

a (cautionary) tale of 4 dead VESCs and question of repairability

6 posts / 0 new
Last post
ligi
ligi's picture
Offline
Last seen: 3 months 1 day ago
VESC Free
Joined: 2020-08-21 08:21
Posts: 24
a (cautionary) tale of 4 dead VESCs and question of repairability

I was happily using a VESC MKIV for long time on my cargo-shuttle. Unfortunately recently I had a cascading failure that was leading to 4 dead ESCs. I was changing spokes on my rear wheel with the motor - while doing so I must have damaged the cable going into the motor and motor phases with 48V where shorted to the HAL sensors. This made my old trusty MKIV go bad. Unfortunately initially I did not know what the root cause was - and just thought the controller died of old age or some freak error. So I attached a second controller as replacement that directly also died - and was leading me to the root cause. Unfortunately I needed to be at a place to help build up a festival so I decided to use another controller (without HAL sensor attached to not destroy this one als) - that was not really matching this ride - it worked for a while - but died of overheating in the middle of the ride. As this was a dual VESC (ripped out of a project with 2 motors) I could continue the ride with the second conroller (this time removed from the enclosure to dissipate the heat better). So I made it the way to the festival - but arriving there this controller also died - ordered a replacement controller there via kleinanzeigen. Was a bit scared to order via trampa as I needed to order it to the festival grounds and was not sure how this works with a parcel from another country. Unfortunately this controller also died (VESC compatible but not an original VESC)

I am writing this as a word of warning - if your controller dies: check the motor first before attaching another controller to not kill this one also.

Also wondering if there is a chance to repair these controllers as they are kind of expensive. Skimming through the forum it looks like replacing the STM might be a solution - I can solder basic stuff - but replacing the controller with so many pins scares me a bit. Also wondering what could have died in the overheated ones - also the controller? Is there any service repairing VESCs? Would be a shame to throw all these controllers away - I think there are still a lot of good working parts on them like the FETs.

Snabelost
Offline
Last seen: 1 month 2 weeks ago
VESC Free
Joined: 2022-08-24 10:05
Posts: 22

I hope it turns out good for you!
It's a nice ride you have there!

ligi
ligi's picture
Offline
Last seen: 3 months 1 day ago
VESC Free
Joined: 2020-08-21 08:21
Posts: 24

Thanks!

Just received my replacement VESC (updated to the 75V version in the process) - also ordered a new motor with some more power and as a direct-drive so I can do recuperation finally.

Still hope I can repair the other controllers somehow - maybe anyone knows someone that repairs VESCs?

---
friendly greetings,
ligi
https://ligi.de

drummerboy5408
Offline
Last seen: 2 weeks 1 day ago
Joined: 2019-05-26 22:42
Posts: 24

I attempted to repair a Dual VESC that had failed due to a similar Hall sensor issue.
A friend of mine had a Trampa board with Flipsky 6374 Sensored motors, and after 2 years of solid riding...the Hall sensors inside the motor shorted and killed one side of the Dual VESC (Only 1 side has 3.3v power now)

I found the issue to be a 3.3v LDO regulator, but this is an Aftermarket Clone VESC (Spintend UBOX). You should be able to hook up Battery power and use a Voltmeter to check if there is 3.3v power to the MCU (F405 on mine). Hopefully this helps!

ligi
ligi's picture
Offline
Last seen: 3 months 1 day ago
VESC Free
Joined: 2020-08-21 08:21
Posts: 24

Thanks for the info - will try this.

Also thanks for reporting I am not the only one with the problem. Wonder if there is a way to hardware-wise protect from this fault. I know over-voltage protection exists in some power-supplies - wonder if would also be possible to integrate something like this in the VESC.

---
friendly greetings,
ligi
https://ligi.de

drummerboy5408
Offline
Last seen: 2 weeks 1 day ago
Joined: 2019-05-26 22:42
Posts: 24

There is a way to protect 3.3v shorts in hardware.
What I ended up finding out about the Spintend UBOX vesc, is that they had to redesign a 2nd Version to incorporate this very fix with 3.3v Over-Current protection and resetting.
Lots of people were having their Hall sensor wires being shorted to ground through the motor mount frames over time, and this can cause a current spike enough to burn the 3.3v LDO's.