FOC configuration error and burned VCC trace

Report your hardware problems here. Include as many details as possible with pictures to get the best possible help and feedback.
Posts: 9
Joined: 07 Apr 2016, 05:14

FOC configuration error and burned VCC trace

Postby jamesshuang » 07 Apr 2016, 05:33

Hi everyone,
I made a small mistake today when I was testing my VESC 4.10, firmware 2.8. I had BLDC mode working great, but then I saw the FOC video and wanted to try it out. I set FOC mode, but forgot to detect the parameters, hence both R and L were 0. When I tried spinning up the motor, it made some horrible noises, and there was the puff of magic smoke.

After close inspection, I noticed that the V_supply trace on the back side of the board going to C40 was burned up. I soldered a jumper cable from the other side and it powered back up correctly. However, it seems like one of the phases seems to be broken now. BLDC mode no longer spins reliably with severe cogging. Two questions:

1) How does an incorrect R and L value damage the VESC board? Is this a result of incorrect assembly, or does it actually damage the system? I'm willing to contribute a bit of code to BLDC_tool to avoid this problem in the future.
2) What components could be damaged? Would it be possible to resurrect the board with a new DRV8302 chip?

Posts: 9
Joined: 07 Apr 2016, 05:14

Re: FOC configuration error and burned VCC trace

Postby jamesshuang » 09 Apr 2016, 19:06

I went ahead and replaced the DRV8302 chip. This did not fix the issue with the severe cogging. I checked the conductivity between drain and source on all the mosfets, and noticed two of them were way higher than they should have been. After replacing those two high-side mosfets, my board is working again!

Now all I have left is to figure out why FOC simply doesn't behave as well, especially with startup performance.

Posts: 2
Joined: 15 Mar 2016, 05:10

Re: FOC configuration error and burned VCC trace

Postby BrettRD » 03 Jun 2016, 08:19

Do you mean this trace? (the parts here suffered no damage whatsoever)

IMG_20160602_165237.jpg (130.31 KiB) Viewed 1030 times

I think this one is caused by massive ripple on the FET power rails, and too much capacitance distributed about the board.
If you get the config just right, you can get the on-board capacitors to ring with the motor windings.
The resonance is clamped by the FET body diode to a bit above twice the supply voltage.
This one was with a 12V supply. The motor squeaked audibly for half a second before the trace fused and the logic lost power.
The FETs and driver IC stayed cold the whole time.
On the IRC channel, we estimated this trace took about 8A AC current at 6KHz.

I replaced the trace and everything was fine.
I just need to do my configuration and add capacitors next time.

Return to “Hardware Problems”

Who is online

Users browsing this forum: No registered users and 2 guests