Washing machine motor control board (CR7195R-6)

[Eheran]

I have a CR7195R-6 (aka HANNSTAR-PENTA-2841610300) motor driver/control board, it is used in Beko washing machine(s) (and driers?). So it creates 3 phases for the motor from 1 phase AC. So far so normal.

One of the 6 IGBTs blew, taking the current sense/fuse resistors of that rail with it. I replaced that. And here is where I am at a loss now. I get ~400 V DC after rectification at the big smoothing capacitor under no load. So up until there everything works. But the inverter circuit does still not work. Maybe the driver is also broken?

But this is what really confuses me: The positive rail does not go anywhere except a small trace for powering the remaining control circuit (with another transformer and a LNK564 switcher IC) + a tiny sense(?) trace with high value resistors in series. There is a big trace from the capacitor to where it should/could go to the IGBT section, but the small connection that could have been there seems to be cut/milled out of the PCB. It does seem to be factory made, it does not look like it is blown up. When googling for this board, all the pictures of this side are of too low resolution to tell if they look the same or not. See the attached picture with my crude drawing of the traces as well as the resulting 3 phases (black, brown, grey). The blue section is not connected to anything, the positive (red) side of the 400 V DC does not power anything significant. Is this the issue? The connection from red to blue was blown from the short but does not look like it somehow? If not, where does the power even come from?

I found these 2 Youtube videos (no subtitle) of someone repairing the same board with the same issue. He only replaced the IGBT on the daughterboard. Every part of the PCB is visible there essentially:





PS: How to only link the video without embedding them? In the preview it is only a hyperlink, but in the post its the whole embedded video.

[Kim Christensen]

The connection from red to blue was blown from the short but does not look like it somehow? If not, where does the power even come from?

I think that trace is blown, but it's a clean looking burn. (Almost like they designed it like a fuse). Appears to be complete in the video that you linked:

[Eheran]

Note that it really is milled out, not just that there is no trace. There is a very clear and defined depression, below the height of the "raw" PCB. There is also no inner layer, that are all the traces you can see there. 

Edit: I removed the big cap, see the pictures. It is thinner there, that trace (if it ever existed) was milled out at the factory. I am 99 % sure.

Edit2: Could I test the IGBT section with some low DC voltage (say 15 V) from a power supply and see if the switches turn on by connecting the gate to V+?

[SeanB]

That trace is blown, it is a fuse. Replace with either a soldered in 10A fuse, or a single strand of 22AWG wire from some equipment wire.

[Eheran]

It took a microscope / macro picture to verify this. It is indeed a blown trace, thank you guys. I will replace it with a copper wire. Absurd that it is THAT clean.

[Kim Christensen]

Good to hear. In your macro picture it looks like your PCB has a thick conformal coating on it. That's probably why it looked like it was milled out, but it's really just the missing conformal coat that was covering the trace that's missing.