auxiliary electric coolant pump repair 1K0965561J

The auxiliary coolant pump, AKA turbo after run electric coolant pump, parking heater, on a 1.8 TSI VAG engine did not work.
part number 1K0965561J
I removed the pump and applied 12V to it but nothing happened.
The contact on flat side of the plastic molding is the positive.

I opened the pump by prying and pulling out the bottom piece where the electric plug is, it needs to be pulled out, it can not be twisted.

I discovered a PCB inside and it was wet and full of corrosion, I cleaned it with alcohol and toothbrush and tried 12V again and the motor jumped but didn't continue rotating, every time I powered it the rotor jumped to a position and held itself there, what does that mean? maybe only some of the 3 mosfets were firing?

I found a 47k resistor that was open circuit,

I didn't have the correct size SMD resistors but I had some smaller 0603 so that's what I replaced it with.
nevertheless the same behavior persisted.

what ensued was a lot more work trying to figure out the problem, and I found two more open circuit resistors, one 47k, another 220ohm, replaced those.
there were resistors that weren't making contact to their traces so I reflowed them, for the trace of the second 47k ohm resistor I replaced reflowing didn't help so I soldered a thin jumper wire from it to the 220ohm resistor it connects to, the first jumper ended up being a very thin copper wire which broke and later I replaced it with a thicker wire.

after a lot of board repair and the pump didn't work I removed the 3 mosfets to test them.
the middle one is a 2N06L64 N-Channel MOSFET and I tested it with multimeter with a method I found someone wrote here and it tested fine https://www.badcaps.net/forum/troubl...lt#post1842389

the two other mosfets are BTS118D and from their datasheet they don't seem to be normal mosfets, they mention Smart Lowside Power Switch, anyway they didn't pass the test and behaved different but that must be because they're special. does anyone know how they can be tested?
in the end all 3 mosfets turned out to not be the problem.

I actually did the PCB repair in one night and the pump didn't work after exhaustive work, the rotor would still just violently rotate to a fixed position and then stay in place, I tried to manually give it a fast rotation and it actually did continue to rotate slowly and very poorly like it was misfiring or firing only on one or two mosfets, so I gave up on it and ordered a new pump.
in the morning I was checking it out again and just decided since I gave up I will just finally assemble it entirely back together, and all of a sudden it started working!
Why didn't it work earlier then? well, when I was repairing the PCB, every time I would find one problem with it and repair it, to test if it now worked I would slide the PCB with the stator that's attached to it, into the pump housing, but I wouldn't push it all the way in, because then it would be difficult to pull it out again.
but after giving up and just wanting to put the thing back together, I of course pushed the PCB/stator assembly all the way in, and evidently that's what the motor needed to work!
I wish I had known that from the beginning, now I'm not completely sure at what point during my extensive board repair the motor would've already worked...
unbelieveable... I never expected a couple millimeter axial offset between the stator and rotor to cause the motor to not work! can anyone explain why it would matter so much?
actually now that I think about it, there is a component on the stator side of the PCB, it has markings U18944, I don't know what it is, but could it be some kind of sensor that senses the position of the rotor? and if so maybe it wasn't able to sense it without the stator pushed all the way in?

The design of this pump is quite interesting, the stator is the inner part of the motor, and it is inside the pump's plastic housing and completely weather sealed from the rotor by the plastic housing, then the rotor or impeller is on the outside of that housing.
obviously the rotor assembly wasn't weather sealed well enough since water somehow got inside it and caused the PCB to corrode...

before taking apart this pump I had removed the main water pump from the engine which then causes coolant to fall onto this auxiliary pump which is in front of the engine lower than the main water pump. I suspect that may be why the PCB got wet.
so in the future I think it's a good idea to remove the auxiliary pump before removing the main water pump to prevent the aux pump from becoming wet and possibly being damaged again like what happened to this one.