Car electric windows ...How do they work?

[Wightsparks]

Or more to the point, how do they stop?

I recently acquired a a very lovely 2002 Saab 9-3 Convertible and I'm having an issue with the rear side window.
It was making a nasty noise at the fully closed position when being auto closed as part of the roof closing. So I stripped it down and found that the teeth on the quadrant were a bit chewed at the end section and part of the motor end support was torn (round bit,middle of the pic). So I fixed/welded that lubed it all up, adjusted the stops correctly (as per the workshop manual) and it worked OK manually, but a couple of auto operations later the same issue is back.



The window will close, get to its stop then continue to be powered until the motor starts chewing up the quadrant again.

So my question is - there are no microswitches, and only 2 wires to the motor, no movement sensor on the motor and the motor will turn constantly so no internal interrupts, so how does the car know when to kill power via the relay to the motor - ie when it is all the way up/down.

a) some mysterious method (load sensing on the relay?)
b) it guesses by time and relies on the mechanism stops to just stop the motor (or self destruct in this case!)
c) ????

The later models have a pinch sensor in all windows so you need to calibrate them but this is a different issue.

I have googled this to death and trawled through the Saab Workshop system - nothing.

 

[ajlelectronics]

Current sensing when the mechanism hits the stop. On a Scimitar, there is a stud welded onto the quadrant to act as a positive stop. Is it possible that you are missing something similar? If the motor can't stall, then it can't shut down cleanly.

 

[colnerov]

Hi, I would have said load sensing because legislation says if a small persons head gets stuck in the way it's supposed to stop and back off a bit.

Colin

 

[Wightsparks]

Current sensing when the mechanism hits the stop. On a Scimitar, there is a stud welded onto the quadrant to act as a positive stop. Is it possible that you are missing something similar? If the motor can't stall, then it can't shut down cleanly.

There are 2 stops at the top of the rails and a big one in the quadrant that can be seen in the pic just above the round pressing part that I welded up.

If it is load sensing either it is not working or the teeth are jumping before the load level is reached - it does stop after a few seconds of nasty noise though so that makes me think the former. In which case it is either a new regulator mechanism .. or I try and hack the quadrant out and flip it over!