ECU holds IAC valve fully opened

[sink]

As mentioned in the subject row my VEMS ECU holds the IAC valve fully opened, 2 months ago when I left the car in the garage everything was fine on that end. Now two months later when starting the car something was odd - it didn't idle well at all it fluxuated between 2000....2600 rpm and the car ran rich, rich off the charts.

It doesn't seem to be a config mistake cause I've not fiddled in that area :D but still I'm uploading mine so that someone could please look it over.
http://uploading.com/files/f23886ee/triggerid_asjad_paigas2011.03.26-23.29.21.vemscfg/

Could it be that I've blown something in the ECU itself?

Edit: everytime I forget this part :-\
The car is an Audi S2 with a 3B engine modified to work on 1.8T coils. Still using the original 2pin L-shaped PWM type IAC valve. Using Firmware 1.1.85

[Ignitec]

If it's always fully open, it's probably dead FET. Did you put flyback diode on IAC?

[sink]

As I said, the car stood in the garage for 2 months and in that time I did put the flyback diode. Now after 2 months when I started the car for the first time this problem occured. Maybe I put it in wrong :-[
Can You show me on a picture or on a schematic what should be replaced? And with what?

The ECU should be a version 3.3 ???

[[email protected]]

Try disconnecting the fly-back mate, see if it still holds it open.
If the fly-back was in the wrong way then you may have blown the FET open.

[Ignitec]

Disconnect IAC and remove diode, switch ignition on and use DVM to see if there is permanent gnd on iac connector. If it is - dead FET. If you install diode in wrong direction you would short FET out to +12V, and something would smoke (fuse, diode, FET...), but IAC would be probably closed...

[sink]

In short terms - problem fixed  :D
In longer terms - how? I really don't have an answer for that.

But I'll tell what I did anyway  ;) I already removed the flyback diode yesterday and there was no improvent on the problem. Today I took the ECU to work and took it apart, checked the MOSFET's and they all seemed OK, atleast checking them with the DVM they all acted as a mosfet should. That didn't satisfy me so I check the hole board under a microscope. I actually found small pieces of metal shavings. One between two legs of the Atmel processor and another also somewhere - don't remember anymore where exactly. My thoughts on the shavings are: because the  ECU is a second hand one and originally intended for a BMW then I suspect that the bloke who re-did it for an Audi wasn't very tidy (lack of a better word :D) There were quite many small pieces of wire on the board when I looked closer.
But hey, who knows:D Well no-one to blame but atleast the problem seems fixed.