That connection you have made to the TDI hot start issue is quite interesting and worth looking into further. I'm wonder if, given the age of the cars now, that a combination of factors are sufficient enough to throw the startup parameters outside their assumed values when the car was new. If you could send the ecu file that would be great. Also , from what I can tell, the car has not been serviced at the dealer since the 3 year warranty expired, I think I will get in touch with a friend who works at a dealer and see if he can check for any service bulletins or ECU updates (have seen this fixing similar issues on Audis).

It's quite remarkable how changing all of those things made no difference. Regarding the PCV system, I believe it is functioning correctly, but haven't tested it fully. I tried removing the oil cap and the idle became very lumpy - which apparently indicates that it is working. It is common for the PCV valve to go bad on the turbo cars, and the fix there is to buy a new valve assembly which is not too expensive. On these engines though, the valve is integrated into the rocker cover - not cheap to replace. I did find where you can buy the membrane only for a reasonable price. Membrane for valve cover VAG 06F103469D 2.0 FSI - Vanos BMW Repair kits for cars This involves prying the circular cap off, and at the moment I'm not game enough to try it, given it is obviously not designed to be serviced. Might give it a go at the wreckers and see what condition the valve is in on the car there.

The other threads on this issue are frustratingly incomplete, and I too am not convinced that carbon build up on the intake valves is to blame (or at least the sole cause). Surely if the carbon was causing enough disruption to airflow to cause misfires on idle, the problem would be evident or even exacerbated by different, more strenuous operating conditions. I cannot say I've noticed any of these, the car seems to pull well and without hesitation at wide open throttle throughout the rev range. That said, one day I will probably clean them as well as the the manifold and check the HPFP all in one go (and with the right set of hose clamp pliers this time). The data from the 1.4 TSI Caddy seems quite different and I can't say I'm too familiar with that engine. Seems like it is using a narrowband sensor for B1S1. The resistance is much lower than any of the wide or narrowband sensors on my engine.