Here'* a link to the data collected so far - it'* in the form of graphs saved to a Word document.
This data was collected with the logger set to 'normal' or 'road test' mode which shows the pattern of open loop to closed loop and then back to open loop behaviour that I've been describing. The closed loop period starts at around 6 minutes and then goes back to open loop at around the 15-16 minute mark.
The short term and long term trim values appear to be compensating for a rich condition which surprises me because that doesn't fit with my vacuum leak theory which would make conditions go lean.
Check out the large number of idle flags being set and not just during idle conditions. I'm not 100% sure about this but I'm thinking that an idle flag will be set as a result of an unacceptable variation in RPM. It may also set for an unacceptable variation in RPM at a constant TPS which may explain the large number of hits over a wide range of driving conditions. In any case, as I've mentioned before, there is a noticeable roughness in the idle but not to the point that it'* threatening to stall - it'* just noticeable enough to be annoying.
Also surprising to me is that the TCC doesn't operate in open loop. I've always know that it doesn't activate until warmed up but I didn't realize that it was tied directly to the open loop requirement.
Going back to the idle data, there are some good examples of the idle RPM variations at both the 8 minute mark (in closed loop) and the 20 minute mark (in closed loop). It might be worthwhile to expand the O2 and INT data during those periods to see if anything obvious stands out. If you're interested in playing with the ranging of the data let me know and I'll set up a link to the data file.
I've also now collected data for a similar run but with the logger set to 'field test' mode. In this mode, the ECM stays in closed loop for the entire period after warm up so it should be interesting to compare with the 'normal' run data. I'll have this data ready shortly.
In any case, the information acquired using the data logger in conjunction with Jonas Bylund'* Winaldl provides so much more detail than the scanner tool that I doubt the scanner will get much use in the future unless Winaldl doesn't have a definiton files for one of my OBD1 GMs.