Okay. Knocking noise in D sitting stationary with the brakes applied.
Thanks for the chart. I didn't notice it in the service manual.
I performed the prescribed test and there is no noticeable change in noise/vibe between D, 1 and 2 (went ahead and checked 2 while I was at it).
I also hooked up my scanner during the lengthy test (about 15-20 minutes) and watched the live data numbers, though they may not be much help here.
I've been having a sporadic rough/rolling idle problem since before the [presumed] tranny issue. The PI intake swap had no effect on the bad idle either way, and I had already changed vac lines before the PI intake swap. The MAF and filter were recently checked and cleaned, as was the IAC. I'm at a loss at to what's causing the bad idle (unless it's the tranny).
When idle is normal/smooth:
RPM in P or N averages 800
in R it averages 760
in D, 1, 2 it averages 575
This accounts for the noise being less noticeable when in R, since the higher the RPM's, the less noise and vibration there is. I very slowly increased RPM and could hear the knock/rattle increase in speed (not decibels, though). Once I reached about 775-780, it was no longer noticeable.
When idle rolls/roughens:
RPM in P or N averages 650-750
in R it averages 600-700
in D, 1, 2 it averages 525-600
It was difficult to keep track of this, even using the graphing function of the scanner, so those figures are guesstimates.
When high fan kicks on and during its operation:
RPM in P or N averages 700
in R it averages 665
in D, 1, 2 it averages 540
These numbers fluctuate and symptoms are worst when the rolling idle starts while the fan is on, since both dramatically drop RPM.
Ignition advance jumps to around +29.5° upon change from any tranny mode to any other mode. It stabilizes after about 8-10 seconds at around +18°.
Fuel trims dance around, but I didn't watch them closely or log them -- I can, if it matters for the tranny issue.
During the test, I punched the accelerator (not to WOT) a number of times in R and D, 1, 2, with brake applied/stationary, and the noise stopped (all modes). Unfortunately, it returned a little later during the test period.
Whatever is happening, it's obviously not permanent (yet).
The other day, I went for an errand and when I returned home, the noise was present in P as well.
Later on the noise stopped in P and was again present only in R and D, 1, 2 (worse in D, 1, 2).