After going back and re-reading this entire thread, I have confused myself. Back on November 11, 2018 while posting about the day the dealer came I said that "I saw the rings before they were cleaned, and both the rings and that whole end of the alternator were covered with a heavy black film, but the rings were otherwise intact and not worn or deformed." Boy I have NO recollection of seeing the rings, and I really wonder if what I looked at back then was even the rings. Two moths later I posted that I did NOT see the rings, but that was all the way back in April 2018 and I just can't swear what I saw at this point, so I am going to give the dealer the benefit of the doubt.
Last week, while waiting for help trying to figure out how to remove the fan, I had time to kill and I put my ohm meter back on wires 0 & 4 at the voltage regulator to see where it stood before I touched the brushes. It read 21 ohms at first, but then I realized that with the top cover now off of the fan I could easily spin the generator around and see if the ohms varied. I was stunned to see ohm readings jump as high as 110 ohms at some points in the rotation. It probably was about equally divided between reading 21 ohms and then reading 110 ohms, and I got these readings consistently over multiple cycles of the fan. I was turning it just a tiny bit then stopping to see the reading, then I would turn it again just a tiny bit and read it again, probably watching it for 15 minutes while I had free time. Seeing this made me feel like I was chasing the right rabbit, that the brushes or the rings were my root cause. After my final reassembly I checked it again and it stayed right at 21 ohms all the way around!
Back in January 78buckshot asked me if the dealer used washers to shim the brushes. After now removing them myself I can say that there were no washers installed. I am not sure that washers would have been needed since I think washers are used to correct brushes not riding on the rings on-center, and my pictures clearly show the black on the rings was dead-centered.
I wish I had re-read this thread before the tear-down, I completely missed 78buckshot talking about how the dealer should have measured ring to ring AND ring to ground resistance. I could have easily done that during one of my THREE teardowns! I am curious though, what would those readings tell me about my generator?
While I was sanding the rings I thought to myself "you know, this could possibly get these things out of round"..... but I just kept sanding thinking I had nothing to loose. I am not sure what Wirefoam is describing as his technique for avoiding the pits. My technique was I cut the sandpaper into 12" long and probably 1.5" wide pieces and I sanded the rings in halves, where I would sand the top half, then turn the paper in quarter-turns to try to keep a continuous amount of pressure to the rings as I traveled around them. I probably sanded them a little too much though, I have a tendency to over-tighten, over-sand, and over-paint stuff! I just hope I didn't get them too much out of round.
After tearing it all back apart to repair my broken rubber foot I borrowed a torque wrench for the reassembly. I was surprised at how tight these readings were, the torque wrench had me tightening them tighter than I would have tightened them on my own.
While tearing it down this last time I found where I had left a bolt out, and this missing bolt would not have contributed to my "thump", so I am not sure what was going on there. After my most recent reassembly the thump is MUCH improved, with only a tiny bit of contact with these two metal sheets over the muffler.
I want to thank everyone who contributed to this thread and read through my manifestos. This was a job well beyond my pay grade, but if in 12 months this thread remains cold it will be worth all the hard lessons learned!