Finally have time to get back to y'all. Work keeps getting in the way.
We took the loco apart and used a toothpick to put a wee bit of oil on the upper bearing; no change. After reassembly the engine still squeaked (both with and without a consist and tender).
Before reassembly we ran leads to the engine and powered it up; while it was possible to hear the smoke generator run, it was minimal and not the source of the squeak.
Next, Just to be sure we left nothing undone, we removed the front truck; the engine still squeaked.
To see if load on the locomotive made a difference we added 4 NYC heavyweight cars to the Polar Express consist. At first it didn't seem to make much difference, but after a few minutes of running at the 3rd level in Legacy it began to squeak a little stronger.
The plot thickens.
So far, we are a team of two IT specialists, a carpenter, a Catholic priest, and a Lockheed-Martin engineer.
The IT guys said, "Turn it off and then back on again to reboot it."
The priest prayed.
The carpenter tapped it with a screwdriver.
The engineer said to "always engineer for design, and then customize to fit", whatever that means. Then he said he was going back to something less challenging, like a F-35 jet fighter.
Finally, another friend, who has never owned or operated an electric train, asked if it could be the track. We scoffed.
Then I set about cleaning the track with electrical tuner cleaner. Hmm! It made a noticeable difference. Still squeaking but nowhere near as much. So I moved it to another loop of track (we have 3 racetracks of Fastrack, about 24' x 15'). The noise almost completely disappeared. If you weren't listening for it, you might not notice it over the noise of the train and cars running.
I put a scale Berkshire on the first track to see if another locomotive would have the same issue, but it behaved itself admirably.
So now we wonder if there be some minuscule gauge inaccuracy in the drive wheel flanges that made them wear enough to begin rubbing on the inner edge of the rail. It doesn't seem logical because the noise doesn't change while the engine is going around a curve. I am going to order a gauge to check that, but for now am just about ready to let her run till something breaks. As I write this it occurs to me to test once more while holding it upside down and see if one or more drivers are wobbling. I didn't notice anything of that sort earlier, but it might be very slight and only having an effect after a couple hundred hours of operation.
Now we wait for the gauge.
Happy New Year everyone!