I got an e-mail from Lionel and my F40PH is being shipped back to me. That's a one week turn around.
My "guess" is a wire getting sanded down by a flywheel on the front motor. I haven't heard the "official" word from Lionel, but that's my best guess not having seen the inside, but having seen three of them fail in the same way. Whatever powers the number boards seems to be involved as the right number board goes in every instance I've seen them fail. Two in the row had a flickering right number board and then "dead". It would also explain some reporting that it was running a bit jerky.
This is all total speculation on my part, it'll be interesting to hear the actual failure reason from Lionel.
Based upon these type of comments I took the shell off my locomotive. There are a couple of harnesses to guide wires to the front of the locomotive and to keep the wires away from the motor flywheel. On my unit a couple of wires worked themselves out of the harnesses, but they were not near the motor flywheel. I secured the wires to make sure they are never free.
The shell wires are separate from the wires of the motherboard. The shell wires connect to motherboard wires by series of 6 or perhaps 8 (I did not count them) plug connectors. Another possibility is that some of these got reversed during the manufacturing. With some AC circuits the wires can be reversed and all seems well for a little while, but then failure.
With some hiccups, my Phase II unit #200 is still going well. The hiccups were rubbing of the coupler wires on the wheels and a broken wire from the pickups. It has about 15 hours of run time.
Most failures that I hear about are Phase III units; thus, it might not be a surprise that this Phase II units is ok.
Needless to say, my experience with these units have been largely positive; especially when considering all the failures many of you have had to deal with. Even so, knowing what I know now would I sill drop $800 on the powered and cabbage units? Probably not. They are nicely detailed and have some cool lighting features and such but also have a cheapness feel to them. They could really use a cast metal chassis and get rid of the flimsy plastic chassis side frames.
Also, why is there 1/8 inch of space between the wheel and the truck side frame? The large gap really detracts from the look of the trucks. 2343 from 1950 has much tighter tolerances and looks much better.
Also of note is that these units go around 048 curves just fine. I have tested every type of rolling stock I have: freight cars, 15" aluminum passenger cars, articulated couplers, fixes couplers, and even the lightest of MPC rolling stock with plastic wheels. Everything navigates the 048 curves and switches just fine without any derailments. It seems that Lionel was overly cautious in calling these units 054 in the catalog.