Electrical gremlins

In part two of the saga of transferring my son's layout to Burlington, I discover that gremlins are real . . .

To recap (or you can read the original post): I took the original 4'x6' section of Victor's HO scale model train layout down to Burlington to install in my sister's basement. Part one went very well, the layout (aside from some scenery) arrived and co-operated when set up in the new space. Nephew's state: very happy. Some time passed, during which I was going to check that the second section, a 2'x4' module containing the turntable, was still operable.

That was delayed an extra couple of weeks, as I couldn't find my multimeter, and there clearly was an electrical issue with the section: no power appeared at track level when applied to the under-table wiring. This is where it got complicated: my nephew had been eagerly awaiting the arrival of the turntable section and was very disappointed when it failed to show up on schedule. And I couldn't even diagnose what the problem was without having a meter to find out where the power was going (I had a brief urge to tell him that there was now a puddle of electrons on the floor underneath the layout, but common sense prevailed).

My meter has gone into hiding, and has not yet turned up, so I borrowed a meter from Jon (my virtual landlord, who, by happenstance, had just bought a new one . . . because his original multimeter had gone into hiding a few weeks earlier). Using Jon's new meter, I discovered the following:

• a loose connection to the turntable bridge track itself — a screw connector, easy to remedy
• a loose connection to one of the six turntable stub tracks — also a screw connection, fixed very easily
• otherwise excellent connectivity on all the rest of the track surrounding the turntable.

The turntable is one of the Atlas 9" models, and the switches controlling power to the stub tracks are all Atlas slide switches. In spite of the abuse they'd received during multiple moves, the turntable still works (manual, not powered), and the slide switches all still work fine. What doesn't add up is that tracks 2, 4, and 6 (counting clockwise) don't run. That is, the multimeter reads the same voltage on tracks 1 through 6, but a locomotive is only able to actually move if it's sitting on the turntable itself or on tracks 1, 3, or 5. There are no breaks in either the wiring or the actual physical rail to account for this. It's the same locomotive (a Kato NW2 — Kato has a very good reputation for product quality), yet it runs happily on 8 volts DC on one track, but refuses to acknowledge the same voltage on the adjacent track.

Gremlins are the only possible explanation.

Still, when my sister's family arrives later today, I'll at least be able to give my nephew a partially working turntable section, which is better than nothing. :-/

Update, two hours later: Apparently, Gremlins have nationality. . .

In first part of this post, I mentioned my frustrations with trying to troubleshoot the wiring. What I forgot to mention, because I'd completely forgotten about it, was that not all the track was Atlas. Some of the track was from one or another of my son's various train sets . . . no brass, but some steel and some nickel silver. It's all been painted and given the beginnings of weathering (but no ballast), so it wasn't immediately obvious which sections were Atlas code 100 (made with nickel silver) and which were Brand X code 96.733333 (made from the bones of imprisoned dissidents).

When I belatedly recalled that, it was a simple matter to swap out the (Japanese-made) Kato locomotive for a (Chinese-made) Bachmann. The original run of the Spectrum GE 44 ton diesel, to be precise. It had no problem running on any of the track . . . in other words the problem wasn't the track or the wiring: it was the frickin' locomotive. Kato NW2's don't condescend to run on mere "trainset quality" track. It has to be brand name or better before the Kato will stop sulking and actually run.

So, in addition to the turntable module, my nephew is also getting an old, noisy, but still serviceable GN 44-tonner. At least I know _that_ will run on the "new" section.