Saturday, November 16, 2019

Trackwork wars, Part 4

It’s been a whole year since I had to report on any “challenges” with my layout trackwork, so no doubt Murphy’s Law suggests I am overdue to report again. And I have indeed been fighting my way through a problem or two. The main current challenge is the same track area and turnout as in the first of the posts on this topic, back in 2016, which you can read at this link: .
     I have rigorously checked the track gauge throughout this Shinohara turnout, which is the lead to Track 7 at the back of my layout town of Ballard, so I believed that there was not a gauge problem. But steam locomotives of any wheelbase length whatever don’t play well with this turnout. I had long suspected that this turnout may not lie entirely in a single plane, as I noted in that previous post (see citation in previous paragraph), and I decided to borrow a friend’s tool for trackwork evaluation, a clear acrylic plastic “floor” with trucks. This has often been described in model magazines, and usually looks like this:

The turnout in question is at the far right of this photo.
     Moving this vehicle through the offending turnout, slowly and carefully in both directions, immediately showed that there is indeed an out-of-plane problem, that is, that the entire turnout does not lie in a single plane, but has one side or the other elevated at different points. That means that if you could make yourself HO scale, lie on the track, and sight along the rails, this turnout would look a little like a potato chip, with first one side and then the other side elevated.
     The best way to evaluate this in detail is to use a small level, which I proceeded to do. At first, I used the track car that you see above, laying the level across it as shown below, and this kind of works.

     But the truck mounting on this track car is not sufficiently rigid to permit dependable and reproducible observations with the level. Instead, the level was used by itself, both on each line of track converging into the turnout, and across all rails of the entire turnout. This quickly showed the “potato chip” problem. I soon learned to make sure that the level is resting on, and only on, the particular pair of rails I was checking.

     After brief reflection, I decided I should also re-check track gauge throughout the turnout area, just to make sure, and it was good that I did, since near the points the gauge was indeed a little wide. An additional track spike quickly fixed this. Here I am using my new NMRA Mark IVb gauge to verify conditions in the affected area.

     With the information in hand about track levels, I used some styrene strip of various thicknesses, and progressively worked through the whole turnout to correct each local area and bring it close to the full level that I wanted. Here is this turnout (the background track is the lead to Jupiter Pump & Compressor). The bridge here is the pile trestle across Oso Flaco Creek (see my post at: ). The shims are white.

     With the track shimmed and close overall to being in a single plane and in gauge, I decided to try and keep it that way. This turnout has had ballast in place, but not glued down, since I had been continuing to work on making this turnout perform better. But now, with things apparently corrected, or at least improved for the time being, I went ahead and used my regular method with dilute matte medium, dripped into ballast wetted with “wet water,” to tie everything down.
     Once everything had dried thoroughly, I powered up the layout and ran a 2-8-0 back and forth through this area. This turnout certainly performed better than before, though still not perfect. And of course I may have to revisit this yet again, one of these days!
     But the real test, as always, is an operating session. The just-concluded operating sessions over the previous weekend (you can read my account here: ) did find this turnout performing pretty well, certainly without some of the problems of earlier days. In the absence of a perfect outcome, I am happy to accept an outcome that is significantly improved.
Tony Thompson


  1. Thank you, Tony, for this post. I have two offensive turnouts and had resigned myself to adding a guide rail to each, but first, I will check the level! They may still need the guide rail, but at least I can eliminate one more variable in the troubleshooting.

  2. You're welcome, Galen. I only arrived at checking the level because I'd thought the gauge was fine (actually, it wasn't quite), and decided to investigate the level. One of my locomotives did rock a bit going over the turnout, which I thought might be tight gauge, but in fact was finally found to be the level. Good luck with YOUR problem!
    Tony Thompson