Really? You break in a locomotive, but tracks? – Let me explain. My CNJ Bronx Terminal has 28 turnouts and 7 crossovers. When I built the individual pieces I always tested a single two- or three-way turnout and when an entire section was complete I’d also test the finished section. But those tests were never exhaustive: I always used the same Micro-Trains 70-ton truck and one of my test engines. When mounted to the layout, additional axial or shear forces would alter the geometry of tracks; furthermore, without the servo and throw bar actuator mounted, the turnout cannot be fully tested. As a result, all turnouts somehow worked, but they were far from perfect or robust enough for an operational layout.
Many of my turnouts I had to fine-tune so that they would work flawlessly. Making sure that a turnout meets the NMRA specs by checking the gauge with the track gage is a good starting point. In N scale, a tenth of a millimeter (.004″) can make the difference between a smoothly working turnout or a turnout where locomotives and cars consistently derail. If that happens, the process is always the same. Understand the patterns why a loco or car would derail and exactly where it does, find the root cause, and once found, fix the problem. Testing with additional locomotives and freight cars of different lengths, from different manufacturers, and with slightly different tolerances for trucks, wheel gauges and couplers (there are many ways for rolling stock to comply with the NMRA Standards!) increases the robustness of the track-work. To better understand where things went wrong I’d sometimes analyze slow-motion video taken on my smartphone.
Common Issues with Turnouts
- Switch-point rails that are not flush with the stock rails (1 in photo below). This was and still is a particular problem when a locomotive with cars negotiates the outer circular track. Both the centrifugal force in effect when running on the circular track and the shear force caused by the following freight cars tend to push the wheels against the outer rail. A slight gap between point and stock rail will cause the loco or cars to jump the turnout. In most cases the cure is to file the (outer) point rail so that it would sit on the stock rail without a gap.
- A similar issue occurs when the track gauge is too tight where point rails rest on the stock rail (even though the track’s gauge conforms to the NMRA standard at this point). This happened to me almost systematically and it was much harder to fix. The only cure I found was to make sure that the foot of the stock rail was completely removed and the point rail was filed down to almost a razor sharp edge. Joggling the diverging rail just is not an option.
- In cases where the point rail’s curvature is too tight (2), the bulge about 1/2″ inch away from the throw bar prevents the point from properly closing. To remedy, the point rail needs to be slightly straightened so that the bulge does not touch the stock rail.
- Another issue that is much harder to detect is a point rail that is too close to the stock rail when open (3). Locomotives or cars would then derail when the have a slightly less than standard gauge or when they have thick wheel flanges. In those cases wheels will be ejected and the loco or car derails; or, the first or second truck will hook up on the open point rail and jump the turnout. Reducing the distance between the two point rails or slightly adjust the curvature of the point rail tp increase the gap between point and stock rail would help.
- In some cases, rolling stock would hook up on the frog and derail. This usually happens when either the wheels are beyond NMRA tolerance or the guard rail would fail to do its job in properly guiding the wheel and avoid it from snagging the frog point. Re-adjusting the guard rail and reducing the gap to the stock rail will fix the issue.
Other improvements included adjustments to the placement of turnout servo brackets when they were not properly centered. This can help increasing the pressure of the switch-point against the stock rail and so remove a tiny gap between the two. Sometimes, increasing the strength of the servo actuator by slightly extending the range would also help. A last option would be to change the bracket type from horizontal to vertical servo shaft. With the vertical shaft, the servo actuator travels a longer distance which translates into a better torque at the throw bar.
Tools Used
For a while, my most used tools (see below) were the soldering iron, the Dremel rotary tool with a cutting/sanding wheel, and files (with my favorite, a nail file for less than $2). When analyzing certain issues I got some valuable insight by taking a video in slo-mo with my smartphone.
After a while and some frustrating experiences, it became obvious to me that not every piece of rolling stock – locomotive or freight car – would be operationally suitable for the layout. Fortunately, the prototype defines a very narrow scope, so precluding six-axle locomotives is not a problem. Some of the weaknesses could be eliminated by adding weights to a car, by replacing the wheels with a type I know that will always work (my favorites are FVM 33″ wide metal wheels #3310 or #3311) or even by changing the coupler. Still, there are cases where a very prototypical design of the model make them unsuitable – and that hurts. Some of my most beautiful and most detailed freight cars with body-mounted couplers will always be prone to derailing and will never make it through my CNJ Bronx Terminal certification program.
Finally, after a lot of hours spent on testing, problem analysis and fixing, things were running smoother. Trains made it from one end to the other without jumping turnouts. And once that this BNSF GP60M from Fox Valley Models was able to negotiate the outer circular track and get over the diamond crossover without derailing, I knew that the track-work had a good level of robustness. And I’ll promise you won’t ever see this loco again on the CNJ Bronx Terminal layout.