I have a Shinohara/Walthers #7 or 8 turnout located near the base of a 2% grade. The diverging route goes to a siding. Over the years, I have rebuilt the switch to look more prototypical. Got rid of the hinge and the metal bridle at the points, and soldered the point rails into a continuous rail to the frog. Recently, one of my solder joints at the points broke loose and I resoldered it to match the NMRA gauge for flangeway and track gauge. It looks a lot nicer.
However, now I find that my Kato RS-2 derails coming downgrade on the mainline route through the switch. Oddly, the rear axle of the front truck is the first to derail. It somehow picks the point and attempts to ride the point rail into the siding.
Complicating the issue is the fact that I discovered the track bed tips outward somewhat. I haven’t checked the wheel gauge on the loco yet, and I will do that before doing anything else.
I AM getting around to the question: Do I regauge the switch point rail to leave more of a flangeway (although not prototypically correct) or should I try to pry the track and the switch up to shim it to at least level, or even to possibly superelevate it a bit? I think I’ve read that it is not a good idea to superelevate a switch but I’m not sure that is correct. I’d appreciate all help and opinions.
The subroadbed is 3/4" plywood, cut cookie cutter style, with WS foam roadbed. I tried unsuccessfully to lever and pry the subroadbed risers to force the subroadbed to a level position but the plywood resisted any such attempt.
I would try and bring up the section of the switch that seems to be causing the problem. You might want to check it first using a 6 inch steel ruler cross-wise on the turnout and checking all parts first to be sure that is what it is (low).
If the loco is picking the point, you could try and bevel the top end of the point rail where it contacts the stock rail. This is a common practice both on the real RR and the model RR. I have used N scale Shinohara turnouts on my last layout and filing a small bevel on the end of the points worked very well. The bevel should be at a 45 degree angle and go DOWN and IN to the stock rail. You should put something between the open point rail and stock rail to help hold the point in place while you work on it with a jewelers file. You don’t need to take off very much, just get the heavy part of the rail head beveled down some.
After all of your improvements, the defect that has caused the change you object to is the planar geometry of the turnout as if viewed from the side, not it’s conformity to gauge issues. So, as the previous gentlemen has offered, I would also favour shimming or taking it out and sanding to a more perfect curvi-planar supporting surface.
Shim first, several times if necessary to give it a good try, and then rip 'er out and do the job properly from below if that doesn’t work.
As a last resort, insert a guard rail just in advance of the point to force flanges closer to the stock rail. I have done this in hidden track (belt and suspender thinking) and wouldn’t hesitate to do so where the track is visible.
I’ve got a number of Peco curved turnouts. I had trouble with one of them, but the problems (plural) were ultimately traced to less-than-perfect trackwork, not the turnout itself. So, I’d also advise shimming it up until it’s flat.
Also, have you checked out the engine carefully? It may be that the rear truck is binding somewhere, and doesn’t pivot freely as it goes through turns. Most of the time, this isn’t enough of an issue, but in this case it’s exerting just enough lateral force to push the wheels up on the points. One quick test would be to reverse the engine and see if the same truck or the other one has problems.
I think the first thing I would check (similar to what tomikawaTT said) would be the outside guard rail to make certain it is pulling the wheels far enough away from the frog. If not put a little shim there.
Check that the turnout is as level as possible at the points and a short section either side of it. Also check the truck swing and wobble on the engine, feel for any light clicking feels or sluggishness.
For the points, check the gauge again, prototype points do not cut into the running rail but lay against the running rail. Yo
The outside rail of a curve must be level or higher than the inside rail. If not, there will be problems. This holds true all the way up to 1:1 scale. Super elevation shouldn’t hurt anything. Widening the gauge in the turn a bit may also help, Maybe around a scale inch or so.
Once I derailed at a switch due to improper cross level. 1:1 scale doesn’t rerail nearly as easily as 1:87. [:O]
I’ve had similar problems. Check that all the components of the switch are in gauge and that they are level. A six inch file run across the whole switch works wonders as it gets everything at the same level. Usually only a few strokes are needed.
I suspect you have bent your rail, (2) voided your warantee, and (3) will have continuous problems, and have to replace it when you find this out.
Curved turnouts are tricky, require a LEVEL grade and ‘chamferred’ points. Elevation changes also need ‘transition’.
Shinohara turnouts reinforce the points where pressure is exerted against the stock rails, for added reliabiity. (You removed this). Next time try a kit.