OK, so yesterday, I finally got the last section “temporarily” laid so that I can run the train for the first time on permanent track. Here’s my problem. I have 2 areas that seem to be issues with derailing…both seem to be at switch machines. On 1 switch machine, the train will go around 4-5 times, no problem, then a car or 2 will just decide it’s going down the other track and derail? I’ve made sure the switch is all the way to one side…no difference. I’ve got a NMRA guage on order, so hopefully, that will let me know if the tracks spread out or come together and I can’t tell it visually. On the other, the train will roll thru the switch, then I hear that tell tale sound of a wheel being off the track, but the car won’t completely derail and stop like the other situation. So, here are my questions…
Is there any tips to making sure the switch machine is laid properly?
I’m using Atlas #4s. Do they have inherit issues that make people go to others like Peco, etc?
It also happens more often when one particular car is at the rear. When I move it to right behind the loco, it doesn’t seem to mess up. I’m planing on guaging the wheels (once the guage comes in next week) to see if there is an issue there, but is there anything else to check on the rolling stock?
Should rolling stock have a certain weight to them? Seems like my heavier box car rolls through everything like a champ, when these other coal hoppers seem to be the issue. I know you don’t want too much weight, from the loco being over-worked standpoint, but is there a happy medium?
Any help from you gurus will be greatly appreciated, as it ALWAYS is![bow]
The only reason for the car to go down the wrong path is that the flange on the wheel running the wrong route has slipped either over or past/through the point tip. It may be related to faulty gauge on that axle (that should always be your first determination when you get one car derailing intermittently or continuously), or it may be that the point tip is not either at the same height as the stock rail against which it should sit, or the point is not sitting tightly against the stock rail. Often, you can use magnification and good light directly overhead and see how the tip sits when you hold it reasonably tightly against the one rail. The tip may need to be filed a wee bit sharper, or you can actually very gently bend it an entire/whole degree and get it that much more flush with the inner web of the stock rail.
Don’t fret. At least one turnout in a set of five or six will cause problems. They must be laid flat and kept from rocking sideways or bowing, and of course their entrances and exits cannot be kinked very much, say about three degrees tops, from the rails to which they adjoin. For longer steamers, three degrees is too much. If they are properly fixed in place and there are no adjoining alignment issues, then the problem is either gauge of wheelsets or the placement and condition of the points tip.
Good luck.
Crandell
Edit: I really should admit that there is another possibility. The points rails may not be in gauge partway along toward the frog. If they are too tight, they’ll pinch the axle the widest in gauge and cause it to lift out of the rails. The rest should be obvious.
If the light hoppers are Life-Likes, yes, they are seriously underweight. NMRA recommended practice is 1/2 ounce, plus 1/2 ounce for each inch of car length. If all cars are about equally heavy, or light, they track better.
You might want to take a look at the switch points. I remember Atlas having blunt, square-ended points. If that’s the case, take a moment to bevel the gauge side of the point tip until it comes to a sharp point. (I hand lay specialwork, and my points are all filed to a razor’s edge. Running fingers along the rails into a facing-point turnout is NOT recommended.)
If your locomotives and longer cars are happy on those #4.5 Atlas turnouts, count your blessings. Some of mine would be actively unhappy on the tight closure rail radius. My smallest number frog is #5, and my longest car is a couple of inches shorter than a US-prototype Pullman or humongubox. Of course, the biggest problem is the long fixed wheelbase of my EF18 and EF58 2-Co+Co-2 catenary motors.
Once you have it, the NMRA gauge is your best friend. One other thing to check is that the flanges of each truck line up - if they don’t, the flanges will be forced against the railheads, and they WILL find every imperfection.
Just my [2c], based on over half a century of experience.
I think you are referrign to the track switch as a switch machine - well, do you actually HAVE a switch motor of some sort attached to move the points? ANd what brand are they? If you are using Atlas without any sort of switch machine attached, just the traisn runnign through will gradually work the points over until a car or loco picks them and goes the wrong way. There is nothign to hold them without some sort of mechanism, either electrical or mechanical, to hold them in place. There is a quicky way to do this using a pieceof music wire with a small bend inthe middle, plus 2 sharp bends in the ends. One end goes int he hole int he throwbar,t he other end goes in a hole drilled a coupel of ties away. This works like an over-center spring such that the tension is least when the points are to one side or the other, you can manually move them with your finger, but there’s jsut enough spring tension to keep them from moving themselves.
Randy, what diameter wire do you recommend in making the over center springs? Oh, and I DO have a pair of hardened wire shears. I ruined a perfectly good set of diagonals cutting hardened wire with them.
While I haven’t used them as Randy described (I need to power live frogs) I have used bent paper clips for springs with KTM double-coil switch machines. They’re softer than music wire, but plenty springy enough for this application.
In the December 1995 MR project layout they used .025 music wire. In January 2001 there is a bit more detail in how to make these, using .020 wire. Basically, pick a tie approximately 3/4" fromt he throwbar and drill a hole in it. Take a piece of .020 wire, bend it in a U shape just slightly more than the 3/4" (use the actual distance) space, and put a slight kink int he middle of the crossbar. Put the two legs through the holes, one in the throwbar, one in the tie you drilled. Done.
I had some trouble with the Atlas turnouts because the frog was too high. Put a straight edge across the across the frog to see if it’s higher than the rails. In my case I used a track nail along side the frog to draw it down.
Get out your files and like Yankee Flyer said check the height of the frog and points. File the high parts down and exaggerate your stroke to make a uniform cut. I had the same problems your describing. Filing the frog and the points worked for me. File a very little bit and check to see if your problems are gone. Don’t try to get it all at once take your time. I had to put in a new switch because i got a little anxious for results.
Make sure the turnout and all three approach tracks are in the same plane. Use a metal straight edge and make sure there are no vertical kinks. Also, check for horizontal kinks where the switch joins up with the tracks on either side. Use a small spirit level to make sure the track isn’t tilted left-to-right, either, as that puts sideways stress on the wheels and encourages derailments.
It’s been my experience that our turnouts get blamed for poor tracklaying as much as for actually being the problem. Of course, I can only blame that on the poor tracklayer…[banghead]
A lot of times the guard rails have too close of a dimension to the frog. I shim the outside of both guardrails when this happens with a .010 or .015 styrene strip long enough to go past the ends.
Undercutting the stock rails sides for the points to nest into helps. Also the point rails may be too low vs the stock rails and so shimming under the point rails by gluing some shim stock to the ties will fix this.
The forthcoming NMRA gauge may ID the problem right away.
Even the lightest car should not be a problem with properly laid track. First check the joints just before the switch, next (as others said) check the height of the switch parts like the frog and points, next check the gauges. That is if it is a switch problem but I think you will find it in the car and is a bad truck or wheels or something as simple as one truck is too loose or tight, which I find is the most common proublem.