What constitutes "bullet proof" track?

[Rossford Yard]

Serious question.

Wyatt (Soo) and I finished the staging level track laying and wiring over the weekend.  I have been testing it before we start the upper level (the real layout, with scenery, etc.)

We started running a Trainman GP 15 and a few cars, through the yard ladders forwards and back, at 2/3 and full max thottle speed.  Cars include a light, 89' flat right behind the loco, some other light MT flats, an MT box, some MT tankers and I think an Atlas box car.  Mix of MT and Accumate couplers.

This morning, I changed out to a six axle loco (IM SD45-2 (BNSF paint, if that affects operations )) and repeated.   Tonight, I think I will try a Kato SD 70 since they have the lower profile wheels.

I have filed, adjusted etc. any problem spots as I found them.

That said, this morning, that consist could back through one turnout at 2/3 throttle speed, but the 89 foot flat derailed when backing through the same turnout at full throttle.  I did file it a bit, and it went through backwards at full throttle five out of six times.

For you experienced modelers, is that bulletproof, or do I keep filing until I run it backwards, full throttle, ten times or more?

[ljudice]

My usual test involves pushing or more accurately flinging cars through places at high speed, I think above what any Kato or Atlas loco will do.  Long cars are a good test, but I'd also try a mix of truck and bodymount couplers nowadays since they are becoming the norm.

The acid test would be those BLMA 89' flats, although I have found that when loaded they work fine even with their original wheels.  Empty is a different matter.

[Coxy]

There is a lot to get right in order to get smooth, derailment-free track, especially in the smaller scales.

Purely looking from the standpoint of the track, it must provide a smooth path for the wheelsets to follow. The rails must be in gauge, have no kinks, no debris stuck to the gauge side of the rail, be without significant or abrupt vertical curves. Turnout points need to nestle snugly against stock rails and be beveled such that any flange that strikes them is smoothly deflected along the gauge face of the rail. Frogs and flangeways need to be within NMRA spec and free of debris. Basically, the flange surface of each rail for all alignments needs to be smooth and flowing so that any flanged wheel rolling along them will have no cause to ride over the rail or drop off the rail head.

The above can be accomplished through inspection, selection and installation of quality track components, followed by filing, shimming and bending as necessary much as you described.

The track is important but the cars, trucks, wheelsets, weight and couplers all play a part as well. It is tempting to continue adjusting the track until all of your cars track smoothly but this may not work out as planned. If you have cars that are out of spec (too light, tight trucks, out of gauge, problematic couplers etc) you will be tuning your track to support those bad actors.

Better to get a representative collection of cars and locomotives and to thoroughly check them and tune them first before using them as test subjects to verify your track. You should also think carefully about whether it makes sense to convert your fleet to body mounted couplers. This will improve backing performance but may take a lot of time and effort depending on the state of your fleet. Low profile flanges look better but may be more inclined to derail than deeper flanges under heavy load conditions.

Your intended operations play a part as well. If backing up grades or round tight curves or through the s-curves of a turnout ladder, or you are wanting to run particularly long heavy trains in general as part of your normal operations, you may need to modify your rolling stock to be heavier than NMRA standards to get things to run smoothly. The faster you want to run things, the more critical all of the above-mentioned factors.

89-foot flats are often lighter than they should be. Check the wheels for gauge and debris, check the trucks for tightness, check the car's weight and check the couplers for freedom of motion as the car moves through the s curves.

Zero derailments is a journey as well as an objective. As you learn more about how your rolling stock behaves on your track, you will be able to progressively eliminate problems. It's critical though, to make sure you isolate the problem to the car or the track. In simple terms, if the same car derails in multiple locations, check the car. If multiple cars derail at the same location, check the track. And always bear in mind that while the problem is likely with one, it still could be with the other.

Hope that helps

[seusscaboose]

Back a freight train through it... That's my definition.
EP

[mmagliaro]

Back a LONG freight through it, not necessarily at full speed, but at "medium" speed, repeatedly.  Then
back it through at lower speeds, a MIX of speeds.  I have had derailment issues that only show up at lower speeds,
not higher, believe it or not.

AND... forward and backward, a long-wheelbase steam loco with a string of passenger cars.   That will expose a whole
new host of imperfections in your trackwork.  Maybe you don't run steam, or passenger cars, but the flaws are there
nonetheless, and will eventually show up somehow.  If you have 89' freight cars, I guess that's a good substitute
for the passenger cars.

My test goes in this order:

1. Kato F unit, by itself, forward and backward.  Those things can roll over a mine field, so this rarely exposes any problems
other than a wiring mistake

2. Kato F unit with string of 40' freight cars (about 10) forward and backward at medium speed

3. Same engine with 20 cars.

4. Short steam: Kato Mikado + 20 freight cars

5. Now it gets serious.  The PRR T-1, the SP&S 4-8-4, PRR I-1.   Those things are nasty.   Long rigid wheelbases, with
drivers that basically say, "Either you keep those rails in gauge everywhere, and make every point and frog proper and smooth,
or I will climb right up and out of those rails."

Backing long strings of cars is a great way to expose issues.  Little spots where a wheel picks a joint, or a curve isn't
quite uniform, or there's a dip or spot where the track has a twist side-to-side, show up when you are shoving
lots of weigth in reverse.

[Rossford Yard]

Thanks for the replies, and yes, they all give me a better idea of what I need to do.  Sounds like I am basically on the right track, no pun, and will refine my tests for a few weeks to see how more cars and locos do in tough, backing and forward conditons.

BTW, I doubt many trains will ever back through my staging.  On a 10 x 24 U shaped layout, I opted not to do continuous run on the staging level.  Trains will pull in any of 20 tracks (10 on either side of the layout) and the loco will pull ahead to the tail track and run around to pull it back towards the helix and then upper level.

[John]

for me - any track that will run BLMA wheels is bullet proof .. 

[Rossford Yard]

I have been using some BLMA spline cars as part of the mix.  Makes me feel better that others have major problems with their running, too.

[Hiroe]

5. Now it gets serious.  The PRR T-1, the SP&S 4-8-4, PRR I-1.   Those things are nasty.   Long rigid wheelbases, with
drivers that basically say, "Either you keep those rails in gauge everywhere, and make every point and frog proper and smooth,
or I will climb right up and out of those rails."

THIS. I've found that once a piece of track has been adjusted for the long rigid wheelbase locomotives, pretty much everything else will run flawlessly.

[pnolan48]

THIS. I've found that once a piece of track has been adjusted for the long rigid wheelbase locomotives, pretty much everything else will run flawlessly.

Almost flawlessly. I can almost always find a combination of cars that will derail, such as ore jennies hooked up to 89' lightweight flatcars or some similarly absurd combo. If I don't bother trying to find that combination, my trains have no problems once the long rigid wheelbase locos are accommodated. Although backing a long train up a grade requires a lot of delicacy and even more prayers.

[jereising]

John's on the right track - pun intended - the BLMA wheels plus long wheelbase cars will show you where your trouble spots are - but make sure the trucks are free with some good tilting movement or you'll go nuts...I would make the test train up using Kato Maxi IVs, BLMA RTTX flats, BLMA spines, and BLMA reefers, all with the original first run BLMA wheels.  If that train passes at various speeds, you're golden.

[LV LOU]

I don't think any trackwork,no matter how well done,will run 100% all the time with all car combinations,it's just impossible.Even cars that are perfectly in gauge aren't going to always run perfectly depending on what they're hooked up too,even track design can cause problems.To me,bulletproof track is track you put down so well,ten years from now,you haven't done much more than clean it,and occasionally replace that gone south switch or two..And,I'm pretty much there,my last RR ran ten years without a single track issue other than some badly designed Atlas C80 switches having the frogs go dead,and the lower level on my new RR was done 4 years ago,haven't done a thing to it,runs like a watch,all ME C70..

[SkipGear]

From what I understand, open up a book looking for the definition of "bullet proof track" and you will find a picture of Todd Treasters layout.

I think it is very possible to make track 99% error free if not 100% if you take the same level of attention to both cars and track so they work together. Perfect cars will not go over bad track and bad cars will not go over perfect track.

[mmagliaro]

A big part of this we've been overlooking is the actual design of the track plan itself.
(and this is one reason that Todd Treaster's layout runs so well, I think).

Do you come out of a curve right into a switch with no straight section in between?  Any nasty "S" curves?
Steep grades?  How about curve radii?   Turnout on a curve on a hill?

Not that any of these things can't be used, or that they never occur in real life.
My point is, that by avoiding these things, or moderating them, you can make the track more "forgiving".

For example, the only turnout I had a problem with in my current layout construction was one where the train descends
a long, straight hill, goes into a right hand curve for 90 degrees or so, and then hits a turnout.   Everything is in gauge,
and the points are closed tight, tips filed.  But no matter.  When a long rigid steam loco comes down that hill and
rounds that curve with a train behind it, it is pushing hard on that outer rail and just begging for any excuse to
climb over it.  It would come to the points of that switch, and beveled point tip or not, some of those engines
would climb right up the inside wall of the point rail and derail.  (I could watch them do it!)
I added some guard rails (rails inside the main rails) just before the switch, to help guide the engine and keep it from slamming
that outer rail so much at the bottom of the hill.

But really, if I had a section of straight track before that switch, even 6" might have been enough, then the whole
problem could have been avoided.

So it's not just precise trackwork and properly gauged wheels.
The trackplan itself has a lot to do with how reliable your track is going to be.

[rogergperkins]

Skipgear, you have expressed my view.  The best track cannot compensate from cars and engines that have problems.
As a big MTL fan, adjustment of trip pins is still one of the main problems I encounter.
In fact, I was running a Life-Like E-6A/B unit yesterday; it came to an abrupt halt!!  Found that a trip pin had dropped out of a car that passed that area of track previously.   Maintainence.....
Even when every element is adjusted, one still needs to maintain the trackwork.
Real railroad in my hometown had a section crew out daily during the week for that very purpose.