OK, yeah, I know, “bullet proof” track work? Am I sure that my track work is bullet proof? Well, let me put it this way. My primary objective when I built my new layout was to be as careful as possible to avoid track work issues, so no humps or valleys, no kinks, soldered 32" radius curves, almost no turnouts on the mainlines. That carefulness has resulted in a mainline with almost no derailments.
Most of my locos, freight cars and passenger cars run flawlessly and consistently. Yet, a few freight and passenger cars derail on occasion, not often but once is too many. Also, a couple of locos that derail, especially in consists.
What I have found in every instance is that a coupler assembly or truck is the cause of a derailment. For the most part, I have been able to correct the problem. But, in a few cases, I continue to struggle to find the cause and to correct it. My temptation is to just toss these few cars. Any thoughts on how to proceed?
What in the coupler appears to be causing the derailments? Trip pin catching? I don’t use magnets to uncouple so mine are bent up more than the minimum to clear the Kadee gauge. Lifting the trailing car under load? Might just be a bad coupler, with the pulling face not flat, or a mold part line.
Wheels all properly gauged? Trucks screwed on too hard? One end should swivel freely, but not rock, the other should be a tiny bit looser so it can rock slightly. Too tight and the truck can’t swivel to follow curves, too loose and the car will wobble like the Toonerville Trolley.
A simple troubleshooting technique: move the suspect coupler to a car that never derails. If the newly equipped car now derails discard that coupler and install a brand new Kadee #148 complete with brand new Kadee coupler box if the coupler and its box were transferred as a unit.
If the same car continues to derail even when a known good coupler is fitted from the correctiy running car then it isn’t the coupler itself.
Do that swap test for each derailing car.
Track induced derailments tend to occur at the same spot on the track regardless of which car is involved.
I am on my way down to the layout, so I will get back to you with a more definitive response to your questions.
But for now, I am not totally sure what in the coupler assembly or installation is causing derailments. I long ago began a practice of clipping the trip bins to avoid entanglements and also to avoid catching the trip bin on obstacles between the rails such as grade crossings. It may be that the coupler is lifting the trailing car under load.
One issue is close coupling on passenger cars where the diaphragms snag each other, but I have pretty much solved that problem by using long center shank couplers, prototypicality be damned.
Regarding the trucks, I will check wheel gauge on offending equipment. I have never heard of the swivel that you mention. I find this comment interesting, “one end should swivel freely, but not rock, the other should be a tiny bit looser so it can rock slightly”. I will definitely look into that issue.
Truck screw torque is also related to car weight. The wheels on truck that is a tad tight can tend to climb up the rail rather than turn the truck.
Reason for one tighter than the other is the car then rides on three points rather than four. One truck supports against lateral movement (countering the wobble effect) while the other truck accommodates tiny amounts of twist in the rails. The longer the car the more this becomes important because of the compressed scales we operate under. Model railroad track just cannot be as level as prototype.
Sometimes it is manufacture error. It can be so slight that most people would never know. BLI has an NW2 that has a problem with the frame, the hole for the truck is slightly off I think (well documented problem but the hole placement is my best guess). Had a truck one time that was slightly off on a boxcar (easy fix) but to see what was causing it you had to look real close to notice the very slight tilt on one side.
On the Scale Rails of Southwest Florida HO scale boxcar layout, there was one turnout where one of the club-owned diesel locomotives would derail frequently.
No other locomotive derailed at this turnout, and the locomotive did not derail anywhere else. No one could figure it out.
I remember the day well when club president, Roger W., through the locomotive into the canal that ran next to the clubhouse.
Problem solved.
I cannot answer this question other than to say that once I took car weight seriously, I had fewer derailments. Of course other factors were addressed the same time as the weight, so I cannot say what had the greatest impact.
I do not follow the complicated NMRA guidelines… my freight cars are precisely weighted like this:
Too light of a car can derail much easier than one with some weight to it. Doesn’t even have to be up to NMRA recommendations. One of my favorite cars when I was but a wee lad had a “weight” in it that was virtually pointless. It was as light as air, I swear. It was one of my lightest cars and derailed constantly. My dad popped it open and melted a bunch of solder on the weight in the car making it much much heavier. It was like a different car altogether. Derailments with it virtually vanished.
It was the only change made to the car, so I was convinced.
Yes the old GSB SD40-2 had endemic tracking problems and one of the Athearn Blue Box SDs did as well, either the SD45 or the SDP40. The old Athearn Blue Box 40’ quad hopper was notorious for needing to be overweighted to run right - just something about the physics.
I thought my benchwork and trackwork was pretty bulletproof too but had issues, some of which were due to learning that some Peco turnouts have gauge and flangeway issues. Also I bought a small level, circular and about as big around as a half dollar, with the bubble in the center. I moved it around the track and the benchwork and learned there were all sorts of variations. Then on a whim I took a larger level to the basement floor and made even more discoveries! Sometimes perfect bulletproof benchwork and trackwork nonetheless are transmitting upwards oddities about the floor.
I tend to measure down from the ceiling in basements. The underside of the floor above is usually perfectly level. The concrete slab on the other hand probably wasn’t even level when it was newly poured.
I also use a level for installing wall mounted shelves or hanging pictures for similar reasons. Level is level. Measurements are all relative in some sense.
I was amazed, too, at how many areas had dips and high spots that I couldn’t see. Laying a good straight edge on the rail head will also find lows and highs.
I agree. It is also handy for aligning track around a turntable:
I suggested a laser-level in one of the threads here a while back and was pretty much told to go soak my head, phoo-phoo. I backed into the corner and left the REAL modeling up to the experts.
I recently bought a dozen of the Broadway Limited P70 coa
Ed: I think that is the base for the Cornerstone Modern Roundhouse in your picture. If it is, can you let me know the diameter, in approximate inches, of your turntable?
I don’t think that level track, or the absence therof, is my problem. But, it is good to raise the issue for others who are following this thread.
My problems seem to be coupler related and truck related. That said, I still cannot figure out what the problems are. The derailment problems are limited to just a few locos and passenger cars.
I don’t know that ‘bullet-proof’ is achievable. It may be 99.9% there, but in my limited experience there will be one engine or car that will derail somewhere along the system. In fact, it’s just a fact of life for me that when I get a new type of locomotive, I expect it to act ‘homesick’. It balks at something and refuses to budge past it. It takes me anywhere from a few minutes to a whole day to figure it out, but often it’s the tracks, and not the locomotive. I usually find that the outer rail on a curve dips just enough over four or six inches to allow the lead axle’s outer flange to slip sideways and outside the gauge. Or, at the back, there’s the opposite, a rise, and that levers the entire frame high on that side, thus causing the same issue for that lead outer flange.
Which is where laser " levels" come in handy in part because they aren’t just levels.
Prototype surveys relative to mean sea level, which interestingly is not a physical reality but a construct based on an imagined reality. It’s real because we say it’s real, like money is real.
So, if you use a level to set up your layout you are using the ultimate prototypical technique. You won’t need a theodolite though, unless you can’t see that far.
