For argument’s sake, let’s say we are using good Kadee couplers and have a layout with modest 2% ups and downs and 40" radius curves. There are other variables I am probably not including, so how many cars as an average do you think an HO train could have before suffering coupler breaks under the weight? Whether the entire train is on the 2% grade or on the flat should make a difference. My tiny layout is too small to conduct an experiment of this magnitude, so, what say you? What would the variables be on a HO scale layout? I am betting someone out there has actually done this test at some point.
1 Like
Sheldon!!!
Calling all forum members named Sheldon.
Rich
Sheldon would certainly have a handle on that, or make an educated guess. Wayne as well, Mark Pruitt perhaps, the late Chuck Beckman, Big John would surely have some empirical first-hand knowledge.
Cuda Ken would be another. Those guys push the envelope to see what’s what. [:P]
I seriously think, Brent, that you’d get stringlining before a Kadee broke, or it would be a defective Kadee. But if we controlled all the variables and relied on good Kadees, no curves, 2% grade, we would have to be into the 150-200 car range, and even then I think the pin wells inside the coupler box would go first.
My thoughts exactly, but I did want to give you a chance Rich. Buck passer.[(-D]
1 Like
I think you may be right, so a straight piece of track may have to be a constant.
1 Like
The longest train that I’ve run on my layout was 71 cars, none of them weighing less that 4 oz.
It was done as a test on my longest grade (about 45’ long and on a 2.9% grade), basically to determine how many locos were needed. There were two opposite horseshoe-type curves (34" radii) followed by a couple of “S”-bends of 40"+ radii.
The main wrench-in-the-works was that the grade ended at the point where the planned partial upper level should have been, but it was not yet in-place.
The three photos below show pretty-well the entire grade…
(The train shown in the second photo is not the test train, but simply one loco taking a fairly long train down the hill.)
I also did a test with a heavy train of 48 hopper cars loaded with Black Beauty blasting medium. It was done on the lower track in the first photo.
The first attempt was done with one Athearn U-boat, modified with two can motors and a weight of 33oz.
It couldn’t move the train up the entire grade, but when a second identical loco was added, went up the grade (all of it on an “S”-bend) easily. I would have liked to try a longer train, but had no more Black Beauty on-hand.
Total weight of the loaded hoppers was 24lbs - no derailments, no issue with couplers, either.
The three U-Boats later went to a friend in the U.S., and I don’t have any other locos with their pulling capabilities.
On the 71 car train, I used a variety of locos, usually 6, steam and/or d
I have pretty much the minimums specified above. Code 83 Shinohara trackwork, visible curves are superelevated. My curves are 32" minimum on the main lines and crossovers are #10.
While not routinely I have run 90 car freights. I presently have a 22 car passenger train made up of mostly BLI PRR P70 coaches but several Walthers Proto B60b and BM70 baggage and RPO cars. Couplers have never been a problem.
Sometimes I let trains like this run “round-n-round” while occupied with other layout chores. still, this isn’t the norm and even prototype passenger trains were kept to about 15-18 cars due to passenger platform lengths and steam heat constraints. Most often my freights are in the range of 45-50 cars.
Why would anyone think Sheldon is the only one who likes to run long trains?
Good Luck, Ed
You only need to test the breaking strain of one #5 coupler with a spring scale or measured weights. There must be hundreds of thousands of Kadee #5 no longer in use now Kadee has issued the #148. Divide that number by the rolling resistance of a single car pulled up the grade with the spring scale. That gives you the total number of cars required to break one coupler. Only one will break in any given train, likely at the head end as for prototype, leaving aside multiple mid or end of train power units.
My guess is the mounting to the car would yield first.
The weakest point on a Kadee is the knuckle pivot and the way it takes force means the strain on that pivot is also taken in part by the casting of the rest of the knuckle, just looking at it. Pretty robust design imho.
It really is not the coupler to worry about. My old club had a very long grade and I used to haul 70 or more cars up the grade. Until one day, an Athearn hopper decided to shed it’s coupler cover and fifty five hoppers and a cabin went screaming down the hill in reverse. Let’s just say the crash beams on the cabin did not hold up to the collision with the hill and three hoppers hit the floor. Now all my couplers are secured by a brass or steel fastener.
In the early 1900s the PRR tried out their class CC1. A compound Mallet type of a 0-8-8-0 wheel arrangement. The problem they had was the tractive effort was too much for the rolling stock. If the draft gear didn’t tear out, the whole car would be ripped apart. It finished it’s life in pusher service.
Pete.
I seam to remember somebody testing the strength and came up with an impessive 11.5 lbs or so.
Shane
It is interesting to note that coupler strength is indeed the limiting factor. This is the reason that the guys from Miniatur Wunderland in their attempt to pull a real locomotive with HO scale locomotives did not rely on couplers, but had harnesses for the model locomotives. Most of this video
https://www.youtube.com/watch?v=vC-dmKHiCnY
is about their first attempt, using 198 Roco HO model locomotives to pull a prototype DB 101 class electric locomotive. Just as an afterthought, they mention that a day later they managed to do the same with just 85 model locomotives. A 101 class loco weighs 83 tons, meaning that each HO locomotive was pulling about a ton. Very impressive!
This is an interesting question. i am an engineer so I’d like to do some calculations. Since Kadee use metals for their coupler, I assume it is steel. For a coupler, we only need to worry about the weakest link, or the place where the coupler has the samllest cross-section area. I assume it is 2mm or 0.08 inch in diameter. Considering the modulus and yield strain (the strain before plastic deformation), you need 250 lb or 4000 oz to yield the weakest link. If the force needing to move a car on a 2% grade is 4 oz (not the weigh of the car), you can move a least 1000 cars… A huge number…
I seem to recall a club running over 500 coal hoppers in a train. It had mid train and pusher helper’.
Rick Jesionowski
The trip pin is a soft steel. The coupler shank and knuckle are pot metal possibly zamack. You can clearly see parting lines and injector pin marks on the number 5 coupler. That is why KD wants you to burnish the shanks for smooth operation. ( How many modelers do that? )
Years ago, I tried the plastic McHenry couplers and quickly switched back to the KD. I never had a failure. But you can clearly see them flex even with a minor load.
Pete.
Just to remind everyone that the rails (and roadbed) hold up the weight. The coupler, and indeed the locomotive moving the train, only overcome the rolling resistance not the total weight of the train.
Indeed, the whole secret to the load carrying capacity of railroads is the low deflection of the rail and roadbed.
Once the initial static friction is overcome, rolling resistance becomes quite low and not meaningfully related to train weight.
Any force will accelerate any mass absent friction. Eventually you’ll notice the mass moving…
Ed!!!
Calling all forum members named Ed.
Rich
On my layout, I ran an 81 car freight train. 4 engines on the lead, 2 Mid-DPU on level grade. Radius varied from 24" - 30" and no issues and a decent speed. All DCC and engines are speed matched for the most part. Cars are weighted and metal wheels on all cars.
If your couplers are properly installed and the train is assembled so you don’t have any type of stringlining occurance, you should be able to run very long trains, even up a grade, as long as the motive power adjusts to the grade as well. I have a 2% grade helix and depending on the engines, more than 16 cars on the train requires a helper engine…
Here’s the video of my 81 car train…
Neal
I have a fairly large basement layout. My steepest uphill grade is 1.5% while the downhill grade is 1.75%. I generally limit my freight cars to 20 cars but early on I tried to see if I could run a 50 car freight. The couplers held up with no problem. The operational problem I had is there is a curve near the bottom of the grade and I superelevated it slightly. I found that all that weight tended to pull the cars over the inside rail.
The other problem is a train that long dwarfs the layout. The locos on the front end would pass through a town while the caboose had yet to reach the town behind it.
I’ve since reduced the amount of superelevation so the rails are near level on that curve so I might try the experiment again just to see if I get better results, but 20 car freight will continue to be the max. My freelanced layout is loosely based on the NYO&W and 20-30 car freights were not unusual on that line. That’s probably why it went belly up.
My former club The Soper Valley RR once ran 120 iron ore hoppers on our portable layout. Five Altlas Engines on the front, no extra power. Wide curves and level track.
I have operated 80 car trains on my home layout, which has Kato 26.375" minimum radius curves, and a 0.4% maximum grade, with no Kadee coupler failures.
Hoppers, especially if empty, are light. My trains include big heavy boxcars and autoracks that weigh more than one typical freight car.
Since the couplers are metal, they are not the weak link. Any of the Intermountain or Accurail style kit coupler boxes that are glued in place via a plastic pin into the floor piece are likely to pull out before couplers will fail.
The other issue is that 80 car trains require careful train handling with the throttle, whether you have superelevated curves or not (I have a mix of both). If you have to do an emergency stop (say the cat jumps on track in front of engines) one can experience a dramatically prototypical train wreck, as I have.
However, couplers have not been an issue. Most of my rolling stock has Kadee couplers, with a few ScaleTrains couplers on their rolling stock, and a few off brand metal shank couplers (Kadee clone). Again, I experienced no coupler failures.
One exception: current BLI Kadee clone couplers just fall right apart even with short trains of less than 10 cars–the ones on the new GP-20’s self destruct after only a couple hours of run time and literally fall apart and stop the following cars in the train. Two GP-20’s, two coupler failures in the last week. They were replaced with genuine Kadee couplers.
John