Coupler Displacement on Vertical Plane Under Load

[tehachapifan]

Here's a phenomenon that I have not been able to get my mind wrapped completely around. I run relatively long and heavy trains on a layout that has around a 2% grade. It is not uncommon for the couplers between the last loco in a lash-up and the first car to begin to displace on the vertical plane under load (one up and one down to an extreme) and, in most cases, eventually uncouple. Strange that this never seems to happen much anywhere else, like between the first and second car, for example, although sometimes the locos nearest to the rear of the lash-up will do this. The other thing that seems odd to me is why couplers would displace like this at all when it would make more sense to me that everything would want to stay in a straight line under load, especially since the displacement gets so extreme. Because of this phenomenon, I can only run a select few locomotives in this position and they are rather random (sometimes only one out of three or four of the same, exact loco will work here, for example). It really doesn't seem like there is excess play in in the couplers either. Anyone else experience this or have a clear understanding of what is going on here?

[bbussey]

Sounds like reverse-draft-angle problems on the inside faces of the couplers.  There must be a slight draft on the faces so the parts eject from the molds.  On KD/MTL couplers back in the early days, the draft was very pronounced so the coupler displacement could occur even with slight load.  MTL fixed that issue by reducing the draft angle on the inside faces of the knuckle component.  But a slight draft may still remain, and there may be draft on the other brands of knuckle couplers as well, so that would be the first area I would look.

[garethashenden]

I think it makes sense that it would be that coupler which has the problems. All the locomotives are pulling forward but all the cars are effectively pulling backwards, so the highest forces will be at that coupler.
If you're not doing a lot of switching you could consider a designated first car, maybe fit an Ultimate coupler. But this may just move the problem to the next car down the train.

[GaryHinshaw]

It's almost certainly a draft angle problem: if the shape of the pulling face is slightly convex, it forces the couplers to displace vertically:



You should be able to tell by inspection if this seems to be a factor, and you can probably gently reshape the face into a concave shape, or reverse draft angle.

[randgust]

you need a really sharp, FRESH #2 or #11 xacto blade, and magnification unless you're as nearsighted as I am.   Trim the 'fingertipss' only of the knuckle so that the taper goes in toward the center of the knuckle, from both the top, and the bottom, in toward the middle.  You don't need to trim much.   You can convert any older MT coupler to RDA that way.

Now if you look at the inside of the knuckle face, you'll see it has that mild outer taper.  So when you fix the 'fingertips', it actually pulls the knuckles CENTERED, instead of vertically, under heavy load.  It's simply magic.

This has been around for a long, long time.  I think it was Jim Fitzgerald that came up with this technique, I know I read about it in an Ntrak bulliten and it changed everything for me.   It's an absolute requirement for long couplers like on the extended frame boxes.   

Note that BOTH coupled cars have to have this feature, or it doesn't work.

[tehachapifan]

Great tips, guys! I'll try it out.

Funny.....I've always heard "reverse draft angle" referred to but never knew what that meant.

[strummer]

Funny.....I've always heard "reverse draft angle" referred to but never knew what that meant.

Me too.

Mark in Oregon

[Sokramiketes]

I bet there is some bending movement of the pilots on plastic locomotives that also exacerbates this problem.

[dave n]

Me too.

Mark in Oregon

Me three .  It's true, you DO learn something every day!!

[PAL_Houston]

It's almost certainly a draft angle problem: if the shape of the pulling face is slightly convex, it forces the couplers to displace vertically:



You should be able to tell by inspection if this seems to be a factor, and you can probably gently reshape the face into a concave shape, or reverse draft angle.

So THAT's what a reverse draft angle coupler is!!  Thanks for posting this, Gary.

[tehachapifan]

OK, I tried trimming the "fingertips" of the coupler into a reverse draft angle (yesterday I didn't know what that was and now I'm using it in a sentence) and....BOOM!....instant fix!

Thanks for the info!! That's really cool!

[randgust]

I've actually broken a couple knuckles before they'd part vertically now, shears them right off at the shank.  Always on an operator-stupid move like dead-stop downhill on a grade in reverse running the slack out on a heavy train.

This solves another problem as well that you may not realize you had - trip pin hang.   Because the couplers now re-center themselves vertically under load instead of even mildly deflecting, the pins don't get pushed down to hang on switches, etc.  So instead of getting in the habit of cutting off trip pins, suddenly everything works as intended.  Like magnetic uncoupling.

The other thing that happens is that now the couplers are no longer a limiting factor in train length.  I'm running 30 cars up a 2.5% grade behind three Kato SD45's, around a 13" radius 180-degree reverse curve at the top.   The problem then becomes potential stringlining to the center on long trains, not couplers popping apart.   Unintentionally, that pushes the envelope on what reduced flanges will take.  Certain flange profiles are a lot worse than others.   I wouldn't have this resulting problem if the couplers weren't hanging in there.

[bbussey]

The answer to that issue is to body-mount your couplers.  Then the load is transferred through the carbody frames instead of through the truck wheel flanges.  Technically, the load is being transferred through the truck frame into the truck bolster and then into the carbody.  But the wheels are attached to the truck bolster.  So if the load is stronger than the combination of gravity and the wheel flange depth, the wheels will be pushed/pulled over the rail top.

[GaryHinshaw]

The answer to that issue is to body-mount your couplers.

Unfortunately, that has not been my experience with body-mounting, at least in some cases (and in general, I am a big proponent of body-mounting).  With 89' flats in particular, I found that body-mounted cars would string-line more easily than truck-mounted ones, as described in this pair of posts:

https://www.therailwire.net/forum/index.php?topic=24108.msg264568#msg264568
https://www.therailwire.net/forum/index.php?topic=24108.msg264710#msg264710

And this is with 18" curves and 2% grades.  I have yet to decide what to do about it, other than to run pushers to reduce the load on the body-mounted couplers.

[bbussey]

More weight and/or lower the center of gravity in the cars.