Bad craft, poor plan or what?

[stuartmit]

No interest in thAt layout other than its perfect operation in 027. Tight curves. Grades ETC. my plan modified to 17 in Marx curves but still lionel 027 swutches. Working better with suggestions from this thread  

[fifedog]

Those that have visited my layout know that I'm a big advocate of gradient curvatures and loops to gain elevation.  I run long trains, albeit pulled by MTH 2.0 engines.  And, I have 2 switches within the grades.  I use K-Line tubular track, and Lionel swithces.

Two constants I've kept to on any railroad I've built: 1) using a minimum O-42 curves on grades, and 2) starting the rise-over-run as soon as possible, stretching the rise as even as possible to the apogy.

Not sure how committed you are to keeping true to this track plan.  I'd be tempted to take the stub ended 2 track yard (right side) and run that off of your "high line" along the back stretch, thus giving the train a "reason" for taking that circuitous route.  Also, I would cut out that little redundant loop of track on the inside, freeing up the "real estate".  I think you'll have a pretty good switching/operating layout, that would keep any engineer "busy" during an evening's operating session.

[ADCX Rob]

stuartmit

If you check this link to youtube, you will see an 027 layout withe curves that I believe are 027 and grades that functions perfectly:

 

 why won't my layout do that?

That elevated loop doesn't have any switches, and the trains are short and lightweight.

 

[tinplatacis]

Is that one of the incarnations of the 1957 catalog layout? That design had some flaws, but there was a Greenburg's book for Lionel maintenance that had a working design in o gauge. it will probably work better with that plan.

[lionelsoni]

I don't recall ever having a toy locomotive whose motor couldn't make the wheels slip.  That says that tractive effort is limited by adhesion, not torque.  For any two wheel and rail materials (like sintered iron and tin-plated steel), static (Coulomb) friction force is proportional to the weight on the wheels and has nothing to do with the wheel diameter.  On prototype railroads, the rule of thumb is a friction coefficient of 1/4 (which they express as a "factor of adhesion" of 4).

For a grade up to about 14 degrees, adding weight to the locomotive should increase the tractive effort up to the point where the motor stalls.

Traction tires greatly increase the friction coefficient above 1/4; and magnetration simulates a great increase in the apparent weight of the locomotive; but how much effect these tricks have varies among individual locomotive designs.

[stuartmit]

Referring to Buckeye (as a Wolverine, I have a lot of trouble with that, BUT HARBAUGH IS BACK IN TOWN!), the reverse banking seems great--about 1/8 from one side of the track to the other seem OK? How does that work with a train heading down the grade around the curve--is the loco more likely to go off the tracks with the outside tilt? 

 

Eliminating any grade on the curved portions forces you to put more grade on the straight portions, but maybe that works out better.  I'll check that out over the weekend.

Regarding Gregc's post re pulling power of different locos, I believe the smaller the drive wheels, the better the power should be, but the slower the train will travel. This is because 1 revolution of the axle on a smaller drive wheel carries the train a shorter distance because of the smaller circumference.  But it pulls better because with the amount of foot pounds of torque being the same at the axle, for a given model of motor, the smaller the distance, the greater the force delivered to the rail. The unknown in the argument is how the different motors Lionel used compare from one model to another . My perception counters all that baloney I just put out  when comparing my 2020 to my 2046; I believe the 2046 pulls better. But I have no idea of the torque delivered by the two different mortors so that could be true even tho the 2020 has smaller drivers. Also the differing weight has an effect I have no idea how to judge; a heavier loco may deliver greater force than the same motor/wheel combination in a lighter engine because to a certain extent weight improves traction, but the heavier loco has to drag itself up the hill as well as the load. Is there a point where additional weight is no longer a help? Is junk in the trunk really a help?  Damn if I know!

[gregc]

stuartmit

why won't my layout do that?

I went thru a similar experience with a folded dogbone in HO.   I couldn't understand why my 0-4-0 could pull a train uphill but not my new 2-8-0.   I eventually gave up.

since then i've learned 1) not all locomotives had the same amount of traction.  That 2) pulling a train around a curve is like pulling a train uphill (one rule of thumb is that the effective grade is 32 / radius).  3) there is a reason for limiting grades to 2%.   And 4) it will be less frustrating to design conservatively than plan for the best.

The derailments caused by string-lining of cars around a bend can be minimized by weighting all cars properly.   Since I model HO, i'm not familiar with the guidelines for O gauge, but in general, cars of the same length should weigh about the same.

best of luck

[Buckeye Riveter]

stuartmit

 Further have frequent derailment where car is pulled over by the tension of adjoining cars.    

This type of derailment is called a string line derailment and it is a frequent problem of model railroads because the center of gravity (CG) of our empty or lightly loaded model train cars is not the same as the prototypes.  For example, most likely your cars are plastic.  The prototypes are not plastic. So that is one major modeling difference that changes the CG.  If you increase the weight of the cars the horizontal force needed to string line derail the car is increased , but the model locomotives traction can not make the 4 percent grade especially with a curve added. 

 So how do you attempt to prevent the string line derailments?  First wider curves are the best way.  Eliminate or minimize the grade around the curve.  However there is another trick the HO people use and that is reverse banking of the track especially in their elevation changing spirals.  Reverse banking puts the model train cars CG farther to the outside of the curve.

BTW,  Magna-Traction doesn't work with stainless steel track.   

[stuartmit]

If you check this link to youtube, you will see an 027 layout withe curves that I believe are 027 and grades that functions perfectly:

 

https://www.youtube.com/watch?v=yNVtwPxwu3k

 

why won't my layout do that?

[KRM]

ADCX Rob

 

 

For your layout, consider K-Line K-275 & K-276 switches, which will give you 20.14" radius turnouts.

 

You could also consider Lionel 027 42" R switches would be the same as the K-Line switches.

6-65167 and 68. I have 14 of them on my layout and they work well.

[ADCX Rob]

stuartmit

I estimate the length of that upgrade at 80" max even with the 18" radius curves...

The backstretch is closer to 100"-110" taking into account the 10" length of straight sections(although the diagram shows CCW train travel).

For your layout, consider K-Line K-275 & K-276 switches, which will give you 20.14" radius turnouts.

[stuartmit]

Rob:

re yr comment

"That track plan uses 18" radius curves, the problems aren't nearly as pronounced as with your layout, which uses 12.37" radius curves.",

I estimate the length of that upgrade at 80" max even with the 18" radius curves. . Possibly the direction of travel was down in that location, limiting you to one directional travel, and the other end of the grade does seem somewhat longer, but I am still feeling ambushed by the advertising guys and disappointed at the changes had to make in my plan.  Looking for a lefthand Marx 034 switch motorized to maneuver in place of offending 1122.

[rtraincollector]

Just one comment about grade on a curve Lionel use to show it all the time especially when they did a over and under figure 8 there is no reason not to do a grade on a curve if you do it correctly. and don't over load the train. I presently don't have any grades on present layout but have in the past with no problem ( using post war trains ( just know your limit your train can handle. ) 

[ADCX Rob]

stuartmit

I 'm sure you have seen this trackplan which I'm sure violates your recommendations 1 and 2: Consider the curving grade at lower right portion of the plan

I may conced on your points 3 and 4

 

That track plan uses 18" radius curves, the problems aren't nearly as pronounced as with your layout, which uses 12.37" radius curves.

[KRM]

sturatmit, 027 curve on the grade is a poor choice for sure. As for a grade on a curve I like them. I made mine with 027 profile 054 and K-line 072 radius and have no problems with the derailment where cars are pulled over by the tension of adjoining cars. Note in the picture the pitch I have on the track that helps. Some would argue, but it works.
Many of my postwar steamers have a hard time on pulling more than seven cars up the grade. My 736 Berk will pull 12 RMT ore cars up but you got to be moving first and then lay the juice to her to make it work.

Here are some shots of ours.

 

[cwburfle]

Over the years, some of the layouts I've built had curves on trestles, either with a grade, or at the max height. I have also put switches on trestles at the max height.

For the curves, I put extra trestle bents under the center tie. Otherwise the track would flex too much. When doing so on a grade, it was necessary to put a shim underneath the base of the trestle.  Certainly had to watch my speed :)

For "O" gauge switches, it took four "A" trestles: one for each track (3) and one to support the switch motor. I cut pieces of aluminum sheet stock to match the base of the switch, and the attached track, out to the first tie. The tracks and switch were attached to the plate to keep everything together tightly. I ran the control wires down the side of the trestle that supported the switch motor.

As a general rule., I put both power and ground feeders on all three tracks that are attached to a switch, whether they are elevated, or on the train table.

[stuartmit]

I think that some of the things you say relate to the bait an switch feeling: I believe that my layout is of a size that many would feel is reasonable, if not their dream layout. In a layout of this size I think it is difficult to come up with a design using grades which are gentler if it you want one train to pass over another. I also believe in a modest size plan which includes grades, you can't avoid a grade going around a curve. Yet I feel completely surprised to have run into the problems I have encountered because of my familiarity with  old advertising materials.

 

I 'm sure you have seen this trackplan which I'm sure violates your recommendations 1 and 2: Consider the curving grade at lower right portion of the plan

I may conced on your points 3 and 4

[ADCX Rob]

1) 4% is steep.

2) Do not include curves on grades.

3) Even worse, do not include switches on grades.

4) Go back to the 1955(when the 110 trestle set was introduced) & later catalogs and pick an O-27 set for your O-27 layout. It will probably run fine on a 110 trestle set equipped layout that does not violate 2) & 3).

[stuartmit]

This is an expression of bait and switch dissatisfaction or a once in a lifetime (well at least a week!) chance to make myself a target  of attacks for poor workmanship, so comment or fire away:

I designed a 5 x 11 layout which I briefly described elsewhere as a folded dogbone where the loop on one end is actually located over top of the other with about a 41/2" level seperation between; maybe 5" depending whether you measure underside clearance to next level or rail head to rail head. Measuring from the top of the L girder itself to the top side of the 1/2" plywood and zero elevation, and comparing to the same measurement at max elevation, I found a difference of 5".  I measured the length of the grade which goes around the 180 degree "fold" in the dogbone and found the grade is 125" long, thus yielding a 4 percent grade. This is more than optimal, but probably not surprising in a layout of this size. My annoyance is that I believe Lionel created the impression years ago that a magnetraction loco steam or diesel (and I believe that may make a diff) would be successful tugging a train up that hill. I would like a 2046 to pull 7-8 cars of mid 1950's vintage and can't seem to have success. The "U" in the curve is made of 027" radius curves, and maybe that is too sharp, but no one tellsl you that! Further have frequent derailment where car is pulled over by the tension of adjoining cars.   An 1122 switch almost at the top of the hill frequently causes derailments (in either direction; heading in the opposite diection, the 2046 really gets up a head of steam, and seems to smack something in the middle of the curve of the switch which causes pony wheel under cowcatcher to jump off tracks).

Is my design asking too much, or am I lousy at putting track together? No question I am not a fine craftsman, but I believe that I am  good enough on this deal which is not exactly wallpapering or clock repair; I just think the real world performance specs of these products, and the historical advertising lead one to unrealistic expectations.

I have improved things somewhat by tearing up track, and moving it in such a way as to accomodate Marx 034 curves, but at a price of other features which I have had to eliminate from my plan, or move in a way which dimishes them or crams them into less space to less scenic effect or reduced clearance.