Superelevating Track

[oakcreekco]

It's been a long time since I superelevated mine, but I used .060 x .060 plastic strips, and glued them down on the outer top edge of my roadbed. I use the vinyle roadbed, and just lay it down with contact cement.

Once the strips were secured with contact cement, I hand sanded the entry and exit of the curves for a gentle transition to the full "superelevation". I also layed out the roadbed for the "transition radius" to the curves.

Once I was happy with the sanding, I filled the top of the roadbed with Duram's putty, and used a small  trowel/putty knife to smooth it out.

Layed the track, put on the ballest, and done.

Transition curves, along with superelevation makes a HUGE difference in looks, and I've had no problems with derailments, track pulling up, or things like that.

Many way to skin the cat, but this has worked out well for me.

[randgust]

I may as well keep trying, if nothing else it pointlessly raises the post count!

This is the most noticeable curve on the entire layout, and it's equipped with spirals and relatively heavy superelevation:



On the prototype, this "S" curve at the west end of the Winslow yard was responsible for a major derailment of the Super Chief, from excess speed, and much of the roundhouse was taken down to reduce the curvature.   But any train rolling through this curve at around 30mph looks absolutely majestic.    Method was stripwood under the tie edges, adjusted and sanded, and ballasted in much later after I was sure I hadn't created an operating problem.   Even my brass Hallmark 4-8-4 negotiates it just fine.

[wazzou]

From that angle, it looks great Randy.

[oakcreekco]

Very nice.

[mark dance]

...super-elevations of about 10-20 thou on the passing siding and 20-30 thou on the higher mainline sanded directly into the cork roadbeds.  There is actually a cross-over half way along this siding, and it is on a curve, which made the transition between the roadbeds uber-complicated however reliability has been great.  The low angle view certainly shows the height difference clearly here.



md

[MichaelWinicki]

Funny...

I asked an old PRR engineer friend of mine, who started with the PRR in '55 and retired from CR in '97, if he ever took a train over super-elevated curves, and he said that he never did.

Obviously he didn't go everywhere on the Pennsy, but I thought his answer was interesting.

[randgust]

Mike, I'll guess that you've got a copy of the PRR "TRIUMPH 7" book, and on page 297 there's a nice shot of PRR train 571 northbound at Arcade, rolling through a nicely superelevated curve right at the depot. 

I think its probably a PRR thing that they just didn't call it 'superelevated'.  And/or the fact that engineering and operations spoke (and still do) two different languages.  Ask Barstow Rick and I about switches vs. turnouts,....LOL!  Maybe 'cant' was the PRR term?

I grew up around the PRR, and never saw much of anything that I'd call extreme superelevation.   I think it was because of the slow speed on the coal and ore trains around here.  But the first time I got out and studied the Erie...WOW!  The north main (that's taken up now through Bear Lake) had a main line curve on it, with a crossing through it, that probably had 6" on the outside rail.   I think it had a speed restriction of 50mph.  Same thing with the Kennedy NY curves, I walked those many times and marvelled at the superelevation.  I have shots of a big K5 4-6-2 on that curve completely leaning.   By Conrail, the low rail was getting beat to pieces, end battered and curveworn, and the high rail was relatively untouched - the difference between running it at 55mph on EL and 40 in the CR days.   I think we took 2" off the high rail there in the rehab.

Here's an old shot of Waldwick, NJ on the Erie....yeah, this is when you don't need no stinkin' speed restrictions, just don't stop or you'll fall off the inside rail:

http://www.railroad-line.com/forum/data/hudsonelectric/2004122921141_ERIE%20STEAM%20COMMUTER.gif

Given that your 'typical' railhead height is around 6", if you think the bottom of the outside rail is higher than the head of the inside rail, gives you an idea.   I think the MAXIMUM superelevation allowed now is 7" per FRA in any case.
http://ecfr.gpoaccess.gov/cgi/t/text/text-idx?c=ecfr&sid=7365b2a00d781ef9d1a95cb9c1f443fa&rgn=div8&view=text&node=49:4.1.1.1.8.7.5.12&idno=49

That translates into .04375 ; think .040 strip under the outside rail.   1/32 stripwood works out to about .030.

[coosvalley]

I think the MAXIMUM superelevation allowed now is 7" per FRA in any case.

I would very much like to see an example of 7" superelevation!

I think Randy is correct......terminology differences...

Another problem could be some see a difference between superelevation and SUPER-elevation......

[BCR 570]

Lemosteam:

Thanks for the tip - flexibility is definitely a requirement of whatever is used for the superelevation.  I dont mind your son's smiling face at all - that is the future of our hobby.  I am also pleased to see that you have something father and son can enjoy together.

Tim, those pics definately show a good amount of superelevation,BUT, a pic from your era would convince me .I, however, do not need to be convinced, you do ! 
A tamper can change track structure very quickly. The track might have looked even better in your era, or possibly worse....

There would be some superelevation on the curves in any case,the question is how much........

BTW, I am VERY impressed by your trackwork......... code 40 wood ties 

Another good question!  Most of my photographs from the 1970s are taken on straight track or in yards, but I did find this photograph of a train climbing the East Pine hill in 1978:




At this location on the 2.2 per cent grade the track speed would have been very low - trains are either grinding their way uphill after crossing the Pine River or are in full DB trying to avoid going into the river at the bottom of the hill.

I am now fairly settled on superelevation of about .020" with graduations at either end.  I think that will offer a little visual effect without too much for the location and era.

Thank you for your kind comments on the trackwork - it has gone well; however I will reserve judgement until it is laid permanently and trains are rolling over it satisfactorily.


Tim

[Lemosteam]

Randgust that math is too much for my brain but I assume it takes into account the gradual elevation of the outer rail at the beginning of each transition curve. 

Anybody, if that is the case at what point generally speaking does the outer rail stop elevating and become a constant height above the inner rail?  For instance, let's say there was a 12" long transition into a 36" raduis, assuming the entrance point into the transition was level.  Should all of the S.E. be complete before the end of the transition or a certain percentage into the radius? 

Are there any NMRA standards governing S.E.?

[randgust]

The NMRA doesn't necessarily apply, but the FRA sure does!

I'll give you the tags to the calculations, then explain in english...
http://www.law.cornell.edu/cfr/text/49/213.59

All mainline curves are entered and exited by spirals.  Even industrial track - for the most part - uses at least some short spirals.   So as the curvature decreases out of the spiral, the superelevation also drops accordingly by speed.   So if you have some spirals and you sand down your superelevation method thorugh the spiral, it should look and behave properly.  The lack of spirals, rather than superelevation, is more of a model problem.

On the prototype, you'll manually mark the curve out into 62' chunks, compute the average degree of curvature, and set the superelevation accordingly.  The newer tampers have an on-board computer that reads the curve with lasers, you input the track speed, and it will compute both the set for the superelevation and runoff.  The 'buggies' on tampers are part of this sensor chain system.

Spirals happen kind of naturally in flextrack, hard not to.  But in sectional, including Unitrack it can be darn abrupt.

If you explore the CFR, you'll find out the runoff limits by track speed - how far out you have to 'taper' any changes in rail elevation by inches in a 31' track section.   Those also apply here, but if you're setting the superelevation properly through the spiral, it's pretty well flat out of the spiral anyway without having to resort to running the superelevation out into the tangent track.

It's amazing how good you can make models look if you learn just a little about this.  Watching a train enter a spiral curve and lean into and out of superelevation is kinda wonderful. 

On the prototype, seeing it done right is equally beautiful; if you study the railheads you can see just how the wheels are riiding; low, high, whatever, and when it's all together the operations and track supervisor people have a justifiable sense of pride.

But that 7" maximum superelevation rule, and the difference between freight and passenger speed...is basically where curve speed restrictions come from.  You can't tip the track any more to compensate and it becomes a compound problem, you simply have to slow down.  It's at the heart of why high-speed rail is so difficult to achieve when you are mixing conventional freight and high-speed passenger.  A loaded unit train coal car behaves differently from an empty autorack from a stack train from a Talgo tilt-train.    In the old days your high-center-of-gravity problem was steam boilers so if you got the locomotive through it everything else was comparatively easy... until train lengths got long enough to start stringlineing on the low side.