Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Super-elevation & roadbed height

6794 views
31 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    April 2003
  • 305,205 posts
Super-elevation & roadbed height
Posted by Anonymous on Tuesday, February 13, 2001 7:10 PM
Can someone please tell me how to super-elevate the curves on my HO layout?
I'm also wondering what is the best way to transisition the height of my roadbed from mainline down to a siding or spur. My roadbed is cork on a foamboard based table.
Any & all advice is very much welcome.
Matt
  • Member since
    January 2001
  • From: US
  • 76 posts
Posted by sumpter250 on Tuesday, February 13, 2001 7:29 PM
superelevation!!! In the prototype, the outer rail of a curve is placed fractions of an inch higher than the inside rail to force rollingstock to lean inwards, compensating for the tendancy to lean outwards at speed. The train can take the curve at higher speed. Model trains DON'T lean outwards on the curves,(unless they're moving at warp 8),and ,in fact tend to be pulled into the curve. Superelevation increases this tendancy, and,if the curve is tight enough, as model curves tend to be, the train can be pulled right off the track! This is especially true of auto racks, and double stacks. Looks good!,,,, not recommended.
  • Member since
    January 2001
  • From: US
  • 76 posts
Posted by sumpter250 on Tuesday, February 13, 2001 7:35 PM
Mainline to siding transition.
Once both rails are clear of the mainline balast, the track can be sloped down the same way you would begin a mainline downgrade. A foam ramp can take the cork roadbed down to where you want it. Just don't make the change too abruptly.
  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Wednesday, February 14, 2001 3:02 PM
There was an article on superelevation in MR a while back (and I believe is part of their new book on trackwork and lineside details) that advocated using stripwood under the outside rail. I tried it on one curve -- all I can say is you need VERY broad curves to have any visual effect, and there is absolutely no positive operational effect. In other words, unless you're set on it, and your trackwork and equipment weights/couplers are perfect, I'd pass.
  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Thursday, February 15, 2001 6:07 PM
Matt,
I've used a Stanley Surform tool to slope the cork down to the subroadbed on foam and plywood with great success,then lay the siding directly on that material[it should be evenwith the ground anyway]. This in N ,but no reason it won'twork in the larger scales also.
Scott
  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Friday, February 16, 2001 1:00 AM
Hmmm, dang the luck with the super-elevation!!
My curves are mainly 22" dia. & something just looks hokey when my trains go through them. I was hoping a little elevation on the outside might give a more realistic look but I guess I'll have to wait until I graduate from a 4x8 layout to use broader curves.
At least now I can give my sidings & spurs a more "prototypical" look. ;)
Thanks for all your advice guys!!!!
Matt
  • Member since
    September 2009
  • 44 posts
Posted by jharrison on Wednesday, August 8, 2018 2:41 PM

sumpter250
superelevation!!! In the prototype, the outer rail of a curve is placed fractions of an inch higher than the inside rail to force rollingstock to lean inwards, compensating for the tendancy to lean outwards at speed. The train can take the curve at higher speed. Model trains DON'T lean outwards on the curves,(unless they're moving at warp 8),and ,in fact tend to be pulled into the curve. Superelevation increases this tendancy, and,if the curve is tight enough, as model curves tend to be, the train can be pulled right off the track! This is especially true of auto racks, and double stacks. Looks good!,,,, not recommended.

Above is a quote from an earlier post, the following is my response:

Prototype super elevation is multiples, not fractions of inches. With broad (48" and greater radius in HO),  superelevation and easements will create very realistic track. This has been the observation of hundreds of visitors to my layout (60" min radius except in staging, which is 48" and not superelevated but is easemented)
  • Member since
    March 2018
  • 688 posts
Posted by BNSF UP and others modeler on Wednesday, August 8, 2018 2:47 PM

I have a question too. For code 83 track, what would be the height of the outside rail prototypically? How high is it on the prototype, and how high would that be in actual inches to get the correct scale height?

I'm beginning to realize that Windows 10 and sound decoders have a lot in common. There are so many things you have to change in order to get them to work the way you want.

  • Member since
    February 2005
  • From: Vancouver Island, BC
  • 23,330 posts
Posted by selector on Wednesday, August 8, 2018 3:25 PM

For almost all applications of super-elevation in HO, about the thickness of cardstock is almost going to be too much.  I use the clear bubble pack plastic in packaging.  I cut strips of it and place it under the edges of the ties on the outside rail on a curve.  It looks good.  I find that a few people, not many, over-do the super in HO.

Here's what mine looked like with no strips...the sub-roadbed was MDF spline that was canted and planed:

  • Member since
    September 2009
  • 44 posts
Posted by jharrison on Wednesday, August 8, 2018 3:39 PM

BNSF UP and others modeler

I have a question too. For code 83 track, what would be the height of the outside rail prototypically? How high is it on the prototype, and how high would that be in actual inches to get the correct scale height?

 

First, superelevation does not depend on what code of rail is used. For the prototype the outer rail is usually limited to three inches above the inner rail. I have found that this amount of superelevation in the model looks very realistic. Too much is worse than too little, as it starts to look "toy like". So for HO scale (87.1:1) the outer rail would be raised 3/87.1" which is  0.034 " (about 1/32") above the inner rail. I lay a digital level across the rails to measure superelevation tilt angle. It turns out that a pretty accurate relationship between the angle, in degrees, the rail height difference, and the track gauge in inches is : tilt angle (deg)=57.3 x rail ht difference/ track gauge. So, for the prototype with a track gauge of 56.5 " a rail height difference of 3" results in a tilt angle of 3.04 degrees. For low speed track such as sidings the superelevation should be about half as much to look right.

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Wednesday, August 8, 2018 4:08 PM

On my eased and superelevated 60" curves, I achieve full superelevation with .040 Evergreen strips under the outside end of the ties.  This gives a 2" elevation to the outer rail.  I use steadily diminishing shims under the eased section of track.

To me, it looks perfect.  It's not so much that it calls attention to it, but you still "see" it.

I can recall no derailments on these curves.

Here, by the way, is a short and interesting discussion on superelevation:

https://www.trackplanning.com/superelevation.htm

As I said, I think 2" is perfect for my trackage.  I will mention that it represents a single track Western main line.  So trains might move briskly, but not blindingly.  I MIGHT raise the rail a bit more if I was running a lot of high speed passenger.  Probably would, actually. 

 

Ed

  • Member since
    March 2018
  • 688 posts
Posted by BNSF UP and others modeler on Wednesday, August 8, 2018 4:12 PM

Thanks!

I'm beginning to realize that Windows 10 and sound decoders have a lot in common. There are so many things you have to change in order to get them to work the way you want.

  • Member since
    September 2009
  • 44 posts
Posted by jharrison on Wednesday, August 8, 2018 4:32 PM

I agree with your choice for the following reasons:

If you are placing the shim under the ties at a point directly under the outer rail then the tilt point of the ties is probably at the far end of the tie, which is not under the inner rail. By my measurement, for a standard 8 1/2 ft. HO tie, this point is 0.947" from the outer rail so I calculate that your tilt angle is actually 2.42 degrees (57.3 x 0.040/0.947) which is about a 2 1/2" prototype rail height difference. This is right at the ideal superelevation from my experience. Plus or minus a half a degree is very hard to notice by eye.

Moderator
  • Member since
    June 2003
  • From: Northeast OH
  • 17,247 posts
Posted by tstage on Wednesday, August 8, 2018 4:39 PM

IIRC, I used 0.02 or 0.03" styrene strips for my superelevation.  Enough to be noticed but not too much if my trains were stopped on the curves for some reason.  I also eased the superelevation into and out of the curves in height incriments of 0.01" at a time for a determined length of track.

Tom

https://tstage9.wixsite.com/nyc-modeling

Time...It marches on...without ever turning around to see if anyone is even keeping in step.

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Wednesday, August 8, 2018 4:48 PM

jharrison

I agree with your choice for the following reasons:

If you are placing the shim under the ties at a point directly under the outer rail then the tilt point of the ties is probably at the far end of the tie, which is not under the inner rail.

Ah, but I am not.  As I said, I placed the .040" shims under the outer end.

 

By my measurement, for a standard 8 1/2 ft. HO tie, this point is 0.947" from the outer rail so I calculate that your tilt angle is actually 2.42 degrees (57.3 x 0.040/0.947) which is about a 2 1/2" prototype rail height difference. This is right at the ideal superelevation from my experience. Plus or minus a half a degree is very hard to notice by eye.

 

(track gage) / (tie length) X (shim thickness) = superelevation in true inches

     .7"          /       1.2"     X       .040"           =                 .023"

 

Convert true inch to HO inch:      .023"   X   87.1    =    2.03" HO

 

I certainly agree the difference isn't stunning.  I just didn't want people thinking I was shoving the shims underneath the rail.  That's too much like, man, work.  According to Maynard G. Krebs, anyway.

 

Ed

  • Member since
    September 2009
  • 44 posts
Posted by jharrison on Wednesday, August 8, 2018 5:00 PM

My mistake, I read that as under the outside rail where you actually said under the outside edge of the tie. Sorry, but we still are in agreement and in fact most of my curves also use closer to 2 degrees than 3, although 3 looks good also.

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Wednesday, August 8, 2018 7:59 PM

For those of you considering adding superelevation, note that it has to happen "gradually".  You can't just be going along flat, and then kick up the outside rail.  For my curves, it takes 18" to go from flat to full superelevation.  Which, conincidentally, is how long the easement is, also.

Other folks I've conversated with use 12".  I prefer the longer distance.

Trains, real and model, don't like physical surprises.  Gentle does it.

 

Ed

  • Member since
    March 2018
  • 688 posts
Posted by BNSF UP and others modeler on Wednesday, August 8, 2018 8:31 PM

Good to know. I was wondering about that too.

I'm beginning to realize that Windows 10 and sound decoders have a lot in common. There are so many things you have to change in order to get them to work the way you want.

  • Member since
    February 2001
  • From: Wyoming, where men are men, and sheep are nervous!
  • 3,392 posts
Posted by Pruitt on Thursday, August 9, 2018 7:53 AM

I use multiple layers of 1/4" masking tape under the outside rail.

The tape is roughly .004" thick per layer, so the six layers I use gives about .024" of height over the inner rail. This looks about right.

The advantage of this method over other shims is that no tapered shims have to be fabricated. I start with one layer of tape, then the next is offset 1.5" into the first strip, and so on for all six strips. It builds a gentle nine inch long ramp that none of my equipment has a problem with. 

  • Member since
    January 2004
  • From: Canada, eh?
  • 13,375 posts
Posted by doctorwayne on Friday, August 10, 2018 1:11 AM

I've been attempting to open a Word document that outlines my method of adding superelevation, with no success.  I copied the following from another thread, which pretty-much explains the process.  The method discussed is for cookie-cutter style plywood roadbed, and deals with track on a grade, but the method works for level track, too: 

"For the riser-supported track, I had all of the risers clamped to the benchwork crossmembers, with a pencil line on each denoting the proper height to keep the rise of the grade constant.  I then selected the riser closest to the middle of the curve, and after placing a locomotive on the curve, carefully removed all of the clamps.  I then lifted that centre riser to approximately its proper height, and pushed its bottom end towards the outside of the curve.  When the angle of superelevation "looked right" on the locomotive, I aligned the inside end of the pencil line on the riser with the top of its joist, and re-clamped it in place.

Because all of the other risers were not fixed to the benchwork, the 3/4" plywood roadbed (about 2" wide) flexed torsionally when that centre riser's bottom end was pushed out, but the amount of cant decreased as the distance from the centre riser increased , automatically creating easements into and out of the superelevation. 
All that was left to do was to lift each riser, in-turn and without changing its newly created angle of deflection, until the inside end of its pencil line was was even with the top of the joist to which it was to be fastened.  It's actually easier to do than it is to explain."

Wayne

  • Member since
    September 2009
  • 44 posts
Posted by jharrison on Friday, August 10, 2018 9:23 AM

Tilting the risers or or cutting a taper on the top (both directions on grades) is what I've done for years but for some reason most don't like it. The beauty of this is that I can change the tilt or elevation if something doesn't look right. With shims you better get it right the first time.

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Friday, August 10, 2018 9:54 AM

jharrison

Tilting the risers or or cutting a taper on the top (both directions on grades) is what I've done for years but for some reason most don't like it. The beauty of this is that I can change the tilt or elevation if something doesn't look right. With shims you better get it right the first time.

 

 

Ballasting is what seals the deal, with the plastic shims I use.  Up until then, I can just pry them out.

Fact is, once a person decides the proper superelevation, it's hard to avoid getting it "right the first time."

My curves are built on flat plywood.  So there are no risers to cant.

 

Ed

  • Member since
    February 2005
  • From: Vancouver Island, BC
  • 23,330 posts
Posted by selector on Friday, August 10, 2018 1:08 PM

Those of us who choose to ballast our tracks soon learn to trial the arrangement using typical consists before applying the ballast.  It typically takes me a couple of weeks of running trains and then tweaking rail heights before I dare to groom and glue the ballast.  

  • Member since
    September 2014
  • From: 10,430’ (3,179 m)
  • 2,311 posts
Posted by jjdamnit on Friday, August 10, 2018 1:38 PM

Hello all,

I use wooden coffee stir stick that are 6/64-inches thick, 5/32-inches wide and 5-1/2-inches long.

On my pike I use sectional track in 15-, 18- and 22-inch radii.

To superelevate the curves I cut the stir sticks into 1-inch pieces and place three pieces, evenly spaced, on the outside of the curve.

For transitions back to level track, on the last piece of sectional track, I don't place any pieces of stir stick, but rather build up the ballast.

Once the ballast is set and dried the transitions are stable.

Hope this helps.

"Uhh...I didn’t know it was 'impossible' I just made it work...sorry"

  • Member since
    December 2001
  • From: Northern CA Bay Area
  • 4,387 posts
Posted by cuyama on Friday, August 10, 2018 5:23 PM

jjdamnit
use wooden coffee stir stick that are 6/64-inches thick, 5/32-inches wide and 5-1/2-inches long. On my pike I use sectional track in 15-, 18- and 22-inch radii. To superelevate the curves I cut the stir sticks into 1-inch pieces and place three pieces, evenly spaced, on the outside of the curve. For transitions back to level track, on the last piece of sectional track, I don't place any pieces of stir stick, but rather build up the ballast.

6/64" equals more than 8" in HO scale. That's a lot compared to what many experienced modelers use for much broader curves. And if you are only allowing at most 9" for the superelevation to run off, that's a lot of twist across a short length of track, not to mention the vertical kinks on the elevated rail.

Are you sure that's the amount? Can you post a photo to show how it looks? 

In an earlier thread, you gave the amount as 1/32".
http://cs.trains.com/mrr/f/11/t/259681.aspx

jjdamnit
In the May 2015 edition of Model Railroad Hobbyist magazine; page 33, there is a Q&A regarding superelevation. The recommendation from MRH is that it should not exceed 1/16-inch (1.5 mm) for HO scale. On my flat curves I use 1/32-inch (0.75 mm) thick wood coffee beverage stirrers on the outer edges.

For such sharp curves, superelevation has been often found to be a reliability headache.

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Friday, August 10, 2018 5:50 PM

cuyama

 

 
jjdamnit
use wooden coffee stir stick that are 6/64-inches thick, 5/32-inches wide and 5-1/2-inches long. On my pike I use sectional track in 15-, 18- and 22-inch radii. To superelevate the curves I cut the stir sticks into 1-inch pieces and place three pieces, evenly spaced, on the outside of the curve. For transitions back to level track, on the last piece of sectional track, I don't place any pieces of stir stick, but rather build up the ballast.

 

6/64" equals more than 8" in HO scale. That's a lot compared to what many experienced modelers use for much broader curves. headache.

 

 

 

.7"      /     1.2"      X       6/64"     =      .0548"

 

.0548"      X     87.1        =     4.77"   of superelevation

 

 

Quite a bit, I agree.  What with my choice of 2".  On a 60" radius curve.

 

 

Ed

  • Member since
    December 2001
  • From: Northern CA Bay Area
  • 4,387 posts
Posted by cuyama on Friday, August 10, 2018 6:06 PM

Don't understand the point of your math. But agree, that doesn't seem like an  amount that will operate reliably, whether it's under 5" in HO or over 8" in HO. On a 15" curve.

  • Member since
    September 2014
  • From: 10,430’ (3,179 m)
  • 2,311 posts
Posted by jjdamnit on Friday, August 10, 2018 6:16 PM

Hello all,

cuyama
6/64" equals more than 8" in HO scale. That's a lot compared to what many experienced modelers use for much broader curves.

I defer to my signature.

cuyama
Can you post a photo to show how it looks?

Still working on how to post photos without going through 3rd party site.

Hoping to post photos to several of my threads soon.

Thank you for all the positive input.

Hope this helps.

 

 

"Uhh...I didn’t know it was 'impossible' I just made it work...sorry"

  • Member since
    December 2001
  • From: Northern CA Bay Area
  • 4,387 posts
Posted by cuyama on Friday, August 10, 2018 6:23 PM

jjdamnit
I defer to my signature.

And your own earlier post compared to this one? One of them must be wrong. Right?

The issue is that newcomers read these posts without an appreciation for the poster’s level of expertise. Anyone can post anything they like, but when it diverges so far from best practices, I think it’s important to note that and ask for more information. But that’s just me.

jjdamnit
Still working on how to post photos without going through 3rd party site.

You're in luck -- you may post them right on this site. One waits a few days for it to be approved sometimes, but no need to go elsewhere.

http://cs.trains.com/mrr/m/mrr-layouts/default.aspx

 

 

 

 

  • Member since
    September 2014
  • From: 10,430’ (3,179 m)
  • 2,311 posts
Posted by jjdamnit on Friday, August 10, 2018 6:36 PM

Hello all,

On smaller layouts sometimes features are abrupt and compressed.

I am limited to a 4'x8' pike.

OH God, how I would love for even a 5'x9'- -but even that is unobtainable.

cuyama
...but when it diverges so far from best practices, I think it’s important to note that and ask for more information.

Again...

"Uhh...I didn’t know it was 'impossible' I just made it work...sorry"

I apologize for not having enough room to build a pike that can utilize "best practices" in the mean time I will run my trains on my sub-standard pike and continue to enjoy this great hobby.

Sorry, this doesn’t help.

"Uhh...I didn’t know it was 'impossible' I just made it work...sorry"

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!