If you do them right, they look terrific, in my opinion. The rub is, who defines "right". Well, you do, but if you show your work, you may get some dissenters.
I think that some of us overdo superelevation. In truth, all you need is an angle of about 2-3 degrees, 4 degrees at most. That is measured from the non-elevated railhead level across the tracks. That translates to about three layers of masking tape, at most, under the outer tie ends and rail....pressed well into each other. Superelevation seems to show up in many layouts, and it can look very good, but it often looks overdone.
What you can do is shim up a trial section where you may want it, and hold it down with some track nails. Place a train along it, and see what you think. It should be obvious, but not, "Oh, oh!!"
The "techical term" is SuperElevated.
Worth it? Depends on what you mean, I guess.
The real ones tend to conteract the centrifugal force of a fast-moving train, so it doesn't derail to the outside. On a model railroad, they won't help keep trains on the track, and in fact may make them more likely to derail, to the inside.
But in some opinions, they look good, so some of us do it. We no doubt over-do it, often, by elevating them more than is prototypical.
How? I use Woodland Scenics foam roadbed, and just put shims made from cardstock along the outside. Since the foam roadbed is pliable, you don't have to be exact. I start with 6" or so of thin stock, then thicker stuff (I'm in HO).
With cork, you'd probably have to be more precise.
Some folks put the roadbed down, then shim between it and the track. I find it easier to ballast the other way, but whatever works is fine. If adjustments are needed, I do sometimes shim between the roadbed and track.
The only things to be careful of are to keep the track as even as you can, and if there are any turnouts, be very careful that they're in the same plane. Meaning, if it turns out away from the curve, it should rise for a few inches before levelling off; if it turns inside, it must actually descend for a few inches. Otherwise you will have Derailment City! And if you have a turnout on the straight, just before a curve, don't change the plane near the turnout -- either start before the turnout, or well after it.
I have a large double-track main, with some crossovers on superelevated curves. I had to raise the whole outside track to keep the whole thing in the same plane. Big hassle, but it finally worked out so there are almost no derailments there.
So be sure all your equipment will go in all directions and in all combinations before ballasting -- you can still shim & adjust, but it's much harder.
Superelevation is the technical term for what I think you are talking about. My understanding is that for the most part we have no engineering need for it on our layouts (compare road racing sets where the banking of curves actually does help the cars stay on the track), so its value is to add realism. I think it adds greatly to realism especially if you have nice sweeping curves and like to run passenger trains. There is something about seeing an engine and then the cars lean into a curve that looks realistic to me -- takes me back to the days of watching high speed C&NW passenger trains go through the curve in downtown South Milwaukee. But my curves are 40 and 42 inch radius. It might just look a bit goofy with 18" radius curves.
I have seen built up applications of decreasing lengths of narrow masking tape, long tapered pieces of styrene or wood, and even superelevated subroadbed be used. On my layout I am using a flex track from PSC that actually comes with molded in superelevation on the bottom of the ties. Where you want to transition to flat tangents I use rail nippers to nip away the plastic nodules. The only problem comes with ballasting where obviously one edge of each tie is sort of hanging in air, which is why superelevating the roadbed or subroadbed makes some sense.
I first learned about PSC's superelevated flex track from ads in Model Railroad Planning. I do not think I had seen it advertised in MR oddly enough.
By the way obviously superelevation creates some complications for things like highway crossings and turnouts, so think ahead where you want to use it.
Dave Nelson
Nothing says, "Well engineered railroad," like a properly superelevated mainline curve. OTOH, doing it wrong, or overdoing it, can be the starting point for a major PITA!
With the exception of dedicated super-high-speed trackage, prototype railroads are only moderately superelevated. I find that two layers of cardstock under the outer end of the ties of HO flextrack makes things just about perfect. The first layer is 16mm wide, and extends just over half the width of the flex (which is 28mm across the ties,) while the second is only 4mm wide. After ballasting, the outer edge of the curve is obviously higher - which agrees with what I saw while railfanning my prototype.
I prefer to combine superelevation with the spiral easements of my curves - zero at the actual point of tangency, tapering up to full superelevation at the start of true curvature. Since the spiral easement is about twice as long as my longest cars, the effect is gradual and gentle enough to avoid problems.
Good looking, derailment- and problem-free superelevated curved track does not lend itself to slapdash, fast-as-you-can-go tracklaying. The time spent in careful construction will be more than compensated when trains run over the line without a bobble or a hiccup.
Happy tracklaying.
Chuck (modeling Central Japan in September, 1964 - with superelevated curves)
Superelevation is generally about three to four degrees, depending on the tightness of the curve. This is essentially three to four inches height difference between the outer rail and the inner one (the actual calculation uses trigonometry, but it works out pretty close to "number of degrees equals number of inches difference" for superelevation of prototype track. At greater than five or six degrees, this approximation breaks down).
Let's go with three inches of height difference. In HO scale, that's .034 actual inches.
I use narrow masking tape under the outer rail for my superelevation, like this:
Each layer of masking tape is about .005 inches thick (depends on brand), so for .3 inch superelevation (corresponding to 2.6 inches full-size), I use six layers of tape. This is the final result:
There a bit more about how I superelevate curves on this page of my website:
http://www.thecbandqinwyoming.com/CM%20-%20Super-Elevating%20Curves.htm
Mark P.
Website: http://www.thecbandqinwyoming.comVideos: https://www.youtube.com/user/mabrunton
The September 1997 issue of Model Railroader has a tutorial on Superelevating HO flextrack on page 90.
I have found superelavation adds realism for trains in curves.
I may not have done this right but I am modeling a branch line switching layout; this has only one "main" line turn (about 25" radius). For my roadbed I am using stacked cork that is used for gaskets (you can get this in auto parts stores) and it is 1/16" thick. To shim for superelevation I just use one thickness of the gasket material; that's about twice the .034" you suggest, Brunton, but it looks prety good; my trains don't look like they're about to fall over on the turns.
Tom