TheRailwire
General Discussion => N and Z Scales => Topic started by: Ed Kapuscinski on September 16, 2019, 03:24:25 PM
-
I'm doing some more track planning and am trying to decide on if I can live with 20" radius "scenic" curves (as opposed to purely functional ones).
What's the thinking on 20" radius curves and things like SD50s and 89' flatcars?
Is it "you can do it where you need to" or "yeah, it'll look fine"?
-
I have approx. 20" min mainline curves and am happy with them even running 89' flats. They don't strike me as being too tight but I wouldn't want to go any tighter. I would also want to make sure any S curves have a straight section of AT LEAST 89' between direction changes. Easements as much as possible help with initial (toy-like) overhang issues when entering curves as well.
-
Depends on space, depends on the prototype if that's the goal. It passes the threshold of "looking dumb" by a fair enough margin personally.
-
It also depends on the terrain being modeled, high speed mainline or branch line operations, etc.
-
It's the section along the top wall. The space between the S curve in the center is only 6".
This is all laid out using sectional track. In reality it'll be built with easements and flex.
[attachimg=1]
And in case anyone is wondering, this is what "playing out doing the NCR from Baltimore to York in my current basement" looks like. You're looking at Baltimore right now, with Mount Vernon in the bottom left and proceeding through Timonium by the bottom right (and yes, if I include Timonium, I'm definitely including the cow palace).
-
Will you have the cow palace in odorama?
-
I would try to bump-up the 6" straight section in the S curve if you want to run 89 footers...especially if using body-mounts. Less is just asking for problems....or at least a visual distraction. Otherwise, nice track plan!
-
What's the thinking on 20" radius curves and things like SD50s and 89' flatcars?
Is it "you can do it where you need to" or "yeah, it'll look fine"?
I wouldn't try it with body-mount couplers.
As for "look fine" that's completely subjective. Perhaps try a flextrack mockup?
Ed
-
...also, the two tracks (or perhaps just the farthest one) on the far right as you enter the yard from the bottom. Those look like they have a potentially troublesome S curve arrangement if longer stuff is going to run there as well.
-
Spiral easements will increase the visual quality, as well as the functional reliability exponentially. The main problem with long equipment is the sudden entrance of a car or engine into a curve, and the immediate overhang on the ends that produces. Spiral easements makes the transition from straight to curved much much smoother.
I was going to recommend 24" minimum cosmetic curves with 18" being the absolute mainline minimum operational radius, but an "eased" 20" minimum cosmetic radius will look and function very nicely with longer, modern equipment, as will your other functional non-cosmetic lesser curves.
Another thing to think about is superelevation. Ya don't want too much, but you want it to be noticeable. I think Kato's Unitrack superelevated track is too much, so something less than that looks more prototypical.
It's a huge rush to see your train with long cars and engines smoothly enter an eased curve, gradually tip slightly into the superelevated curve, then straighten out when the curve becomes straight...with the rest of the train still superelevated behind the engines. A superelevated "S" curve is at least twice as nice, with the front of the train leaning one way, and the center/end leaning the other way as the train negotiates the two curves....YEAH! :D
Although it's tempting to make all of your cosmetic curves the minimum radius, especially if you have broad 20"+ curves. As John Armstrong has suggested to us, extra-broad curves in a couple of places (or more) really add to the looks of your layout. He called them "Photography Curves", and I've incorporated several in my layout designs that exceed 130" in radius. I'm not talking about half circles of track, but just a 1/8th or 1/16th slice of a circle.
Also, in "S" curves, there's no reason to make both of the main radii the same radius. One at the minimum radius, with another at a larger radius will also add to the looks of your trackwork.
Time to get back to work!
Cheerio!
Bob Gilmore
-
I wouldn't try it with body-mount couplers.
Ed
Why not? I would think 20" would be fine (in HO scale that would be 36.8" radius).
My next layout will have 21" minimum radius on the main.
Mark
-
...also, the two tracks (or perhaps just the farthest one) on the far right as you enter the yard from the bottom. Those look like they have a potentially troublesome S curve arrangement if longer stuff is going to run there as well.
Yeah, I've been watching out for those. That track is just a helper pocket, at best, and should only ever see slow speed movements onto the non-diverging route. Unless you're talking about the one coming out of the curved turnout heading into the year. I might want to reevaluate that one too.
-
Another thing to think about is superelevation. Ya don't want too much, but you want it to be noticeable. I think Kato's Unitrack superelevated track is too much, so something less than that looks more prototypical.
Interesting thought with that, but I think I'll skip it. The NCR was decidedly NOT high speed railroading (thanks to being laid out in 1830). But that's part of what gives it it's charm!
-
Yeah, I've been watching out for those. That track is just a helper pocket, at best, and should only ever see slow speed movements onto the non-diverging route. Unless you're talking about the one coming out of the curved turnout heading into the year. I might want to reevaluate that one too.
This is the S curve I was referring to...
-
20" should look fine... at touch of superelevation will help things look nice as well. I work up to about five layers of masking tape on the main line. Branch lines, like the one described in the next paragraph, don't get that treatment.
I have one 's' curve and I simply let XTrackCAD do the heavy lifting for me. It is a 20" radius, into an easement, about 1.7" of straight, easement out the other side into a 29" radius. I designed and planned it to run body mount 89's next to 40' cars. I laid it out on a table prior to committing to the plan to ensure things would work. Thus far, the only problems I've had are the very light Kato 70 ton hoppers next to +85' cars. Other short cars - both body mount and truck mount - haven't had any notable problems.
-
This is the S curve I was referring to...
Oh yikes. Yeah. That'll be on the revision list.
Thanks!
-
20" should look fine... at touch of superelevation will help things look nice as well. I work up to about five layers of masking tape on the main line. Branch lines, like the one described in the next paragraph, don't get that treatment.
I have one 's' curve and I simply let XTrackCAD do the heavy lifting for me. It is a 20" radius, into an easement, about 1.7" of straight, easement out the other side into a 29" radius. I designed and planned it to run body mount 89's next to 40' cars. I laid it out on a table prior to committing to the plan to ensure things would work. Thus far, the only problems I've had are the very light Kato 70 ton hoppers next to +85' cars. Other short cars - both body mount and truck mount - haven't had any notable problems.
Sweet. 20 it is.
There are a few places where I use 18 or so, but generally I'm trying to keep it to 20+.
Thanks!
-
Ed, If you are going to use easements, I suggest that you do your planning with the easements drawn-in. While an easement doesn't really change the necessary space for the radius of the curve, it has a big effect on the amount of straight track leading to a curve. For example, that (now-famous) S-curve at the top of your drawing has a 6" piece of straight track between the two 20" radius sections going in opposite directions. But, an easement into a curve needs to be at least as long as your longest car, preferably a little longer. Since you are talking about 89' cars in N scale, that means that your easements will need to be about 7" each, and you will need 2 of them where you now show only one 6" straight section. Keeping that straight section, you need to push those curves apart by more than an additional foot! So, "planning" with sectional track is not going to give you a good idea of what you can do with flex track in a restricted space.
There are track planning programs that have "easements" available in their planning tools, but my experience with those in AnyRail is that they are too short, and not really proper spiral easements. So, they will neither "look" right nor have the full benefits of a correct easement, and will fool you into having too little room for a correct easement. However, there are several discussions of how to lay out easements in model railroad track planning books and on-line. You can use those to get an estimate of the length and angle for say, a 7" easement to a 20" curve, and just draw one in your track planning software to copy and paste wherever you need it. Actually, you can spiral your easement down to shorter radii, and mark the lengths for shorter and longer radii so that you can use that same easement drawing for any curve on your layout, just including more of it's length for tighter radii and less for looser radii.
-
Well, I even have a few 18" spots (where I couldn't avoid it) and they handle 80' passenger cars with body mounts fine.
But I did want to ask, looking at the curves in question along the top of your plan, there are long stretches of straight track come in and out of that "S". Couldn't you get rid of some of that dead-straight track and make broad sweeping curves in there, which should allow you to flatten out your 20" "S" into much broader radii? Maybe you want to preserve the look of those parallel straight tracks in the passing siding, but I think they'd look just as nice if they were curved.
20" is nice. But the places I have that are 24" are definitely a little nicer looking with long cars (and long engines) on them.
-
I would think 20" would be fine (in HO scale that would be 36.8" radius).
I remember a pic from a while back of a long flat car (might have been by @GaryHinshaw, but I couldn't find it) with body-mounted couplers tipping considerably (tho not derailing) as it ran around a curve that (again IIRC) was in the 18"-20" range. It's the overhang that causes the trouble, esp. when coupled to a car with a shorter overhang.
Looking at the drawing, I would consider perhaps using some kind of ultra-large radius curve instead of an S-curve (especially if superelevation will be considered). The effect would be something like this:
(https://www.railpictures.net/images/d1/1/6/0/4160.1162000800.jpg)
One other thought WRT an S-curve, the scenery perhaps should have something like a hill, river, etc. which suggests that the railroad was built that way to avoid the obstacle, as opposed to just because someone decided to put in an S-curve for no apparent reason.
Ed
-
Ed... BTW.... that is a serious yard lead you've got there.
I like it.
-
I have 18" radius curves. 89' cars run fine with no problem -- and some of them, like my Autoracks, all have body-mounted couplers.
Superelevation was mentioned -- I use .020" Evergreen styrene strips under my outer rails. Looks and works perfectly.
-
Atlas SD50s run fine on 16 inch radius. I have one curve that sharp, and my SDP40Fs, kitbashed with SD50 chassis, ran fine. MT 89 ft flatcars will also run on it, with the stock couplers body mounted. For those, I did have to narrow the "swing arms" for flange clearance. If you cut them off completely, and screwed the coupler to the car, it wouldn't be a problem. My way I didn't have to drill holes in the car, just glue the arms to "carrier bar" on the underframe to keep it from moving.
As mentioned, the biggest problem is what they're coupled TO. Even the prototype has limits on coupling long and short cars on sharp curves.
-
I also have 18" curves with .020" superelevation as well, but most of my mainline is on a 2.2% grade which is what complicates operations. Light cars with body-mount couplers and short wheelbases (e.g. 89' flats) do tend to string-line (https://www.therailwire.net/forum/index.php?topic=24108.msg264568#msg264568) if they're near the front of the train. That's when it's time to call out the pushers. :lol:
-
The tradeoff with 20 inch plus curves comes up when you design the layout. A loop that size against a wall means you can't reach the wallside track if the loop is against the wall, leaving you with tunnels or access hatches and fewer options for what else will fit on the layout. Published HO track plans start to get very useful.
-
Ed, ditch the sectional track in your plans. Go with flex, so you can plan easements properly--you may otherwise wind up with some nasty surprises when it comes time to nail down the track geometry.
-
I'm an old HO guy, and I also 'upgraded' from a 3x6 HCD, so when I did my next layout.... the thought that I would have minimum 13" curves on the main (26" HO!!) seemed perfectly wonderful in N when I'd had an 18" world in HO.
So the current N ATSF layout is primarily double-track with 15 on the outside and 13 on the inside; hidden storage tracks and a reverse loop are 11. Maximum grade is 2.5%. Visible curves are all superelevated, with easements.
That was by necessity, not choice, to fit the layout in a small room (which it still is) established that.
And, I'm running 30 car trains with 85' passenger cars, autoracks, and a full 15-car 89' Super C. And a mix of body-mounts and truck mounts on both the passenger cars and freight cars.
Other than the Atlas piggyback flats, it's all worked reliably. Those stringlined, and I also had to take all my 85' standard passenger cars to bodymounts from truck mounts. Works fine, even on the 11" reverse loop.
Now, I envy the guys that can even consider 20" radius and have the space to pull it off. But for appearance, here's what I've observed.
If you are on the INSIDE of a curve, and the layout is relatively high (mine is 52" off the floor) you don't notice it much at all. If you are on the OUTSIDE of the curve (and I only have two situations of that on my entire layout) it's a lot more noticeable. If I had the option of doing it over, I'd deliberately max out radius where the curve was outside and not stress about inside curves to an aisleway/view. And not stress about the couplers/mounting much. You're going to notice the curve radius a lot more the lower the layout is to eye distance, too - my HCD layout was 36" when I first built it and when I raised it up to 48" it was like a whole new layout. Also, the higher the layout the better even a little bit of superelevation on the curves looks, it's really good.
One of the reasons that T-trak looks as toylike as it does is directly related to the height issue being tied to the height of a standard folding table. Think about that one, and also the height difference of N-trak. Different visual impression entirely. It's not just radius.
The only piece of equipment I've had that about drove me nuts was my Hallmark 4-8-4, and that wouldn't even do 15" reliably with four flanged drivers and a long fixed wheelbase. I blinded the center axle flanges and both after about 5 years as a shelf queen, and now it handles everything just fine, but that was a tough mental sell to do that stunt on a brass locomotive.
-
But I did want to ask, looking at the curves in question along the top of your plan, there are long stretches of straight track come in and out of that "S". Couldn't you get rid of some of that dead-straight track and make broad sweeping curves in there, which should allow you to flatten out your 20" "S" into much broader radii? Maybe you want to preserve the look of those parallel straight tracks in the passing siding, but I think they'd look just as nice if they were curved.
I think the actual implementation would be very different in that regard. Definitely all flex and less straights like that. It IS the NCR after all. But for now I'm just doing that for expediency and to kinda "prove out" things like grades, what can fit, etc...".
-
Ed, ditch the sectional track in your plans. Go with flex, so you can plan easements properly--you may otherwise wind up with some nasty surprises when it comes time to nail down the track geometry.
Yeah. If I decide to go forward with this I'll revise it all with ACTUAL stuff before cutting the first board. For now this is all a VERY rough draft.
Meanwhile, I'm also contemplating the "Tudor Branch" on some HCDs in the middle of the room based on something like the Reading Cluster.
-
18 inch with easements would be preferrable to 20 inch radii without. I found a few problematic situations with 18 inch mins on the Oregon Belt, though they were also related to grades. Having a curve start where a grade begins sometimes causes some finicky equipment to act up. If you are body mounting couplers, or thinking of using Z scale ones or the MT scale couplers, 18 might not be a cosmetic curve for all equipment.
On longer engines (SD 60s, C9-44s, etc....) you might find you need the long shank coupler to get the engine around a superelevated curve on a grade without pulling 89 foot cars with body mounts off the tracks. The equipment needs to be in perfect shape.
If I had to do it again, I would shoot for the NTRAK 24 inch minimum radius for my mainlines. It definitely helps run longer trains with fewer issues. Course I can't widen my basement, but I might get the minimum up to 21-22 inches.
And watch out for those curved turnouts. :D
-
Despite my tolerance for sharper curves, I think you're pushing it on the distance between curves on that top "S". By the time you add easements and superelevation, looks like trouble to me. You've got to get enough room in there to get an 89-footer off the curve and flat before you start into the next one.
And is that an actual spot you're trying to model or just throwing an S curve in there to be interesting? You do have that around Seven Valleys for sure.
My ATSF layout has a specific spot I fell in love with decades ago, a reverse "S" curve on the main at Chalendar, AZ (between Williams and Maine), with the snow-capped San Francisco peaks in the background. Slamming through an S curve with heavy superelevation on both main line tracks and no speed restriction, downgrade into Williams, through the Ponderosa pines. The spot was the subject of the cover on "Canyonlands and Super Chiefs" book by McMillan, a Howard Fogg painting for UTLX, and lots of calendar shots. That was my excuse for throwing an "S" curve in for no particular good reason and learning to live with the consequences over the years.
https://cdn.shopify.com/s/files/1/0072/5972/products/Canyon_Lands_book_large.jpg?v=1332612194
I'll admit that pretty much the entire layout was designed to make that 'the spot', like PRR guys set up the room to do the Curve.
The telephoto shots on that curve are just fantastic to this day and the curve-to-curve distances are surprisingly tight if you look at the cars into and out of the superelevation area. Glad I did it, but when you look at the track plan it does look like I made a mistake somewhere.
-
I suggest making both curves of the "S" 30" radius or more. You can adjust the straight sections on each end to accommodate this. I'm not familiar with the NCR, but even if you're trying to duplicate a scene I seriously doubt the RR used there the equivalents of 20" radii, so you'd probably be making the model look more realistic.
Instead of a long tangent along the front of my layout I built the track to a very wide radius -- just guessing, it's an 8 foot or larger radius (I laid it by eye). The trains look really cool going through it. Most of my other curves are 20+" R. but I have a double curve with a 15" outer and 13" inner curve. That's only on part of that loop tho.
-
Light cars with body-mount couplers and short wheelbases (e.g. 89' flats) do tend to string-line (https://www.therailwire.net/forum/index.php?topic=24108.msg264568#msg264568) if they're near the front of the train.
Yep, that's the pic, thanks!
(https://lh4.googleusercontent.com/-9WF9F0CJ9rM/T7ExxLjKiKI/AAAAAAAAEEg/3HE2Elo7OrA/s640/DSCN1159.jpg)