I have been experiment in HO to find out how long easements should be with a 2% grade. They were very important when I did a 4% grade on last layout but with all my experimenting on 2%, you don't seem to need much. I run short stuff both steam and diesel and the longest car (flat car) at 50', all others 40' or less and shot trains. Also I am talking functional, not visual (I will give it all the extra room I have for that.
IMO, this is one of those problems that you have to defy common sense to create. If you keep joints in the subroadbed away from the vertical curve, use 1/2" or thicker plywood, and firmly attach it to at least 2 risers before the vertical curve, you can't bend it enough to cause a problem.
PS - real plywood - not OSB. OSB is satisfactory when used in sheets, but not in strips as in cookie-cutter subroadbed.
PPS - it should go without saying but no turnouts on the vertical curve.
I have the right to remain silent. By posting here I have given up that right and accept that anything I say can and will be used as evidence to critique me.
I used OSB on my last layout and didn't have a forumula. But I needed to go from no slope to 2.9% so I did it over approx 16 feet I'd guess. I started with slight grade of maybe 0.8% for around 4 feet, then 1.4% for another 4 and then 2.1% to maybe 2.5 and finally 2.9%. That way the grade change was fairly gradual but controlled. I didn't have any issues with couplers mismatching or anything so it seemed to work fairly well. What I listed isn't exactly what I did but from memory as it was about 5 years ago.
Rio Grande. The Action Road - Focus 1977-1983
I am using Woodland scenics inclines on a foam base with cork on top, works flawlessly.
I used cut-out 3/4" plywood as subroadbed, and it forms its own vertical easements as the risers are added. With ground zero securely fixed in place, and likewise the highest point of elevation, I simply add a riser at the mid-point of the grade, putting the height of the roadbed, at that point, at one-half of the total rise. All that's needed then is to install intermediate risers that meet the lower face of the subroadbed at its then current height, neither pushing it up nor pulling it down.
Wayne
doctorwayneI used cut-out 3/4" plywood as subroadbed, and it forms its own vertical easements as the risers are added.
But naturally the elevation of the risers is going to form/force the easement and grade so the riser height needs to be figured and deliberate. I measured the distance between risers and was able to calculate the grade needed to transition, over a distance, from one grade to the next to reach the maximum grade.
riogrande5761I did it over approx 16 feet I'd guess
That's way too much. 16 inches would have been a better choice. If you had a 16 foot transition at both ends, your grade in the middle would have been significantly steeper than required.
From March 79 MR:
Well I have done some experimenting. Now this was all done with stuff I plan to run so my observations will not apply to larger stuff. I tested with 40' boxcars and small engines like a Proto 0-8-0. You can accually get away without easments on a 2% grade (very small coupler height change and looks fine, small easement looks a bit better with a very slow moving train). Now with a 4% grade it still works but coupler height change is more noticable, dosn't look as good. Add a 2" easment and it looks ok and functions fine, 4" look great. So I will go with about a 2" easment at the bottom and top and will give it more if I have the room on my 2% grade. Never had problems with room in my old house, downsizing dose have its cost but I do have a heated enviorment controled room now. Would not be asble to work on old layout for alot of winter and don't have tgo worry about noise bothering tenents above a garage.
carl425That's way too much. 16 inches would have been a better choice. If you had a 16 foot transition at both ends, your grade in the middle would have been significantly steeper than required.
I could be over estimating from the transition distance from memory, but I was being very conservative.
OTOH, I'd think 16 inches would be way too little distance to change from 0% grade to essentially 3%, especially with the coupler off-set that would happen with 89' long freight cars. Heck, an 89' freight car is about 12.5 or maybe 13" across the pulling faces so I fail to see how 16 inch transition is going to keep couplers from mismatching at the transition point. Even modular layouts with no grade transition but not well leveled can cause long cars to uncouple.
riogrande5761I'd think 16 inches would be way too little distance to change from 0% grade to essentially 3%, especially with the coupler off-set that would happen with 89' long freight cars.
Sorry, my point (less than clearly stated) was that 16 inches is way closer to the "right" answer than 16 feet.
shouldn't the length of easement depend on the change in grade
and the purpose of the easement is to limit the change in grade to avoid coupler problems
greg - Philadelphia & Reading / Reading
gregcshouldn't the length of easement depend on the change in grade
Yes, check the formula in the MR article I posted.
gregcand the purpose of the easement is to limit the change in grade to avoid coupler problems
That's one of the issues, but you also want to keep all your wheels on the rails and the flanges between them. A 2-10-2 might lift the center drivers off the rail or a long car with a small amount of play in the bolster might lift wheels off the rail. Also, a locomotive pilot or coupler air hose might dig into the ties.
I was being very conservative on my last layout with the grade change. The truth may be somewhere between 16 feet and 16 inches. I would think several feet transition would be needed for long cars to avoid mismatches, at minimum.
carl425 gregc shouldn't the length of easement depend on the change in grade Yes, check the formula in the MR article I posted. gregc and the purpose of the easement is to limit the change in grade to avoid coupler problems That's one of the issues, but you also want to keep all your wheels on the rails and the flanges between them. A 2-10-2 might lift the center drivers off the rail or a long car with a small amount of play in the bolster might lift wheels off the rail. Also, a locomotive pilot or coupler air hose might dig into the ties.
gregc shouldn't the length of easement depend on the change in grade
gregc and the purpose of the easement is to limit the change in grade to avoid coupler problems
riogrande5761 doctorwayne I used cut-out 3/4" plywood as subroadbed, and it forms its own vertical easements as the risers are added. But naturally the elevation of the risers is going to form/force the easement and grade so the riser height needs to be figured and deliberate.....
doctorwayne I used cut-out 3/4" plywood as subroadbed, and it forms its own vertical easements as the risers are added.
But naturally the elevation of the risers is going to form/force the easement and grade so the riser height needs to be figured and deliberate.....
I actually do have a curved # 7 1/2 L.H. turnout on a 3% grade with 32" radius overall curve and it works fine. I used a Shinohara turnout and a Tortoise switch motor. I will say it was the most difficult turnout and took the most "tweeking" of all the 21 turnouts on my 12'x8' HO layout.
I think that you are actually saying "do not place a turnout along the transition from horizontal to vertical." And you are right! That would not work. I have read, however, in the past to avoid turnouts along a continous curved grade.
jgraffiI actually do have a curved # 7 1/2 L.H. turnout on a 3% grade with 32" radius overall curve and it works fine.
Turnouts, curved or otherwise are fine on a grade, but you don't want them on the vertical curve that is the transition for a change in grade.
Back to vertical easements. I've transition from no grade to 1.8% in about 48 inches.
First from 0 to 0.5 over 14", then 0.5 to 1.0% over13", then 1.0% to 1.4% over 14", then 1.4% to 1.8% over 26". Then maintain 1.8% for about 8'.
Main concern is long car coupler offset at transitions. Seem reasonable?
riogrande5761First from 0 to 0.5 over 14", then 0.5 to 1.0% over13", then 1.0% to 1.4% over 14", then 1.4% to 1.8% over 26". Then maintain 1.8% for about 8'.
That seems overly conservative, but fine if you have the room. Clients and friends seem to be OK with about one longest-car-length per percent of grade ... or even a little less in a pinch. Just make sure that you have another longest-car-length beyond the transtion before any turnouts.
Byron
Layout Design GalleryLayout Design Special Interest Group
My new layout will require a 5% grade to do what I want in the space I have.
I mocked it up with Kato Unitrack and scrap plywood I had on hand to verify it was possible. Ran FA locomotives and 40/50 foot freight cars with no problems and no special calculations.
Maximum train length with th FA was four freight cars and a caboose.
-Kevin
Living the dream.
So looks like I'm going about half % per longest car or less transition. Yeah prefer conservative.
Think maybe a broad turnout may be ok at the 1/2 percent tail end being so conservative?
riogrande5761Think maybe a broad turnout may be ok at the 1/2 percent tail end being so conservative?
Personally, I wouldn't put a turnout in the middle of any change in grade*. That's where I am conservative. If you are talking about right at the end of the vertical transition, I'd personally still want a bit of space between the end of the transition and the turnout.
* RG knows this, but just for the record, turnouts are fine in the middle of a grade; it's just where grades are changing that it can be a problem.
I found that if I lower the track being crossed over, I can get the necessary clearance with a more gradual incline. I have a 7" difference in height but lowered the lower track by 1" gradually where the upper track crossed over. I had 25' feet to create my incline. Worked great!
Marlon
See pictures of the Clinton-Golden Valley RR
riogrande5761So looks like I'm going about half % per longest car or less transition. Yeah prefer conservative.
I agree with Byron that you're being overly conservative. You can easily calculate what is acceptable for any car.
Consider a worst-case where the grade starts immediately with no vertical curve. If car 1 has his back wheels barely on the grade, and car 2 is on level track, the couplers will be at maximum deflection. Being in N-scale I don't have one on hand to measure, but as I recall the face of an HO coupler is about 1/8". Since grade = rise/run, take the coupler face as rise and the distance from the coupler face to the truck bolster as run (3" for a long car maybe?) and we see that .125/3 = 4.17% required to force a decoupling. I think limiting the grade change to less than 1/4 of that would work 99% of the time. So the 1% per car length that Byron suggested works out pretty well.
I don't mind doing such calculations when necessary, but the method I used, as outlined earlier, worked partially due to the 3/4" plywood roadbed, which wasn't forced into an abrupt rise (or drop at the top). It also worked because I had a pre-determined amount of rise necessary (15.5"), and a set length (45') in which to accomplish it. Because the 45' is laid out over a number of curves of varying radii, the risers could not be spaced at regular intervals.
However, I did have the track laid and fastened to the 3/4" plywood roadbed for the entire climb, so simply created a 10' long train to use as a measuring device, which yielded both the total length of the grade and a method to easily find its mid-point. Once the riser was added at the mid-point, with a height equal to half of the total required, the rest was simply common sense, as the stiff-ish plywood formed very acceptable vertical easements at both the bottom and top of the grade. All I needed to do was add risers without altering the grade.I also added superelevation in the same manner, using the 10' train as a measuring device to determine the mid-point of each of the curves.
All of the fastened-in-place risers were marked with a pencil line denoting the top of each open-grid crossmember to which they were attached, then the screws holding them in place were removed.Next, the riser closest to the mid-point of each curve was lifted so that its pencil line was aligned with the top of its respective crossmember. The bottom of that riser was then pushed towards the outside of the curve, until the look of the train sitting atop the curve seemed to be realistically superelevated.The riser was then re-screwed into its new position, with only the inner end of the line matching the top of its riser.All of the other risers within the curve had also been deflected outwards, to diminishing degrees, as they moved away from the mid-point, so it was a simple matter to lift them (without altering the deflection) until the inner end of their pencil lines matched the top of their respective crossmembers, then fasten them in place.
I don't run any high-speed trains, so the superelevation is purely for esthetic reasons, but even slow trains look good working around the multiple curves, and I've had no issues whatsoever with unwanted uncouplings.