First off, I understand that a helix is evil and that looking at one might turn me to a pillar of salt. But sometimes there just is no other way to get there from here.
If, as I have assumed from reading every helix related post on every forum I can find, the critical factors for loco performance with a given number of cars are grade, and “curve induced effective grade”. It would seem to me that an oval helix would exhibit the same effect on the train as a circular helix with the same length and rise.
So… if I have an oval with a 24" curve on each end with 18" of straight track on each side, the total length per lap would be 187", which should be equivalent to a circular helix with a 30" radius.
Does anybody other than me believe this? Anybody tried it?
Many people have tried, with varying degrees of success – unfortunately, you still don’t get a free pass from physics. So this still works better with broader curves than you are contemplating. Here’s why:
Assuming HO and 4.25" railhead-to-railhead through each lap, your 187" lap length would be about 2.3% nominal grade. But the 24" radius curves will add roughly 32/R (where R is the radius) of effective grade due to friction through the curves. That equals an additional 1.33%, so on the curves, your train will be dragging as if it is a 3.6% grade.
In the failed oval helixes I have seen, that variation in effective grade around the loops increases the tendency for the cars to stringline (derail across the center of the curve), especially with mixed length and/or longer cars.
A more reliable approach would be to redesign for 28" (or broader) curves in a traditional circular helix so that the forces are more consistent. Even broader would be better, of course.
Successful oval helixes tend to have broader curves and (significantly) longer straights between curves. This allows room for compensation, reducing the grades through the curved portions to better match the nominal grade through the straight portions. These must be carefully engineered to provide transitions. I’ve seen this done a few times successfully, but it’s a significant engineering project.
In my experience, there is almost always another way to achieve the operational goals without compromising reliability. [:)]
OK, I think I get it. The difference in the drag between the cars in the curve and the cars on the straight would create the same type of effect that you get with cars of different weight which Armstrong has told us for 50 years is a significant problem.
Any chance I could get away with 26"? The difference means the fascia of the helix would be 8" from a track that needs to be visible vs 4". I’ll be putting the helix track directly on 1/4" MDF (with extra supports) so I only need to climb 3 3/8’ per turn. My yard tracks, passing tracks and staging are all sized to support 3 SD’s, 15 hoppers and a caboose.
BTW, I read in another post that the LDSIG has done some testing and found 28 to be a better factor than 32 for the effective grade calculation. Is that correct?
LION uses foam in many different ways. This is the “Gold Street Interlocking”
Bottom deck is scrap plywood with a 1/2 Celotex deck. All of the roadways are 1/2 Celotex, none have additional decking under the Cellotex. Various kinds of foam are used for the risers. Note the number scribed on the riser 11 3. That is 11 and 3/8 inches. Foam is used for the tunnel walls.
This blob is the first thing people see when they enter the room. I call this tangle the “Gold Street Interlocking” not that I have any interlocking here, but between DeKalb Avenue and the Manhattan Bridge, NYCT has many levels of tracks twisting about over and under each other. I will put more decoration on this complex so that it will look like a cutaway view into a subway tunnel.
From Bottom to top, the first track is the South Ferry Loop, next the lower level of the Nevins-Lenox helix loop, above that is the upper level (Smith Street-Lenox Av level)of the same helix. Above this are the two single tracks of the Smith-42nd St helix, and the top level is the 42nd -Coney Island level of the same helix.
The foam is easy to work with, especially for construction elements and risers. The Celotex product is no longer available. Celotex is still in business, but this particular product went out of production with the introduction of fire codes. Fortunately, the layout is finished as far as construction goes, and so I’ll not need to worry about whay I will use to replace this product not that I have used up the 10 sheets that we had in our warehouse.
Speaking of warehouse we have over 100 sheets of some sort of extruded white foam about 1.5" thick, but it used to be the roof of the library building, It is not smooth any more, but I can use it for other parts of the construction.
From my reading and viewing of other layouts - disclaimer: I have never built a helix myself - those that have built a 24" radius helix in HO have been unhappy with the result in the long term. The small radius helix gets replaced with a bigger radius or gets eliminated after being under-used. It seems to take a 30" radius or better to bring long term happiness. Even then, operational problems such as speeding up in the helix because of the train being hidden, are real. Will a 26" radius work and give you the desired results? Only you can answer that, and probably not until after you build it. [:)]
The divisor of 32 for the effective grade calculation came from John Allen in the late 1950s, IIRC, and was based on the rolling characteristics of the trucks available then. A change in the divisor based on current day truck characteristics does not surprise me. However, I believe there is a flaw in the formula/model - that there should be a knee in the curve around the 21" +/- radius point to account for RP25 wheels starting to slide rather than roll with the outside wheel having an effectively larger diameter. Above that radius - actual radius depends on suppositions about actual track gauge and how the wheel climbs the outer rail - there is much less impact on rolling friction by the curve. Below the radius, the impact of the curve is much more severe due to the wheel(s) doing a combination of sliding and rolling rather than just rolling.
Quite apart from all the foregoing, a small radius imposes a critical problem for the helix builder. That problem is that with a smaller circumference (due to your intended short radius), you get less ‘run’ in the rise-over-run formula that yields your height gain. The small radius you appear to prefer, and to resist enlarging to what we recommend, means that you will only achieve the height you need, but more importantly the clearance between the decks of the helix, by having a very steep grade, something in the order of 5% or even more.
The way the physics works, as you dangle the cars trailing the locomotive closer and closer to vertical on an ever-steeper grade, you have gravity taking over in a big way against all the couplers holding the train together. You also force the cars to want to adopt a straight line dangling behind the coupler of the tender or diesel. The result is stringlining. On a shallow grade, say below 2.5%, resistance against the wheel sets rolling in their trucjs and the flanges against the rail head are the greater concern, but they get reduced as your grade reduces, and your grade reduces when you enlarge the radius comprising the helix.
The inescapable conclusion from everything you have been provided in this thread is that you must have a helix with a sufficient diameter AND grade to clear above itself between rounds or decks, but also to allow your trains to both roll without stringlinge and actually make it up to the height you need by the time the helix ends above it all.
For helices:
Short radius = tight curves with very steep/too steep grades and problems for the locomotive.
Larger radius nearer to 30+ inches = greater ease in deck separation, ease of train movement due to easier grades, but a much larger module and footprint. Also means more time the train is likely to be unseen and unappreciated if it happens to be in a mountain…as mine is. FWIW, mine is twin tracked w
Possibly. Personally, I’d consider a redesign rather than risk reliability issues. Your call.
Since the elevation gain is railhead-to-railhead, that may be a bit tight. I’m not sure that MDF is the optimal choice for maximum stiffness with minimum thickness.
Empirical testing is currently underway for future publication in the LDSIG’s Layout Design Journal that suggests 32/R is more accurate for HO, particularly for radii below 30". Beyond 30", it may be that some adjustment is needed in the formula owing to lower friction wheels and trucks today. Free-rolling trucks don’t help much with the flange edge sliding friction against the rail in tighter curves, as Fred alluded to.
I have built a double track helix with 30" and 32.5" radii. Personally, I wouldn’t go much smaller on the radius. Yeah, they are space hogs…
There is enough evidence gained from actual helix construction to make an informed decision that will result in a helix that is trouble free. As Byron notes: your call. I checked the evidence and followed the suggestions (everyone is saying the same thing) and I don’t have problems with the helix.
My helix climbs 3.75" per revolution and the sub-roadbed is 1/2" plywood. No roadbed. This leaves me with 3" of clearance… The foot print is a 65" square in the layout room. I used all manner of tricks to get the most out of the helix, so that the space was as well utilized as possible. I have a PDF on helix building - send me a PM if you would like me to send it to you.
As for the r32 thing, there is a lot of debate on this and little empirical data, In my own experience I have found that free rolling wheels sets and equipment that operates well makes a huge difference. I can pull a 25 car reefer train up my helix with a single cab forward (no traction tires). IMHO The whole issue is pretty much moot unless you are pushing the edges of the envelope - large layouts with very long trains or tight curves.
My engineering mind say’s it could be possible to camber the curves with reverse super elevation by raising the inside rail so the train slopes to the outside to minimize string lining.
Unfortunately, folks have tried that and failed in actual practice. The build-up of forces through the couplers and the relatively high center of gravity of cars still causes stringlining – but the cars do eventually flip over quite spectacularly.
Gents: While I don’t have and don’t intend to have a helix, the discussion was really fascinating. I never realized Helixes (Heli?) were such a complex and well studied subject. Thanks to all of you for the tutorial. OWEN W
First off, thanks to everyone that shared their experience.
Secondly, while my goal is certainly to consume as little real estate as possible on this helix, my intention was never to go forward with a configuration that I wasn’t convinced would work (assuming I executed it correctly).
So the bad news is that I’ll have to dedicate more space to the helix than I had originally hoped. If I had gotten a “probably” on 26" I might have gone for it, but “possibly” wasn’t very encouraging.
The good news is that in redesigning to accommodate 28" I realized that 30" was not significantly worse. Additionally, expanding to 30" made a 2nd track a possibility. I had originally planned to have 2 helices in the same space. One going up to staging above the upper level, and one going down to staging below the lower level. With the second track added, I’ll go all the way down to a single staging yard under the first level. This will free up more space above the helix on the upper deck than I’m losing by expanding the radius.
I’ll have a 30" radius “UP” track and a 27.5" radius “DOWN” track.
So, “In my experience, there is almost always another way to achieve the operational goals without compromising reliability” holds true.
Thanks again for the help.
Oh, BTW, What is the optimal material for maximum stiffness with minimum thickness?
I have not built a helix, but have built two 2-level layouts with winding grades to go from one level to the other. If I had the room for a helix, my thought is that - for best performance - it would need the widest radius curves I could use, along with the lowest percent grade I could manage.
All things being equal, having a short sided oval would not make any real difference in what a given loco could pull. A long sided oval, the length of a typical train, would be of help - but at the cost of a lot of space.
I’ve seen some helixes in action, and they are really cool, but they are a major project for sure.
in another thread, an analysis showed that the minimum grade was dictated by radius, clearance height and plywood thickness. In other words, the minimum grade is determined by space (i.e. diameter).
So what options are there if you’re limited in space (e.g. 48") and you can’t operate on the grade dictated by a conventional helix design?
a very long oval, more like a bathtub or barbells. While clearance height and plywood thickness are still fixed, radius is not. The length of the straight ways primarily determine the minimum grade, and yes, they are long (i.e. ~optimal helix radius x PI).
it’s an option. at some point it’s no longer a helix
Of course the thickness of plywood/roadbed, etc., come into play in designing a helix. If one had a two level layout and had the room for it, that would surely be the way to go. But, they do take up a lot of room and are not the easiest thing to build.
For my 11x15 two level layout (and its predecessor), I built a 2 percent grade throughout the length of the transition trackage. It winds down, makes a left, makes another left, goes about 10 feet and makes a 26 inch radius 380 degree turn, goes back along side the way it came, until it is at the final lower grade altitude. This one, and the previous one, work just fine, but there are limitations on single engine powered train lengths. Thank goodness for F, E, and GP consists…
A number of friends and clients have laminated overlapped sections of tempered 1/8" hardboard with good results. Laminating 1/8" material can make for less material waste. Personally, I like 1/2" quality plywood spaced by nominal 1X4 stock (actually 3 1/2"), since only the lowest turn need be measured precisely – but that does not create the thinnest possible roadbed, obviously.
Several years ago I built a rather large (four track) helix for my layout. Minimum radius was 36 inches. I had 24" straight sections between each semi-circle to reduce the grade. Here’s a link (with lots of photos) to the construction page: http://www.thecbandqinwyoming.com/CM%20-%20Behemoth%20Helix.htm
the table below indicates the straight length required between each semicircular end of an oval helix to provide the indicated grade for the specified radius for a rail-to-rail change of 3.5" (3/4" plywood). For example, 2% grade with 22" radius curves requires 69" straight sections, so the total size is 44" x 113".
