I would use Masonite or bender board instead of foam.

Just make your sub roadbed on the helix wide enough for clearance and install the Masonite or bender board to the sides to make walls to keep the trains from falling off of the helix from either side.  You just need to make sure that the sides are low enough to fit your hand between the top of the side and the bottom of the next level to allow access for re-railing the trains.  Also allow enough clearance between the rails and the sides to allow your fingers to fit for re-railing.

I would also use something like Atlas re-railing sections for the straight tracks.  That way, if a train derails anywhere in a curve, you have the potential of it re-railing itself when it gets to the straight.

Sub roadbed thickness

The biggest disadvantage I can see to using spline material for the subroadbed is overall thickness of each layer.  You have about 2 inches of sub-base with the 1-5/8" clearance.  This means you MUST have a steeper minimum grade for a given circumference of helix.  Check out how Charlie Comstock did his helix - I can't see any advantage of using splines over his method, and it would just increase complexity.

  Just my $0.02  

not very efficient

This design is very wasteful in terms of thickness and resulting in a much steeper grade than could be accomplished otherwise.  I use 1/2" plywood to make my trackbed as thin as possible, so one of my HO helices has a trackbed thickness about half of what yours will have.  A 22 inch radius helix like yours could have a grade of just over 1 percent for N scale, but doubling the thickness doubles your grade. 

The problem is you need crossbearers to support the spline, which would not be needed if the trackbed was wider, or your struts were attached to the sides of your roadbed instead of under it.  I would also dispense with the cork; it's pretty pointless in a helix, and again, it wastes space.

Will your "full scale" hand fit in 1 5/8 inch?

Even when we model in smaller scales, our hands or other appendages are still full scale, and may need to fit into the space for maintenance or re-railing trains.

Concept for N-Scale Oval/Elliptical Spline Hybrid Helix 2

I appreciate the comments. 

My N-SCALE engines and cars are all within the 1 1/4" to 1 3/16" high range except for the recent Micro-trains giraffe gondola which will be placed in a scene someplace on the layout.

I will use re-railers on the straight sections.  Thanks for that idea.

The total thickness of the hybrid/oval helix from the cross piece (for the threaded rods on each side), the spline sub-roadbed, cork roadbed, and track is 1 inch and the railhead to railhead separation is 2 5/8".  If need be during the building of this concept, I will raise the separation between the lower track and the bottom of the next turn plus lengthen the straight section to keep the virtual grade.  I measured my NMRA N-scale clearance gauge today and the minimum clearance is about 1 5/8 inches. Once I build a test section over the next week or two, I'll see what needs to be adjusted, especially for my fingers and hand which aren't N-scale.

I have calculated the circumference of the oval (which is an approximation technique) using a couple of different methods and each time got 14 feet per turn.  Using the chart on page 48 of the John Armstrong book, "Track Planning for Realistic Operation," 1979 edition, or page 82 in the 1998 edition, a rise of 2 1/4" in 14 feet is pretty reasonable.  The fact that this is for a hybrid type or oval/elliptical helix cause the virtual grade to increase to about 2.6% but when the straight sections are added, it brings that down to around 2.2%.  If I stretch the straight sections out even more, then I believe that the grade should be reduced accordingly.

I just want to try this because after 30 years of using plywood for curves, I'm ready to try something else. I just have never really liked the way 3/8" behaves on curves as well as all of the waste.  I still have some in my garage from the last layout that I built. 

I will take some pictures when I get something built and let you know how good or bad it went.

Good ideas

Adding the straight sections to reduce the grade of the helix sounds is a great idea. Adding a few rerailers sounds good, too.

The spline sub-roadbed seems like an unnecessary use of time and materials, unless you're planning on putting the helix on display. If you are planning on using it for sound deadening, I think that there are other options that would be thinner and cheaper. Automobile sound insulation can be used under the cork roadbed with minimal additional height. Adding rubber grommets/washers between the nuts and structural roadbed would also help. Simply adding a fabric "curtain" inside or outside the helix would help, too.

Sometimes doing extra handiwork is a nice escape, but if you're just trying to get something done, why add the extra costs and time.

If you want to go low-tech to prevent equipment falls in case of a derailment, consider a simple canvas disc snapped to fill/cover the inside of the void of the helix. It won't prevent the fall, but it will provide a cushion to avoid damage.

The Economics of Building a Spline Helix

Thanks for the comments. 

The economics and cost of the helix built from spline is definitely less than it would have cost me if I built it from a good cabinet grade plywood.  The 1/8 inch sheets of masonite or hardboard are anywhere from $8-$9 each and I need 5 to build the five-turn helix so the cost is $40-$45 per helix.  A good sheet of 1/2 inch or 3/4 inch cabinet grade plywood costs between $30-$40 per sheet.  The helix would require at least 2 to 3 sheets because of the way you would have to lay out the curves for the five levels.  There would also be a lot of waste. In comparison to using spline, say you used 2 sheets of good grade plywood, that would cost roughly $60.  Sure you could use a cheaper grade of plywood, B-C, with lots of knots and even thinner, say 3/8 inch.  You get what you pay for in plywood.  I have seen may helix installs on the web that use thin plywood and you can see a lot of flexing.  Some even have used laminated sections of masonite hardboard or even 1/4 inch plywood (such as a luan) to stiffen the helix but then you've added cost because you double the amount of material.  I used 3/8 inch plywood in my previous layout and I really did not like the way it performed especially when it was cut into curves.  That is the main reason why I wanted to use something different.

The first helix that I built is documented in this blog. The second one (for the other end of the layout) is now under construction and will be as strong as a 3/4 inch piece of plywood with minimal waste during its construction.  The only waste that results from using the hardboard splines is a pile of hardboard sawdust.  Sound deadening was not a factor as to why I switched to using spline roadbed for the helix but rather my experience over the years in using plywood on previous layouts.  I do use plywood in the city and staging areas but not in the areas between where I believe that spline will work better.

Remember this is a hobby and not a job.  I am not in a hurry to build the new layout so I take my time to enjoy it while trying to learn new techniques that I have observed being used over 30 plus years of being a model railroader.