Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Clinchfield & Interstate, June 1952

14 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    January 2008
  • 23 posts
Clinchfield & Interstate, June 1952
Posted by jaartrip on Tuesday, January 25, 2022 5:19 PM
Hello all.
This is my first post on what will hopefully be a long and successful build thread.  My apologies if this is in the wrong location on the forums.  I'm finally embarking on my layout construction after over 20 years of wanting to build, and a few messy false starts along the way.
I've always been in love with Appalachian mountain railroading, particularly in east Tennessee and southwest Virginia.  In my mind, there are no two railroads that I think better embody this type of railroading than the Clinchfield and the Interstate roads. 
I think at this point, I have nailed down the vast bulk of my build constraints.  I'm planning on a postcard, semi-prototype build.  What I mean by that is I plan on remaining as faithful as possible to the prototypes, except where it's impractical or unfavorable to do so, and taking “postcard” scenes of famous locales on both lines.  There’s also a few technical, construction driven reasons for the postcard approach, which I’ll discuss in more detail below.   I'm an engineer by trade, so attention to detail and accuracy are a big deal to me, and while I don't consider myself to be a "rivet counter" (yet), I've been planning meticulously for quite some time in what I want the finished product to look like.  I'll be working in N scale for my build.  While I would love to do HO, there are several constraints preventing this:

·                     The room.  The room is nominally 146" wide by 177" long; with a 30” x 52” "chunk" taken out of one of the corners.  Being a second floor room, it also has sloped walls due to the house roof line that start to slope in about 70" up off the floor.  The lighting is also especially poor in the room, there is one sad low output fixture, and given the roofline issues, the lighting will be something that will need to be dealt with as a build in in the model, maybe in the form of a valance or something of the like, rather than modifying the room.   On the bright side (no pun intended), there is a generously sized 47" x 49" closet with a 24" door attached in one corner of the room, which will serve as a helix area, which brings me to my next constraint.

·                     Levels.  To fit everything I want in such a small space, even in N scale, I'll need two levels.  I'm planning on the lower level to be around 36" up, and the second level to be around 45-50".  I would like to go a bit higher with my level heights, but I think the sloped rooflines will preclude any additional height.  That's something to be confirmed.  The Clinchfield will occupy the lower deck, and the Interstate the upper one.  I plan on having several negative (below track) features on the CRR so, it naturally has to be the lower deck. To preclude it causing any viewing issues.  If was on the upper, that would hang down into the lower deck and cause problems.

·                     Radiuses. I have figured out I can do an around the walls, with a folded inverted peninsula and leave enough room for 30" aisles.  My wife has affectionately nicknamed it "the toilet seat" as that is the general shape of the aisle way cutout.  While 30" aisles is a bit on the tight side, it will only be me, no other operators, so, a bit tight is ok with me.  We don't have children, so, that's also one less concern in this build.  I'm planning on my Clinchfield deck maintaining at least a 20" radius on the main line.  More typically it will be 22-24". This is driven by the locomotives I am planning on running, which I'll discuss more on below.  On the  Interstate I will probably go a fair bit tighter, as the real Interstate was extremely tight in some locations, and ran Alco RS-3s which can be a bit more tolerant of curves.

·                     Era and motive power.  As in the title, I'm planning on June of 1952.  This is specific for a few reasons.  One, the Clinchfield had begun the process of heavily dieselizing, so I can run EMD F bodies, SW and NW switchers, GP units; and while their big 600 series articulated steam power was in its final legs, they were still hammering the rails hard up into 1953.  Two, this is also when the Interstate was in its heyday, and they had their gorgeous Creamscicle RS-3s.  Three, it also was when the Interstate interchanged with the Southern, L&N, N&W, and the Clinchfield, and the Clinchfield interchanged with the Interstate, N&W, the B&O, the C&O, the Southern, and the L&N. This gives me the opportunity to run a massive array of motive power, at least on the upper Interstate deck, under a broad range of banners and still remain honest to the prototypes.

·                     Grades.  Ironically, despite being a mountain railroad, the only grades will be in the helix.   This is due to the double deck configuration. I want to keep spacing of the levels as consistent as possible, for the best viewing, as I want to be able to pull visually stunning coal drags with the big articulated CRR steam and grades will negatively impact that.

·                     Key Locations.  There are several key locations on both prototypes that I want to model.

o   Elkhorn City, KY.  On the CRR deck, on one side of the peninsula I want to model the north interchange at Elkhorn City, KY.  While I have had some trouble finding pictures of Elkhorn City in the early 50s, I think I can piece together an accurate model of what it looked like.  I specifically want to model the route 80 car bridge over the Russell Fork, the car yard, and the coal loaders.  I don't have space for the wye, which is unfortunate as it was one of the most important parts of the interchange, but, it is what it is. 

o   Pool Point, KY.  Out of Elkhorn city, I'll highlight the Pool Point crossing over the Russel Fork.  It's only about 3 miles out of town, but the grade is steep, and the river that was at about the same elevation as the rails in Elkhorn City is now about 167' below the track.  Given that I'm not doing grades, this change between the rails and the river poses a challenge.  My plan is to faithfully recreate all my intended focus points on the railroad, so, to get around this issue, rather than trying to flow from scene to scene, I'm going to make very deliberate, abrupt "postcards" on the layout.  This will be made easier by the mountain scenery, which will allow me to chop up the scenes and isolate them from one another.  I'll utilize tunnels through large hills and other natural features to help break it down into scenes.  The CRR had several tunnels on its main line, which makes even this approach somewhat prototypical.  This will also serve a structural purpose for the benchwork which I will get into in my next point, but, first off the rest of the plan. 

o   Dante/St. Paul, VA.  Coming out of Pool Point, the next focal point will be Dante/St. Paul, VA.  I'm going to compress the two towns/areas together, for a couple of reasons.  One is the location of the closet in the room.  The closet is the helix point between levels, and St Paul is where the two railroads handed off (technically it was Miller yard, but, very close enough...) so that point is pretty well physically dictated in the room/layout.  Dante will allow me surge capacity for cars, and was a very important stop for the CRR on the north end of its line for locomotive servicing.

o   Copper Creek Trestle, Clinchport, VA.  On around the lower level will be the next focal point, which is Copper Creek Trestle.  Again, I'm planning on modeling the bridge full scale, which will be a magnificent 7' long and 18" high.  Interestingly, the Pool Point bridge, and the Copper Creek Trestle are about the same deck height, but the Copper Creek Trestle is almost 4X as long as the top deck truss bridge over Pool Point. 

o   Kingsport, TN.  The last major focal point on the lower deck will be Kingsport, TN. This is an area where I am still rolling around various concepts and ideas.  Kingsport in the 1950s is pretty well documented, and there was a fairly large yard along with several excellent industries, so, unfortunately this area will also have to be heavily compressed/cropped.  Elkhorn City and Kingsport will be on opposite sides of the peninsula in the middle of the room, and I will use the postcard approach and some clever hiding of the trackage around the end of the peninsula to make the lower deck a continuous loop run, where trains will roll from Kingsport around the corner and start the run over again in Elkhorn City.  I should also point out at this time that I am going to deviate from the prototype too to make the CRR main line a double track.  This will allow me to run significant coal drags continuously in opposite directions on the lower level with no input from the operator (me) and be visually impressive.


On to the upper deck. 


o   St. Paul, VA.  Coming out of the helix, on the top deck, I'll compress the distance between St. Paul and Norton, VA. I'll also have a wye on both levels at St. Paul.  There are double wyes in St. Paul, one of them in the east where I’m pretty certain the N&W interchanged with the IRR, and one in the west where the IRR interchanged with the Clinchfield.  By modeling one wye in each level, and basically splitting St. Paul in half between the levels, in accordance with the prototype, I can turn around locomotives on either level without having to travel up or down the helix.  The wye on the second level also allows me to split the tracks and build a bonus run of the N&W/L&N in parallel with the IRR.  The real prototypes laced back and forth over one another's tracks as they ran the hollows from St. Paul to Appalachia, VA, so this allows a fantastic level of prototypical modeling. 

o   Norton, VA.  In between St. Paul and Appalachia is Norton, VA. I'll model this area somewhat compressed.  In reality, the IRR split at Norton with another wye, and there was a very important spur that went north, the Dixiana and Critical Fork run, but I don't have the space for it, so there will be a staging track peeling off that will represent that major point.  There will most likely be a small, compressed yard in Norton as well. 

o   Appalachia, VA.  From there, the rail line will come into Appalachia, VA.  Appalachia was a major hub, with all the major railroads of the area interchanging there.  There was also another wye there, which I may or may not model, depending on my available space.  Appalachia will be on one side of the peninsula, and I will actually have a gap in the mountains/divider backdrop back where it meets the room wall.  The rail line will travel through this gap into Andover, VA, just like the prototype, where it shot the gap in the mountains and headed north up the hollow. 

o   Andover, VA.  Andover was a major yard and service point on the IRR, so I'm going to try to fit as much of that in as I can.  That will happen on the opposite side of the peninsula from Appalachia. 

o   Derby/Stonega/RodaThe line will split coming out of Andover and one side of the split will be the Derby leg, and the other side will be the Stonega/Roda leg. All three of these points were major mine load out points on the IRR, right up there with Critical Fork and Dixiana.  While small areas, they provide a tremendous modeling prototype, with the mines, the coke ovens, the company towns/stores/housing, the creeks, and the tight, winding scenery.

·                     Benchwork.  One major challenge in all this planning and design is that I don't want to make any structural modifications to the room.  I don't want to have to repair walls if we end up potentially moving in the future.  To that end, I don't want to mount any benchwork to the walls, and don't want to make any changes to the electrical/lighting.  I've been using my engineering skillset to design some innovative, free standing benchwork.  My plan is to utilize 3/4" and 1/2" plywood, in conjunction with foam board for the benchwork and Masonite board for the backdrops.  Given the complex shape of the room, and the need for freestanding benchwork, I'm utilizing my 3D printers to design and construct a 1/6 scale model of the model railroad before beginning any construction.  I've drawn up the room and the benchwork in SolidWorks, and am using 3D printed models of the floor plan, benchwork and backdrops, in connection with foam board for the walls to model the entire room and layout before I cut the first single stick of benchwork plywood.  This model will also help me identify the needed negative steps in the benchwork, and if any of the benchwork will land at any switch points on turnouts and require modification.

·                     Track.  I'm planning on hand laying the track.  I'm in no rush to pound out poor track; operation and appearance are critical, and I personally think done right, handlaid track is a cut above in these regards.  I plan on a #6 turnout as the bare minimum. For the Clinchfield main lines, I'll most likely step up to a #8 or a #10.

 In mynext post, I'll update on the CAD models, 3D prints, and 1/6 scale model.
  • Member since
    March 2013
  • 427 posts
Posted by Colorado Ray on Tuesday, January 25, 2022 6:53 PM

Sounds like a very well thought out and detailed plan.  Should make for a great layout.  Looking forward to following your build.


  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, January 25, 2022 6:53 PM

Ok, now that I have the photo hosting/posting/linking thing sorted out... on to some good stuff.

Here are the CAD renderings of the room and the double deck benchwork.  I am still working on modeling the helix, and the support legs, but, I have designs in my head for both of those.  I've modeled both levels of benchwork, as well as a model of myself, which I can move through the model to get an idea of clearances and how things will look.  I also modeled a typical perspective view, looking from my viewpoint, with my head tilted downward at a 15 degree angle.  Dragging my "humanoid" around the layout will show me exactly what I will see.

Here's a cutaway section view giving a little better look at the peninsula:

And a cutaway section view to show the persective viewpoint:

  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, January 25, 2022 7:07 PM

And this is the above models, starting to be translated into something more tangible and visual...

The upper deck backdrop and benchwork plans, 3D printed at a 1"=6" (1/6) scale:

Here is the room that I am working with, looking in the door "full scale":

And here is looking in the door of the 1/6 scale 3D print and foam board model:

A few more views of the model:

The Clinchfield system map:

And the Interstate map, overlaid with the other roads that worked the area as well:

  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, January 25, 2022 7:17 PM

I started translating the track plan over to some foam board, where I could do 3 dimensional concepts for the scenery and start planning where to add/hide supports for the upper deck, but I quickly realized that I should probably hold on any more development work on the foam board track plan before I finish the 3D printed parts.  As those are based on the room constraints I really need to let those drive the limits on the track plan planning.  I have about another 4-6 days worth of printing to do to finish all of the components needed, but, it was a fun exercise, and showed that the approach is going to work, as well as providing some much needed visualization and encouragement.  I scaled and photocopied FastTracks templates to use as cutouts for my turnouts to ensure that they were to size.

  • Member since
    July 2006
  • From: west coast
  • 7,496 posts
Posted by rrebell on Wednesday, January 26, 2022 8:02 AM

First off don't worry about the walls, very easy  to patch sheetrock and touch up paint. Next comes the plan, looks like a duckunder, if so, bad idea unless you have removable sections and even then a pain, only ones I truly ever liked were ones with a bridge, best was a bascle bridge. As far as lighting goes, my entire train room ceiling slopes, put in track lighting and used LED bulbs, can sometimes be picked up cheap on the secondary market, also repair is as few screw holes to fill.

  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Saturday, January 29, 2022 9:41 PM

Thanks for the reply.  It would certainly make things easier to attach it to the wall.  Im still rolling things around on that front, and the lighting.  I fully agree the duckunder is much less than ideal, but, it causes a major disruption to my track plan to split it out into just a simple bridge.  Maybe I can figure a way to make it a swing away or something of the like.  Still some improvements to be made there.

I've made more progress on the model.  The last of the 3D printed parts finished, so, I've started working on the lower deck track plan and 3D elevations.  I've made it about half way around the lower level.  A note on the real world pictures below:  None of them are mine, they were found in an image search, and were not marked with a copyright, so, fair use with no intended financial gain or infringement.  I have some that are marked, but will not be posting those they will be for my own private use reference.

The 3D printed base:

The track plan taking shape:

Elkhorn City, KY:

Looking south/Southeast, through the Elkhorn City Yard.  That's a quick and dirty mockup of the US-80 Bridge over the Russel Fork.

The real US 80 bridge, built in 1912.  Abondoned now, used as a foot bridge.  I'm not sure when it was taken out of service.

Interesting riveted footing details:


The real Elkhorn City yard, in the 70's:

Quite the mix of power on display, and notice the tracks!  All that sand and oil and coal, no ties to be seen anymore!

Leaving Elkhorn, on the way to Pool Point:

Pool Point, Ky,

And the real deal, in the late spring.  River is very high in this picture:

Looking west, back towards the tunnel/Elkhorn City:

It's hard to gather just how big the bridge is, but, here are some pool jumpers standing on the bottom cord, notice how the gusset plates are chest high!

Dante, VA Yard and Turntable:

Side view of Dante Yard, Notice the riverbed in the edge of the scenery, in front of the tracks:

Dante towards St. Paul (The Helix in the closet):

And some shots of the model with the upper deck frame sitting on top, a virtual walkthrough:

Elkhorn City:

Pool Point:

Looking towards Dante Yard:

Looking down the aisle, Elkhorn City on the right, and Dante/Pool Point on the left:

A little further back, on the other side of the central backdrop.  Kingsport will be on the immediate right, and back in the rear left in the Dante turntable/St. Paul wye.


  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, November 14, 2023 11:42 AM

Hello all...

Well, its been a bit.  My apologies for vanishing into thin air for a while, life threw a few curveballs along the way that brought progress to a complete standstill for quite a while.

I do have some major updates, however!

I took the past feedback to heart (thank you to those that replied), and started looking at methods of attaching the deck directly to the walls in the room, however, a few issues came up with this.  The wife still wasn't too keen on me poking a ton of holes in the wall, and, when I started looking at how structural the walls were in the room (it's a bonus room over our garage) I didn't get a warm and fuzzy about having a bunch of weight cantilevered off the studs, so, I reverted back to my original plan of self supporting benchwork.  This turned out to be a huge benefit for a few reasons that I'll go into as I move forward with more posts as I share actual construction progress.  

The above delay in my timing was actually a blessing in disguise as well, as it gave me time to sit down and really critique my benchwork design as I still wasn't sold on it, and I ended up making a major change.  It occured to me that I was wasting a ton of space by having one central peninsula down the middle of the room.  Not only did that waste a bunch of space creating aisles, it also made for very long, straight scenes, which, for anyone that has visited southwestern VA/eastern TN and followed a rail line, that is nowhere near prototypical.

To that end, it dawned on me I could do 2 peninsulas sideways, and be much more efficient with my space, so, I went back to the virtual drawing board, and came up with the below:

Npt only does this fit great in the room, it breaks the layout up into distinct short  scenes that i can isolate, and am calling "snapshots", and lends itself really well to the prototype track plan.  When I saw this come to life on the screen, I knew instantly that was it... The shape I had been wanting.  

Even with the serpentine aisles, it never dips under 30" of width, and the corners create viewing pockets where you can stand and allow someone else past comfortably.

  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, November 14, 2023 12:00 PM

On to design details!

Self supporting benchwork...sort of.  I broke the layout up into what I called benchwork modules.  Originally, I had planned to actually make the layout in sections, so if we moved (something we have had a habit of doing a lot of) I could in theory take it apart and take it with us.  In reality, this hasn't panned out, and I quickly gave up on that once I realized just how difficult that would be to do in reality.  We also have been where we are at for almost 3 years now and love it here, so, I don't forsee a move in our future.  

In the below 3D, each module has a backer board, made out of .703" thick furniture grade plyood, and you can see the seams between each module.  this backer board is the "spine" of the railroad, and supports each module.  The modules all interlock together and are bolted to one another, so that when everything is fully assembled, it is all pressed tightly to the walls of the room, without a single screw being driven into the walls, providing tremedous rigidity without attaching anything to the house itself, much like a keystone arch becoming self supporting once all the components are together.

All the benchwork with the exception of the peninsula support structure got made from plywood.  I went with plywood for a few reasons.  1, its very dimensionally stable.  2, it's flexible, as long as you don't need a piece bigger than 48" x 96" you can make pretty much whatever shape you want.  3. It's cheap... Kind of.  When I started adding up the cost of individual 1x4 boards vs the cost of sheets of .703 thick furniture grade plywood and then ripping them down to dimensional lumber (3.5" wide) the plywood actually came out cheaper.  

I went through each module, designing each part and assigning it a part number, building a bill of materials, cut list, and manufacturing sheet.  Each plywood sheet got a cut sheet, so I could optimize how much of each sheet got used...  I factored in the kerf of my blade in each sheet cut list to keep everything straight and organized, and ensure that I didnt come up short at the end:


  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, November 14, 2023 12:29 PM

The next step was to figure out just how to attach all these boards together.  One of my requirements was that I did not want any sort of supports blocking the view of the railroad, so all of the support of the shelving had to come from the attachment of the benchwork lumber to the backing boards, so the layout was floating, essentially.  I initially loooked at using metal shelving brackets, but, didn't like the astronomical cost or lack of flexibility for locating them where I wanted them, so, I determined that the wood would need to be the primary material.  One big drawback of making plywood boards is that because it is laminated together, it doesn't have a high tolerance for screws run into it on the "end grain". 

I bought a small 2'x4' sheet of the .703" thick plywood and started playing around with some concepts, both for ripping the boards out of the sheet, as well as for  different methods of attaching the benchwork boards to the backer boards.

My first test was to confirm that the end grain would split when a screw was run into it...  Yup.  No good:

Then, I thought about doing a pocket of sorts, where I could use a large headed fastener to clamp the board directly to the backer board.  I used my 3D printer to create some fixture tooling to help me make the geometry:

In the end, this was a ton of work, and didn't provide any better support than what I ultimately settled on...

My next attempt was to simply drill a hole through the board, and use a small section of pipe to act as a clamp, with T nuts sunk into the back side of the backer board to thread the retaining bolts into.  This was very strong, but was difficult to put together cleanly and square, as well as being expensive when the cost of the pipe and anchoring hardware was factored in.  One benefit of this was it provided a pass through for wiring, though!

I then tried pre-drilling the end grain and running large head screws into it from the backside of the backer board.  While this worked well, I didn't like the idea that the screws would ultimately end up captive behind the backer board.  If (when) I inevitably need to make a modification to the benchwork, It would be impossible to remove the screw, and would get messy fast.

So, what did I end up settling on?  A glued joint, with (4) Kreg pocket screws.  Each type of joint was load tested to failure very scientifically...  I had a bathroom scale I was standing on, and would bear down on the end of the test joint with my weight until it broke, noting how much of my weight the joint was carrying based on the scale reading,and how the joint failed.  Ultimately, A glued T point with 4 Kreg pocket screws (2 on each side of the board) was just as strong as the pipe clamp joint, while being much easier to manufacture, and since the screws are in the benchwork board vs the backer board, I can in theory take them out, break the glue joint, and modify the bechwork as needed in the future. 

Another thing I noted as well was that with all the joints except for the initial test of screws direct into the end grain, I got a significant amount of deflection in the joint, on the order of 1" downward 24" out, and could apply over 75 lbs of downforce before joint failure.  With a joint every 12" in my design, that translates to hundreds and hundreds of pounds on each shelf distributed out over the area of the railroad, something that gave me more comfort in the Kreg pocket screw design.  My deflection scale test was significantly worse than anything the benchwork will ever see in reality.

  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, November 14, 2023 1:43 PM

Once I had a good plan for how to assemble, I whipped out the credit card and plunked down, making the first major investment into the railroad.  Getting serous now!  I ordered the plywood, as well as the foam for the scenery, and the tempered hardboard for the backdrop and fascia panels.

I knew that A) this was going to be a process, turning 13 sheets of plywood into a 12'x17' room full of double decker benchwork, and B) that the wife would want the garage back ASAP, so, I organized everything as best as I could before I began to make it as efficient as possible:

I knew that the key to cutting and assembling all the benchwork and get it to fit precise in the room so that the backboard-module assembly approach would work would be to fixture as much as possible.  One big hurdle I had was that I did not have a table saw to rip the sheets of plywood into the boards, so, I devised a jig out of 3D printed clamps bolted onto a piece of 1" square aluminum tubing that acted as a fence to guide my circular saw and allow me to repeatably rip 3.5" wide "boards".  The four 3D printed clamps fit around the edge of the sheet, were clamped tightly with screw clamps, and created a fence exactly 3.5" wide that my saw shoe would ride against as I made my cut. 

I used the stack of foam as my work surface, putting a piece on top that got beat up in transit, so I could cut through the plywood into the foam and not have to worry about damaging another sheet of plywood underneath while supporting both the sheet and the part being cut off securely.  I ripped several 3.5" x 8' long "boards" off my first sheet and lined them up to see how well my fixturing jig worked...  I could hold between .005 and .010" tolerance.  Nice!

You might have noticed above in my cut sheets for the plywood dimensions that I was working in decimals, which is odd for woodworking, but, not odd at all in the CAD world, or for the level of accuracy I was looking for.  In order to precisely mark the lengths of my boards for cutting once the 3.5" strips were ripped, I created 3D printed T fixtures, and used cheap modified plastic calipers to allow me to mark off the location on the board for the cross cut.  I had a 6", 12", 18", and 24" plastic tee that would connect to a 6" caliper, giving me a 12", 18", 24" and 30" measuring set that could mark accurately and repeatably down to the .01":

Next, I had several small blocks that would need to be manufactured as part of the support leg assembly, so, I 3D printed some fixturing for both ripping blocks to size, as well as for locating the screw holes in both the blocks as well as the legs themselves.  The legs were made from a 3.5" board that was full width at each end, but ripped in half lengthwise, then those two sections glued and screwed together at a right angle to form an extremely strong but compact and lightweight leg.  I load tested a leg to several hundred pounds and got zero buckling or deflection. 

In the end of the leg, I used a knob with a 1/4-20 threaded stud, threaded into a T nut pressed into the end of the leg to allow for support and adjustment of the module height, and to ensure that it didn't damage the carpeting in the room:

Drill fixture for leg blocks:

Fixtures for locating screws and holes in leg assembly:

And an assembled leg:

Once I got rolling, It went quickly:

And the time spent planning out each sheet of plywood paid off.  This was the waste I had left over after several sheets of plywood.  I was utilizing about 96% of each sheet:


  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, November 14, 2023 2:21 PM

I worked through the lumber module by module, building them as I went, starting at the back of the room and working my way towards the door.  The first one, althought one of the simplest, was also one of the most challenging, as I was figuring out how well my plan would actually work:

One thing I unfortunately didnt think to take good pictures of was how the boards interlock.  In the above picture, the boards on the floor you can see where I notched them, with some notches facing upwards, and some downwards.  Each board was notched half way through, with the cross board carrying a mating opposit notch, so, they would interlock.  The joint was then drilled and glued/screwed, creating a super strong intersection.

The weight of each module is supported by the legs under it, the height fine tuned with the threaded knob at the bottom of the leg, and then the benchwork boards clamped together, holes drilled, and bolted to one another using 1/4-20 bolts and T nuts.

  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, November 14, 2023 2:44 PM

As I mentioned in my introductory post, the intent is for the upper deck to be the Interstate railroad, running from Appalachia, VA to St. Paul, VA, where it interchanged with the Clinchfield.

My room has a closet in it, which is where I've hidden the Helix that will interchange the levels, and represent St. Paul, VA.  Im not a fan of Helixes, at least in the traditional sense.  As I see it, round stacked helixes have two inherent flaws which causes two issues, 1: a significant amount of drag, as the train is always turning as it is climbing, so you have a lot of side force in the wheels which translates to load on the loco, which means short trains, and 2:since the levels are stacked directly on top of one another, there is a minimum physical radius and rise per rotation of the helix. 

My closet is 47" wide by almost 60" deep, so, using a traditional helix, I would have had a radius of around 23", depending on design, and would have had to rise at 2.25% per loop in order to provide sufficient clearance for the locomotive, the track, the plywood, and any supporting structure.  2.25% was quite a bit more than what I was wanting, so, back to the virtual drawing board, I devised this, what I am calling: "the rectangular, double decker behive helix":

As I see it, there are lots of advantages to this design:

  • I was able to follow the shape of the closet, drastically increasing the length of track between the levels. 
  • It allows for a significant amount of the climbing to be done in a straight line, only turning when I have to.  
  • It allows each set of turns to be cut from one continuous strip of plywood, eliminating the need for any joints except where you "reset" back to the outside diameter... more on that below.
  • It fully utilizes a sheet of plywood, Vs half circles for a traditional helix.
  • The longer run means my grade is a constant .65%.  Yes... .65 percent!
  • The trains are easily reachable from above on any level, no derailments getting trapped between levels
  • If i have a derailment, there is a high likelihood the level below will "catch" the train and minimize damage.
  • By "resetting" back out to the outside turn when you reach the minimum inside radius (possible because the lower outside diameter is now several turns below), you can effectively stack as many of these on top of one another as you want.  In my case, I needed 2 to keep my grade to .65%.  I was targeting a grade of .75% or less.

At this point, I have the helix frame built, the risers all cut out, and the plywood bed cut.  Once I get the final elevations set for each shelf, I'll poke holes through my backdrop and connect it all up.

  • Member since
    January 2008
  • 23 posts
Posted by jaartrip on Tuesday, November 14, 2023 3:14 PM

And all of this brings me to the final post for now, where I am at today.  

I have all the foam base rough cut, with about 75 percent of it glued down, with 95% of the backdrop installed.  Once it is all glued down, I will go back and trim the front edge to the final shape and profile. 

One big advantage of having plywood backer boards is that it does a great job supporting the backdrop hardboard, and I can run screws into it literally anywhere I need to to get the perfect backdrop shape.

I installed the foam in two different ways between the levels, and the lower level was somewhat of a mistake.  I installed the backdrop board on that level first, then realized I had to meticulously trim each piece of foam with a knife edged blade in a jigsaw to fit the foam to the curve of the backdrop.  I use a circular template while installing the backdrop boards to try to keep the radius of the corners in the backdrop the same, but, there are a few that ended up needing to be larger or smaller to fit correctly, and those ended up being a real pain to cut the foam for, as I had to basically hand fit the foam piece with no jig. 

For the upper deck, I did a little better and installed the foam first, then for the backdrop, I simply set the board on top of the foam, effectively elevating the backdrop 2" above the plywood backer board. 

In the end, this is going to be a happy accident, as, one of the big problems I need to tackle quickly is the poor lighting in the room.  The lower deck will be illuminated by lighting mounted to the underside of the upper deck, so, it will not be too much trouble to light well, but, illuminating the upper deck was going to be a challenge, as installing a hanging fascia above for lights with no support from the walls, and the angled sides where the walls meet the roof line of the house would be tough.

Luckily, with the stepped backdrop on top sticking up 2" above the plywood supporting it, I now have a little locket behind the backdrop that I can install lights to, and run wiring for them in, all very clean.  I am working on the lighting design, but, I don't want it to draw too much attention from the upper deck, I like how open and flowing it is, so, my general concept plan is to have lights on curved stems, reaching out from behind the backdrop in an inverted "U" shape, with spacing sufficient that the lighting is hopefully pretty uniform.  This is a problem I need to tackle quickly, as the single window in the room is more or less blocked at this point, and the one dismal light fixture above does a very poor job of illuminating. 

One other thing I want to point out is that I only have any grade on the railroad at the helix.  I know that seems counter-intuitive... Building an appalachian mountian railroad and not having any grade.  I did this for two reasons: 1) I spent a bunch of time and mock-ups getting the two decks the perfect height.  The lower one is the perfect height for my roll-around stool, and the upper is the perfect height for standing, while having lots of vertical clearance between them for the scenery, lighting, and effects I have planned.  Grades would invariably mess with that.  Secondly, I want to run looooooong trains, big coal drags that the Clinchfield and Interstate were known for.  I have purchased a couple of N&W Y6b Mallet locomotives that will make "guest appearances" on the Clinchfield deck, as well as planning to buy some Clinchfield Challengers.  I want big trains to go with big locomotives, and grades would detract from that.  That is the reason for the base foam layer, to allow for negative scenery features, below grade, to give the illusion of grade without the need for it.

At two points on the Clinchfield deck, the scenery dips way down.  The smaller of these is Pool Point, Ky, and the larger one is Copper Creek Trestle, VA.  These dips will allow me to model both these magnificent bridges at almost full scale (I had to cheat the length on the copper creek one, due to the space I had available by about 150 scale feet, so it is ~1150' long, instead of 1300), and will be focal point "snapshots" on the lower deck. 

  • Member since
    February 2021
  • 1,057 posts
Posted by crossthedog on Tuesday, November 14, 2023 8:03 PM

Wow! Just... wow!
I'm gobsmacked at what you've put together in the time that it took me to assemble my simple, unscenicked kidney bean of a layout. Amazing. Can hardly wait to see this with track on it.


Returning to model railroading after 40 years and taking unconscionable liberties with the SP&S, Northern Pacific and Great Northern roads in the '40s and '50s.

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!