Trees are bigger than you think

I’ve been doing some research into realistic scenery for my planned layout, and will be testing some of these concepts on my mini-layout.  Since I’m modeling the nearly-completely-forested Southern West Virginia and Eastern Kentucky regions, trees will play a major role in the design.

I’m going to need a good looking, but fast way to mass produce trees, for sure!

I found this website about Virginia trees (which are rather similar to West Virginia trees), and also this PDF from the WVU Extension Office.  From them, I was able to deduce some of the following typical tree heights for common trees in this region:

  • White Pine: 50-60ft, 1-2ft trunk
  • Shortleaf Pine: 100ft, 4.5ft trunk
  • Black Pine: 50-75ft, 1-2ft trunk
  • Hemlcok: 60-100ft, 2-4ft trunk
  • Cypress: 80-130ft, 5-10ft trunk
  • White Walnut: 70ft, 3-ft trunk
  • Black Walnut: 100ft
  • Bitternut Hickory: 100ft, 2-3ft trunk
  • Shellbark Hickory: 60-100ft, 1-2ft trunk
  • Birch: 70-100ft, 2-3ft trunk
  • White Oak: 60-100ft, 2-3ft trunk
  • Post Oak: 50-80ft, 2ft trunk
  • Elm: 60-70ft, 4-5ft trunk
  • Sycamore: 140-170ft, 10-11ft trunk, head can be 100ft across
  • Sugar Maple: 100ft, 3-ft trunk

This is only a partial list of trees common to Virginia.  Clearly most of them grow to well over 50ft, and many as tall as 100ft.  The Sycamore can be positively huge!

In scale modeling, folks often fail to create realistically tall trees, and commercially available tree kits fall woefully short.  A 100-scale-foot tree in N scale would be 7.5 inches tall – as tall as a ten story building!  A single fully grown sycamore would take up more than a third of the center of a 9.75″ radius circle of track.  Prototypically sized trees would tower over the landscape of a model railroad.

If we look at the common commercially available trees, we find they are quite small by comparison.  I recently picked up the Woodland Scenics Tree Learning Kit.  Don’t get me wrong – this is a pretty nice kit.  It include plastic armatures for about 20 trees of varying heights from 1 to 8 inches, two colors of foliage, and Hob-E-Tac glue to build them.  It’s a nice place to start.  Still, unless you’re modeling a nursery or a newly-built subdivision, only about half of the trees are more than 50 N-scale feet in height.  In HO, none of them would be fully grown. Thee tallest deciduous armatures available from Woodland Scenics are 5-7 inches – again, OK for N scale, but only half-grown for HO.

They also provide some tree kits for “big old trees” and “hardwood forest trees” which fall in the correct range, but they are quite pricey and not well suited to massive forest construction.  The “big old trees” kit has an MSRP of $19.99 for just two trees.  That’s right – they are $10 each!

There are alternatives, of course.  One method that produces very nice results for individual trees is to build armatures from florist’s wire (keep digging through that thread – there’s some fantastic looking trees by user Scotchpine, as well as some other techniques submitted by other users).  This produces some very realistic trees at very low cost, but can be time consuming.  Further, there are ways of making trees from real plant material.

There are also a couple of techniques for quickly making entire forest canopies.  These largely rely on building a perimeter of individual trees to make the “front” of the forest look right, and then filling the center with either foam or an elevated mesh of some kind, topped off with puff balls to model the forest canopy.  These methods rely on the fact that beyond the edge of the forest all you can really see are the treetops anyway.

Here’s one article by Model Railroader with a foam-based technique for this (you’ll need a free registration to read it), and another thread at showing how to do it with a wire mesh.

A complicating factor in modeling trees is the idea of forced perspective.  Often our models have very little actual depth to them… 18-24 inches being typical.  In order to make things look farther away, we can make them smaller.  If done right, this tricks the user’s eye into thinking the object is farther away than it really is, giving our layouts the illusion of depth.

This could be one use for the smaller trees available.  Placed in the back of the layout, some distance behind more prototypical foreground trees, they would appear to be distant, not small.  I think this would work best for isolated trees or lines of trees, rather than a whole forest, as you need visual separation between foreground and background to make the illusion work.

I believe for my layout that I will be using something similar to the “Forest in a Flash” method where possible, by modeling realistic trees in the foreground and filling in the intervening space with massed puffballs at the appropriate height.

Stay tuned for results!!

Half an Inch

That’s how far off my layout is.

Half an inch may not mean much in most cases, but when designing something that has to fit in a room and wraps all the way around said room, half an inch can mean a lot.

In this case, I neglected to account for the width of the Fast Track verticals on which the layout would hang.  They extend 1/2 inch out from the wall.  Because the layout fits snugly all the way around the room, the width of those verticals means the layout is one inch shorter in all directions than the room dimension.

Carving an inch out of a layout design that is already tight is a challenge.  What I did was, I found a “cut point” along the two walls where the layout extends the full length of the wall – the left and top walls in my overview drawing.  The “cut point” is a point where I have one or more straight track segments parallel to the wall, and no curves, angles, or turnouts along that line.  I could then shorten those straight segments by an inch, and that pulls the overall design of the layout in by the required amount without changing any of the angles or connections.

I did also have to adjust the town, streets, and scenery a small bit to fit the new track positions.  However, since those are merely drawn in for illustration and concept purposes, it is OK if they don’t line up precisely.

Had I not caught this, it would have been a significant “Uh-Oh!” moment while building the benchwork!!

Just another example of how attention to detail during the planning phase can really help prevent large headaches during construction and operation.

Concept Naming Conventions

So far, I’ve gone through three complete from-scratch redesigns of the layout.  Each redesign has gone through several versions before being scrapped for one reason or another.  To keep track of all this madness, I have been using a naming convention.

Each basic design is given a “Concept Number” to idenfity its basic design.  Within a given Concept, incremental improvements or variations are given a Version number, which may have a sub-letter for minor variants.  Here’s a brief history.

Concept 1 was the original design.  It was an extended “U” shape, with fairly wide shelves (18-24″ in most places).  This version had the yard on the right side and the town on the left, with a “scenic” area cross the narrower section behind the couch.  One of the key features of Concept 1 was the engine service area at ground level and the right-side return loop on an elevated trestle around the roundhouse.  It was a very nice layout, but operationally it simply did not work.  Concept 1 went through 9 versions and was essentially complete when I finally learned how to run virtual trains in XTrackCAD and figured out that it wasn’t going to work.

Concept 2 was an attempt to improve up on Concept 1.  It had the same benchwork design, but a rather mangled track design with a huge yard across the top above the couch.  I decided rather quickly (at version 1!) that the design was basically broken, and so that was that.

Concept 3 is essentially the current version.  I narrowed most of the benchwork to 12-15 inches, extended it to the fourth wall, moved the yard over to the computer wall, and put the town above the couch.  Concept 3 has gone through 7 major versions, with version 7 going through 7 minor revisions (7a-7g).

I recently started what I thought I might call Concept 4.  However, in retrospect, it’s really just version 8 of Concept 3, since it’s almost exactly like Concept 3 V7g, except it’s been done in Atlas Code 55 track instead of Code 80.

I’m thinking my versioning could use some revisions, though.  So, what I’m going to do is use something a bit closer to common convention for software versions.

I will retain the “Concept” number, since this differentiates what are effectively completely different, unrelated layouts.  However, I’m going to introduce a major / minor / tweak versioning, starting with Concept 3.

So, all of the old Concept 3 major versions (1-7) will now be minor versions of Version 1 (that is, V7 is now V1.7).  The letters will be changed to sub-version numbers.  The scheme will be like this:

  • Version 1 => 1.1.0
  • Version 2 => 1.2.0
  • Version 7a => 1.7.1
  • Version 7g => 1.7.7

Note that there is no 1.0.0, nor is there a 1.7.0.  This is intentional, and is meant to make the mapping back to old version numbers easier (because I’m not going to go rename the files they’re archived under.

The latest – Code 55 – layout, will be Concept 3 Version 2.0.0.  Subsequent revisions will be 2.1.0 and so on.

I hope this ends up being less confusing than the older scheme.

Concept 4, Version 1 released

Tonight, I finished converting the layout design from Atlas Code 80 track to Atlas Code 55.  The conversion required some adjustments since the turnout numbers (angles) and curve radii available in Code 55 are somewhat different from Code 80.  Fortunately, in most cases, that resulted in a better, not worse design.  Here’s the overview shot:

Concept 4 Layout, Version 1 Overview
Concept 4 Layout, Version 1 Overview

Why change from Code 80 to Code 55 track?

Code 55 has a lower, more prototypical rail profile, closer spaced more prototypical ties, and, well, just looks better than Code 80.

For example, here’s a comparison of a Code 80 crossing (on the left) and a Code 55 crossing (on the right)

Atlas Code 80 vs. Code 55 Track
Code 80 (left) vs. Code 55 (right)

Code 55 is more expensive, however, so it’ll just take that much longer to build the layout.

The new version still has a few problems.  There’s a spot along the front edge in the upper right corner where the track is way too close to the edge.  I’ll need to either redirect the track or build a bulge into the benchwork at that point.  I also don’t like the alignment of the tanker tracks in the chemical plant.  I want them closer to the front and angled more around the corner.  The lone stub going into the plant is also a “TBD” item, but I’ll likely not be able to finish that until I start on the plant itself.  That track will be a “plant team track” used for bringing in supplies and large repair parts.  There’ll be a platform or something, but I don’t know where it will go yet.

I like the town layout right now.  The rear track section will be at a somewhat higher elevation (maybe 1 or 2 inches) than the front where the passenger station is.  This will give the impression that the town is on a hillside.

Anyway, that’s the new version.  The Layout Page has been updated with detail drawings and descriptions if you like.  I welcome your comments and critiques!

What’s in a name?

You may be wondering where the Chestnut Hill & Frost River railroad got its name.  Well, I’ll tell you… in a song…

No, really!  The name came from a song…

I first got re-infected with the model railroad bug during the Christmas season of 2009.  I will detail that story in a separate post.  Anyway, the layout idea started as a Christmas decoration – an N-scale “around the tree” layout.  That proved to be somewhat impractical, so we quickly modified it to be a small oval on a 2′ x 3′ board.  Since it was a Christmas Layout, I wanted a name that was tied to the holiday season, but that sounded like it might be a real Railroad name, not a silly holiday name, or a corny pun.

All railroads need to have a name, and many of them are geographical.  It’s common to name the line after its main terminal points:  The Louisville & Nashville, Chesapeake & Ohio, Baltimore & Ohio, Atchison, Topeka and Santa Fe, Chicago, Burlington & Quincy, and so on.

One of my favorite secular Christmas tunes is “The Christmas Song”, and its first line reads:

Chestnuts roasting onan open fire

Jack Frost nipping at your nose

This, I said to myself, had a nugget that was subtle enough a reference that it just might work!  However, there aren’t any place-names in those lines.  On the other hand, Chestnut trees are (or at least were) quite common in the region I am modeling, as is frost… and mountains and rivers are inescapable in West Virginia…

So… roasting chestnuts and Jack Frost morphed into Chestnut Hill and Frost River!

Now, creating a backstory for how those names were created in the model world should prove rather interesting…

Hello world!

Welcome to the Chestnut Hill & Frost River Railroad!

This blog will be my “official” documentation of the creation, development, history and operation of the Chestnut Hill & Frost River Railroad, an N-scale model of a fictional short line railroad set in modern times in Southwestern West Virginia and Eastern Kentucky.  The CH&FR provides rail service to the towns of Chestnut Hill and Frost River, named for two prominent geographical features in the region.

CH&FR Railroad Services provided include:

  • “Shortcut” interchange service between CSX and Norfolk Southern
  • Coal service to the <name?> Coal Mine
  • Materials in / product out service to the <name?> Corp. Chemical Plant
  • Industry and team track service to smaller industries in and around the town of Frost River
  • Passenger service between the towns, with periodic links to the Amtrak station in Charleston, WV.
  • Rail support and services to the Chestnut Hill Historical Preservation Society’s Railway Museum
  • Light/Medium MOW, engine and car repair services as required for other lines in the region.

As of this writing, the layout is in the planning stages.  The current version is planned to be a shelf unit totaling approximately 35 feet of the perimeter of a 10′ x 10′ room.  Nominal width is 12-15 inches, with two sections at 24 inches.  Base track elevation will be approximately 52 inches above the floor.

A fictional Appalachian Short Line Modeled in N Scale.