Model Railroad Benchwork:
Evolution and Innovation in Design

by Bruce A. Metcalf

For a version with metric dimensions, click here.

Introduction

"Before we had children, my wife and I were intellectuals. This means that when we wanted to do something everyone else just went and did, we went to the library and read books about it."
--Dr. William H. Cosby, Fatherhood [Cosby 1986]

By Dr. Cosby's definition, I guess I would have to consider myself an intellectual, as I began benchwork construction of my new layout by going to the library. Now, Bill Cosby was speaking of childbirth, and I do not mean to suggest that model railroading is as painful as pulling your lower lip up over your head, but it can be filled with just as many regrets, as I have discovered on six previous layouts.

In the first part of this paper, I will review the evolution of model railroad benchwork that I discovered in my studies, and give credit to those responsible for innovations wherever possible. In the second part, I will explain how I came to adopt a particular set of these innovations--plus a few new ones--in the construction of my present layout.

The design of subroadbed, roadbed, backdrop painting, lighting, layout height, and general layout room esthetics are beyond the scope of this paper, but as each of these has a bearing on the benchwork itself, I will touch lightly on each. I have tried here to be an objective reporter of benchwork history, except for a few clearly identified personal opinions. Be warned, however, that the comments in the second part on my current layout's design are highly opinionated, and further, are subject to continual change and development with or without justification.


The Past

In the beginning … there was the rug. The sophistication of model railroad benchwork design (and it's elevation from the floor) has been rising ever since.

"Benchwork" properly began when the first modeler moved his track from the living room floor to the dining room table. Flattop design hasn't progressed much since then. Ping-pong tables--and for that point-to-point line, wallpaper pasting tables--provided easy, prefabricated solutions all too often [Armstrong 1954]. The basic form of a flat sheet of wood or plywood, with or without supporting timbers, is still very much alive, and for the smallest layouts it is still quite adequate.

By the 1940s, open table designs were being used when tables grew too large to reach across [Van Ness 1949], but basic flattop construction was little changed. Large layouts were often of the spaghetti with pop-ups plan, and a limber spine was a prerequisite to maintenance and operation, and sometimes just to viewing such layouts.

However, open-pit design affected only the table's shape, not its method of construction.

The great simplicity of tabletop designs is also their greatest limitation. While providing a flat and stable platform for trains, they do a very poor job of representing our non-flat world. The first effort at providing a vertical dimension came from the tinplate concept of bi-level trackwork [Morgan 1944]. This was not the multi-level benchwork of today, but a second altitude at which one track could pass over another by bridge or tunnel (or sometimes just track in mid-air). Sometimes the two levels were interconnected by ramps or toy train trestle sets. In other cases the lines were simply left unconnected.

Cookie-cutter design allowed for a much more graceful transition from one level to another, but it was seldom used to produce sloping ground other than right of way, and it was still very difficult to depict scenery below the base level.

Grid benchwork design, while it was the same framework as used below tabletop or cookie-cutter layouts, implies strip roadbed used only where needed, and the use of risers on even the lowest track level. At last the hills--and especially the valleys--were not constricted by plywood and could take on a natural shape without hindrance.

Many consider L-girder construction to be the epitome of grid design. There exists some doubt as to whether it was invented by Linn Westcott in late 1963, as many think [Westcott 1963]. Two years earlier that author described it in a general article in another publication under the name "jackstraw method", and promised further study [Westcott 1961]. (Note too the similarity to a drawing done for an article and book by A.C. Kalmbach ca. 1938 [Pete 1938].)

One is given cause to wonder how much Westcott was influenced by the stacked joist designs previously published in the same publication over a decade earlier [Houghton 1948], especially as they show up in his own writing very soon thereafter [Westcott 1951].

While L-girder design can be used to build mundane designs, such as a 4×8-foot tabletop, its great strength lies in its unlimited joist lengths that permit freeform table shapes. It has two signal disadvantages: more complex construction is required, and a greater depth of joist and girder than the grid design, which results in more restricted space under the layout, especially for low benchwork designs. This point is often overlooked until wiring begins, but can be the source of great aggravation.

Walkaround control and reliable automatic couplers in many scales (read "Kadee") spelled the end of spaghetti and pit table design, and inspired equal attention to aisles as well as tables. Rooms were filled to the walls, pits were joined into continuous aisles withe few duckunders, and most importantly, table width was limited to arm's reach.

While any benchwork design can be built to a 24-inch width, and many can be simplified by the use of the wall for support, narrow benchwork against a wall is an excellent candidate for cantilever construction. Construction with wall anchors and diagonal supports continues to be popular, even when used with L-girder or other modern elements.

Cantilever designs that did not use the floor were promoted as "standard construction practice" as early as 1941 [Taylor 1941]. Frank Taylor's early design suggested a design using 1×4-inch hangers suspended from ceiling joists, with cantilever joists. Bracing could either be above or below the joist as suited track arrangement and scenery.

Ed Ravenscroft demonstrated the use of triangular plywood cantilever joists in one incarnation of his Glencoe Skokie Valley RR [Ravenscroft 1949]. A fascia board provided longitudinal stability and, with a notch in the front end of the joists, a cable runway.

Among the benefits of cantilever construction is the elimination of legs. This greatly eases access under the layout, encouraging work on hidden switch machines and wiring. Storage, always a problem for model railroaders, is also much easier in such an environment. While some, like Ravenscroft, believe in fixed cabinets, the ability to quickly and easily remove portable storage units so as to provide free access separates cantilever design from cabinet-top construction. Just about any method of support that eliminates under-table obstructions can and has been used.

Despite some good early work [Booth 1939, Houghton 1942], and a few contemporary efforts [Hamilton 1989], lightweight design is a relatively unexplored concept in permanent benchwork. Most modern activity in lightweight design revolves around modules that also require extreme ruggedness and limited dimensions.

Early benchwork designs that used 2×4-inch legs and 3/4-inch plywood tops were based on the premise that modelers would often be standing on top of the layout. A recent example of this tendency to over-design uses 2×4-inch legs to support a 24-inch wide table. Considering the upper level was built at nearly a 60-inch height [Hediger 1983], one must ask, "Was the builder was really planning to stand on top of this yard?"

With the high and narrow designs now popular, it is not necessary to have man-rated benchwork, and very few locomotives weigh as much as the average modeler, even in 3/8-inch scale. Each of my layouts has been built with lighter construction than the last, and I have yet to have a train fall through to the floor.

A mixture of grid, L-girder, and cantilever construction was proposed by one writer, but it suffered from duplicate joists, doubled girders, and redundant legs, using more lumber than strictly necessary, and its rectangular shape failed to take any advantage from varying the lengths of its stacked joists [Larson 1972].

Innovative materials is another topic but lightly touched upon in the literature, and most reports concern either lumber shortages or creative reuse of found materials. Galvanized iron pipe has been used for legs and major girders [Page 1948], Irv Schulz used steel brackets for the cantilevers under his wood benchwork [Schulz 1979], and some have tried corrugated paper and insulation foam for support and scenery, but none have gained general acceptance or widespread use in permanent layouts.

No mention of benchwork engineering would be complete without comment on Tony Steele's painstaking design for his Upper Second Subdivision [Steele 1989a, 1989b, 1989c]. L-joists and low-profile U-joists are cantilevered from posts or studs with steel shelf brackets, each is given a length appropriate to the specific benchwork width and track arrangement, and backdrop supports are considered longitudinal structural members. While much of his design effort is focused on crowding the maximum amount of track into limited space while maintaining prototype civil engineering standards, his innovative use of a variety of materials has valuable lessons for any layout builder.

A more modest approach for those who don't have engineering degrees, but one that still requires pre-planning of both track and scenery, is offered by Jack Burgess' design of his Yosemite Valley layout [Burgess 1982]. Burgess was striving for minimum benchwork depth due to his layout being a multi-level design, but the simplicity and economy of his method are yet worthy of careful study.

In all of these designs, it is rare that much attention has been paid to human factors. From the backbreaking duckunder designs of the 1940's, to the deep benchwork of stacked joists and L-girders, to the obstruction of permanent cabinetwork, the presence of people in the layout room--especially those under the benchwork--has generally been restricted to the question of aisle widths and arm lengths. There are, obviously, a few other considerations that have garnered (insufficient) attention, and the principle among these is elbows.

Elbows are an unavoidable consequence of the design of the human body, and they have a natural tendency to come to rest where the maximum damage can be done. Elbow-bending, or the consumption of beverages, also presupposes the existence of beverage containers, and the need of a resting place for same. Finally, tools and supplies--both for construction and for operation--have a tendency to blight the model landscape like giant weeds unless some other provision has been made.

Houghton was the first to design a layout for observation by humans by the provision of a wide "elbow rail" [Houghton 1952]. It was over twenty years before the hobby press again took up this theme. Burgess suggested both elbow rails and beverage holders--separate designs [Burgess 1975]. However, his elbow rails were an addition to his cantilever benchwork--not an integral part of it.

By the 1980s, a number of writers had adopted the practice of adding narrow shelves to the edge of the layout. John Olson's Jerome & Southwestern used intermittent shelving, interrupted by scenery and control panels [Olson 1982]. W. Allen McClelland's V&O features a near continuous shelf with beverage holders, operation card holders, and telephones added [McClelland 1977].


The Present

The proper mix of all these elements must, of course, be blended to match the space, scale, budget, and taste of the individual modeler. I can only identify the ingredients I have used, along with my reasons (where I know them), with the understanding that different conditions and different modelers will almost certainly generate a different mix.

The site for my layout is one end of a blessedly dry basement measuring about 25 foot wide. The walls along three sides are cinder block, and there is no ceiling installed--the joists of the main floor are exposed, along with the plumbing, heating ducts, the random nail that missed a joist, and some 2000 crickets.

In order to seal out moisture, insects, and possible radon gas (common in my part of the world), the walls and concrete floor were given a thick coat of white latex ceiling paint. The overhead joists, plumbing, heating ducts, and the top two rows of cinder blocks were painted flat black. With careful attention to lighting only the layout and floor areas, the black paint will render the rather busy ceiling nearly invisible, and lighting fixtures can be raised nearly to the level of the floor above. This also saves the substantial cost of a drop ceiling, which would have to be too low to clear all of the heating ducts. Some have suggested that drop ceilings would block dust falling through from the floor above, but my experience shows that a good paint job will seal out most dust, and I know that composition ceiling panels shed their own dust at a horrendous rate.

Phase one of this model railroad consists of a 24-inch wide shelf around three walls of the basement, with a string of modules across the fourth side to provide space for a continuous main line about two scale miles in length. Subsequent phases will build connecting shortlines or branches onto peninsulas in the middle of the area. Should I live so long, right-of-way has also been secured to the rest of the basement for an expansion that would double the lenght of the mainline.

The benchwork is of a composite construction. Against the walls, 1×2-inch hangers are suspended from the overhead floor joists on the same 16-inch centers. Short joists are installed between the last regular joist and the top of the foundation wall on the wall that is parallel to the joists. This provides the basic foundation to which all else is attached.

Cantilevered from the hangers are 24-inch long triangular joists made of 1/4-inch plywood, stiffened by a 1×1-inch board that extends an additional 3 inches. While a triangular shape is preferred by physics, it isn't necessary, and it does not have to be angled on the top--I have done this only because I am modeling the Western Pacific in the Feather River Canyon, where there is precious little flat ground to be modeled. The angle could as easily be on the bottom of the joist, and the stiffener could be attached at the top or middle, as suits conditions, so long as it is horizontal.

The stiffener extensions on the joist assemblies support a continuous 4-inch wide cup rail made from 1/4-inch Masonite® composition board. This provides an attractive and a more appropriate site for elbows, tools, and beverages than the track and scenery. The cup rail also serves as a longitudinal stiffener and is an integral part of the structural design.

The bottoms of the hangers are connected, and a cable trough created, by attaching plastic rain guttering just beneath the joists. (Much thanks to Robert Metcalf for this flash of brilliance.) The guttering does not need to be connected, or even close, and for those planning to use zip texturing on their scenery, it's better if it does not hold water. It may prove necessary to provide additional supports for the gutters if the weight of the wiring grows too great, but this can be provided by a length of coat hanger wire and two screws. Again, the guttering is a structural element that provides additional strength to the benchwork system.

Diagonal bracing for the hangers is provided by the backdrop material. It is attached to the face of the hangers between the bottom of the overhead joists and top of the cantilever joists by nails, screws, or caulk adhesive as best suits the material and your preferences.

My original plan for the backdrop material, based on previous good experience, was to use surplus linoleum flooring turned to face the wall, and attached with carpet tacks at the top and bottom of each hanger. Time and technology have taken their toll, and "linoleum" has been superseded by "vinyl flooring". The vinyl does not form as easily, tears more readily, sags willingly, and all too often, the texture embossed on the top is visible on the bottom. I am now in the process of removing it.

Other materials widely touted that could also work are 1/8-inch Masonite® composition board or 1/4-inch wallboard. These thin materials permit fairly easy curvature around corners, and are readily available. These require more joint patching than a continuous roll, but are easier to handle--especially for one-person crews.

Several other materials include enameled aluminum gutter stock [Schulz 1979] or trailer skirting material. The latter is made under the trade name of Trim Coil® in widths up to 24-inch in 50-foot rolls. While sky blue isn't available, a white pre-painted surface should take paint easily. For taller skies, steel appliance stock can sometimes be had, especially in the form of damaged coils--just be ready to rent a good size truck to move it, a full roll can literally weigh a ton. Like steel rail, such backdrops are not as easy to install, but they will easily outlast you. Finally, photographic backdrop paper is available in rolls as large as 8×150' and it is avialable in sky blue. I plan to try this next, although I will admit to being concerned about its ability to provide diagonal bracing.

When it comes time to expand the layout onto peninsulas in the center of the room, I intend to build "stud walls" from 1×2-inch "studs", and continue as before. For tables with no backdrop, a low "stud wall" will be placed in the center of the table and double-length joist assemblies will be attached near the top of the "studs". The amount of anchoring and bracing these will require is yet to be determined, but I presently favor shelf brackets attached between the stud wall and the floor.

For peninsulas with a backdrop, the stud wall will extend up to tie into the overhead joists, and either single- or double-ended joist assemblies, cup rails, cable troughs, and backdrops applied as with wall-mounted benchwork. Note that these stud walls do not need to be straight, but can adopt any curve needed to match the shape of the peninsula.

While it would be nice to eliminate the use of legs on peninsulas, it is not appropriate to use hangers away from the walls, as there would be too much sway from the weight of the average elbow.

Peninsula backdrops should be double-sided, even when there is benchwork on only on side, for cosmetic reasons. The back side backdrop may need a cup rail for convenience, but it will not require one for strength as the front cup rail and cable gutter will provide ample stiffening. Black or dark paint on the back side of the backdrop will help it to vanish from your attention, but you should consider using a painted sky or scene, especially when there is an operating or viewing position that faces it.

Depending on your ventilation situation, it may or may not be advisable to extend the backdrop on the back side below the cable gutter or cup rail. If the stud wall is installed directly below and parallel to a joist, it may also be necessary to stop the backdrop a foot or so below the joists to permit air circulation, otherwise the space between the joists should suffice.

This benchwork design is surprisingly strong and rigid, and can be used with nearly any type of roadbed and scenery. It is not attached to the wall, except where the modules connect, as gravity holds things in place with high reliability. Californians and others can easy add the occasional anchor if this were felt insufficient. Stud walls should be firmly anchored to the floor in order to prevent shifting when kicked, and to the overhead joists when full height.


This paper has attempted to expose you to some of the more innovative concepts in layout design technology, their evolution, the developers behind it, and my personal response to the design challenge. The benchwork design presented here represents a potpourri of these ideas gleaned from the past sixty years and dozens of friends and correspondents--like most areas of hobby design. While my particular mix of design elements may not be the answer to the design of your next layout, If I see any of these ideas put to use on future layout tours I will be well satisfied.



This paper was prepared with the support of the Kalmbach Memorial Library, National Model Railroad Association.
Originally published in Larry Kline, editor, The Columbus Clinics (National Model Railroad Association Convention Proceedings), pp. 33-41. Discoverail '92 (National Model Railroad Association Convention), 1992. Copyright © 1992, 1997, Bruce A. Metcalf.
Updated 30 November 2002 by Bruce A. Metcalf, bruce.metcalf@figzu.com, who would very much appreciate your comments.