A Tale of Two Depots

The whole issue of plausibility, and its tenuous relationship with reality, informs a lot of modeling. The problem of getting the 4-4-2 that I'm building "right" spurred me to go back and take a look or two at a couple of other interesting models that have grappled with the same problem - one a model of an actual station, and the other, a model of a model. Since my next planned project after the 4-4-2 is, likewise, a model of a model, I've been giving this point some thought.

Sometimes the prototype is perfect - just what you need. Other times, the prototype comes close to what you want it to be - but you want to modify it just a bit. One good case in point is the Colorado Midland Railway' station in Granite, Colorado (station is unfortunately just out of view to the left of this photo). This was built in 1887, and at the time, Granite was the seat of Lake County. The station was a two story split-level affair, and when I first saw photos of it I never associated it with this.

The model you see in the picture is a 1970s-era craftsman kit, known as "Granite Station," manufactured by a small and apparently defunct company called Timberline Models. I've been partial to this model ever since I was a kid. My dad picked up a copy of the old "Narrow Gauge and Short Line Gazette" for me when I was ten, and it contained a piece on sprucing up one of these models; the image stuck in my mind (as the work of a lot of 1970s-era modelers like John Olson and Mal Furlow did), and when I came across it decades later, I started searching online in hopes of finding it, and discovered the kit, the maker, and the prototype.

I was surprised to find, however, that the real Granite depot was a very austere structure by comparison - those marvellous windowframes, for example, were a product of someone's imagination. The real Midland was too cash-strapped for fancy ornamental windowframes, and settled for simple double-hunge windows - and it didn't bother with a canopy, supported by fancy posts and the elaborate bracketing for the eaves. When I realized the Timberline model was unaffordable and went back to look at Dan Abbott's Daylight Through The Divide in hopes of finding plans, I was surprised at how simple the prototype really was. It was like seeing someone famous without the makeup on.

Similarly, I've always been fond of the logging diorama John Olson built and photographed for Model Railroader in May, 1976 - his "Mule Shoes Meadow" layout. I've always been partial to the high mountain parks and valleys, and I thought the Mule Shoes layout is a very fine mix of the railroad and the deep woods. It's a little modern for my taste, but backdate it a few years and you can plausibly imagine something of the kind as a junction for a branchline on a slightly larger railroad at the turn of the century. The original photo essay captured several shots of the depot, from various angles. It was a less adorned building than the Timberline structure, and so far as I know it was scratchbuilt, but I always liked the look of it - a big,rambling frame building, one third depot, one third office, one third cooking and living space. Apparently I'm not the only one, because Wiseman Model Services seems to have used it as a rough prototype for their Horseshoe Meadow Depot, which is currently available (and worth the price, I might add).

I bought one on the strength of the initial resemblance, and did a closer comparison once I had it at home. I was surprised at some of the variations the kit builders introduced - the pitch of the roof, for example, is noticeably steeper on the mass production model. I had not initially noticed this, but something about it strikes me as noticeably unnecessary - the steep pitch makes the building a bit higher, but doesn't provide much in the way of additional covered space. I understand that buildings in snowy areas need a steep camber to shed the snow, but the camber is so steep that it would require a lot of additional roof space, but would not generate much in the way of usable additional floor space - if you look at the second floor, for example, you can imagine that the interior's basically a cathedral ceiling. They also added a shingled roof on the bay window, which protrudes out just a bit further on the Wiseman model than it did on Olson's original, and of course, the cast chimneys are more ornate.

It's not a bad-looking result, but it does sort of raise an important question: is it worthwhile doing something just because "it looks good?" And does it look good because it's well-engineered, or does it look good because it conveys an impression you wish for it to convey? I'm open to either answer, actually. The fact of the matter is that I like both manufactured models. I also like the Granite depot as it was built, and I like Olson's depot enough to have retained an enduring impression - for I first saw it in the library at Oak View Elementary School at the age of six, when I paged through a bound volume of MR during library time, and I remembered enough to recognize the photo essay when I saw it again decades later. Something made each of these structures memorable, and beautiful. Plausibility is a part of that beauty, because it contributes to the suspension of disbelief, without which there can be no effective modeling. There's a tension between brute reality and beauty, sometimes, and the trick lies in the resolution. I've been working at this on my 4-4-2, and hope to have a photo up in a couple of days; we'll see then how well I've resolved that tension!

Next project

So in true model railroad fashion, it's off one project, and on to the next before the first one's wholly finished. My next project is the Model Die Casting ATSF 4-4-2 you see here. I just bought it a few months back, but I suspect it's fairly elderly, because the boiler is wholly die-cast Zamac, rather than the plastic-coated Zamac MDC went to on some of its models in the last years of kit production. I don't know whether they ever retooled the Atlantics; I suspect this kit dates from the early '90s, since it does have a can motor. I'm frankly glad it doesn't have the boiler detail in plastic; I find metal a bit more forgiving. Plastic needs a lot of care and non-abrasive treatment, so you can't buff your errors out, at least not so easily, and I tend to be heavy-handed.

I'm also going to customize it a bit. As it is now, four things mark this engine - almost irrevocably - as Santa Fe property: the trailing truck, the train indicator boards, the cab windows, and the front end. I'm going to leave the front end as it is, because I like the number plate at the center of the smokebox, but the trailing truck has to go. The Santa Fe used an early built-up truck, and it has a clunky, dreadnought-era look to it. I got a Precision Scale Hodge trailing truck that's essentially the same as the trailing trucks used on the USRA engines, and since I have a couple of those already, the different truck will, I hope, impart a slight look of mechanical commonality. My Bowser 2-10-2, for example, has a similar front end and an identical trailing truck. The cab is another matter; I have given some thought to fabricating one out of brass or just picking up a PSC cab kit, but I haven't quite made up my mind yet. It's not ugly, and it certainly has an era-appropriate look, but the paired arched windows are distinctively Santa Fe. The same can be said for the indicator boards, but those are easily disposed of. It's amazing how different an engine looks when a prominent feature like a train indicator board is removed - the additional advantage, in this case, being that there's no piping to rereoute or alter, as there would be with most appliances. Pictures will be posted as construction progresses, and we'll see how my notions work out in practice.


One of the problems with freelancing is that at some point, you start to notice the divergent lineages of the different pieces of equipment, and more annoyingly, the provenance of the models. With a model like this one, it takes a frustratingly large amount of work to avoid that problem, particularly since the piping and a lot of the detail is cast into the boiler. Big changes, like switching the trailing truck, at least distract the eye away from the stuff that's eradicable only with excessive amounts of work, and I think when bashing an engine like this success depends on correctly identifying those big things and changing them to fit your own design.

28' boxcars

So this is the project I’ve just completed – a set of five BTS Pennsylvania Railroad XA-class 28’ boxcars. Twenty-eight foot cars were very common in the period between the end of the Civil War and the First World War, and they were state-of-the-art in the 1870s, when the XAs were built: they have a definite “shorty” look, even compared to the 34’ cars that replaced them. Like a lot of modelers, my freight car fleet began with the injection-molded MDC 34’ and 40’ boxcar kits that have been on the market since time immemorial, and I never thought much about anything else until I started doing some research on the 1870-1910 era; I had sort of mentally walled off small boxcars as Civil War/Central Pacific-era equipment. My reading on the Colorado Midland confirmed this prejudice: the Midland was built in the late 1880s, and its boxcar fleet was all 34-40’ cars, and you don’t see much small foreign road equipment in the pictures, probably because the Midland restricted the number of non air-braked cars that could be included in trains. I was therefore very surprised when I found a pair of tables that catalogued the numbers of box cars on the Philadelphia and Reading at the turn of the century and shortly thereafter, and found that a substantial number of 20-22 ton cars (which were probably about 28’ long) were still in revenue service. But this seems to have been fairly typical, particularly for roads that had been around for more than twenty years before the turn of the century.

This turned out to be great; I had always liked the look and the proportions of the shorter cars, particularly the original Central Pacific equipment (I still have one of Rio Grande Models’ CP ventilated boxcars on a shelf awaiting my attention), and they have certain aesthetic and operational advantages. They’re shorter – here’s a picture showing how five 28’ cars fit in the space occupied by four 34’ cars – they tend to emphasize the age of the layout, and they don’t dominate the surrounding scenery and buildings. I was looking for cars that could, once built, become a sort of rolling scenic background. Not everything can stick out on a model railroad, and not everything should. I was looking for decently unobtrusive equipment that would be era-appropriate without distracting the viewer away from the structures, locomotives, or specialized equipment that deserves a prominent place. I also wanted a group of cars that looked the same, to provide the “fleet” feeling you get on a real railroad.

All of these considerations made the BTS cars a natural choice. They were also economical: they are sold individually for $22.95, and you can get a five-pack for $99.95, which is a $14 savings - almost enough to order a sixth car. They don’t come with trucks, but I wanted a 4’6” wheelbase truck with a high diamond arch bar, and Bitter Creek had just the thing. You can put bigger trucks under them, of course, but at some point you run into a proportionality problem, because they look too big for the car – some of the early BTS cars on their website, for example, were built with the generic-looking MDC arch bar trucks from those 34’ boxcars, and they look just a bit too big. I have used Rio Grande Models CP arch bars (as seen on the ventilated boxcar) in the past, but I decided to try the Bitter Creek trucks, and I like them. They’re the perfect length, and they come assembled, so you just have to paint them.

The BTS kits are laser-cut, and they build up very well. When you do five of them at once, you can assemble them all together, and this saves a considerable amount of time. At some point I’m going to letter them, but it may be months before I do that. I recently saw a nice picture of some NP boxcars in Bob Lorenz’s book on NP steam, and the consist included several cars that were obviously of the same class, decorated in the same way. In a nice “period” touch, the “Northern Pacific” was painted as a sort of arc on one side of the car, and I really liked the introduction of the curve into a surface that’s otherwise wholly linear – boxcar lines, after all, are pretty much linear or angular.