What To Model?

When it's time to design a layout, a crucial question will quickly arise and which must be addressed frankly: what should we model and what shouldn't be modelled. As you can expect, there is no recipe to make such choices which will be based on various conditions we have to live with (resources, space, time, skills and personal interests). However, it would be wrong to assume you can't replicate the general characteristics of a specific train run if you don't have all the pieces. As long as the story you tell make sense, your brain will fill up what's missing.

I previously compared layout design with a song "chorus-verses" structure. This analogy stay relevant at this next step. Just think about it, you probably know familiar songs by heart but often discovered less famous verses rarely used. They can be omitted without altering that much our appreciation, in such a similar way a song radio edit versus an album edit can vary in length.

In the case of a layout, it means you don't need to model everything to get a coherent and somewhat "complete". Many people will select a few locations, a particular industry and others bits they like and put them together. However, this can be tricky. If the selection is carefully done, you'll end up with a decent layout bolstered by a logical concept that link everything together. The missing parts will be similar to our edited song: barely noticeable. On the other hand, if the selection is whimsical and based on anecdote rather than what's typical, you get an incoherent patchwork of scenes which are extremely hard to logically link together with a coherent story. Unfortunately, most of the time, we go that route.

CFC network with main locations (1994-2011)

I'd like to use the CFC Murray Bay Subdivision to illustrate this with a case study. You probably recall a few days ago I said Villeneuve (cement plant) and Beaupré (paper mill) were equivalent in term of operation and interest. Superficially, they both are small yards where the local freight set out and pick up cars that are switched by industrial locomotive to feed large industry. But in fact, the comparison stops there. Upon closer examination of old timetables, pictures and videos, it has become clear that Beaupré was a much more important location on the railway line.

Original 1994 CFC track plan

Villeneuve is basically a siding where you pick up loaded cars and set out empties. Then you move forward to your next destination. On the other hand, Beaupré was where trains were broken or assembled. To save on fuel and keep trains shorter before entering the rickety topography of Charlevoix, cars picked up en route to Beaupré were left in the yard. This was because most customers had east facing sidings. It was useless to bring these car up to Clermont when their destination was Limoilou, thus they were left at Beaupré and picked up by the next westbound train (523). Now, add this small operation feature with serving the plant in Beaupré and you understand it was a location where a lot happened. In fact, you don't have to physically model some customers because their traffic is actively handled in Beaupré, which is quite a bonus!

Minimum useful trackage to keep operation coherent 

But better than than, a 1994 timetable indicates that it is possible switching at Beaupré isn't done yet by the paper mill and CFC trains must wait until 14:00 before leaving with these cars. This is quite interesting because it means an operator could run his train up to Beaupré, then control the switcher and prepare a cut of cars to be shipped to Limoilou. Then the local train leaves Beaupré. And to make things more interesting, a busy grade crossing was located in the middle of the yard, thus extra care was required to not impeded the important truck traffic serving the plant. It means cars must be blocked in certain ways and consists may have to be broken depending on condition.

Car blocking and train movement per location.

Why does it matters? The reason is simple, as things stand, the layout area depicting Villeneuve is quite underwhelming in term of efficiency. Beautiful, realistic and prototypical... but nothing really happens there. Most of the time, we run the train directly to Clermont which takes about 2 minutes and don't make a stop at Villeneuve most of the time. Picking up cars in Villeneuve isn't a great activity too. Thus, you end up with a 14' x 11' room (about 40' of main line) which basically only exist to take a train out of staging before moving it as soon as possible to the next room. Let's call this a serious design caveat! Villeneuve is nothing more than a vignette... a long one, but still a place where nothing really happen. And from a historical point of view, by 1997 the plant was closed and since the very early 1980s, it no longer received gypsum and coal by rail cars. Not a very impressive customer...

In the case of Beaupré, the mill was still strong up until the late 2000s. Railway employees had to take decision there depending on condition and various train movements made sure it wasn't a pit stop in a F1 race. And that's the key element: at Beaupré you take decision... in Villeneuve you do as per instruction like a robot. From an operator perspective, the choice is easy.

And that's the important key feature of Beaupré in regard to how to decide if you model or not a location. Beaupré and Villeneuve handle the same amount of cars, but they don't have the same relevance. If I had to make a choice, I would choose Beaupré because the layout room would be a large comprehensive scene where an operator can be kept busy at his "operation station" for a while. The room and scene would have a purpose, which is dramatically lacking in Villeneuve. And it makes sense since real railroads don't give a damn about stuff that no longer as a purpose. It should be the same with a layout.

Another interesting aspect of choosing Beaupré over Villeneuve is about replicating the essence of a particular prototype. I've often wrote about my admiration for Tom Johnson's Inrail layout based on an Indiana branchline fully dedicated to serving feedmills and grain elevator. Tom's layout isn't the greatest our there in term of scope, size or goals, but it shines because it sings a song about the harvest season. All the towns and empty space between share are similar tone that create a strong feeling of travelling a particular part of a country. Each grain elevators, which are basically similar, makes the theme even more stronger, creating a sense of immersion, of realism and of purpose rarely equaled. In that regard, focusing Hedley-Junction on paper and lumber products is probably the best way to convey what characterized the line up until the effective end of freight service in 2011.

Having two paper mills, both similar yet different in the way they interacted with the railroad, give a powerful sense of purpose to the layout while making it highly coherent. It also provides two large and mostly prototypical scenes instead of a collection of poorly grafted together locales. Each scene immerse the operator in a different world where multiple tasks must be carefully handled... And each room end up with very scenic elements such as large rivers and bridges that are exquisite railfanning spot.

At the end of the day, after almost two intense weeks of research and thinking, I must admit modelling the Murray Bay Subdivision as a comprehensive layout is easier to achieve when done as a modern railray. Like anybody, I started with the 1950s as a benchmark for optimal operation opportunities. This wrong assumption was based on the idea since trains were still dominant in that decade and served virtually all industries, the potential and interest was higher. With the time, I have come to understand potential and interest aren't related to quantity. The early 2000s Murray Bay may have been seen as a plain branchline without redeeming factor, but in fact, it was much more interesting that trying to figure out what happened in the 1960s, 1970s and 1980s when nobody cared about documenting the railroad except for a few rare steam excursion.

That brings me to my last point about choosing what to model. At one point, you've got to do what you can with what you have. Up to 1994, it's quite hard to understand what really happened on the Murray Bay in respect to freight operation. We have glimpses here and there, but not a general portrait of what really happened. From a cosmetic point of view, that may be fine, but it's clearly not enough to create a coherent story - a song - about the railroad. I struggled a lot trying to do so... and failed more than once. Only the Donohue plant worked because the available data was comprehensive and we knew exactly what happened there. Jack Burgess - which have been quite vocal recently about his excellent Yosemite Valley Railroad layout - stressed how having access to a lot of quality data convinced him it was possible to replicate the railroad as it was. The same applies to our project; after 1994 we know exactly what happens. We look at a train and we can determine which customer a specific block of cars is for. We know each locomotives and their operation history and have a good grasp of how things were handled. It's not a matter of being lazy, but this is convenient and makes layout design and operation far much easier. Knowing so much how things happened enable us to make definite and coherent choices about what to model and what to not model... And at the end of the day, you can even look at the official track diagram and select the best verses and keep them in your layout "radio edit"