The recent design exercise with the Hereford Railway paid a lot. That small layout project is plagued by a detestable 18" radius out of the staging. My first concern was to bump the curve radius to 22" or 24", even 28" to make it looks not too much silly. I know it won't never look "good", but I know the best weapon against that ugly geometry is to use real railways' weapon of choice: easement.
Easement is that transition curve that makes a curve blend seamlessly into a straight track. A lot of engineering logic and calculation is needed to determine an easement, but most model railroaders used the good old bent stick trick which works wonder.
I tried it on a 18" radius curve and the result was more than pleasure. It got rid of the toy-like appearance and if one can hide the sharp part of the curve behind some scenery, it does the trick perfectly.
On the other hand, while we were cleaning the club alyout after radical lighting improvements, I quickly fount out many curved I design first lack any easement and looks bad. Truly bad to be honest. Villeneuve yard is particularly unsightly and the fact the curve is at the end of a 15 feet stretch of straight track doesn't help to conceal the annoying optical effect.
At this point, before doing any paint and ballast job, we will have to relay the track carefully with easements to make it look good once for all.
By the way, I'm still thinking over the Hereford Railway project from a prototype stand point. I had the chance to find pictures of typical two-storey station on the line, flag stops and freight sheds. The railway now as an architectural face. What can I say, I'm an architect and such things really bother me when not done correctly. But that's only part of the problem as I'm really starting to think I should go back in time and model the line during the late 1920s when it was still operated as a complete and coherent system by Maine Central. There is a lot of advantage doing it including smaller locomotives and smaller cars. I've got more than enough rolling stock and engines to populate the layout anyway.
Reassessing Operation Fun VS Track Plan
I had a good discussion with Jérôme about the track plan, which is now completed at 100%. We observed Villeneuve's cement plant is rarely switched while Clermont's paper mill is the most active part of the railway.
I recall Lance Mindheim's advice about industry selection: large industries requiring to shove long string of similar cars can quickly become boring. That's so right.
That leaves us with a strange conclusion: our Murray Bay subdivision layout could be better off with only Clermont and Donohue and a long scenic mainline most of the time.
Jérôme's regular operation sessions made it clear to me the paper mill is the soul of the layout and it can take a lot of consuming time. If we were to model Clermont correctly with a realistic and credible team track, it would made an operation session far more interesting. And should I note this layout works better with only one freight train?
Am I saying we should rip off everything we did so far? No. But it put in perspective, after a few months of regular operation, what pays off in term of model railroading fun and what is a waste of time.
It also means the cement plant will need a radical new pattern of operation to be truly interesting. One way would be to do things like to prototype and weight each car. We are already do something similar while unloading the coal and gypsum hoppers, but it will have to be better handled since it can become boring in no long time.
Once again, it makes it clear slapping lots of track and a big industry isn't a guarantee for fun. To be honest, I've got more hours of fun out of my short-lived and diminutive Quebec South Shore switching layout (a one siding layout) than Villeneuve.
If I had only one advice for anybody starting a new home layout, it would be to do one or two location weaved together by a logical flow of cars. Do less, but take the space available to do it right. I feel some industries work well to feed a mainline with long strings of cars, but most people can afford the space and implication of building an empire big enough to make it works properly. At this point, if you love long coal, grain or whatever commodity trains, you are probably better staging them as through trains instead of wasting your time trying originating the traffic on a small or medium sized layout. On the other hand, if you just love these industries, go for it, but be aware of the limitation and design your operation and track plan in such way you'll have fun and meaningful interaction with your favorite cars.