Already I have seen a few potential changes that could be made.
- I think it will be possible to eliminate the 3-way turnout at the entrance to the station (which is good because I dread the prospect of having to build it!) by splitting it into two standard turnouts - a LH B6 into the bay platform and goods yard (pushed as far up against the baseboard join as possible while still allowing room for the point motor underneath) followed by an A5 RH into the loop. Measuring this out, there would actually only be about a 5cm reduction in the length of clear road on the main platform between the two loop turnouts, a small price to pay for a lot less hassle.
- While I have yet to find a suitable prototype to base it on, I think that if the goods shed can be kept reasonably small, there would be room for an additional siding in the goods yard adjacent to the bay platform and of a similar length, which would add significantly to the operating potential. My one concern is that things don't start to look too cluttered - with just the one siding there is quite a spacious look which I have to say I do like. However, examining photos of Combwich (still a signifcant source of inspiration), the goods yard is quite cramped widthways but still looks ok.
- I'm undecided about the coal siding. While it definitely adds operating potential, and I do think coal facilities are a necessary part of a branch line terminus, I just wonder if having it here works. I do want to retain at least some illusion of the railway in the landscape as opposed to simply covering every available space with track and railway related cameos. The alternative is to have the coal stacks situated in the goods yard itself, as on Combwich - which saves space but I'm not sure quite how realistic this is. Ideally, different types of goods traffic should have seperate sidings, otherwise you run into problems with coal wagons obstructing access to the goods shed etc. I need to look into prototypical goods yard operations a bit more to find out what is feasible, and also think about what works from an operating perspective. What might work in a prototypical situation with a twice weekly pick-up goods is not really so attractive for a model where I might want to run several goods trains per day (whatever a 'day' in model operating terms is!).