Hi all!
Firstly, my route's been saved by the grace of RW tech support, only downside is that all the track gradients are jumbled. The main line is easy to fix but at Gifford station the track is quite complicated and I don't fancy my chances of being able to re-lay it again, so since the gradients are menial (1 in 85,000!), I'm just leaving it be. Above is the latest development picture, there's not much change, but the run round loop has been completed.
Anyway, all this talk of gradients brings me neatly the article on heights and gradients I wanted to do. My problem with the route is, first and foremost, the fact it no longer exists. Secondly, even in its prime, the route was very obscure (In one of my first posts, I described it as a branch line of a branch line of a branch line of the ECML) and so there's not much information availible on it. All I have to go on, is Google Earth, following the scars in the landscape, and OS map, the railway plans and a book on the railways of Midlothian. This may seem a lot, but theres pro's and cons to each resource. For example, the OS maps offer high detailing of railways, yards, infrastructure and surroundings, but lacks information on signaling and gradients. The railway plans weirdly lack information on the railway (It only covers the main line and not yards or stations) but has excellent gradient information.
With all my resources, I have a good idea of what the surroundings are and where the lines and yards should be. My problem is gradients. The OS maps show spot heights on roads and rivers but not railways as well as contour lines, making accurate gradients difficult. The plans for the railway show excellent gradient information, but only for what was expected of the route, so the actual gradient information is lacking. In the book (The Haddington, Macmerry and Gifford Branch Lines by A.M. Hajducki) there is a gradient diagram from after the construction of the line, the only issue is that the diagram is rather wonkily drawn and so I doubt its accuracy.
The issue list extents when you use all the information together! For example, the hill just outside Gifford is shown as 1 in 50 on the diagram, but only 1 in 53 on the plans. Also, the plans show the whole railway, which was originally supposed to run from Ormiston to Garvald, on which Gifford station is on a 1 in 50 hill, however the line only completed as far as Gifford, there the gradient was leveled to accommodate the terminus.
My solution, is to use the gradient from the plans, then compare it with the information from the OS maps. If the railway ends up 10m above the ground or is otherwise unreasonable, try switching to less accurate diagram and compare results. If it is favourable then use that instead. If it is still unacceptable, then you just have to use a combination of the plans and the diagram, validating it with the OS information, to create an in-between that works. For example, the problem on the hill outside Gifford, I originally tried the 1 in 50 gradient, but this left the railway mid-air, so i tried again with 1 in 53 gradient and it fit rather well!
Well somewhat pointless, but thats my article on why conflicting 110-year-old records are somewhat of a headache to use. More updates soon!