Swin Did


I’ve been wanting to do a steam-powered, semaphore-signalled mainline layout in R3D for a long time. Sadly, my imagination has always been lacking - I come up with half a dozen miles of railway, then run out of ideas and get bored.

Swin Did”, or “Swindon-Didcot”, originated as a plan to model the gwr between those two locations. However, I have since got carried away and the idea just keeps expanding! The current planned extent is from Purley, just west of Reading, to Bristol or so. The route is that via Didcot, Swindon and Bath, though I see no reason for not adding in the South Wales Direct (Wootton Bassett - Severn Tunnel via what is today Bristol Parkway) towards the Bristol area as well.


I have chosen to do a post-ww2 layout for several reasons. Firstly, photographs and first-hand accounts are much more plentiful, making it easier to work out the layout. A lot of interesting additions and trackwork alterations took place during the war to handle the increased traffic - Hinksey (also known as “Oxford New”) Yard is a prime example - and as many of these changes have stayed, it’s easier to go with them then try and work out what was pre-war and what wasn’t.

The wtts I’m using are the 1948/1949 br(wr) versions. Many services are marked as “Suspended”, presumably due to the effects of the war. However, having compared to wtts from the late 1950s/early 1960s, many of these suspended services were reinstated with little or no change to their timings, and so I intend include all of these services, if for no other reason than adding interest to the layout!


Clearly, for a project this large, lots of information is needed!

Google Earth

The main routes have been traced out in ge, and converted to Rail3D by Roberto Ceccarelli’s converting tool (I may also have done a small edit to it allow it to work out track node angles before importing, too - I’ll have to investigate and suggest the change to him if so!). As the gwml hasn’t changed very much at all in terms of route over the last century, this has made life quite easy. More difficult are some of the abandoned branchlines, but these have generally stayed quite visible on aerial images, as the trackbed and land boundaries have stayed untouched in many places.


This site provides access to many older maps. Tracks are generally shown in fairly good detail (pointwork and signal posts are indicated), although the maps aren’t always totally reliable on such things - for example, the 1933 map of Didcot shows a strange combination of the 1910s layout and the redesigned 1920s layout! However, these are especially useful for trackwork that has since disappeared - many yards and sheds have long gone or been completely altered, such as Moreton Cutting Sidings.

Signalling Record Society diagrams

The srs has a fairly comprehensive selection of signal box diagrams, available on cd. These are incredibly useful for signalling the layout, and some even include details such as interlocking and lever leads, making accurate lever frames an intriguing possibility!

Photographs and books

The gwr always had an eager following, so information and photographs are plentiful - and it’s amazing what sort of information can be gleaned from a simple photograph.

gwr/br Working Timetables

Michael Clemens has a lovely collection if gwr/br(wr) timetables available for download on his website (http://www.michaelclemensrailways.co.uk/?tok=34Swin Did). I intend to use the 1948/1949 versions.

In addition, the National Archives contains a huge and comprehensive collection of wtts, dating back for much of the last century, accessible for free.



The layout began in summer 2012, but I was without proper internet access for much of that time, hampering progress. I now have both free time and internet access again, and have picked up from where I left off.

The line from Purley to Bristol, as well as Didcot-Oxford and several branchlines, were imported from ge. I’ve started putting in trackwork and signalling and have got as far as Didcot, where I have run into some problems with the signalling limitations of R3D - it can’t handle closely-spaced signalboxes very well, and slotted distants are especially tricky (possibly impossible) to make work correctly.

Didcot, as with any large station, contains plenty of interesting signals - route indicators, backing signals, centre-pivot signals and impressive brackets litter the station. 4 signalboxes control the station - Didcot East Junction, with the Avoiding lines for Oxford and the junction for Newbury; Didcot West End, with the Chester lines splitting off for Oxford; Didcot North, with the junctions between the Avoiding lines, the Chester lines and the West Curve; and Didcot Foxhall Junction, at the junction of the West Curve and the mainline, also controlling access to Milton Ordnance Depot.

The layouts and signalling from Purley as far as Didcot are mostly complete.


Having started copying out the wtts into a spreadsheet (probably multiple - by 9am, down trains between Paddington, Didcot and Oxford already number almost 300!), I’m becoming more and more tempted to extend eastwards towards London, and northwards towards Banbury. This is mostly because the London area adds some really interesting traffic - several strings of express trains chase each other on five-minute headways as far as Reading, and the interconnecting lines add some fun routings (London-High Wycombe via Maidenhead and London-Oxford via Princes Risboro’, for example). Of course, with complicated train services comes complicated layout building…