Train Sim World


We spoke to the team working on the complex signalling of the upcoming Spirit of Steam: Liverpool Lime Street – Crewe, finding out what research is needed to bring you such a vital aspect of railway logistics, and how they are conquering this mammoth task.
The Signalling Team give us an amazing insight into the lengths they have gone to, to bring you a realistic and immersive experience, and the challenges that come from trying to reproduce something from a bygone era when little or nothing remains of the item they are trying to reproduce.
We find out more about the overall research gone into the route, a detailed look at the numerous unique signal boxes that are dotted along the route and how this variety is going to be represented in-game, and the complexity of modelling signals that bridge the gap between the transition from semaphore to colour light in a busy junction area with complex architectural features.

Researching the Route

ST: The route, initially set in the 1950s, presented something of a challenge both in its scope and in its era for research. Documentation is harder to acquire from this period than present-day material, and many subtleties or local arrangements can get lost to history, especially if they were informal. Being able to collate comprehensive detail for a route the magnitude of Crewe-Liverpool in this era was no small affair, and my colleague pulled in many personal favours and undertook literal months of research to end up with something cohesive. It is highly likely that the information for the route as a whole has not been seen together since it itself was current, and this is very exciting. The research pointed us to a date of 1958; a year immediately before the start of largescale track/signal works in preparation of the electrification of the route, but at the end of a period of relative stability for the routes’ infrastructure.
Sources used include the National Library of Scotland’s Historic OS map collection, many researchers’ personal collects of plans and documents, railway photographic archives, old postcards, and countless other collections both public and private.

Modelling the LNWR signal boxes

ST: The 1950s was an era where mechanical signalling on main lines was still ubiquitous, though this wouldn’t last much longer with the BTC’s mainline modernisation schemes beginning in earnest. On this route, there are about 50 signal boxes, ranging in date from the mid-Victorian period to almost brand new (at the time). The largest family of these are from the LNWR period of ownership, and though of differing size to suit requirements, come from a family of modular parts the LNWR developed to enable what would nowadays be termed ‘rapid deployment’.
The kit consisted of standard size/arrangement windows, timber frames, brick plinths, doors, stairs, etc – very much modular from a building/builder’s perspective in the real world, but not necessarily as useful a breakdown from a blueprinting/game perspective. To significantly reduce the number of unique complete buildings of this style required to be made (over 40), a modular approach was again specified, but this time in terms of volumes rather than individual building elements.
A Pictorial Record of LMS Signals, L. G. Warburton
ST: This had to accommodate:
  • The differing lengths of windowed sections on top
  • Two different styles of the brick on the lower level (one inset and the other flush)
  • One or two brick levels underneath
  • A mismatch in the repeating lengths between the top and lower levels
  • The implicit need of how the top and bottom arrangements of what might be differing total length could be matched.
A review of the majority construction configuration of such boxes on the route eventually revealed the need for 9 distinct meshes, including a ladder, that could be combined to best represent the individual boxes as they appear on the route. Only two boxes were wooden in both base and top, so for brevities sake, they have been omitted (though the modular approach does not preclude these being made and used if needed in the future). This would allow the top of boxes to be assembled to match the actual construction, the dimensions of window widths and other elements of the arrangements being explicit, whereas the bottom layer(s) could be represented by a spline of scalable length to match the specific arrangement on top.
For the top, these consist of end pieces for the box, and two different central sections that two and three windows wide.
Images ©Dovetail Games
ST: For the bottom, these consist of ends with an access door or a window, and central sections either wide with inset brickwork or narrow with flush brickwork, that can be stretched to match the top.
Images ©Dovetail Games
This will enable a good representation of almost all LNWR boxes on the route to be built up, and of others from the same family in other projects, where needed.

Implementing the complex signalling of Ditton Junction

ST: Ditton Junction station eventually closed under Railtrack in 1994, but in the project’s era was still a busy place with some express trains calling, in addition to various local trains, with an active set of goods sidings and a rail connected works nearby. It was on the cusp of being thoroughly modernised, with its Victorian station building and waiting rooms being demolished and replaced by brutalist concrete structures.
As it will be represented in-game, Ditton Junction is in an early stage of transition. Formally a stronghold of exceptionally tall LNWR lower-quadrant signals, by the late 50s these signal posts were due for replacement and had become a maintenance liability. Such was the urgency of replacement that new 4 aspect (albeit working as 2 aspect temporarily) colour light signals specified as part of the modernisation/electrification project, were installed to replace arms shorn from the signal posts. A series of pictures from magazines and collections demonstrate the transformation.
British Railways Illustrated (Vol 28, Nov '18)
ST: In this 1957 dated picture from British Railways Illustrated, (above), the transitory state is seen with the posts still standing by this time without arms, but colour light replacements with feathers cut in. As can be seen, the replacements are further back tight against the foot overbridge.
This gives good points of reference for where the signal links should be placed and the visuals sighted, however; the station buildings and overbridge/staircases, along with the huge signal posts, were subsequently demolished. The platforms were extended/rebuilt, and the colour lights were removed from a cantilever structure and put on posts, leaving no modern-day reference point that existed at that time as well. The incredibly tall signal posts present a spatial challenge to arrange in 3 dimensions being near the end of the platform but not clipping into the station buildings above and over. In addition, from this angle it's not entirely clear what the relationship of the staircases, overbridge, platform ends, and signal locations are.
This presents an interesting puzzle where multiple different photos from different angles of wildly differing quality need to be consulted together, and educated guesses made to fill in the blanks.
British Railways Illustrated (Vol 28, Nov '18)
ST: This photo from the same magazine show how this platforms’ end relates to the staircases, also showing the position of the big signal post as being about halfway along the staircase (stay wire and post being to right). Other views taken from a completely different angle show that the bottom segment of all staircases to each platform were in alignment at the platform level.
British Railways Illustrated (Vol 28, Nov '18)
ST: The final angle we have is this image showing that the end of platforms 5 & 6 were both in line with the end of 3 & 4, but also further away from the road overbridge. This is because the station building is over the tracks, creating a deeper bridge for the pair of tracks on the right of the image above.
NLS (National Library of Scotland) 1905 OS map excerpt
ST: A quick check of the 1905 25 inch=mile OS map confirms this is the case, and doubly confirms the footbridge position as being in line for all spans, and with a gap between it and the road. S.Ps stands for Signal Posts, which are represented by the larger black dots.
So, we can now see how everything relates to each other and can visualise this from multiple angles. But how can we use this in-game if we still lack a definitive reference point? The Editor has a Google Earth plugin that overlays on the terrain a decal of the current satellite imagery for the area; however, the buildings and staircases we are looking for were demolished in about 1960, and external overlays of this station with current imagery do not align particularly well with the old OS maps of the area.
We can speculate, however, that since the constructional/structural arrangement of the station building and overbridge were similar before and after the rebuild, that the newer building and bridge likely take up the same footprint as the old. With this in mind, we can now directly use the current closed building and footbridge as a reference for the previous building by using current images - this should aid us in figuring out how the platforms and signals are placed in relation to each other.
Image ©Dovetail Games
Image ©Dovetail Games
Stay up to date with Spirit of Steam: Liverpool Lime Street – Crewe, and everything Train Sim World 2 by following @TrainSimWorld on Twitter and Facebook.
Check out the Railfan TV teams recap of all the exciting progress we have made on Spirit of Steam here!
Screenshots and images displayed in this article may depict content that is still in development. The licensed brands may not have been approved by their respective owner and some artwork may still be pending approval.
Train Sim World
29 Mar