“Towards the end of July 1837 I heard that Mr. Brunel wanted some one to take the post of locomotive engineer on the Great Western Railway, and I at once went to him, on July 20th, preferring that department to railway making.”

 

Thus wrote Daniel Gooch about the event that changed his life when he was just 21 years old. As a result, he left Manchester and went to London, beginning his duties with the Great Western Railway at West Drayton on the 18th August 1837. Because no engines had actually arrived at that time, he recalled that his first work was to prepare plans for the engine-houses at Paddington and Maidenhead,

 

In such circumstances, one might have expected the engine-house designs to be rather perfunctory affairs, something to fill in the time until some ‘real work’ on engines became available, but my investigations into the design of the engine-house showed this to be far from the case.

 

For Paddington, Gooch designed a ‘round house’ shed which, I suspect, was probably the first example of its kind. He was severely constrained in both timescale and costs. The accounts show an extraordinarily low figure of £1,402 for ‘general construction costs’, plus £216 for ‘other items’.

 

An early site plan, probably dating from 1837, shows the engine shed as below:

 

Paddington Site plan, 1837, showing the Engine Shed.

 

At first glance, the octagonal form seemed to match the track plan but I was surprised to see that the locations of the tracks corresponded with the vertices of the octagon, rather than being placed centrally in the sides, as I would have expected. Indeed, an illustration of the interior in ‘Measom’s Guide to the GWR’, dated 1854, appears to show such a layout – a reminder not to trust all those early engravings!

 

Looking more closely at the plans, however, I saw that the vertices were ‘squared off’, with entrance doors, where required, or short segments of wall. Thus, I started my modelling of the shed by copying the above drawing as a ‘canvas’ into ‘Fusion 360’ and tracing the foundations of the perimeter walls. That was the easy bit but where to go from there?

 

I had a couple of sources to draw on for guidance:

 

An early photograph of the demolition of the original station includes a glimpse of the engine shed roof in the background. This gave me a slope of 15 degrees for the main roof. Incidentally, several of the iron columns from the original station can be seen lying amongst the debris in the foreground, while a very tall disk & crossbar signal appears to the right. The sheer-legs appear on early plans of the original station layout.

 

Paddington Demolition works, c 1854-5

 

The other source was a much better engraving than the one in Measom’s guide. This one shows the clerestory roof over the centre of the shed and a rather bewildering forest of wooden supports for the roof. After studying this illustration for some time, I decided to take a ‘Stonehenge’ approach to the design of my model by concentrating on the locations of the upright pillars and hoping to fill in the rest of the details later!

 

Engine Shed Interior c.1846 (Firefly-class ‘Ganymede’ was new in 1842)

 

From this illustration, I deduced that there were two rows of pillars along the sides of each track between the entrance doors and the clerestory. Taken together, these formed an inner ‘ring’, supporting the edges of the clerestory, and an intermediate ‘ring’ between the perimeter. walls and the clerestory.

 

Creating a 3D Model

 

As I laid out the locations of all the pillars in ‘Fusion 360’, a very ingenious geometric pattern started to emerge, reminiscent of the vaults of a mediaeval cathedral. This seemed to me a far more sophisticated design than I had expected from a young man who had joined the GWR to do work on engines! It was a portent of the illustrious career that lay ahead for young Daniel Gooch.

 

Once I had determined the locations of the bases of the various pillars, I had to adjust their lengths to correspond to a slope of 15 degrees in the roof that they were designed to support. Rather than attempting to describe the process in many words, I have made a series of ‘screen shots’ of each stage, as my construction progressed in ‘Fusion 360’

 

Steps in my ‘Stonehenge’ approach to Modelling Paddington Engine Shed

 

Now that I’d worked out the overall layout, I decided to go back to the beginning and start again!

 

This time, having worked out the geometric pattern, I used the ‘Pattern on Path’ tool in ‘Fusion 360’ tu ensure that all the pillars and their linking ribs were aligned in a regular, symmetrical pattern around the central axis.

 

Bird’s-eye View showing Symmetrical Design of Shed

 

The various ribs connecting the tops of the pillars, in order to support the roof, were all copied from an initial 3D drawing of a square-section bar. After placing the first pair of ribs, I could use the ‘Pattern on Path’ tool to complete the rest of the pattern around the central axis.

 

Adding details

 

As an initial check on the general validity of my structure, I set up an internal view of my model in ‘Fusion 360’, to compare with the contemporary engraving shown above. I couldn’t match the perspective exactly but it gave me a good idea of what needed to be done to capture the ‘atmosphere’ of the original interior.

 

I started by adding the diagonal ties at the tops of my pillars. This was a fairly easy ‘tweak’ that I could apply to one pillar and then copy around the complete array, again using ‘Pattern on Path’ to maintain the alignment around the central axis. The result at this stage is shown below. I have included my model of ‘Vulcan’ to indicate the overall scale:

 

Interior view of my basic model of Paddington Engine Shed

 

I could find no information about the entrance doors but then I remembered some drawings that Gooch created for Maidenhead depot. These included shed doors (NRCA161482) so I extruded my model from the Maidenhead design and found that these doors were a perfect fit into the doorways of the Paddington shed. An early example of standardisation!

 

My enhanced version of NRCA161482 showing Doors for Maidenhead Depot

 

I didn’t have much to go on for the design of the interior walls, except the engraving shown above, which indicates vertical planking and window openings with vertical bars. I took some further design cues from the Broad Gauge Transfer shed at Didcot Railway Centre, which suggested diagonal cross-bracing on the large panels.

 

I started from a single rectangular panel on which I incised one ‘slot’ to represent a single vertical plank. I then used ‘Pattern on Path’ to complete the planking of the whole panel. Next, I created the various frame pieces and attached these to the inside face of the panel. Finally, I cut the rectangular openings for the windows and added an array of vertical bars within each opening.

 

My model Wooden Planked Wall with Bracing

 

The result looked pleasing and made me loath to add the roof, which would hide all this from view! Nevertheless, the roof was necessary to provide an overall impression of the interior scene.

 

Adding a Roof

 

For the roof, I could again take advantage of the circular symmetry of the structure, to reduce the amount of design work that was needed. I had only to design one triangular roof panel and one rectangular one, which then alternated around the ‘ring’. The clerestory roof is even simpler and only involves triangular panels.

 

I started with what seem the easiest one to draw – the rectangular panel. I checked the lengths of the support ribs with the ‘measure tool’ in ‘Fusion 360’, to ensure that panel would be a correct fit. After that, the detailing followed similar methods to those I used for the sides, with a pattern of rafters as indicated on the contemporary engraving shown above.

 

The triangular panels were rather more difficult. I established the lengths of the sides by again using the ‘measure tool’ and I also checked the angle at the top vertex, remembering that, because of the slope of the roof, this is not simply a fraction of the circle defined in the ground plan. I then added the intermediate supporting battens and finally the numerous cross-battens towards the apex. I wanted to use the ‘Pattern on Path’ tool again but realised that, if I started copying from the bottom edge, the battens would become too long as the roof segment tapered towards the apex.

 

My solution was to create an array of same-length battens, which I then cut by applying the ‘Splitting’ tool to the plane defined by the edges of the panel, as shown below:

 

Applying the 'Splitting' tool in 'Fusion 360'

 

The triangular segments of the clerestory roof followed, by using the same methods, so I now had three different type of roofing panel, as shown below:

 

Three Types of Roof Panel for my Engine Shed Model

 

I should probably have done lots of trigonometry to work out exactly how to place these panels on the roof, sloped at 15°, but I opted for an empirical approach and used the ‘Move’ and ‘Twist’ tools, while checking from different directions until they were seated neatly over the relevant rafters.

 

After that, it was back to the ‘Pattern on Path’ tool to complete the circular symmetrical pattern of the complete roof. The end result is shown below:

 

External View of my model of Paddington Engine Shed 1840

 

It all looks rather plain from the outside – a bit like a Victorian Gaol – but for a more interesting view, I could use the ‘Camera’ in the ‘Render’ section of ‘Fusion 360’ to look inside.  Here is an internal view with ‘Vulcan’ on one of the turntable roads:

 

Internal View of my Model Engine Shed, showing panelled walls and roof

 

I think the overall impression is in reasonable agreement with the contemporary engraving shown above and is probably as good as can be achieved on the surviving information. Another view, which includes the clerestory roof is shown below:

 

View into my Model of Paddington Engine Shed, through open Doors

 

 

Overview of Paddington Depot c.1840

 

Finally, I have placed my model within the context of the overall site, as it was laid out in 1840. See my previous post for details of the other models shown.

 

Overview of my Current Collection of Paddington Models

 

There are still several more structures to consider, although with very little information to work from. One item, listed in the accounts, is ‘stables’ but these are not marked on any of the drawings that I have.

 

I am currently thinking that they might be the extension on the side of the Engine Shed. I have not completed this part, except for the outer walls as marked on the overall plan. There would perhaps be a 19th-century logic in grouping together all the available sources of motive power within one overall building. Horses still played an important part in manoeuvring rolling stock between the various small turntables.

 

If anyone has any other suggestions, I shall be interested to receive them.

 

Mike