MikeOxon

I can't resist this photo including a well-decorated 'coffee pot' at the opening of the 2nd Tay Bridge

Good to see a subject like this given airing in RM

perhaps they took their cue from tramway wagons like this one seen at the Forest of Dean Heritage Centre:

Don't do that - you'd be depriving lots of people the opportunity to display their knowledge 🙂

Thank you for commenting, Mikkel. I took that header photo with the model lying on its side and the frame and wheels are just resting on the upper side. The other side is just shadows! I need to think about how to support the boiler and firebox between the frames as there's no guidance from the drawing. In this case, I rolled my own brass tube and then realised that I didn't actually need to solder the join, since the natural 'spring' in the split tube holds everything firmly together! Previous models roll along quite freely, so the weight seems sufficient but they do need something motorised to push them. I'm building these out of curiosity and they'll probably remain static. That front axle in front the of the smokebox looks odd and Sharp's standard gauge engines had the more usual arrangement. They must have thought that a longer wheelbase was appropriate for the broad gauge. A pity that some of the early carriage and wagon builders didn't think that way too, since some vehicles had 6' wheelbase on the 7' gauge!

Great to know you are feeling much better, Annie. We need your 'cheer up' pictures and glimpses of your trainset world, over here, more than ever at present. Pleased to know you enjoy 'narrative' dreams - when I describe mine, people around me are surprised by the scenarios my mind creates from somewhere 🙂

In my previous post, describing my design method, I wrote: “This method also means that I do not need to make any of my own measurements but simply rely on the drawing being accurate and traceable” A lot of early railway drawings were far from accurate! I found the following in a collection belonging to William Strickland, an American architect: According to Wikipedia, “William Strickland's Reports are the starting point of American railway engineering, and represent the state of knowledge as the first railways were planned in that country." - so that’s where they went wrong 🤠 I should also add a caveat that the drawings I use need to be complete. In the case of ‘Eagle’, there is only the side-elevation by E.T. Lane, so a lot of details that might be be revealed by plan and end-elevation views remain as guesswork. This guesswork can, however, be informed by other sources of information. There are fortunately more drawings of the contemporary engine ‘Aeolus’, albeit from a different manufacturer. The general layout of items such as smokebox doors and back-head controls can be assumed to have been reasonably standard at that time, so I drew on my earlier model of ‘Aeolus’ for these details. I also know that ‘Eagle’ was based on similar engines built by Sharp, Roberts & Co. for the standard gauge. This knowledge enabled me to make further deductions about the layout of the frames, which were inside the wheels for the broad-gauge. In addition, there is the curious omission of springs on the rear axle in the ‘Lane’ drawing, although springs are clearly shown on similar standard-gauge engines from the same builder. Even more curious is that the much later drawing by G.F. Bird (Locomotive Magazine 1901) of the sister engine ‘Lion’ also omits rear springs.! I strongly suspect that, although purporting to be ‘Lion’, this Bird drawing was actually based on Lane’s earlier drawing of ‘Eagle’, although the Sharp, Roberts valve gear, shown on the Lane drawing, is missing. ‘Lion’ – drawn by G.F. Bird – Locomotive Magazine 1901 I made the decision to add rear springs to my 3D model. Splashers To return to the problem of modelling splashers … I have tackled similar splashers on other broad-gauge engines but those shown in the Lane drawing pose additional problems. There appear to be no backs to the splashers, which removes any potential means of support in a model. The drawing does not indicate how these splashers were attached to the frames of the engine, either. After some thought, I adopted the following method: I first traced the outline of the outside valance from the Lane drawing and used this to extrude into a solid component of 4 mm depth. Splasher Component based on Lane drawing This gave me a robust structure to place over the wheels but which needed thinning on the underside to clear the wheel flanges. I moved this component into position relative to the rest of the model and added extensions on the inside, to butt up against the main frames of the engine. I also reduced the thickness by using the ‘offset’ tool on the underside faces, so that the splashers would be clear of the wheel flanges. Finally I added some tabs to the extensions which I intended to fit into matching slots cut into the frames. I have often complained about the absence of alignment tabs on commercial kits, so I felt that I should not make the same omission on my own model! Underside of splasher after adding extensions Now, after adding various details, including springs, buffers, and lines of rivets, which are easily created with the aid of the ‘pattern on path’ tool, I had a set of components ready for printing. My 3D model of Eagle in Fusion 360 One further detail that I have considered is the ‘box’ that appears behind the smokebox in the Lane drawing, where the front axle might normally have been expected to be placed. According to Ahrons (‘British Steam Locomotive’, p.39): “The exhaust passages of Sharp's engines of 1839-1842 were peculiar. Both exhausts were discharged from the cylinders into a cubical box placed midway between the inside cylinders, and the blast pipe was connected to the top of this box, which was intended to act as a form of " air vessel " to equalise the blast. It caused considerable back pressure in the cylinders.” I am guessing that this may be the box shown in the Lane drawing behind the forks of the valve gear. Printing That’s the hard part finished, so now it’s a matter of exporting the files for each of my components from ‘Fusion 360’ into the slicing program ‘Cura’ I use this program to align each component in an appropriate orientation on my printer bed and I also check the ‘preview’, after slicing, to ensure that all the parts should print correctly. At this stage, I found that the driving wheel spokes were a little too narrow, so went back to Fusion 360 and used the face offset tool to increase the spoke widths by 0.1 mm. After these minor adjustments, it was then simply a case of loading the GCODE files onto a memory card, inserting this into the printer, selecting a file on the touch screen, and pressing the ‘Print’ button. After several short printing sessions, my ‘kit of parts’ looked like this: 3D-printed Collection of Parts I was especially interested to see whether my planned method of assembling the frames and splashers would work. The slots in the frames needed a little cleaning out with a scalpel but, after that, everything fitted together neatly. The splashers themselves are very fine and not too robust – this will not be a model for the grandchildren to play with! Frames and Splashers after assembly I used my now-standard method of applying a low-temperature soldering iron to ‘weld’ the splashers to their associated frames. Using a soldering iron to join the splashers to the frames I was pleasantly surprised by how well my printer managed the boiler fittings, including the safety valve housing, which is only 1.6 mm diameter. 3D-printed Dome and Safety Valve Cover The boiler, smokebox, and firebox fit together over a 14 mm diameter brass tube, which provides weight to the model and aligns these parts on a common axis: 3D-printed Boiler, Frame, and Wheels Chasssis Next, I need to think about assembling the chassis. Because I only have a side-elevation drawing, there will have to be some informed guesswork, especially as the drawing gives no information about boiler supports or other structures. I shall pause here and do some more research before continuing with painting and final assembly. Mike

When I edit an old blog post, for example to restore missing images, the time of posting now defaults to 'immediate' and the original post date has to be re-entered manually. This leads to edits of old posts being bumped to the top of the blog list, which is usually not appropriate. Also I no longer seem to be able to award star ratings to blog posts

You seem to be on a roll with your printing, Duncan. I'm not sure about U23 having a clerestory though - http://www.gwrcoaches.org.uk/index.html suggests it was a 4-wheel version of U21. Mike

Much as I enjoy your posts, Mikkel, this one seems to be bumped to the top whenever you make an edit! The same thing happened to me when I restored some images to an old post and I had to re-set the date back to the original. Is this the effect of a recent software 'upgrade'?

I wanted to give this 5 stars but the software wouldn't let me - is this the result of a recent software 'upgrade'?

I agree that carriages, especially when panelled with clerestory roofs, are a serious modelling challenge. And after the build, there is the painting, too! Your D14 looks wonderful and stands up to photographic scrutiny with flying colours. Excellent modelling!

Jubilación indeed! I've enjoyed 14 years of retirement now and did lots of world-wide travelling until covid struck to cramp my style. Your photos look wonderful - somewhere that I've not been but within fairly easy reach.

As an avid reader of Arthur Ransome in my younger days, I recall that, even in those pre-war days, the Norfolk Broads weren't always havens of peace and tranquillity. Put a portable gramophone on the roof of that cabin cruiser, playing at full blast, and you have a good portrayal of the 'Hullabaloos" from the book 'Coot Club'

Quite possibly but I've not found it yet - there's always lots more to learn about any software package. I blush when I look back at some of the contortions I went through to achieve things that I now know can be done relatively simply 🤫 I expect that in the future, I will look back on some aspects of current posts in a similar way. A few people have pointed out better methods in the comments and I always welcome these.

I agree and I should have said that's what I do! When I move parts to the side, I move them by a standard distance - usually 50 mm or 100 mm depending on the requirement. After fettling a component, I can then place it back exactly where it came from.

Yes, that's exactly the message I wanted to get across. I had to go through a fair bit of pain at the start of my 3D modelling, until I realised that most shapes can be produced by using either the extrusion or revolve tools. The automation tools make repetitive things like spokes, planks, or bolts/rivets very easy too. Mike

i doubt whether anyone knows but i doubt it. These engines were based on the 'standard' Sharp, Roberts engines which, for standard gauge, were outside framed. For the broad-gauge order, they moved the wheels outside their usual frames, which I suspect would have been iron plates.

Thank you, @Lacathedrale. I align the parts by eye in most cases - firebox and smokebox centred to boiler and butting up at each end, etc. The wheels are aligned with the gaps in the hornblocks and their lateral separation is set to match the broad gauge track. The alignment in 'Fusion 360' is merely to check that the overall model appears as it should. I convert each component separately into an .STL file for printing, so the 'real world' alignment is made after printing. More about that in my next post. Mike

I notice that the photo is captioned as taken at Tal-y-Llyn lake, Snowdonia National Park, Gwynedd, North Wales. Does this suggest that slates from here were shouldered?

Where do you buy your chips? When I followed your link to Amazon, the price was £145

Over the summer, I had little time for modelling, although I did continue with my reading, especially about the early engines of the GWR. Although these have been widely dismissed as ‘freaks’, this term only really applied to a very few examples and most were simply too small, because the manufacturers were attempting to meet Brunel's weight specifications, which were virtually impossible. At that time in the late 1830s, locomotive design was still in its infancy but a few manufacturers were working hard to raise standards of construction. One that stood out during my reading was Sharp, Roberts & Co. and especially the efforts made by their co-founder, Richard Roberts. Although little known in the locomotive field, Roberts was a prolific inventor in the field of textile machinery, which was the back-bone of the industrial revolution. He had an excellent grounding in engineering from his time at Henry Maudslay’s works, which was at the forefront of contemporary engineering practice. One of Roberts’ many innovations was the use of piston valves, which he applied to three engines supplied to the GWR – named ‘Lion’, ‘Atlas’, and ‘Eagle’. Unfortunately, these were not a success, possibly because of differential expansion between the valve components, which led to excessive steam leakage but, otherwise, these were very sound engines and performed well, once new cylinders had been fitted. Two of these engines survived in branch-line use until 1872, by which time the broad-gauge was in full retreat, with only the West Country main line lingering on until 1892. . I was interested to note that, like ‘Aeolus’, which I modelled earlier this year, ‘Eagle’ also worked on the Abingdon branch, which happens to be my home town. I thought, therefore, that ‘Eagle’ made a good candidate for my next early broad-gauge model. The spare part for my Geeetech printer arrived from China more rapidly than I had expected, so I was ready to go but, first, there is the task of creating the 3D model in ‘Fusion 360’ My Modelling Work-flow It’s a while since I have created a new model, apart from re-furbishing a carriage, as described on my pre-grouping blog. During the interval, I seem to have almost subliminally developed a fairly rapid design work-flow. As I mentioned in an earlier post, ‘Fusion 360’ is designed around the concept of starting from an overall design and then breaking it down into components - a ‘top down’ approach. In my earlier attempts, I used to create the individual parts separately and only bring them together after printing. Now, especially with these early engines, where prototype information is limited, I start from an existing drawing and then extrude each component in turn from the drawing, before bringing them all together within a single overall project file in ‘Fusion 360’. What I have found is that working in this way doesn’t need much computer modelling expertise, beyond what is needed to create a two-dimensional (2D) drawing. This is because most engine components can be created either by linear extrusion from a 2D drawing or by rotating a profile from a 2D drawing. There is rarely any need to consider more complex shapes involving double-curvature. This method also means that I do not need to make any of my own measurements but simply to rely on the drawing being accurate and traceable by means of the ‘line’, ‘circle’, and ‘arc’ drawing tools in ‘Fusion 360’ With this simple concept in mind, I find it surprisingly easy to put together the main elements of a new locomotive in a remarkably short period of time – and it can all be done while sitting in an armchair 🙂 My Modelling Method, Step-by-Step. This post describes the steps I have taken in order to create a 3D model of ‘Eagle’ in ‘Fusion 360’, although the methods I shall describe have general application. In the case of ‘Eagle’, all that I had to work from, apart from some references to cylinder and wheel sizes, was the very good side-elevation drawing by E.T. Lane, the young apprentice who provided us with most of our knowledge of early GWR engines. Sharp, Roberts & Co. ‘Eagle’, drawing by E.T. Lane, 1849 My first step was to import this illustration into ‘Fusion 360’ as a ‘canvas’. This canvas represents the ‘top level’ concept of my model, which I then break out into individual components. Fusion 360 – Top-Down Assembly I could now start to create ‘components’ by using the ‘Fusion 360’drawing tools to trace over elements of the drawing by Lane. I started with the frames, which are of an interesting design, as used in many of Sharp, Roberts & Co. ‘standard’ 2-2-2 engines. The sloping frames appear to have been used to avoid the need for excessively long supports for the small carrying wheels. My own first drawing of one of these frames in ‘Fusion 360’, produced by copying over the ‘canvas’, looked like this: Drawing the Frame over the ‘canvas’ Working on the imported ‘canvas’, I selected the ‘draw’ mode in ‘Fusion 360’and drew a line from the marked start point to the first curved section. Then, I changed to the ‘3-point arc’ tool for the next segment, then back to the ‘line’ tool, and so on, until I had completed the circuit of the frame. With practice, this can be quite a rapid process, as each segment follows the end of the previous one. Once a ‘closed’ outline has been created, it is only necessary to click on the enclosed area and select the ‘push-pull’ tool (shortcut key ‘Q’), to extrude the frame into a solid 3D ‘component’. I moved this first component off to one side, then created a duplicate for the opposite side of the engine. Correct positioning of the two frames can follow later. Next, I created the boiler. Information from the RCTS booklet on broad gauge engines told me that the diameter should be equivalent to 3’ 6” (i.e. 14 mm in 4 mm scale), so I used the drawing tools to create a 14 mm diameter circle. The cladding lies outside this diameter, so I used the ‘offset’ tool to create another circle 0.75 mm (or 2.5” on the prototype) outside the first. Then I clicked ‘Q’ to extrude the enclosed space between the circles to the length of the boiler – in this case 8 feet (32 mm in 4m mm scale). Then, I used the ‘move’ tool to align the boiler between the frames previously created. Next came the smokebox, which has a larger diameter than the boiler. I took account of the difference by using the ‘canvas’ and created another circle. I added a couple of tangential vertical lines each side of the circle, to create an enclosed space corresponding to the front profile of the smokebox and extruded this to the length of the smokebox. By working over the ‘canvas’, there is no need to make measurements – simply extrude until it corresponds with the ‘canvas’, as shown below: Extruded 3D Components All the coloured items are ‘solid’ (3D) bodies, with their dimensions determined by extruding over the ‘canvas’ – no measurements needed apart from the diameter of the boiler inside the cladding. The firebox was created in exactly the same way as the smokebox and so, after a remarkably short time, I had created all the major components by repeated copying from the ‘canvas’. As mentioned above, I found it easier to move each completed component to one side before starting the next one. After moving them all to one side, the 3D model in ‘Fusion 360’ looked like this: Major Components brought together in ‘Fusion 360’ Creation of the boiler mountings needs a different tool – the ‘revolve’ tool - but the method is much the same. For the chimney, dome, and safety valve cover, I drew lines and arcs over half of the profile, as shown below: Creating the Chimney A touch of the ‘revolve’ tool and I had a ‘solid’ chimney, complete with flared top and bands between the sections of the prototype. The dome and safety valve followed by exactly the same process, so that my model in ‘Fusion 360’ now looked like this: . Assembled body in ‘Fusion 360’ So far, the ‘bones’ of the engine had come together very rapidly but, as always. The ‘devil is in the detail’. Fortunately, ‘Fusion 360’ provide several tools that automate repetitive adding details. One example of this automation is provided by the spoked wheels. I started by drawing circles over the boss and rim of the driving wheel shown on the ‘canvas’. I then used the line tool to trace over the profile of one of the spokes, as shown below:. Drawing a Wheel – First Spoke – in ‘Fusion 360’ Completing the wheel is then simply a matter of using the ‘pattern on path’ tool in ‘Fusion 360’, to make the required number of spokes, all following the circular path of the rim. This automated process ensures a pattern of identical spokes, evenly distributed around the wheel – all at the click of a mouse! One carrying wheel followed, by the same process, and I then copied the required number of wheels and moved them into their correct locations around the other completed components. I used the same automated process to create the planked wooden cladding around the boiler and firebox. I simply had to create a single plank and then use the ‘pattern on path’ tool to produce a complete array of planks all around the boiler and around the top and sides of the firebox. The model was now looking like this in ‘Fusion 360’: Wheeled Model in ‘Fusion 360’ Apart from detailing, the model is now almost complete except for one glaring omission – splashers! These peculiar bicycle-like splashers, as used on many broad-gauge engines can be a nightmare for modellers because of their complex shapes and close tolerances around the wheels. Again, I think the difficulty is actually less when using 3D-modelling than when using brass or card but it still needs careful thought. I have, however, created splashers of this type before, on my models of a Gooch Standard Goods and of ‘Aeolus’, so I shall finish this post here and continue with the detailing next time. Mike

Be careful, we might be holding out the begging bowl soon - though I gather currencies aren't doing too well in Scandinavia either.

Another post that reaches places others cannot reach 🙂 How come you have such an extensive knowledge of British history? I doubt many school children here know anything about such matters. Beautiful modelling, as usual Mike

so they had Bank Holidays on Saturdays in those days.