TangoOscarMike

I'm starting a new thread for this topic, rather than further derail this thread: This first post is copied from that thread. I will start with excuses: my workflow is a little convoluted. I tend to make components with OpenSCAD and assembly them with Blender. And I quite often write Python scripts to modify STL files or, in some cases, to create STL files for a particular exotic object. This will only make sense if you install OpenSCAD and try the examples. The starting point for a piece of coal is a low-resolution sphere - the resolution ($fn = 5) is so low that it is actually rendered as two truncated pyramidal cones. coal_r = 1.5; # coal lump "radius" - a large piece of coal if units are mm and the scale is 1/76 sphere(coal_r, $fn = 5); If you enter this into OpenSCAD and render it (F5 or F6) you will get this. This is a little bit like a piece of coal, but it needs a bit more randomness in its shape, and of course we need many different pieces. This is achieved by making two such spheres, each of which is stretched/squashed by a random amount in the X, Y and Z directions. Each sphere is then rotated randomly about all three directions. The final result is the intersection of these two spheres. If you enter the following into OpenSCAD... coal_r = 1.5; module piece_of_coal(){ scale_first = rands(0.6, 1.4, 3); // produces an array of 3 random numbers in the range 0.6 to 1.4, for stretching scale_second = rands(0.6, 1.4, 3); rotate_first = rands(-90,90,3); // produces an array of 3 random numbers in the range -90 to 90, for rotating rotate_second = rands(-90,90,3); intersection(){ rotate(rotate_first)scale(scale_first)sphere(coal_r, $fn = 5); rotate(rotate_second)scale(scale_second)sphere(coal_r, $fn = 5); } } piece_of_coal(); ... then every time you render it you will get a different piece of coal. The coal load is made of a collection of these. These have to be pseudo-randomly placed, but it's necessary to exercise some control over where they go. Obviously they need to be inside the bunker or tender, but one might also wish to impose some sort of contour over the top surface.

Indeed - I would expect that a technique similar to mine (intersections of randomly distorted low-resolution "spheres") would also be possible in Blender.

Soon I'm going to post another OpenSCAD script with multiple pieces of irregular coal, which I hope you will find more natural-looking. I think I'm going to start a new thread, since although @nberrington hasn't objected, it might be good to have a thread title that matches the content.

I haven't done this for a while, so I will review what I did and reconstruct it here in installments (@nberrington - let me know if you'd like me to take this into a separate thread, or indeed if you'd like me to explore stone textures along similar lines). I will start with excuses: my workflow is a little convoluted. I tend to make components with OpenSCAD and assembly them with Blender. And I quite often write Python scripts to modify STL files or, in some cases, to create STL files for a particular exotic object. When I write OpenSCAD objects, I try to keep dimensions in a separate file, but I often slip into the bad habit of hard-coding dimensions deep inside my script. In this case, for example, the starting point for a piece of coal is a "sphere" of radius 2 (i.e. 2mm for a 00 model). This will only make sense if you install OpenSCAD and try the examples. The starting point for a piece of coal is a low-resolution sphere - the resolution ($fn = 5) is so low that it is actually rendered as two truncated pyramidal cones. sphere(2, $fn = 5); If you enter this into OpenSCAD and render it (F5 or F6) you will get this. This is a little bit like a piece of coal, but it needs a bit more randomness in its shape, and of course we need many different pieces. This is achieved by making two such spheres, each of which is stretched/squashed by a random amount in the X, Y and Z directions. Each sphere is then rotated randomly about all three directions. The final result is the intersection of these two spheres. If you enter the following into OpenSCAD... module piece_of_coal(){ scale_first = rands(0.6, 1.4, 3); // produces an array of 3 random numbers in the range 0.6 to 1.4, for stretching scale_second = rands(0.6, 1.4, 3); rotate_first = rands(-90,90,3); // produces an array of 3 random numbers in the range -90 to 90, for rotating rotate_second = rands(-90,90,3); intersection(){ rotate(rotate_first)scale(scale_first)sphere(2, $fn = 5); rotate(rotate_second)scale(scale_second)sphere(2, $fn = 5); } } piece_of_coal(); ... then every time you render it you will get a different piece of coal. The coal load is made of a collection of these. These have to be pseudo-randomly placed, but it's necessary to exercise some control over where they go. Obviously they need to be inside the bunker or tender, but one might also wish to impose some sort of contour over the top surface. It's not my intention to leave on a cliff-hanger, but I know that I tried a few different techniques for this, and I can't remember which I preferred in the end. I will go delving in my files tomorrow.

That is indeed very similar, and almost certainly better (less laborious!) than my technique.

To achieve the flat inside the bore of the gear wheel, I: Filed flats on either side of a piece of brass tube, until the flats became slots Opened out the slots with a file, then concentrating on one side, filed the surviving metal from inside, until it was just below 0.5mm thick Tinned the outside of this portion Pushed the tube into the gear, supported by a steel rod up the middle Applied heat to the outside of the boss, to solder the shim to the inside of the gear Removed the steel rod, and cut away the surplus brass Well, that was the idea, anyway. Steps 1 and 2 went pretty well. But around about step 4 it started going wrong, with the thin brass buckling, followed by the usual shameful blundering that occurs whenever I wield a soldering iron. But in the end, after some recovery effort with the broken bits, I have a result that works - the gear fits snugly on the motor's shaft with hardly any play.

From time to time I stick my nose into these threads to mention OpenSCAD. If you don't mind a little programming (or a gentle opportunity to learn!) then you might be able to use it to generate pseudo-random surfaces procedurally. This is a coal load for a 3D printed tender. Let me know if you'd like a description of how it was made. I expect that similar techniques could be used for irregular stone surfaces.

The gear has no grub screw. I did toy with the idea of drilling & tapping, but I think that might be a bit too fiddly for me.

Earlier in this project I made a botch of enlarging the 2mm hole in a gear wheel to 3mm. This time I took a bit more care. Lacking a lathe or a pillar drill, I made a jig to make a jig to properly align the drill bit with the bore. I am satisfied with the resulting enlarged hole. I took the gear wheel from the back axle for this - fortunately I have some extra gears, with bosses that are too fat to fit under the motor. The back axle isn't under the motor. The motor shaft has a flat, and I want to make a corresponding flat inside the bore of the gear wheel. Plan A was to file a flat channel, and solder a brass strip into it. Plan B was to attack from the outside of the boss. I think that Plan C is better. I'm going to solder a sliver of brass tube inside the bore. And I'm going to contrive this sliver such that when I solder it in place it is still supported by the original tube, and by a steel rod up the middle. Upon reading, that all looks like gibberish, so I'd probably better take some pictures.

That's the sort of thing I had in mind. If this extra part had tabs on the sides, as well as the top, then the square end of the bogie could be gripped on three sides. If you made it snug enough (or slightly oversized with the tabs "sprung" inwards) then it would not need to be glued or otherwise fixed at all.

I share your taste for making things that can be dismantled. And I try to avoid modifying proprietary mechanisms, by designing my additions around them. I would be inclined to have parts of your frame "locate" onto features of the bogie. The ends of the bogie look square-ish and there is a step at the top upon which parts of the frame could rest. And there holes on the ends might accept a stubby peg. In fact, I would side-step the need for craftsmanship by designing a 3D printed frame that could clip onto the bogie. And I might also incorporate the cosmetic frames into the design.

After some reflection, instead of buying a wheel puller (this time), I made a wooden support so that I could rest the whole chassis on the gear wheel in question, and then I used a nail to drift out the axle. I then did something I should have done before. Checking with a multimeter showed that neither electrical contact is connected to the metal motor body. This means, I think, that it's safe for the wheels on one side to come into contact with the motor, which means that some of my extra insulation was unnecessary. Removing some of this plastic allowed me to assemble the whole thing with a back-to-back not of 14.5mm, but below 15mm. The movement (before I put the motor in) was a little stiff, which I attributed to friction between the wheels and the pickups. Fully assembled, and with power applied directly to the motor, the wheels turned at a satisfactory speed. On a piece of test track it moved a little bit, and jerkily. Pushing it down with a finger (to experimentally improve the electrical contact) broke it in an unsurprising way - the hot glue collar between the motor shaft and the first gear wheel failed - the motor rotates freely now without driving anything. From an optimistic point of view (not really my forte) it is good that the worst bodge is the thing that failed. So before I take the lessons I've learned and make an improved design, I will try this design again with the first gear properly attached to the motor shaft.

Well, I think a delicate wrought-iron spiral staircase would be about right. Perhaps there are suitable photo etches. Easy for me to say, since I'm not making it.

I too use a piece of wire to unblock these nozzles. I've been using (and trying not to lose) a single strand from some multi-strand wire, but I reckon a guitar or violin E string would do.

Ah. Apart from some ancient and decayed Hornby stuff, my track is Fleischmann HO. The OO rolling stock of my childhood is happy on it, but I don't know how similar it is to track that is actually sold as OO. I think I will have to make some sort of pulling tool in order to get the wheels off, but I reckon it's all good education. And next time round I will measure the back-to-back with toothpicks for axles first.

I've been overwhelmed by work, and I'm a natural procrastinator, so when I hit a snag I stopped modelling altogether. And now I would be grateful for advice before I try to move forward again. This blurry picture shows the middle pair of wheels in place. They turn smoothly, and the gear wheels mesh together as intended. But the back-to-back measurement is 15 mm, not the 14.5 mm that I understand to be correct. And if I push the wheels closer together then they no longer turn freely. Maybe if I take the wheels off and drift the axle out then I will be able to get it down to 14.5 mm by filing back the ends of the bearings. 0.25mm each side doesn't seem like much. But I'm wondering if this is necessary. There is still considerable side-play when I put it on OO track, so I wonder in what way the 15 mm back-to-back would go wrong. Could it cause derailment or worse disasters?

Here are the four assemblies of the front wall. And here they are dry-fitted. So: It might end up being a little grander than I intended, but that doesn't matter. There are some errors in the brick courses that must be my fault, although I took care to avoid them. I shall press on regardless. The porch is wonky. I think I might remake it in-situ. For the time being, I've glued everything except the porch together, and I'm going to disguise the seams with clay next.

Here is the beginning of the front wall. These pieces, and the connecting wooden strips, were glued together face down on a piece of glass, with the window pieces in place to enforce the spacing and alignment. From behind it looks even more haphazard. Here is the left hand set of projecting pieces with windows, ready to glue into the wall. The window pieces do not have mortar lines on their edges, so I have scratched these in using a piece of piercing saw blade pushed into a broken paintbrush handle. The large and the skinny plain brick pieces do have mortar lines on their edges, but alas they are wrong - the line of bricks does not continue cleanly around corners, but instead there are (effectively) double-sized brick slabs on top of half-bricks. I'm not going to try to correct this - instead I will be content with the general impression of brickiness.

This is the assembled porch, with a door at the back and paint on the inside. The wooden stick is a temporary measure to prevent movement. There's a risk, I think, of too much uniformity in linka buildings. To avoid too much uniformity in the proportions of this building (and becuase I like to break symmetry anyway) I'm casting rather a lot of this piece. Here, prior to trimming, are pieces for the front wall. The arch in the middle is a placeholder for the porch, which is slightly off-centre. The windows will not be glued in the plane of the front wall: they will be projected out a bit.

I'm defying the old instructions and putting mortar on last. This is to take advantage of the terracotta plaster, and because I want to fill the mortar lines a little bit - they are far too deep to be a realistic texture (except for buildings that need emergency repointing). I won't, however, fill the gaps all the way up, because they are tapered. Doing so would result, I think, in excessively wide mortar lines. I masked the edges (they will be finished after the walls are joined together) and then applied two coats of gloss varnish. My mortar is a paste mixed from varnish, acrylic retardant, various shades of brown paint, and talcum powder. These are mixed to the colour and texture of gunge. I tried cornflour first, but it was too granular. After applying my mortar over the whole surface, I scraped it off with a thin piece of plastic, and let it dry. This is the result (photographed indoors and in natural light). The bricks are considerably dulled by a thin film of the mortar paste. I had a little go at this with 1200 grit wet and dry, and then changed my mind and stopped. I will touch up the paint around the windows, and there will be a coat of matt varnish later. I'm reasonably happy with this. The joints could be better disguised, the overall effect could be more realistic. But I'm not trying to win the Turner Prize, so I will crack on.

A few more bricks coloured in. I kept a tally, to help achieve a similar colour-density on the other walls. And a bit of paint around the windows.

Certainly older, I'm not sure about wiser. Maybe slightly less ignorant. And currently 451 miles from home!

Bricks picked out with felt-tips: pink, red, orange, brown and yellow. The pink one gave up the ghost early on, so there aren't many of those. I might colour in a few more before the next step. I've got a night shift tonight. I should be getting some sleep, not indulging myself on RMWeb.

Ten feet? Gadzooks! Indeed - very sharp detail can be captured. Bubbles/voids are a curse, also with metal casting and I daresay resin. For my most recent set of Linka castings I sprayed the moulds with a watery mist (having previously just sloshed water on them), and this was a huge improvement. Oh, absolutely. And perhaps the most serious objection to Linka is the fact that one doesn't get a choice of authentic brick bonds. But I lean somewhat towards the toy end of the hobby, so that's OK.

I remember a casting of Peter Rabbit, probably made by big brother. My opinion of such things these days is that it's not nice to give children craft projects which produce ornamental nick-nacks. Nobody is going to play with a crudely painted plaster cast of Peter Rabbit, even if it doesn't fall apart (which it will). I think it would be (would have been?) better to give children rubber moulds in which (say) 54mm knights can be cast, to go alongside figures from Britains, Airfix etc. Obviously the figures would have to be in very compact poses, but I don't think this would spoil the idea. The Prince August moulds are pretty close, conceptually. I sculpted and cast (in metal) some knights and princesses when I made a toy castle for my daughter some years ago. And I experimented a little bit with figures suitable for casting in plaster. And then time flew by and now we're here.