justin1985

This is what the design looks like I usually print them as just a bit of a "filler" on the build plate when printing other things, and probably don't pay as much attention to washing off excess resin as on other models, so they often lose a bit of the detail visible in the design. But as a replacement for the whitemetal ones, they seem to do the job for me! Justin

Certainly agree that there are an awful lot of myths repeated without basis. YouTube "experts" generally seem a bit clueless, and posters on the Facebook groups are even worse. (Not to mention the fact they're usually trying to print Trump branded gun accessories or something ...) The angel statute might not a fair comparison for most 2mm printing though! The resin process seems much more forgiving with organic shapes, so this statue, or a "mini" of an Orc or dragon, rarely needs much in way of support at all, as you say. But it's flat and / or straight edges, which are pretty common on trains, that can be difficult to print without a lot of support. These tiny T gauge coaches really did need this much support - basically a support at both bottom and top of each bit of underframe gubbins - otherwise they turned out banana shaped, or lost detail on the underframe to peel lines. Too little support overall and they got suction tear lines through the middle at the point of largest cross section (they're solid). The 22 degree orientation is intended to work with the pixel size and layer height (0.02mm) to minimise visible banding - that is one tip from the Photonster experts that I do follow. In fact I'd say the Photonster group on GitHub are the only real experts who I trust for advice. Their FAQ is very well researched and backed up by thorough experimentation https://github.com/Photonsters/anycubic-photon-docs/blob/master/FAQ.md One thing I would say about people overusing supports though - people don't seem to print flat on the build plate anywhere near enough. If something has a flat bottom and doesn't then have a significantly larger cross section higher up, it's almost certainly better to print flat on the plate and prise off with a sharp scraper. This is how I do the axleboxes, for example. Lots of people default to doing everything with supports, which definitely wastes resin, as well as introducing risks of distortion if the supports aren't sufficient or well aligned. Justin

Many thanks Marlyn! Longer term I'd definitely like to put together a scenic layout, perhaps with Culloden viaduct as a focus. It would just be "watching trains go by" though, perhaps with some kind of automatic shuttle for trains to alternate. I got stumped on coming up with a coupling that got the coaches reasonably close together without locking the bogies together too solidly to negotiate curves, then got distracted back to my projects in 2mm scale. I've just come up with a new design for a pivoting 1mm magnet coupler (similar to the "Hunt" couplers available in N) which I'll try printing soon though. If I can get that sorted, I'll move on to the other coaches for the ScotRail sleeper! Justin

I have a slightly rough and ready design that I knocked up to replace the white metal versions I had been using on Association RCH chassis kits. It's far from a precise rendition and definitely chunkier than the etched versions - in fact it's more of a copy of the old white metal ones than a model of the prototype! (the cardinal sin of a model of a model!) Happy to share the STL with Association members who'd like it, but I imagine the shop would prefer a proper scale design! Justin

I think balanced support is much more important than number of supports - but thinking through how that will play out through the print process can be tricky on some shapes (especially closed shapes at angles) - so it can be tempting to just pepper hidden surfaces with as much support as possible. I certainly have had prints I've rejected because insufficient supports led to flat elements of designs coming out far from flat (e.g. bottom edge of wagons looking banana shaped). I certainly tend to cover the hidden bottom of wagons, for example, with lots and lots of supports - but on an object that doesn't have an obvious hidden face that would provide much support, I'm much more discerning in placing fewer supports more carefully. J

This looks amazing Gareth! How did you do the lettering? Justin

Yup, understood Jim. It was the Templot forum post linked by @£1.38 advocated putting a slight set into the blade itself: "Put a set (that's a bend to you and me) in the rail at the planing length so that the end of the rail is aligned where the running face was (the picture should make this clearer). This location of this set is important since this affects the switch angle." [My emphasis] I'm sure this is written with larger scales in mind, but the principle sounds sensible? J

Many thanks for the tips chaps, really appreciated! I'm always learning with 2mm! Putting a set (bend) into the switch rail to put the planed section back into alignment had completely passed me by, but makes a lot of sense when you think about it! So far I've only got as far as the vee and the straight stock rail. I did put a very slight joggle into the stock rail - using the slot in single etch thickness method from the Track book, but then largely flattened back out again, so it's very subtle indeed. I'll give the switch rails that I've prepared a bit more attention to thin them down further. J

I'm sure I saw some discussion about turnout blade joggles in 2mm recently, but I can't find it again now. I seem to remember it was concluded they're not often prototypical. However I've always ended up using them so far, as it seems the only way to avoid wheels hitting the end of the blade. Does that mean I've never filed my blades thin enough? Justin

Many thanks for everyone's help! It was definitely that knuckle and flangeway. I managed to fix it by replacing the wing rail / knuckle was at the top of the previous picture. New one fitted using offcuts of rail held with bluetack as a spacer. That turnout now works nice and smoothy! However ... (there's always a however) ... I noticed a very slight jump on the plain PCB turnout at the end. This one was actually a very early effort recycled onto the layout as it was going to be buried in cobbles. Closer inspection and I realised one of the blades was actually sat at an angle, which was causing the problem. Of course, when I tweaked it upright with pliers and soldering iron, it was now out of gauge. My attempts to then tweak it to gauge introduced a load of kinks. So ... I've printed a new template, and I'll make a start tomorrow. I'm sure a new turnout will be much better, with everything I've learned since building the original one! J

Hi Nick, The drawbar is the original Farish one and pivots on the screw between 1st and 2nd axles, although the pick up wipers do constrain it a bit. It is the leading driving axle that jumps up when hitting the vee going forward. Looking very carefully the Jinty does the same very slightly, but I suspect being heavier, it isn't anywhere near as noticeable. J

There is definitely something out about the knuckles, although the distance can only be tiny ... The 4F, and other long wheelbase things, jump up as they hit the vee, then get pulled back by the checkrail. Whereas short wheelbase things pass through fine. J

So, I FINALLY got my Farish 4F conversion working after scratch building a new tender chassis frame. Really pleased - this is my first scratch built part of a loco! However, running an even longer wheelbase loco through this little layout has exposed more of a problem with the central turnout. I've tweaked this several times already after each larger loco I tried revealed problems I hadn't noticed with short wheelbase wagons. I suspect I've reached the limit of bodgery. I seem to remember I built this turnout with salvaged / recycled bits, and I suspect that's coming back to haunt me. Looking at it afresh, I think the gap through the common crossing is too big ... and it seems like that is because the vee is set too far back (it should be aligned to the front of the sleeper, right?). With the knuckle supported on the next sleeper, the gap is too big - and to make it "work" with shorter wagons I'd clearly bodged the angle of the knuckle so it's quite a way off if I compare with a fresh knuckle bent in a 1:6 jig. I suspect I could repair by replacing the knuckles with fresh ones bent to an accurate angle, and spacing them accurately from the vee as it is. That would mean the knuckle floating in the gap between sleepers. But I wonder if I should do it properly, and rebuild the vee too, so both can be lined up and supported on sleepers. It also seems like the rail height of the vee itself is a touch lower than the knuckles (perhaps because it was salvaged and had already been polished down quite a bit in its previous incarnation?). Frustrating, but I think I'm talking myself into doing a bigger job? Any thoughts? Cheers Justin

Thanks Simon - yes good idea. I had thought CA for ease of removing the motor if I ever need to again, but went with a thin bead of epoxy, as you suggest. Justin

I was just caught out by "dry" Evostick causing a short across a motor. I took the motor out of my Association chassis Jinty to hack a bit more off the worm end of its shaft, so I could shuffle it a few more mm forward, to hopefully make room for a DCC decoder behind it (without resorting to more body surgery). The motor had been stuck on top of a narrow plasticard strip on top of the etched/PCB chassis using Evostick. So when it was ready to go back, on with more Evostick. I left it for about 8 hours to set, tried power, and got a short. After a lot of fiddling, including taking the brakes out of their wire insulation joiners to make sure they weren't creating the short, I concluded it was the Evostick that had oozed out from under the motor and touched the folded over parts of the etched chassis on both sides, and was then shorting via the motor case. So, I guess Evostick that is neither "wet" or "totally solid", is conductive! But I think Evostick takes months, if not years, to end up totally totally dry rather than gummy / stringy? In which case I must have been lucky first time around! Before I reinforce the motor mount with some thick CA, I just wanted to check that isn't also likely to be conductive? I guess the real answer is I should have made sure the chassis was totally insulated around the motor body, with a sheet of paper or plasticard etc. I don't fancy prizing the motor out again on this one now though! J

Wow! I'd say that's roughly equivalent the operational requirement of Copenhagen Fields at any one time! (and a fair number of CF's locos and coaches are RTR derived - Dapol B17, A1, Farish J39, Dapol Gresley coaches) It looks like a great long term goal, but what about a shorter term goal to get a layout operating with a fleet requiring a bit less scratch building? An MVP, you might say. I'm sure at one point we discussed a "c.1920" scenario, making good use of Farish SECR birdcage stock, C class and N class, maybe Dapol Schools and Maunsells, and Worsley EMUs. Then you can work back adding more LCDR trains as a longer project, but with an enjoyable way to operate the layout in the meantime, rather than a massive load of work required to get a viable "output". Just a thought - don't want to pour too much cold water! J

If you have access to a 3D printer, or someone who does, this little gadget could be just what you need: https://www.thingiverse.com/thing:3954248 Basically a printable "pen" which you load with a thin strip of wet & dry, which threads across the top of a little nib (different sizes included). When the paper is worn, unscrew, pull the paper through a bit, and use a fresh part. It does need a carefully calibrated 3D printer to make the fit of the nib and the threaded screw on parts accurate - mine turned out a little "over exposed" and therefore tight, but it does still work. Very handy little gadget that I'm surprised no one has made commercially! J

I've never actually taken an etch design through to production (have started a few and abandoned), but it seems like the main difference from designing for laser cutting or Silhouette cutters is the need to worry about fills/hatches, rather than lines. Chris, just to make sure I'm following properly, you're saying you complete the design entirely using the outline layers, then activate the "surface" layer to create the fills, so they are derivative of the shapes you click into from the outline layers? If I've got that right, it is quite a distinct CAD workflow, which is very clever - the outline and the fill remain separate, but linked (in Fusion360 this kind of workflow is called "Projection"). I don't think that wouldn't really be possible in a vector graphics app like Illustrator / Inkscape / Affinity: I'm not aware of being able to link together shapes between layers in those kinds of programs. For laser / Silhouette designs I always use AutoCAD (2019 for Mac) because I find the snapping, mirroring, arraying etc options, and the ability to work directly with dimensions and angles MUCH more intuitive than in vector graphics software like Inkscape and Affinity. Also, one I learned Blocks, that was a game changer - being able to store a reference design (e.g. a window, W iron, coupler pocket) and replicate it across a design, but then change the original and have all the instances of it also change, is amazing, and so much better than just copy/pasting (this is a very basic CAD feature since the dawn of time, but I only recently got to grips with it). I'm lucky to have free access to Autodesk products through their education license scheme (all university staff and students can register for free access), although of course this is strictly non-commercial. To be honest though, I hardly scratch the surface of the functionality, and the free version of QCAD probably does everything I need. Some people have said the AutoCAD interface is not intuitive compared to other CAD tools - yes it does have a command line at the bottom (I think virtually ALL CAD applications do, at least as an option?) - this is super useful for working with precise coordinates, lengths, angles etc. However at least the most recent Mac version feels like a perfectly normal (albeit technical) application, with perfectly normal menus and tool panes. You can definitely create the right kind of designs and do the kind of precise work on shapes and sizes etc in Inkscape etc., but it seems much harder work to me. You also have to worry about the difference between dimensions including or excluding line widths etc in a graphics program, and which things snap to, etc. I've always struggled with Inkscape, but I do sometimes use Affinity Designer (successor to the old Serif DrawPlus which was a staple on PC magazine CD-ROMs in the 1990s, and is a bargain at about £50 one off payment, not subscription, from the Mac or Windows app stores) and I find that to have much more intuitive snapping and guidelines than Inkscape. It really is horses or courses though - I think people will EITHER take to CAD and its precision, OR to vector graphics and its more organic approach, but most people would struggle to get to grips with the other when used to one. J

This looks really good Dave! I absolutely loved Maid's Morton - the train through landscape feeling was absolutely spot on, as was the WCML 1990s atmosphere. I wonder about a car park at the front though - this is pretty common on layouts, but it seems difficult to get right, especially in N. It seems easy to end up with too many cars that look too new (Oxford) - or modelling 1980s/90s cars in N becomes really hard work, with fiddly fettling of Shapeways prints. Graeme has done some amazing work modelling 90s cars from 3D prints, but I wouldn't fancy the work of filling a carpark with them finished to a high standard. One thing you don't often see on layouts though is an empty carpark! Or perhaps just one or two cars left at the far end of a station car park on a Sunday ... A lot of the beauty of Maid's Morton, in my eyes, was the "less is more" look - it looked so realistic because it wasn't busy with detail - so that might be a nice way to apply that look to a station setting? Justin

These boards look great! Where did you get them laser cut?

This sounds really helpful! I recently opened a tin of 33 matt black that had the plain metal lid with sticker and Sandwich address, stirred it very well with a battery powered stirrer, and it still dried gloss. On the question of "safety standards" though, I can't help but think that's a red herring. Everyone complained at quality when Hornby moved Humbrol production to China, but then there still seem to be complaints since it came back to the UK, so the conclusion seems to be that the ingredients must have changed because of the fabled " 'elf and safety". But the whole time Revell's enamels, which have to comply with the same regulations, seem to have been absolutely fine! J

I've got some Intercity coaches on order, so haven't seen them in the flesh yet. But I'm inclined to agree with this ... I think there is a danger in seeing a Kinematic Coupler mechanism as a tick-box essential, like etched grilles on a loco, that doesn't necessarily add very much, or might even be worse than the conventional alternative. I don't recall actually seeing the close coupler on my Farish Mk1s or Dapol Gresleys really moving much at all in practice on all but the very sharpest settrack curves - there often seems more flex in the NEM socket itself than in the mechanism. Has anyone tried them with the short Dapol non-magnetic basic buckeyes? Justin

I really like your approach of building a train at a time, Jerry. Really worked brilliantly with that rake of PO wagons. I wish I had the same discipline - but I'm far too magpie / Mr Road like! Ooh what's that shiny new etched distraction ? ...

This coach caught my eye - looks like LNWR livery? (or Caledonian?) But has an MxxxxM number, suggesting BR era. Is it an LMS (or Midland?) coach masquerading as LNWR for filming? Or some other early BR 'heritage' operation?

As well as the models being fascinating, I'm finding this exhibition programme fascinating as well! The graphic design here is amazing - the bold colour and subtle use of Gill Sans type font makes it a thing of beauty to my eye, and it must have looked very cutting edge and fashionable for its day. I even like the diagonal arrangement of the stands - much more interesting than the plain rows we're used to today. I guess back then, there was little alternative than leaving graphic design to professionals, as there simply weren't the means to do it yourself. Or did the MRC have a design guru amongst its midst at that time? Are the place names, "Manchester" etc, clubs attending with layouts, or showcases of models etc? Its interesting to see the NMRA there - I hadn't imagined there would have been much interest in American modelling so early on! J