NScaleNotes

They try to according to their info: https://support.shapeways.com/hc/article_attachments/360057027814/mceclip7.png

The curved brick wall of a road tunnel through a viaduct?

True but isn't the OP printing wanting others to print what is basically an elongated cube with a curved top in many cases. I suppose you'd still have to compromise between the angle for the roof and the end walls or print the body and roof as two separate pieces; you could optimise the angle for both then but I think that'd need a redesign. Ouch, that's a lot of redesigning.

I know most of the internals of most 3D printers actually come out of the same factory but was wondering about this because @JimFin posted a video about orientation of 3D prints on another RMWeb thread a little while ago which made it seem a little more complicated than X degrees is the best. Here's the link: https://www.rmweb.co.uk/topic/181985-dodgy-resin-3d-print/?do=findComment&comment=5307999 With a variety of LCD sizes now available in consumer machines I wonder if it is actually getting a little more difficult to provide generic advice.

It must be well known that home resin printing requires you to use a support structure. It's visible and discussed in all the promo materials for resin 3D printers. A slicer program is used to create a support structure for your 3D model; the slicer usually comes with your printer and you can still add this structure to a model/STL file you have purchased. The support structure can be created automatically and usually works well but depending on your model, the printer, your experience, you might find you need to alter this manually; as everyone else has said this is part of the learning process. As someone else mentioned, home resin printing allows you more leeway to really exploit the capabilities of your printer with thin wires and overhangs i.e. things that commercial printers would refuse to print because of the likelihood of damage at post-processing or shipment stage. So home printing can actually produce finer models with fewer limits than you'd get at commercial printers all things considered. Sometimes the person printing may also need to consider orientating a part to overcome problems with suction that can occur with resin printers but there are printers that can do fancy things as the build plate lifts to help avoid suction issues but they cost a lot more than most consumer 3D printers. Again it's a learning curve. Of course there is hype, there always is and always has been. Having worked in IT I've seen bucket loads of it but I've always found you need to spend time learning and practicing to make any new, hyped, no-experience-needed, AI wonder-system work well especially if you want to produce quality results! The design i.e. the STL file IS the key in all this though. Home 3D printing technology has advanced so much that you really shouldn't need to be modifying designs that print commercially to get them to print using home resin printing technology unless there are flaws in the models/STL files that the commercial services are correcting on your behalf (for instance I believe Shapeways can do some rudimentary fixing of meshes such as surface hole filling, correcting flipped normals). Now if we're talking about scaling of models: if a model prints in resin in N scale then it would usually print in other scales assuming it's supported correctly and your print bed is big enough but you are likely to end up with compromises as you go up the scales i.e. walls become too thick (which are unsightly and use too much resin), detail may have been omitted for N scale etc. So that means, if you're not willing to modify your models then you'll have to be satisfied only supplying them for the printing service/technology and scale they were created for. There's no way around that and I doubt technology is on the horizon to fix that. In case anyone just starting out in 3D printing/CAD is reading you can create 3D models that will allow you to change the scale of a model at the click of a button but you need to plan for that in advance or face a lot of work reworking your models later. If that's not what you meant then I'm not sure what this service you say we need going to do? 🤔 If you have an STL that you want printed then there are services you pay that do the support creation, orientation and 3D printing for you or you do it yourself with automatic/manual support generation in the printer slicer software? If by commercial you just mean Shapeways then there is already a bridge if you have an STL that you need printed: any 3D printing company that offers resin 3D printing that isn't Shapeways! There are plenty of providers of such services at much more reasonable prices as an added advantage. I upload an STL, they print it, I don't have to worry about supports or orientation, I get a resin printed 3D model. It really is that simple. However that assumes I have been provided with an an error free STL file that has been test printed to begin with and we come full circle. Erm, if you're not willing to do it for them, then yes it is if that's all that's available! Buying STLs is a bit like buying a plastic kit, the manufacturer of the kit can't guarantee that I'll put it together right or even has the skills to do so when they sell it to me but as a consumer I expect that the model has at least been put together to test it and that it's relatively accurate. The more expensive the kit the more I as a consumer expect in terms of quality. So if the analogy is not obvious, if I pay a fair price for an STL that the maker says I can print via a commercial service or on a home resin printer then I'd expect that it's at least been test printed on those technologies and produces a result that looks like the STL that I'm buying but it's still my responsibility to know how to print it or get it printed by someone that does. If the seller can offer advice on the equipment/settings that worked for them even better, maybe I'd pay more.

I've used iMaterialise in the past. Very good prints, no visible layer lines and they printed in a water clear resin which did open up all kinds of possibilities. Much cheaper than Shapeways too. Unfortunately iMaterialise then upped their minimum wall thickness and started to refuse to print some items despite me using their "print it anyway" option (they'd also printed them before with zero issues!) so I found a company in the UK called SGD 3D. They had no issues with minimum tolerances, the prints were very good and so was the price. The parts I wanted had long narrow 'wires' so they suggested a glass reinforced resin which is much tougher than other resins I've seen.

That could be your problem, I've found its often best to start fresh.

FreeCAD has it's quirks like any program. First thing to check is that you have the latest version, it's changed a lot in recent years. Not sure exactly what your issue is, you'd need to provide a bit more info but I've always found FreeCAD has had some issues with sketches mapped to faces, particularly if you go back and edit the underlying parts, as part C ends up using part B as a base and part B uses part A as a base etc. If you change A it can often then break part B and C, when I started out many version ago I'd often end up wrecking hours of work and there didn't seem to be a way to unpick the problem as things got so messed up and flipped around. For this reason I learnt to, as much as possible, to create parts with independent geometries/bases. So instead of mapping new sketches to the face of an existing part, I position the new sketch for things like hinges on your box using their own X,Y,Z co-ordinates. Works much better. Perhaps things have improved in newer versions of FreeCAD but I've learnt to do it this way now. Fillets and chamfers always caused problems, these are best left to the very end of any creation process. What is the very small rectangle on the top of your box? Seems odd. Is that the base/top of the small protrusion on the top of the door? Perhaps there's an issue there, like I say hard to tell for sure without more info. I think FreeCAD is the king (in terms of features) of the free CAD software on Linux, least it was last time I looked, keep working at it, it's a very powerful package but like any software, particularly free software, you've got to learn to work around a few issues.

I use Freecad which was originally designed for subtractive processes like CNC machining so the way I learnt to think about it is this: You've got the plans of say a flatbed wagon, so now you have a good idea of the overall width, length and height of the wagon body and the shape, size and position of the various features. I'd recommend that you first scale these measurements, that way you'll have a better idea of any features you might have to leave out or perhaps make oversize for whatever reason. Also you probably only want to start with the wagon body rather than trying to create bogies etc on your first foray into CAD. Now imagine a block of solid material that your finished wagon body will fit inside. You then use the side elevation to remove material from the sides of the block, the plan from the top of the block etc, until the block looks like your wagon. The bit you'll need to learn is how the different CAD programs perform the various material removal steps.

Weird, perhaps it's a browser based issue, I'm using Firefox and no issues. Just drop them an email, I've found they are helpful and very quick to respond (info [@] sgd3d.co.uk) Wow, I should get a discount on my next order

I've used Shapeways and iMaterialise. Both good in their own way but I have always have hassles with going under their minimums even though they subsequently print without problems. So recently I experimented with some different companies with higher resolution machines with even lower minimums and different resins; the one based in the UK was SGD https://www.sgd3d.co.uk/. Now a very good choice as most of the big 3D printers are based in mainland Europe with all the potential shipping issues. SGD produced some very finely detailed, very strong models for me at very good prices, worth checking out if you want something UK based. My only complaint was the packing of the final objects but I'm sure that can be worked out. I think I'll be sending my next models there for printing. I've also tried an Italian company called Fama3d https://www.fama3d.it/en/. If you want multiples of the same (known good*) model these guys are brilliant as they have very low prices for more than one copy of something. Perhaps not so much since Brexit unfortunately as they don't take Italian VAT off so you risk being double taxed plus all the other possible shipping costs. Interestingly though, even if I'd been double taxed all of the models would still have been cheaper than being printed by anyone else. * I say know good because it'd be a very expensive way to test/prototype a model I was going to try FacFox but even though they offer amazing minimum part and wall thicknesses on their ultra-high resolution printers at fair prices they won't let you use them because they say it'd probably break in transit. Oh well, their loss plus that led me to find SGD so it didn't work out badly.

Oh, very nice. I asked because I'd only ever see that peach coloured resin from that company. Looks like a good setup you've got there.

Very nice. Are those FacFox prints?

Yep, it's odd. I have had parts rejected because the sprues weren't thick enough, once I changed that it went through and that was two sets of parts to make two models so perhaps it wasn't so obvious. I suppose the people checking these things don't always understand what we're making or have time to visually check a model carefully so perhaps a duplicate kit of parts will slip through whereas a large set of identical parts sprued together will be easier to catch perhaps even automatically via software. I also think if your parts are sprued efficiently i.e. not eating up bed space unnecessarily, it's more likely to get through.

There are a lot more than two technologies... and I think you might also be confusing your 3D printing technologies: As far as I knew Shapeways no longer offer any filament type (FDM) 3D printing. You can get some amazing results with this tech in the larger scales if you're committed but I'd still say it's not suitable for small scales and curved surfaces in most cases. Check out this thread on a German model forum to see some amazing HO FDM work: https://www.stummiforum.de/viewtopic.php?f=180&t=171433. Then there is SLA. There is liquid resin based stereolithography (SLA) which uses legs/supports to print overhangs and then there is wax-based support SLA this is what Shapeways offers. Wax based SLA is quite rare commercially actually (I've only seen a couple of places offering it) but is a good option in some cases! The wax support structure is melted away once the print is finished so no stress on the model when the support is removed. Also means there is no 'extra' charge for this as it's an inherent part of the printing process just as removing excess powder is a part of the SLS/MJF process. Prints probably just take longer with the SLA process as the layer heights are smaller than those used to print the WSF of the SLS/MJF process. I suspect the build plates are smaller with most SLA printers (even the commercial ones) so less models per print run too. If you ever have a print made using liquid resin based stereolithography then it's likely it will arrive with the supports still in place. You'd pay a fortune to have someone remove them for you, not only that but in most cases they'd probably get it wrong on a complicated model and as far as I'm concerned the supports make the printed object stronger while it's in transit to you. It's a simple task to remove them with a good pair of cutters and sand down any remaining high spots. Resin is a tiny part of the cost of SLA printing from what I've seen using programs like Chitubox. For any commercial operation time is money and SLA takes longer because layer heights are usually much smaller than with other technologies so printing takes longer. There usually also needs to be some post production washing and UV curing of the model which adds more steps that take time; I'd imagine much longer than just sieving the model out of the powder and blasting it with air. Model volume is another major factor with SLA printing not because it uses more resin but because it uses more cubic mm inside the printer. If they can only fit a couple of other models on the print bed because your print is large they are going to pass that opportunity cost onto you! This might all seem irrelevant but understanding the tech you are intending to print with can make a big difference to your designs and outcomes. Your WSF models need to be much chunkier due to the nature of the material and to withstand the air blasting used to remove the excess powder post-production vs. a similar SLA print. If you ever design for SLA you need to make sure you're not inadvertently designing pockets that don't drain the support wax properly or voids that end up filled with uncured liquid resin because light from the UV curing process can't reach it. Print orientation can have a big influence on part strength too so that's another thing to factor in. In any case once you understand the processes you can also start playing around with the minimums that these companies recommend for the different technologies. For instance I'm finding I can almost halve most of the recommended minimums suggested for support based resin SLA printing. You can then force a 'non-conforming' print through using the 'print it anyway' option most companies offer and still get functional and robust SLA prints. There's not so much leeway with the wax supported SLA from Shapeways despite the gentler process as the resin used seems to be much more brittle but again that does depend on the orientation of the part in the printer.

Thought the title of the thread might give it away but you presume right... Yes, it's a tricky balance to strike between being true to scale and incorporating detail/features that the eye expects to see and I'm aware plastic will lean more towards including features that the eye expects to see rather than being absolutely true to scale. For now I want to experiment with plastic but I remember seeing some 2mm/finescale buildings built with brick papers at an exhibition that were very effective and I'm sure your results are similar. In a similar vein I think creating effective scenes is also down to having a similar or complimentary colour palette throughout the layout to help blend everything together which is easier to achieve if you're doing your own artwork and printing; sometimes I find the colours of unweathered commercial products sat together clash and at least for me that can break the illusion. Do you have any pictures of your work you can share?

Of course as usual, right after I posted I remembered something: Redutex sheets. These look pretty good, anyone have any experience of using these? They look pretty scale to me (they have rulers next to the pictures of the sheets on their website, other manufacturers take note ). Expensive but might be worth it; can anyone know how thick they are?

Hello I'm looking to do some scratchbuilding of brick buildings and I'd like to use plastic/styrene. I know there's loads of good paper products but painting/weathering those isn't quite as easy. I've got some Slaters brick sheet but it's only available in English bond pattern and most buildings I'm looking at use Stretcher brick bond. So... does anyone have experience of using Kibri/Vollmer plastic brick sheets? Are the bricks to scale? How thick is it? I've seen some American stuff by a company called N Scale Architects but it's quite expensive once you include shipping from the US and it's a bit over size (picky I know but like I say it's expensive). Thanks Simon

This might not be the 'proper' way but a quick way of achieving that would be to create a new sketch alongside the tube. So, create a new sketch inside the body of the tube and use the attachment properties on the left to move it to one side of the tube. Draw the shape of the cutout in the sketch. Cut across the tube. Pictures speak a thousand words: Edit: Hmmm, just occurred to me from the way you described your work. Did you build the cylinder as a part/from the part workbench? You might need to switch to the part design workbench. I tend to do everything in the part design workbench where you build all your shapes from sketches as shown above. So the tube above is a pad of a sketch of a circle of the right diameter. The hollowed centre is achieved in a similar way by creating another sketch with a circle of the diameter of the cutout and so on. It's definitely the way to go as you start to build more complex models.

I also just realised no one has answered one of your opening questions either. I use sprues when I have parts that would print better as separate parts (because I can control the orientation) or that would benefit from being attached to model after painting. They are also very important in preventing your parts becoming lost during printing/processing. You have to remember that when you have your model printed commercially it won't be the only item on the print bed and things can get lost during processing. Sprues generally need to be at least 1mm in diameter at the big players like Shapeways/iMaterialise. However, sprues are frowned upon if you are simply joining multiples of the same model to save on processing costs. If they catch you doing this they say they'll probably reject your model and tell you to resubmit it as multiple prints of a single model.

I've never needed to use Python in any of the 3D models I've produced. Not saying you don't need it but I'd hate to see someone put off using such a great open source resource. I might be able to help if you're stuck with something.

So like this: Red are half-etches one side, blue on the other. Now I hope this makes sense: I've made 'receiving' end half-etch (red) slightly longer than the 'tab' end half-etch (blue) because I know the etch liquid won't create a perfect right-angle in the corner of the 'receiving' half-etch. I think this is called cusping. It should be no more than about 0.0225mm in 0.15mm sheet according to the guide. But I suppose that also means I need to make the whole of the section with the red 'receiving' half-etch 0.05mm shorter to account for the cusp in the corner of the blue 'tab' half-etch. Should all line up nicely then... hopefully

Now that's an idea, could have half-etched tabs to help connect it together too then. Thanks.

Hi Nigel I see what you mean but I think it's a slight error in the way the measurements are presented in that image. In the file I'd send to be etched the central section of that part is 5.7mm and the tab above the notch is another 0.15mm plus the 0.15mm of the red half-fold. Thanks Simon

Ah yes, thanks Frank. I think that's because I got used to leaving a 0.05mm gap between interlocking parts from when I've been working in 3D printing. Suppose it's not necessary when you've got a consistently thick sheet of etch. I'll change that. Simon