In the first post of this series that you can read here, i talked a bit about my background in 3D modelling and 3D printing model railroad parts, and a “high level” intro. I’d like in this second post, to dive a bit deeper into some of the concepts of 3D printing and how it fits into the process of making a model, as well as how to get into 3D printing if you are so inclined to investigate it.
One thing that seems to be overlooked when i talk to people about 3D printing model railroad items, is that you don’t just sit at a computer, type what you want into Google, and send it to print on a 3D printed. It isn’t like searching for a picture online. While there are repositories of 3D models out there, they are either not designed to be serious scale models, or are aimed at people doing 3D animation and are not designed such that they can actually be printed or completed as a model. An obvious exception to this is the Shapeways marketplace, where countless modellers are selling items in a variety of scales (and a variety of qualities of model accuracy/detail). But there you are still tied to what other people have designed and put on sale. You can’t get something you want unless someone else wants it enough to have drawn it and put it up for sale.
Restored Canadian National/Canadian Pacific joint “Don Station” at the Toronto Railway Museum. A structure I wanted to model, but which there was no drawings I could locate and no commercial kit available to construct.
But, the point of this, is how do you get to there if you want to make a model of something, and have decided that 3D printing the is the way to go? There are many ways of getting from having an idea and taking that through to having a 3D printed part. You can do it yourself (which requires learning how to use a 3D Cad/modelling program); you can have a friend who knows how and is willing to work for free/cheap do it; or, you can pay someone to do the computer model for you. These are not necessarily cheap options, but they are no different than your options if you want a model of something that isn’t available, and can’t scratchbuild it using other techniques. Getting the computer model is only the first half, getting it printed is the second, but I’ll deal with that later in the post.
I learned CAD and 3D modelling as part of my undergraduate degree. That cost $30k mumble dollars and took four and a half years. Not a reasonable option for a modeller to get into the 3D space. Fortunately, there are lots of options out there to learn, look at evening adult education courses, libraries and the like, or get a program and start experimenting. It’s not going to be easy, and likely not free, but it can be done. I’m no teacher, and i won’t pretend to be, i can only offer insights from my own experience. It’s not for everyone, and some people won’t be able to get it. I can’t draw on paper with a pen/pencil to save my life, but i can create complicated structures and designs in a computer, everyone is different.
The good news is, that if you do want to experiment yourself, there are many free options out there for you to try if you want to look at a self taught option, some that are online, some that are downloads. A selection of your options to go and play are below:
- Tinkercad – https://www.tinkercad.com/
- Sketchup – http://www.sketchup.com/
- AutoDesk Fusion 360 – http://www.autodesk.com/products/fusion-360/overview (Free for hobbyists/etc making under $100,000 US)
- Form Z Free – http://www.formz.com/products/formzfree.html
I use Form Z, but I have access to a licensed copy of Version 6.1.2 at my office (the current is version 8.x). This is a luxury as its a program i have 15+ years of experience with it. I have dabbled with Sketchup, it definitely has potential, but it does some things in opposite ways to what i am used to with Form Z. This probably isn’t an issue if you are learning from start, but it causes me headaches. I’ve also recently been told about AutoDesk Fusion 360. I’m going to download it and take a look at some point in the future, and see. I use AutoCAD 2D in my day job regularly, and they make solid programs that are good at what they do, which gives me a lot of hope for Fusion360 being a viable option going forward for me. I’ll hopefully report back on that in a few months time after getting a chance to really try it out. I’ve placed screenshots of TinkerCard and FormZ 6 (the current version looks a bit different, 6 is a bit dated), to give a sense of the level of complexity of each program
Tinkercad (top) and Form Z 6.1.2 (bottom). Showing a web based program aimed at beginners, and a more complicated modelling suite.
Tinkercad is perfect for designing little things to get a feel for modelling in 3 dimensions, and creating items that are best printed on 3D printers like the Makerbot that use coils of plastic wire to produce the parts. These printers are rough, and items have the big stepping that is a downside of 3D printing. They are however useful for cheaply testing that your design concepts work, something I’ll write about in the future.
Once you get into the modelling world, you still have to have plans of what you are modelling. Either through finding plans from archives or other sources, or by going out and physically measuring a building or item and drawing it yourself from your sketches. This is what i have done for many of the items i’ve drawn for 3D printing. Having the luxury of wanting to model buildings which i have access to at the Toronto Railway Museum, such as Don Station, has meant that i can literally go to the building with a tape measure, and take measurements and lots of reference photos. This obviously doesn’t work for something where you only have photographs (of a long demolished building for example), but there are techniques for getting measurements from a photograph and converting that into information to make a model from. Strangely enough, all the people who think 3D printing is going to destroy the hobby, use the same techniques to get dimensions to build a model by “traditional” methods of looking at plans and photographs.
Sketch of measurements for windows and the ticket counter of Don Station.
I’ve had people tell me i should learn to make masters and resin cast the parts, and my response is why? I’m creating a master, its a digital master, and the 3D printer does the work of me casting parts. That argument of course assumes that i was then capable of making a master from wood/styrene, creating molds, casting parts, etc etc. And yes, i realize that’s the exact same situation someone who doesn’t know computers would have with 3D printing. Its not a panacea, its a tool, and there is room for everything and different techniques. There are absolutely things that a 3D printed can’t do, just as it’s possible for a 3D printer to create shapes which would be impossible to cast in resin. For an end user, receiving a resin model or a 3D printed model, should be no different. There may be some slight differences in how you handle the material, but its a kit you get to build and finish!!
I don’t look down at someone because they make resin or etched brass kits, i think that’s great, they are using their skills to benefit themselves and the community of modellers. In fact, they are using skills and techniques that i’ve never been able to wrap my head around to understand and try myself!! I’d like to think that instead of thinking that 3D printing is cheapening the hobby, they’d see the opportunities it affords to broaden the hobby by bringing in more people into the hobby with different complimentary skill sets. There are so many different things that people model, and are interested in, our hobby is a big tent, there should be room for all, and instead of looking askance, we should be looking for the opportunities to enhance what we do, to learn, and to share knowledge. I’m trying to do that with this blogging thing after gently being nudged into it by reading the blogs of my friends in the hobby, and their well thought out posts on the benefits of being a part of the 21st century and the ready exchange of information (i’ve been accused of being old fashioned or resistant to change by some of my non-modeller friends for not being on Facebook or similar, which in some ways i can be, but it’s always funny when I’m often in my mid 30’s and one of the younger people in a room of model railroaders and get accused of being too modern for 3D printing!).
Back to the point of my post, assuming you’ve reached the point where you have a completed model that is ready to be printed, you need somewhere to print it. There is a huge range of machines and producers out there, some local to you, some international companies that offer online services. I’m not even going to try and give a rundown of them all, but what i can do, is comment on the three places i have gotten printing done, Shapeways, i.Materialise, and the Toronto Public Library (using Makerbot and Lulzbot Taz printers that are “affordable” enough you could conceivably buy one).
Shapeways offers a range of different materials, all with different requirements for what will and won’t print, and how well the material is finished. The two groupings i have found most useful for modelling are the “Strong and Flexible” and “Detail Plastic” groups. The Strong and Flexible are a laser sintered nylon, this means the model is built up of layers of powder which are then blasted with a laser, wherever the laser hits, becomes solid, where it doesn’t, stays powder and becomes the support for layers above. The material has a rough surface texture, and some stepping depending on the shape being created. I find it to be good for very small items where the low surface quality won’t be noticed. The Detail Plastics are a UV Cured resin, which means the printer lays down a layer of material, then blasts it with a UV light, and lays the next layer. The material is supported by a non-resin goo for lack of a better description that holds up layers during printing. This is cleaned away mostly before you receive the part, but needs to be completely cleaned when you receive the parts. I’ve also used the transparent acrylic to make light fixtures for ceilings to hide tiny LED’s and disperse the light, the material isn’t truly transparent, it’s kinda a cloudy see through, but it works nicely for the purpose.
Similarly, i.Materlialise offers a range of materials. I’ve only used them once, for an item in their “prime grey” material. This is also a resin product. I found it to be very similar to styrene in feel and appearance. I frankly only used them once as i found Shapeways material standards allowed for smaller pieces and more intricate details, and their pricing/shipping is cheaper to Canada.
The public library is the cheap and dirty option, their machines are good for what they are, but because they print using coils of PLA plastic, the quality of the print isn’t really good enough to use in a finished model, but what i’ve been doing is taking parts of models there to look for obvious problems. Like the span bolster i designed for a project that supported two trucks with their centres each located 6′ from the central pivot point. I discovered that i had drawn the bolster with the centrepoints of the two trucks 6′ apart, rather than 12′ apart, but that cost me 2 hours and $5.00 to find out at the library, rather than 3 weeks and $30 to discover the error if i found it after ordering from Shapeways.
There are also other private companies out there, The Printing House in the Toronto Area does 3D printing, and i know there are makerspaces and private printing companies, i just haven’t looked into them as i am comfortable with Shapeways and like being able to easily sell things through their store.
So, when you get a 3D printed item back from the printer, the material types which are most likely to be useful to a modeller are unicromatic parts, potentially with lots of detail, but no colours. They may take forms that you could not easily generate in other means, but when you buy a station for example, it looks like this when it arrives:
3D Print of Don Station, in Shapeways “Frosted Ultra Detail” UV Cured Resin material. A blank canvas if ever there was one.
As you can see, it doesn’t look much like something you would want to display proudly on your layout!! While making the shapes of the station is doable either way, it comes down to your skillset. I can scratchbuild buildings from wood and styrene, but i am better at drawing them in a computer than i am building them, it means that things like specific windows or doors can be created with widths and details, vs. using the sometimes limited range of commercial options available. As you can see from the print above, the roof is just a wireframe, this reflects another reality of 3D printing, cost. It isn’t cost effective to print large flat surfaces that can be built out of styrene, brass, metal or card. The day where the price drops may come, but at the moment, for buildings and structures, anything much bigger than a shed or shanty needs to be a “hybrid” model using 3D and traditional materials to create the base that then gets painted and detailed as any other project does.
Once you’ve gotten to this point, finishing the model is no different from any laser cut, styrene, resin cast or card kit. You need to know how to handle the material, paint and glue, and can detail as much or as little as you like beyond the kit. There’s a diversion in the course of how you get from Point A of having an idea of something you want to model, to Point Z of having a completed model, where the computer model and the 3D print replaces a period of time where you’d be at your workbench building sub components, but if done right, it can open doors.
And a final thought on this post, if you haven’t figured out, its not necessarily a cheap process. Between the cost of learning, potentially buying software, your modelling time, and printing itself, i don’t for one second suggest this is a “cheap” way to get things. If you want models, and you want them done well, you should expect to pay for them. Through this post i’ve used images of the HO Scale model of Don Station. I don’t know how many hours i spent on the 3D model, between the time spent physically measuring the building, taking pictures, creating sketch drawings in 2D CAD from the rough measurements to go back and get better ones, creating the 3D model, and then finishing the model once it was received (and technically, I’m not actually done finishing the model yet). If i had to guess, I’m probably 200 hours into a station that is about 4 inches by 6 inches in size. Half of the time is the measuring/3D modelling, and the other half the traditional modelling to finish it. The 3D print cost is $120.70 from Shapeways. I’ve put it for sale in my Shapeways shop, and one kind modeller (thank you, you know who you are if you ever read this) has bought it for my selling price of $160.00. That means I’m -$80.00 on just the physical print of my own model after getting the $40 in markup, before i even attempt to calculate how much my time time cost to do the 3D model. I’m no business man, this is my hobby, and anything i sell on Shapeways isn’t meant to put food on my table, but to defray the costs of my own prints in the future (this math should also hint as to why I’m reluctant when people come up and ask if i take commissions!!). I’m not in the hobby to make money, but to relax and enjoy myself, but i want people to understand the realities of the cost of 3D printing.
I hope this helps shed some more light on the 3D printing and model railroad world. This is a topic I’ll be coming back to regularly to fill in the gaps and expand upon areas i’ve only briefly touched on here.