Well I sure didn’t need another hobby, but a few months ago, my son asked me if we could try out 3D printing. I had given it some thought myself already, just as a matter of general curiosity and with the idea of printing miniatures for D&D in mind. Wouldn’t it be neat, I thought, to need a miniature of a goblin, say, and just push some buttons and have one spit out of a 3D printer?

Yes, that would be neat. But it turns out that that’s not how it works. That’s not how it works at all. It turns out that 3D printing takes a lot of patience and troubleshooting and tedious manual work, and even then, your prints don’t necessarily come out looking very good at all. I’ve mostly printed abortive miniatures and a few things like throwing knives for cosplay for my son. I recently played D&D with a guy who had some cardboard miniatures that you put in these little stands. I ordered some, but mine came without the stands. This was the perfect time for a functional print!

So, I found a design online and did the steps necessary to get it ready for printing. I printed the 10 little stands that were part of the design. They were ok, but they didn’t clip my minis tightly enough, so the cardboard figures slipped out of the bases. Surely I could improve on the design. I also wanted to have more than the 10, for large combat. And I wanted to build them in such a way that I could also print bases with a bigger diameter, for bigger creatures who might show up in combat.

So, I went into this online free software called Tinkercad and designed a mini by making the component shapes and fusing them together. My first try looked like this:

It’s a little rough. I didn’t even notice initially that the original bases I had printed had nice little rounded corners. The vertical clip pieces were also wider on the original. I needed to try again. My second version looked like this:

It’s closer! The corners are rounded, and the clips are wider. This looks a little nicer. But the corners are still sharper than in the original design, and I liked the original design. So I made another attempt, and at this phase, I refined my process. I had been making the base and then the clippy bit, which I then copied and spun 180 degrees to make the second clippy bit, like so (this is sort of a scratchpad I was working in in Tinkercad):

But this made it hard to position the pieces precisely, and precision — making sure the width between the clippy bits was exactly right — was what prompted me to undertake this project to begin with. So I tried another design, pictured here from the side:

Here I made a single wide block centered on the base. Viewed from another angle (not pictured), you’d be able to see the rounded corners, which required that I fashion the main orange block here out of two rectangles and two circles, all fused together. Once I had that, I used negative space (the gray boxes) to cut out a center section of precisely 2.4mm wide and to round the bottom bits and make the clip pieces narrower. Once you’ve got all these things laid out, you group or sort of merge them within the software to get a single shape (like the orange base in the image prior to the one just above). The printed result:

This is very nearly identical to the original design, but with a more precisely measured negative space between the vertical bits. My minis fit perfectly, with no slipping. Obviously I still need to do some more work with the X-acto knife to clean this one up a bit. But now I’ve got my own file I can manipulate and refine further (e.g. to make larger bases).

Once you’ve done the work to create the 3D model file, you pull it into software known as slicer software that lets you tune the settings for your specific printer and the plastic you’re using. You can change all sorts of settings, from filament temperature to the speed the nozzle moves to what pattern is used to print the inside bits of solid pieces (depicted in one of the printer shots below). Pulling the model into the slicer looks like so:

You can see here that I took my original base and duplicated it 15 times, changing the numbers. This gives me a single print job of 16 numbered bases, which lets me use up to 16 identical cardboard minis in D&D combat and to keep them distinct from one another using the number (this makes more sense if you’ve played D&D). With my settings tweaked, I export a file that the printer can understand, which basically offers some config info and a bunch of essentially coordinates and short-hand instructions for how to move the extruder and printer bed around and what to do with the filament. Here the things are being printed:

That honeycomb shape is called an infill, and you can (in your slicer software) define different methods of infilling. Using an infill rather than printing the pieces solid saves both material and time, while providing support that other layers can be extruded on top of. Here’s the print job finished:

But you’re not done at this point. You still have to pop the pieces off, trim off the little brims, take an X-acto knife to the many little burrs and imperfections, and, if you’re more industrious than I am, prime and paint them. This print, which I ran three times to get three sets of bases, takes about 8.5 hours a pop to print, plus the time to set up the printer and trim the final pieces. It’s hardly plug and play. The end result is kind of neat, though:

And, with the minis clipped into a couple of them:

I’ve gotten pretty decent at printing the smaller pieces. There are still some things I need to troubleshoot about the bigger ones, which you can see here still have a lot of imperfections. I’ve tweaked several slicer settings (and even tried a couple of different slicers) to try to improve those, so far with little success. It’s frustrating because you’ll get an imperfect print after sometimes hours of printing, tweak a setting that you hope will fix it, and then run another potentially flawed print. So, again, it’s very different from what the process I had envisioned in which I’d push a button and have a beautiful printed item a little while later. It’s neat when you do get a good print, but it’s pretty frustrating the rest of the time.

I’m using a Creality Ender 3 printer (a pretty low-end one) with PLA filament. I’ve tried a couple of different slicers and mostly use to Slic3r; I initially used Cura to slice, but it’s got some bugs with using support materials (which I didn’t even get to in this post — it’s a whole other level of annoyance and tedious manual after-work that I’ll spare you further details of for now).