End-to-End, Making A First 3D-Printed Product

I’ve made all sorts of stuff over the years, but finally had a project that was begging to be 3-D printed, so I bought a printer and here’s the story, to help others understand the learning curve and costs.

• 4 + 1/2 months
• $550 printer
• $200 filament
• 500 hours?

The product was initially to be a pot for houseplants that enables you to note the date on which you last watered your plants. This is helpful when you have several locations in your house with plants… and they get different sunlight… and you‘re not the only one who might water the plants. It arose from a series of cast stone products that you can see at GardenersStone.com. In fact I tried making the pot entirely of stone, somewhat hand sculpted, but quickly realized that my concept — a rotating indexing mechanism — really required the precision of 3D. And thus the journey started.

One’s aptitude for both computers and mechanical work is critical in understanding your journey. If everything goes perfectly and nothing ever breaks and your project is easy, perhaps you’ll get by with no relevant aptitude, but those 3 planets have never aligned before in the history of the universe, so be forewarned. My experience is 50 years of almost every conceivable workshop tool; making stuff in about 15 crafts; and 30 years in the software world including doing some bad programming… but enough to even make 3 apps that I got into the Apple store. The 3D retail world has advanced to the point that you won’t need that level of skill… but you will need to figure stuff out.

• I bought a $550 printer, Creality K1-C because a friend was satisfied with that brand. I chose the “C” since it looked like the newest technology, capable of printing even carbon fiber (which really means small fibers of CF). You can get un-enclosed printers for much less, but the enclosure helps with noise and dirt.

You can buy and use a 3D printer without ever drawing or designing your own objects. I’ll venture to guess that only 1 in 10 owners of printers ever draw an object. Instead they go online, type a word, such as ‘hat,’ and download a model. As with all things that one sees in our miraculously ‘social’ online world, the web then returns to you your choice of about 400,000 models of hat that you can print, which other people have either generously or unwittingly shared with the world. But I wanted to make a product of my own imagined shape, so the first step is to draw its 3D picture, which is called CAD design.

I learned and used the totally free online pseudo-CAD tool called TinkerCAD. Like everything in computers it is simultaneously a staggering miracle what it accomplishes (whether free or not) and a pain in the ass or worse. I was deliberate in adding “pseudo” to the name because that was my first difficult — and very time-consuming — learning experience. TC is not a full-fledged drawing tool, in that 1) you cannot freely draw your own fully 3D, fully manipulatable, shape; and 2) it does not have the efficiency of editing that becomes critical once your project is more than trivial in complexity. As is often the case, there are those who will tell you I’m wrong about both claims and it’s because I just don’t know the right methods. There’s some truth to that for “1” but I don’t believe there’s any truth to it for “2.” For instance to edit a complex drawing, at some point you’ll have to do something like “Undo, undo, undo, undo… adjust… redo, redo, redo, redo.” I’m now learning FreeCAD (can you guess how much it costs?) which seems full-featured but the base version is not very discoverable. There’s a freemium add-on to FreeCAD, called Ondsel, that is purported to surmount the unfriendliness, but I’m a few days away from that item on my to-do list yet.

My first step was to make a set of mirror-image discs that you see in the photo, with 22.5-degree tetrahedrons rotated around in a circle. I thought at least the tetrahedron would be childishly simple but that’s where I hit my wall on TC. So I managed to get the owner of a Philadelphia “maker shop,” Hive76, to draw it for me (with his own tool). To do so, one receives a file called an STL file. I successfully printed his STL file on my printer and felt like a real idiot about it for a while. But that’s the strength of computers: they mike us feel stupider every day, yet somehow more powerful.

I started to learn that TC could sort-of draw a tetrahedron, but I was in no mood to waste time with the next step so I enlisted a freelancer on Fiverr.com, partly for the experience and to add such a capability to my repertoire… the whole world of engineers at my fingertip. It was amazing. A guy in Pakistan drew me the following combined shape that is probably child’s play (even for me now in TC, but the tetrahedron is still tricky) for US$45.

The experience of trying to cast this concept (the full 6-inch tall plant pot) convinced me to change my concept to just a tray/dish/plate… with numbers around the edge. I would make the tray’s “master model” using 3D, but the production trays would be cast in stone from a rubber mold. The date indicator gizmo would be prototyped AND production printed on my 3D printer since I can print many at a time at negligible cost. This new approach, part cast stone/part 3D-printed, is perfectly predicted by my law that “hybrids always win.” Here’s the finished product:

My 3D printer was incredible. I think it printed perfectly the third time I pressed go, which is a miracle, to me at least. Maybe the nozzle was a little clogged. (I used to tell people that the typical cheapo drone helicopter that they can now buy for a few dollars represents almost the entire body of science known to mankind… all in one cheap toy. That mantle is now equally worn by every 3D printer.)

The first thing that may people print is the little tugboat you see on the left, below. It has acquired the name “Benchy,” as in benchmark. So if you know what someone means by ‘benchy,’ you’re halfway to being a 3D whiz, or at least representing yourself as an expensive consultant. The other items you see below were an oval that was my first attempt to print something that I drew in TC. The other ovals were some experiments to improve legends on a stone artwork of the Ernest Shackleton voyage. I’m still working on that. And the little bar at the bottom is an attempt to do a jewelry piece that I’ve stalled out on, the Antiques Roadshow Necklace.

It took me about 50–100 experimental prints to get the 3D design of the dish perfect. Some of this was the inefficiency of editing in TC, some my learning experiences.

So, a few dozen prints later and I’ve got the base right. The main learning experience was that for test prints, you want to print the smallest representative piece possible because 3D printing is very slow. The segments you see above take about 10–30 minutes to print.

The next learning experience was something called “slicing.” When you draw a picture in a CAD program, like TC, below, it produces a 3-dimensional drawing. But the printer needs a whole stack of 2-dimensional instructions like so many sheets of paper. That’s the work of a program called a slicer. There are many free slicers including the one that your printer’s company provides. Creality lets you slice your drawing in their integrated cloud-based workbench, which is great for beginners, but your slice jobs might be in a long queue with others.

The base, below, took a few print attempts to get the lettering right. Then I made a rubber mold from Smooth-On Vitaflex and cast the bases in CTS CementAll, a high-grade casting concrete.

With the base done, it was time to move on to the little sliding date indicator. Remember, these would be printed with my 3D printer even for production — albeit small scale — purposes. I had to try a few designs, then a few materials, and finally got it after about 20 test prints.

The black and white pieces are the basic beginner’s “filament,” called PLA, which is polylactic acid… a totally plant-derived plastic. Woohoo! But there are over 100 different filaments now. The green is “TPU,” an attempt to try a flexible one, to snap over the lip of the base; but it was too soft. The PLA broke when flexing, so I settled on “tough PLA,” or in my case a brand called PLA Pro, which bends a little to snap over the base, but doesn’t break in doing so.

I’ve bought about 8 rolls of filament, for a total of about US$200.

This is where TC’s weakness in editing was nuisance. Now that I know enough technique to edit more efficiently, I could have gotten the size of the indicator (relative to the base) correct with less editing (and trial-and-error printing). I already started learning Freecad and definitely plan to try Ondsel on top of it when I get a chance.

I see that I bought the printer in September of 2024 and it’s now February of 2025, so my job took 4+1/2 months, including… learning TinkerCAD, learning to slice and print, and learning the various materials.

Good luck on your 3D project!