3D Printed Concrete Planter

I made a custom concrete plant pot and did it in the normal Waggy Online procedure: I waaay over researched each step and then screwed up actually following each step. But what turned out unusual was this: I’m pretty pleased with the result! So I documented each step and wrote up some instructions. If you want to see what I did, read on:

... or watch this walkthrough.

1. Computer w/ Blender/Fusion360 & slicer (Cura)
2. 3D printer (& filament)
3. Waste board
4. Glue (Silicone caulk)
5. Gaff tape and/or clamps
6. Mold release (Vaseline/cooking spray/WD-40)
7. Rag/cotton/cloth for cleanup
8. Respirator mask & rubber gloves
9. Concrete mix
10. Colander or sifter
11. Bucket
12. Electric drill
13. Mixing paddle
14. Spoon/chopstick
15. Straight edge/putty knife
16. Water spray bottle
17. Sandpaper
18. Paint & paintbrushes

The section of the process I find most enjoyable is the first: designing the planter. I tend to conceptualize the design in nearly its entirety and then struggle to create the model for 3d printing.

The design is generally guided by the inspiration and the constraints.

The inspiration in this case is Slime Rancher. I’ve been playing it with my daughter. She loves the cute slimes and chickens and toys for the slimes. I thought the “default” pink slime might be the easiest to design and the least likely to cause problems for printing the mold and demolding the result.

I am slowly learning that most artistic endeavors begin with blocking. So with 3d modeling, I start with primitives and booleans and then minimal refinements. (I’m not yet good at all at sculpting but I’ve found success with lowpoly manual edits.) A slime is a good subject for my strengths. We can start with a UV sphere, add some subdivisions and proportionally edit it to have a squashed shape like a slime. Two more spheres squashed into shape can be the slimes eyes and large “jelly bean-ed” sphere can cut out its mouth.

The first physical constraint for the design is the 3d printer’s print bed size. (Clever design can overcome the print bed size if you just plan on making the mold out of more pieces but I wanted a very simple design for my first attempt. The design I made uses only 3 parts.

The second physical constraint for the design is the size of the hole for the plant. I wanted to be able to fit a 6” plastic plant pot completely in the design. A 6” pot is about 4.5” high and has a 4.5” diameter at the bottom and a 6” diameter just below its lip.

The final crucial physical constraint for the design is that it needs to take into account the requirements of its medium. I’m casting the planter out of concrete. Therefore it’ll be heavy, inflexible, brittle, and initially thick or viscous. So I want the details to be large enough they won’t snap on demolding, walls and bases thick enough they won’t break, but I also want the volume of the design to be small enough the result isn’t as heavy as a wrecking ball and I need to make sure that overhangs and cutbacks won’t make break upon demolding. Because I want the bottom of the casting to be flattened, I actually designed the mold upside down.

After I have a design I’m happy with, I make sure to save a copy so I can always revert to this step. Then I plot out the pieces of the mold. If the design is very complex, I might plan for the mold to have squared or rounded perimeters completely unrelated to my design. I could design loops or slots and could even engrave instructions into the parts of the mold. 

In this case my design was relatively simplistic and if I had to iterate, I didn’t want to waste filament on my first attempt. Therefore, I created a shell for my design with a solidify modifier, some booleans, and a lot of cleanup. I knew that the pieces would need to be able to stand flat on a waste board and clamp together so I added some flat additions. I knew that I probably wanted to go overboard for strength so I’d want 3 shell layers some flexibility, and then 3 shell layers so the mold design was made 3mm thick.

After I finalized the designs, I exported the parts individually as STL files, imported them into CURA and confirmed that the planned print size matched the designs in Blender. (I was particularly paranoid that I’d waste a good chunk of filament just to discover it barely didn’t fit the 6” pots.) I printed the mold in PLA because it ended up being strong enough, is cheaper, and is immensely easier to print. If you’re super concerned about dimensional accuracy, you could design a shape in Blender to any specific size, slice it in CURA, and print the calibration shape. If the real-world version is any fraction the wrong size, you could adjust the design in Blender or CURA to compensate.

I printed the parts in black Hatchbox PLA at .2mm precision, 15% tri-hexagon infill, 3 exterior shells, support material beyond 45%, and no bed adhesion. They took several days to print each and look pretty good! I barely did any prep after removing the support material. If you use thick enough mold release in the coming steps, you won’t see any stepped printing lines in the final result.

After the parts were ready, I covered the interiors with mold release. I chose vaseline because I knew I also wanted to glue the pieces to the waste board with caulking next and wanted the caulking to cure before I poured the concrete. I’ve seen that other mold releases work well though with simpler designs. I also tested the vaseline on a test piece from a prior printing project and the vaseline didn’t pool to the bottom after 2 days. Once the pieces had generous but even amounts of vaseline, I “glued” the pieces to the waste board with silicone caulking. I chose silicone caulking because I wanted the pieces to be firmly attached to the waste board (I didn’t want the concrete to leak out the bottom and I was going to vibrate the waste board in a later step) but I want these mold parts to be reusable. I knew the caulking could just peel away later.

A day or two after the pieces were glued, I figured the caulking was cured so I started the next step: mixing and pouring the concrete! I gathered safety glasses and a mask. (You don’t want to get silica dust poisoning from the concrete mix.) I also collected buckets, a kitchen scale protected by a ziplock bag, a sifting method (I had an old screen for a window and a large spoon we don’t use for food,) an electric drill and a mixing paddle, a putty knife, the concrete mix, and some water. I also queued up music in the background to satisfy my ADD.

I’m sifting the concrete mix so that my cast pink slime guy doesn’t have the coarse gravel ingredients in the aggregate because I suspected that the gravel might interfere with the mix filling in the narrower sections of the mold. I’m not sure, but this decision might affect the strength of the result and definitely affected the amount of water I had to add while mixing. I think in the future, I’ll just ignore the prescribed amount and just slowly add water to the mix until it reaches an acceptable consistency.

If you want to add pigment to your medium, now of course is the time to do so. I didn’t want too many added variables in my first attempt, so I decided to paint the result instead of dying the concrete. Concrete pigment powder or regular acrylic paint apparently works just fine.

Once I was happy with the mix, I started pouring (or spooning) it into the mold. And as I did so, I also used the spoon to make sure air pockets in the mold weren’t preventing the concrete from filling every part of the mold. 

Next, I tapped and vibrated the entire mold by tapping and hitting the waste board with my hand, gently dropping the waste board after raising it an inch, and running a de-bladed “saws-all” against the waste board edge. (People suggested random oscillating electric sanders work for this.) I’m doing this in the hope that large bubbles in the concrete will migrate to the top of the pour.

Because it settled lower after vibrating the mold, I added a little more concrete to the “top” of the mold and then used the putty knife to flatten the concrete level with the pour-hole.

I left the concrete to cure and cleaned up all the tools.

After 24 hours, the concrete will likely be hard enough to demold. An easy metric is the “fingernail test.” If you can leave a mark in the concrete by pressing your fingernail into the casting, you probably should let the concrete cure further before demolding.

I demolded the casting at this point and sanded the worst crease created from the seams in the mold while it wasn’t completely cured. Then I let it cure for another week, spraying it occasionally with a spray bottle filled with water. When concrete cures it actually uses water to form the crystals that give concrete its strength so a moist curing concrete is a stronger concrete.

I have a few observations about the painting process that I hope to improve next time. I put the planter on paper towels (I should have used newspaper as it would be easier to remove if it got stuck in the paint.) I started painting the planter right-side-up and should have started it upside-down. Finally, I had a section of the planter (the interior) that I planned to keep unpainted. I should have elevated the entire planter using that section. Maybe I could have set the planter upside-down on a large overturned vase or glass or something. I also know now that I should have mixed larger batches of paint at a time and I know to use less black paint than started with to darken a color.

Once the paint dried, I turned it back over, put a plant in it, and took photos to share with my family and friends! Oh also, I might still paint it with a concrete sealer later…