So it all started when I walked by the 3D printer at work. The object being printed was square castle tower designed to be a dice tower. And that got me thinking. ”I wonder if there are any science fiction themed dice towers shaped like spaceships?” So when I got back to my office, I opened up Google and started looking. The answer: Nope. There were a few sci-fi themed dice towers but for the most part they were just variations on the simple block tower with sci-fi art on them. I wanted something a little flashier.
Now I’ll be the first to admit, I’ve never used a dice tower in a game and don’t really feel the need for it but I can understand the appeal. So I figured It was time to sit down and start modeling.
Square castle towers work well because they are uniform in size from top to bottom, it’s easy to get the dice in the top and straightforward to get them to come out the bottom. A spaceship dice tower was going to be a bit trickier. Most ships taper to points at the top and the bottom, restricting the entry and exit areas for the dice. It also needed to have room in the middle to allow the dice to tumble around inside. I also wanted to go for simple as I was going to be 3D printing this and overly complicated shapes are harder to print.
So I settled on a simple design inspired by the now “retro” ships of the mid 20th century pulp science fiction stories and illustrations. I started by sketching out a simple body design in Inkscape. The goal was to have a somewhat short and squat ship that flared out large in the middle to allow the dice to tumble through. The main goal here was to get a basic shape for the body of the ship as that was the most important part. Additionally I wanted to get the engines roughly proportioned to the body size. The details of the fins connecting them could come later.
Initial 3D Model
The next step was to build the initial 3D model. This began by exporting the cross section of the body and the engines from Inkscape in a format my CAD program could read. Once that was done, I was able to import the 2D cross sections and rotate them into a solid to form the body and the engines. I then created some simple wedges to be the fins to connect the engines to the body of the ship. This gave me the basic body of the ship.
The next step was to hollow out the middle, add obstructions for the dice to tumble against, and an entrance and exit for the dice. I started with the basic ship body and scaled it down 5%. The plan was to subtract this off from the interior of the model to form the central hollow space. From this I cut out three wedges to be the obstructions the dice tumbled against. When this shape was cut from the main model, it would leave material in the area of the wedges. This shape was then subtracted from body. to hollow out the center.
For exits, I simply created three openings in the back of the ship between the fins for the dice to come out of. I realized that the flat bottom of the ship wouldn’t quite work and so I added a cone to provide a slope for the dice to slide off of.
Next, I carved an elliptical opening out of the top of the ship, just above the first wedge as an opening to add the dice. I realized that this was a little hard to feed the dice into so I added a bulge on the side of the ship that, when cut away by the opening, provided a bit of a platform to drop the dice on to and get them into the tower.
Finally, I needed a dice tray. I thought it would be cool if the ship was poised over it’s landing pad so I created a circular landing pad for the ship to rest on. Originally I had thought about having the ship just rest on the walls of the pad but in the end decided to have the base of the engines in the wall.
This gave me a first design and I printed a small 3″ prototype with scaled d6s to test it all out.
First Test and Iterations
All things considered, the prototype worked better than I expected. The dice mostly tumbled through and came out into the dice tray without too many issues. However, there were a few problems.
First, the dice would occasionally get stuck right at the exit. It turned out the pitch of the cone I placed in the back of the ship was a little too steep and the dice were just barely clearing the opening.
Second, ship was too low to it’s landing pad and when the dice tumbled out and went under the ship it was really hard to read the values.
So for the second iteration, I made a few structural changes and added in the cosmetic details that I wanted.
On the structural side, I lowered the pitch of the cone at the back of the ship, nudged the height of the exit apertures up by a millimeter, and raised the height of the base of the ship by about a centimeter to give more clearance and visibility under the ship. I also tweaked the interior wedges to allow more space inside the ship and accommodate larger dice.
Cosmetically I added a number of features. First, I created an aerodynamic cross-section for the fins and extruded that shape into a 3D object to use in place of the block fins I had used in the first prototype. Next, I added some rings around the engines to them a little more style and added portholes down the sides of the ships to allow you to see the dice tumbling through. Then I added a second bulge on the “bottom” of the ship opposite the one I had added to act as a dice tray. Finally I added “landing rings” and cross-hairs to the dice tray as sort of a target for the ship to land on.
With this version of the model ready, I printed a second 3″ prototype along with a set of polyhedral dice scaled to be 22mm for the d20.
Second Test and Iterations
This version of the prototype almost worked. All the dice went through without an issue except for the d20. The exit apertures still weren’t big enough and the die would get stuck just before coming out. The solution was to lower the pitch of the cone in the back of the ship and raise the size of the opening just a bit more.
Additionally, I wasn’t happy with printing the ship and the landing pad together. There is support material need to allow the base of the ship and the fins to print properly and it wasn’t coming off the landing platform cleaning. I didn’t want marring the final version and so I decided that the landing pad and ship would need to be printed separately.
Since everything else was working fine and looked good, to test these change I decided not to print a full prototype at a small scale but rather to print a full-sized prototype of just the pieces in question. For the exit aperture this meant just printing out the back of the ship up to the point where the top of the aperture was printed. That way I could pass actual dice through it and make measurement to determine the exact tolerances in the final dice tower. For printing the two pieces separately, I printed a portion of the engine and the bit of the tray that it was supposed to slip into to make sure the fit was loose enough for the engine to go into but not too loose. This was necessary as 3D printing results in pieces being a little big bigger than specified (by just a fraction of a millimeter) and I wanted to make sure my tolerances were okay.
After printing the full scale parts, I was happy with the way things were working. The final step was to print a full-sized complete prototype to test out. I started by printing out the ship, which took about 12 hours.
Next I printed out the landing pad/dice tray which took another 3 hours. At this point I noticed a problem. Somewhere along the way I had accidentally scaled the ship down by about 8% from it’s full size. I had thought it looked a little small (the openings weren’t quite the same as the partial prototype) but it wasn’t until the dice tray was done that I realized exactly what had happened. I still don’t know how it happened, just what did happen. Rather than reprint the larger, more expensive ship, I reprinted the dice tray scaled to match the ship. Luckily, I had built enough tolerances into the design at this point that my prototype works just fine.
While I like the texture that 3D printing gives the parts, I wanted to test out my Dremel that I got for my birthday and see about sanding down the body of the ship to make it smoother. At a first pass I used my Dremel to sand down all the surfaces on the body of the ship. I then went over that with a finer grit piece of sand paper. I decided to leave the fins and engines alone for the contrast.
After sanding the ship I applied a coat of silver spray paint to the entire surface. At that point you could still see areas where it was a little rough. I sanded the entire body once more until it felt smooth to the touch and then applied another coat of paint. There were still some areas where it looked like it wasn’t smooth but felt smooth to the touch. I probably could have gone one more round of sanding and painting to get an optically smooth surface but I liked it the way it was. Finally I added some color by and painting the bulges, the rings around the engines and the edges of the portholes.
The dice tray was painted black with red in the landing rings and cross hairs.
Final Testing and Model Tweaks
In using the final dice tower I discovered that by raising the ship up in the first iteration, the walls of the dice tray were not quite tall enough and the dice would occasionally bounce out of the tray, especially if there were already a few dice in the tray that they could bounce off of. Additionally, I felt that the opening in the top of the ship was a little small and so made it slightly bigger to make putting the dice in the tower a bit easier. These are minor tweaks and I haven’t printed out a model of these changes but they should not cause any problems.
Traditional dice towers typically have the dice spill out into a tray in front of the tower, not directly under it. I liked the tray under the tower stylistically as the landing pad but realized that it might be more practical to have the tray on the side. To that end I made an alternate model with only one exit aperture and a ramp instead of a cone at the base and tweaked the dice tray to have slots to fit two engines instead of three. In this model the dice tumble out into the tray that is now offset from the ship which makes viewing the values on the dice a bit easier.
Would you like one of your own?
I was sharing the progress of this project on Google+ as it was going along and several people expressed interest in having one of these towers. They are quite unique and I’m quite happy about how it all turned out in the end. In looking at the cost of printing them at work (the cheapest 3D printer I have access to), I realized that I could print them much cheaper myself if I had my own printer. Given the interest in the project and the opportunity to save people some money, I’ve decided to do a Kickstarter for the dice tower. You can find the full details in the Kickstarter Project page.
In addition to the full scale version which stands 8.6″ tall, the Kickstarter includes the option for a “super-sized” version that is a full 12″ tall. The cost for the towers through the Kickstarter are $40 and $60 respectively including shipping in the US. If the Kickstarter doesn’t fund, I’ll still make them available but the cost will be at least $50 and $80 respectively due to the higher cost of printing them at work. So if you would like one of these cool dice towers, consider supporting the Kickstarter (which runs through mid Decmember 2016). And if you have a 3D printer there is a support option for just getting the files to print your own in the Kickstarter as well.
This was a lot of fun to make the prototypes, test them, and tweak them to the final design. I have some other ideas for alternate variations that I might pursue as part of the Kickstarter or afterward that might get another post later. My kids have had a lot of fun just playing with the ship itself. My 5 year old like to take it and fly it around the house. I can’t wait to try it out at an actual game.
If you have any comments, thoughts, or suggestions, feel free to let me know below.