For my final Tinkercad project, my magnum opus if you would, I decided to design an instrument. In the initial stages, I had no idea what instrument I could realistically design in a couple of weeks. I first considered designing a trumpet like this trumpet from Thingiverse. I decided that this and most other instruments I could design would be extremely difficult to design in Tinkercad. I almost gave up on the idea as a whole when the perfect instrument for Tinkercad came to me:
The Alphorn.
The Alphorn is a type of “brass” instrument related to the French horn. I say “brass” because they are typically made of wood. A standard Alphorn from Switzerland (this instrument originated from the Swiss Alps) measures about 3.7 meters long. That’s about 12 feet. For obvious reasons, I can’t make the first print the life-size model, so I would first need to print a prototype model that’s a 1/50 1/25 scale model. Before I can design the model, I have to do research.
Research
Before I can start modeling the design, I first had to figure out what the dimensions of a typical Alphorn would be. After searching around on Google, I found this website that gives exact measurements of an Alphorn. Theoretically, it seemed like all I needed to do was to take these numbers and plug them into Tinkercad to get an Alphorn, right? Not quite.
Small-Scale Print
Design
Wow. This turned out to be a lot more complicated than I thought it would be.
First, I knew that I wouldn’t be able to print the whole thing on one 3D printer, so I would have to print it in multiple parts. This is fine though, as normal Alphorns come in pieces as well. To figure out the number of pieces it would need to come in, I took the total length of the Alphorn, 3,679mm, and kept dividing by 2 until I got the length under 215mm, about the biggest an Ultimaker can print. This turned out to be 32 times.
That’s a lot of printing.
This isn’t even the most challenging part of this. In Tinkercad, you can’t exactly just cut a shape into 32 different parts. You have to design the individual parts.
This was the point that I had to start relearning everything I learned in Algebra in middle school and high school and break out my old TI-84+ graphing calculator. In order to find the dimensions of each individual part, I had to figure out the slope of the cone to get the diameter at exact intervals. After much trial and error, the following equation gave me the results I needed:
Keep in mind that this is for the scaled-d0wn prototype model; I’ll need a new equation for the full-sized version.
After several hours of crunching numbers and creating all 32 of these shapes, this was the final prototype design:
It doesn’t look like much, but when put together, it’ll be an Alphorn in miniature form.
While I got what I needed to know for the full-sized version, this print did not turn out like I thought it would.
I later added smaller hollow cylinders to connect all of the pieces together, but this print was on such a small scale that they never showed up. Additionally, the bent pipe object that I am using to curve the bell at the end turned into a mess of spaghetti due to the same reason that the connectors didn’t print. While this print did not turn out that well, it told me enough of what I needed to know in order to design the full-scale versions, as everything appeared to line up how I expected.
Full-Size Print
Design
I designed the full-size version in much of the same way that the small-scale version was designed, but with a slightly different equation:
[insert equation here]
as well as splitting up the design into multiple files:
This, much like the small-scale version, took a while to do, but not as long as the small-scale version since I had a better idea of what I was doing.
Slicing the Print
It seems odd to give the slicing process its own section, but it will become apparent as to why.
So I open up Ultimaker Cura on a non-3SPACE computer just to see how long it would take to print a sample part. I decided to go for the largest part first, but that’s when I came across another issue:
It’s too big. I rotated this part as many ways as I could and positioned it any way that I could to try to get it to fit on the build plate, but this piece is simply too big to fit on the build plate. This also wasn’t the only issue; I took the smallest piece and put it in Cura to see how long it would take to print:
At this point, I was about to give up and just print a half-size Alphorn. Between the size and the time it would take to print, this project just didn’t seem feasible. Before completely giving up, I had one last idea I wanted to try to make this object print the way I wanted it to.
Back to the Drawing Board: Meshmixer
At this point, this overwhelming project is past due and the class is starting on working with Fusion 360. During our first design with Fusion 360, we used a program called Meshmixer to finalize the details of the object. After digging up some information about the piece of software, I found out that not only could the program cut objects in half, but could also add grooves or joints to hold big prints together. This was my last option at this point, as it wouldn’t be feasible to use a 3D print service (the cost of one part alone would have been over $300!).
I imported the STL file of my biggest object and got to work on it, using a tutorial I found online for this very purpose. In the end, even the biggest object would be able to print on the Ultimakers. They would still take a very long time to print, but it is at least possible to print now.
Printing the First Batch
The hardest part about this section was finding the time to print that would not conflict with a class. I planned to have it all printed out within a week, but the only opportunity I had to print with enough time available was on a Tuesday afternoon, when the next class to meet was the next day at 2pm. This left me with printing objects that had to be less than 24 hours. This left me with limited number of objects I could print, especially when you consider making sure each printer has enough filament. In the end, I was able to print 14 of the pieces and 13 of them turned out fine. Unfortunately, one of the halves of the bell took much longer to print than estimated and used up more filament than estimated, so not only did it conflict with the 2pm class, it ran out of filament before it could finish, so it will need to be reprinted.
While the other 13 prints were successful, there was a design flaw that I discovered; I made the male end of the connectors too wide to actually fit all of the pieces together. To fix this, I decided to sand down all of the connectors I printed with a belt sander. Even though the friction from any electric sander can make the PLA plastic start to melt, the connectors won’t be seen once put together, so this wasn’t an issue for me. For the future prints, I will make the adjustments necessary to ensure a proper fit.
