It’s been a few months since I published a post, as I’ve been pretty busy this summer (you can check out what I was working on here). Luckily, I’ve gotten back to school and I’ve been able to do some more experimenting with coprocessing! As a recap, co-processing involves altering a print job as it is being printed to produce more integrated 3D printed content. Once example that I’ve created and reference a lot is the impossible bolt:
By screwing the nut onto the bolt when the print is paused mid-print, we can create a seemingly impossible object. This is a pretty cool representation of what you can do with the co-processing technique, but I wanted to take it a step further and take co-processing to another level. I made it in metal!
To be clear, I made this in metal, but I did not make it on a metal 3D printer. The bolt is aluminum, and is not printed nor welded. The nut on there is the same type of nut used for the original impossible bolt, just a bit less clean. So how did I make this impossible object? Well, first things first, I started with the normal 3D printed impossible bolt, and went through the usual procedures (you can see them here). After that came the fun part!
One of the great applications of 3D printing is its use for investment casting. With a 3D printing material like PLA, you can go through the process of what is usually called “lost-wax casting”, except instead of using a sculpted wax mold, you use a 3D printed PLA part. The process of lost-PLA casting is pretty simple. The only difference between casting an impossible bolt and casting a normal printed object is that you have to ensure that the heating furnace never reaches the melting point of the nut. In this case, the nut was steel and the bolt was aluminum, so everything was ok.
After I printed the part (remembering to coprocess partway through), I went through the standard investment casting procedures of making a mold. I first attached the printed bolt to a wax tree and secured it in a casting flask, then wrapped the flask in cling-wrap, poured plaster in it, and used a vacuum to draw out all the air. In my print, there was enough tolerance between the bolt and the nut that the plaster was able to wind its way in. This meant that the bolt and nut wouldn’t accidentally fuse together when cast!
The next step was to melt out the PLA. I put the flask in a furnace and programmed a heating cycle that would melt out the PLA. After this cycle, all that remains is a negative mold of the part in plaster, and the steel nut, now fixed inside the negative mold. For the actual casting step, I heated up ingots of aluminum in an induction furnace. Once hot enough, the aluminum was poured into the plaster mold, and then cooled. The plaster was dissolved away in a bucket of water.
The wax tree left a long stem on the end of the bolt, which I had to cut off with the band saw and grind down. The result? The impossible bolt, now in metal! The nut was a bit tarnished because it had been heated up, and the threads are a bit loose, but the object still shows the advantage of the coprocessing technique.
While my previous experiments have been direct examples of coprocessing, this experiment shows an application of coprocessing as I explored how it can be used in tandem with other manufacturing processes. By casting a coprocessed object, you can again broaden the content that can be produced using the 3D printer as a manufacturing tool. As long as the coprocessed part has a relatively high melting point, some really interesting things can be made!
As always, if you have any ideas on how to push this process further, please let me know!