Chocolate-making and metallurgy have many similarities, from phase diagrams and eutectic points to continuous cooling curves. A self-driving chocolate-making lab is a great “frugal twin” of an autonomous alloy discovery lab.
This is one of the items from the AC Training Lab.
Major thanks to Juan Pablo (JP) for working on this! (JP is from Christine Allen’s lab and is working under the auspices of the AC Training Lab over the summer)
The hardware is now built. Next step, software!
The Cocoa Press Chocolate 3D printer is a simple but impressive machine; taking about 2 weeks to build and calibrate, the Cocoa Press was a challenging robotics project for me to take on, yet extremely gratifying in the end. Averaging out at around 3 hours of daily work, building the Cocoa Press was very feasible despite some of the more laborious wiring and calibration tasks. And, more than anything, came with an excellent tasting reward!
The main benefit I found with the Cocoa Press is its customizability; this 3D printer allows for complete adjustment of a wide variety of its parameters for whatever project you want. These include the nozzle and body temperature (ie: the temperature the chocolate is heated to and extruded at), the printing speed, the flow of chocolate, the steps per millimeter that the base moves, and even the current at each of the three motors in the printer. Adjusting these has allowed me to print significantly more complex designs and actually treat the Cocoa Press like a fully equipped 3D printer. It also comes with all of the necessary cleaning and maintenance equipment, which I found to be incredibly useful when approaching clogs or incorrect printing. I will note that the design chosen for chocolate extrusion is not ideal given its limited chocolate-loading capacity, but many of the other included features (such as self-leveling the base, SD card functionality, etc.) compensate for this fact and allow it to be used for so much more than making nicely shaped chocolates.
Though the Cocoa Press has an overall easy-to-build design, it is not quite as intuitive to wire or adjust as one might expect. The electrical portion of the printer seemed to be my main issue with the project, as the manuals had little guidance on when certain connectors needed to be fed or where to connect them, partially delaying the building process. Mainly, powering the digital touchscreen and probe wiring took the most time, but were still doable (even by a novice like myself). I would also note that building it from scratch is the best way to know how to calibrate the machine; knowing which components were involved allowed me to correct a lot of the issues encountered, much more than following any troubleshoot guides. I mostly saw this when height-adjusting the chocolate extruder (a critical part of calibrating the machine), as it required tweaking specific hidden screws that would otherwise be overlooked. Calibration and adjustments alone did take around 4 hours, which considering the amount of rewiring was not too intense.
Despite some of its potential complications, the trial and error construction process of the Cocoa Press pushed me to be significantly more methodical when approaching technical issues, and overall, I would still say that the construction of the printer went smoothly (especially for someone who has never built a 3D printer from scratch). The learning curve you may experience if you have no experience with the many, many different types of screws can feel daunting at first, but ultimately leads to a robust understanding of mechanical assembly and structural integrity.
I’m curious - How good is the chocolate?
It’s pretty decent! You’ll have to try for yourself sometime 
Here is a milk chocolate benchy print made with the Cocoa Press! Benchy prints help to test 3D printers and calibrate them properly.
Just to provide a final update on the Cocoa Press:
Building the chocolate 3D printer in of itself was the least difficult part of this project. Calibration, testing, and correcting the filament speeds/printing rates were the more complex parts of the process. Due to the fact that the Cocoa Press relies on a touch screen to operate and change many of the printing parameters, it may be difficult to integrate it into an automated system. It also relies on an SD card to actually input any new printing designs. Additionally, it is not a particularly easy machine to take apart and put together, limiting its transportability.
The next steps for this project would be to:
Hope to see this project automated soon!
