TofuPresto

Finished prototype top view
Internal gear box of the device, held in place by acrylic ceiling.
Finished prototype isometric view
External view of the device, with user interface on top.
Gear box before assembly
The team

Inspiration:

Compared to meat, tofu is much more environmentally friendly source of protein. Unfortunately, individuals are often hesitant to cook tofu because the preparation process can be tedious: before cooking, tofu must be incrementally pressed for 1+ hours to drain the excess water and achieve the right consistency. We developed TofuPresto for ourselves and all those who are looking for a more convenient way to eat tofu. Specific markets that we’re aiming to address are college students, single parents, vegetarians who are currently eating meat out of convenience, and people who are desiring to eat tofu in order to have a smaller environmental impact. We hope that this device will encourage individuals to reduce their meat consumption by lowering the obstacles for eating tofu. By popularizing tofu, we can help the world make a more environmentally friendly choice.

What it does

General overview:

TofuPresto stores the tofu in a user’s fridge, keeping the block soaked in water to preserve its freshness. Once the user anticipates returning home to cook tofu in one hour, the user sends a message over WiFi to initialize TofuPresto. TofuPresto begins to lift the tofu upward in its container, draining the liquid as the block is pressed with increasing strength against the lid. The process continues for 1 hour without any more effort from the user. When they finally return home, the user is welcomed by a pressed, firm tofu that is ready to be tossed in a skillet and enjoyed.

A more detailed look:

After purchasing the tofu, the user places the tofu in a container with water added. The tofu rests on TofuPresto’s porous, removal tray that suspends the tofu slightly above the bottom of the container, yet still resting in the water.

Once the user desires to initiate the pressing, they send a signal via Twitter over WiFi (for the prototype) or mobile application (for the final product). This turns on TofuPresto and starts incrementally rotating the stepper motor. The motor steps at a rate of 30 rpm, equivalent to the recommended rate for manual tofu pressing. When the motor steps, it rotates a gearbox that is attached to four corner gears and their connected threaded rods. The rotating threaded rod causes the tofu tray to travel upward and press against the container lid. Excess water drains out of the tofu and through the porous surface of the tray. During the pressing process, a screen on top of the system displays the remaining pressing time.

This process gradually increases the pressing force for 1+ hours then notifies the user and holds the tofu in the top pressed position until the user retrieves the tofu. When the user is ready to cook, they press “Stop” to lower the tray to a non-pressed position then pull out the tofu on the removable porous tray. After dinner, the electronics box (top compartment) is easily removed and the rest of the system can be washed in the dishwasher.

How we built it

TofuPresto is integrated on top of an existing $4.00 food container from Target. Two 3D printed compartments stack on top of the container’s lid; the top compartment houses the stepper motor, motor shield, Adafruit Feather HUZZAH, and screen. When connecting to the middle compartment, where the gearbox is mounted, the top compartment matches via the stepper motor shaft and complimentary cutouts in the two compartments. The stepper motor rotates the central gear which transfers rotation to the four corner shafts. By rotating the shafts, the tofu platform travels vertically, pushed up by captured nuts below the platform.

Our build and assembly process for MakeMIT is different than if it were to be mass manufactured in the future. Because of our limited time and resources, we had to partition each part so that it could be printed by multiple machines at the same time and without the use of support, which would elongate the print or mar the cosmetic surfaces. We also laser cutted some flat parts to save time on production.

Challenges we ran into

Like any hackathon, time presented itself as an obvious challenge. Since we chose to prototyping our design in primarily 3D printed parts, a huge tension was the pull between careful CAD design and the need to send it to a printer as soon as possible. We ended up finalizing the CAD around 7:30 p.m. and then found out that each compartment (intended to be printed as one piece) would take 15 hours to print, which was a length we could not afford. For MakeMIT purposes, we printed our parts as slices and glued them together which allowed us to utilize multiple printers for less time. However, to fix this issue in the future we would design the parts for mass manufacturing through injection molding if possible.

We also ran into challenges with driving our stepper motor. There was a documented bug in the Adafruit motor shield V.2 library, specifically with the Adafruit Feather HUZZAH, that was resolved internally in the library but was repeating the bug in our own motor’s library. Therefore, team members had to think on the fly to adapt the other library to our motor’s library.

Another electronic challenge that we faced was in connecting to WiFi. We had success connecting to a team member’s iPhone hotspot, but the system does not connect to any other phones or public WiFi networks.

Accomplishments that we're proud of

Identifying a currently unmet customer need The current solution to draining tofu involves manual presses that are tightened every ~15 minutes over the course of a couple hours. The most popular products are simply two lead screws driving parallel plates toward each other when butterfly nuts and rotated on the system’s top. This is an inefficient and highly attention-demanding process that discourages many individuals from cooking and enjoying tofu because other food options are less time consuming and preparation can begin immediately. Our team was surprised that a product like TofuPresto does not currently exist and are happy to have started filling that technology gap.

Intentional food-safe design with separate electronics box Our team quickly identified that, when dealing with food products, regular cleaning is a must. The product must be designed to be machine washable or else the barrier of use may be too high. Therefore, we designed the

User-friendly design choices We focused on making the device as user-friendly as possible. We designed the compartments to be as compact as possible, limited majorly by the height of the available stepper motor. We also designed the tray that the tofu sits on to be removable, making it easy for users to have quick access to their block of tofu and remove it without difficulty. A third intentional design choice was to make the electronics sit on the very top and have their compartment be removable. That way, no tofu water leaked down onto the electronics (which would likely happen if the electronics were below the food container). This organization also allowed the device to be machine washable once the electronics compartment has been removed.

What we learned

Our team gained much experience in the mindset of prototyping. Although our team members had participated in other fast-paced product design cycles, we did not utilize the prototyping instinct that had been pressed onto us. Instead of iterating on rough designs, we instead polished a digital design and hoped that all went well in printing and assembling. This added much pressure to the team members designing in CAD as no mistake could be easily and quickly corrected without completely redoing the print. However, we realize that this type of digital prototyping may be more realistic for industry applications where products may be too large or complex to be prototyped.

We also gained better instinct in designing for prototyping. When completing the CAD, team members were seriously underestimating the print time that would be necessary to product the parts. Our initial estimates were around 5 hours, therefore making a 6:00 print start time adequate for completing the print by MakeMIT’s close (midnight). However, the prints ended up lasting between 10 and 20 hours, which was infeasible to print in the scope of the hackathon. In the end, we had to slice the parts into segments to later glue together. Now, our team better realizes the inefficiencies of 3D printing in prototyping and has a better benchmark for estimating future print times.

What's next for TofuPresto

There are many possible improvements and extensions for TofuPresto. To produce a more refined final product, we will begin by further developing the communication function. We will replace the WiFi capabilities (via Twitter) with a mobile application that will display the pressing options and pressing time. We will also improve the seal of the container to ensure that the tofu is kept as fresh as possible. The prototype materials will be replaced food-safe materials that are injection molded to improve manufacturability. Once resources are available, our team will integrate more compact components (such as a smaller stepper motor) to decrease the height of the compartments.

Possible extensions for TofuPresto include cutting attachments (resembling cookie cutters) that will dice or specially cut the tofu block for the user. We would also explore marinating and seasoning options.