Modular Street Food - Saeed's Falafel Cart

Falafel is a delicious meal but it takes a long time to make and because of this, having it in a food truck, and having it run efficiently, is nearly impossible. Our solution was to make a Falafel cart that automated the tedious tasks of making falafel. There are many steps in making falafel that could be easily automated to speed the production process. We would design a food cart that would takes these repeated tasked and automate them making a fast and efficient food cart that served fresh falafel. We split up the falafel making process into smaller jobs that had to be completed. We were given the frying job and more specifically we would automate the process of balling, frying, and flipping the falafel.

We felt that we could make an amazing food cart that could make fresh falafel with less time and human involvement. We would accomplish this by automatically extruding, frying and flipping the falafel. In the end we found the actual extruding process difficult to automate and over all not necessary for our cart. So the main thing that our group focused on was the Frying and flipping. We accomplished this by designing a series of frying baskets that all sat in the same tub of oil but could move independently. This would accomplish a few things but mainly making it so each falafel sandwich would be made separately so we knew each sandwich would get the same amount of falafel and they would all be cooked the same amount. One of the main challenges of this project was the create a good design for the basket and get all the friars to move independently. Along with the design challenges, it was very difficult to work in a small team that was a part of a much bigger group. Many things we did were thrown to the side because of the lack of communication with the other groups.

We had two main iterations of our project.

The first involved an extruder and one large basket. Very early on in the project, we felt that as much of the falafel production process should be automated as possible. This meant that in the beginning our group wanted to make and extruder. The extruder would we filled with pre made dough and then pushed into our shaper. Our falafel shaper was a spinning cylinder with falafel sized disks cut into them. The idea was that when the disk circled around, the compressed falafel dough would extrude into the disk cut out, and then as it spins, it would cut the dough off and drop it into the fryer. Another part of this iteration was the fryer. The fryer was one big fryer with one flipper. The flipper was basicly a basket  that pivoted when the falafel was cooked and flipped it onto the conveyor or whatever was the next step. We liked this design because it was simple and it could make a lot of falafel really fast.

Our next and final iteration had no extruder and multiple baskets. After a lot of talking with our group and other groups, we felt that an extruder was not necessary. It was hard to make, might destroy the dough, and over all not necessary. At this point in the design, we had already decided that there were would to at least be a few people working at the cart, and it would be very easy for one of them to either ball the dough, or put pre balled dough into the fryers. Also in this iteration, we also added multiple baskets. This was done so that each falafel sandwich was made separately. We did this so that we did not have to stockpile falafel and that each sandwich would get the same amount. The challenge of this is that each basket had to pivot separately and keep track of its own time. At the end of the project, we have a working quarter scale model of our fryer. It has multiple baskets that all pivot separately and dump into our funnel that bring all the falafel into one point before being dropped onto the conveyor.

As a studio, we were tasked with building an automated falafel cart. Furthermore, our group was tasked with creating an automated folding mechanism. The mechanism had to perform three specific folds that would make the open sandwich a wrap: a left fold, a smaller back fold, and finally a right fold. Our other objective was to wrap the falafel in tin foil.  Our solution was to move the wrap back and forth across a stationary, fluted, curved ramp that would take the edge of the bread and fold it in. The wrap will move via a plate that is attached to a track that runs beneath the upper level, a.k.a. within the box. To make different folds with a stationary ramp, the plate will rotate 90 degrees using a gear attached to the rod and a few teeth attached to the underside of the upper level. This mechanism will serve two purposes; not only will it fold the falafel, but it will also fold the wrap in paper by running it through the machine again.

         Even though folding by hand would be more efficient, we are not aiming for efficiency in this machine. We want to make an automated system for folding that looks interesting, and can serve as an attraction for the cart. After all, this is the final step of the semi-automated falafel process, so we wanted to end the show with something mechanically compelling. Our folding machine works by using a plate to drag the edge of a lavash bread up a curved, sloped ramp, so that the edge of the ramp gradually brings the edge of the lavash to the ⅔ mark. Then, a gear fixed to the rod is turned 90 degrees by a strip of teeth on the end of the track. Because the track is ovular, the lavash is now folded 5 cm shallower. It then turns around once again to make the third fold on the other side (another ⅔ fold). To wrap the falafel in foil, this process is repeated again, but with paper underneath. A person will both operate the plate with a handle, which is attached to the rod fixed to the plate, and place the paper under the folded wrap.Compared to former projects we have worked on, this mechanism has gone through a lot of changes. Over the course of the folder’s development, we have fundamentally changed its design several times, and have talked about fundamentally changing it many more. We listed three iterations, but in reality, those versions are just major turning points in a vast network of small iterations. The track that guides the plate has redesigned many times. The rod has been held by all sorts of sockets with bearings, and our final design does not include a socket. Everything from the turning gear to the size of the box has been changed at one point. The only thing that still remains from the original design is the ramps, and even they have had their shape changed twice. Even though making all those changes was stressful, (especially the changes made a day before the deadline) I believe we had a great process. Every change we made led to a deeper understanding of our mechanism and our vision for the mechanism, and a deep understanding of those things is essential, especially in a mechanically intensive project like this.

If we were to continue working on this project, our next step would be to use a bristle like material to make it easier for the plate to rotate and turn, hopefully eliminating the need for gears. We would imagine the bristle materials would help prevent crumbs and falafel remnants from falling down into the box. We would also look to use materials such as stainless steel instead of wood as we need a material that is “food safe”. Finally, we might look to create an even quicker folding mechanism that does not need to go around the track twice.

The objective of this studio was to make a falafel cart based on Saeed's recipe and offer a contemporary take on traditional designs. In doing so, we had to focus on how to make something that would offer a memorable experience to anyone who would interact with it. One of the first things we did was figure out what groups we could split up into by looking at Saeed's recipe and our process maps, this was especially beneficial to us as this was such a large project. Once groups were decided, they would each design their own facilities within the constraints of an upper hierarchy. This system worked well with the modular aspect of our cart, simple specifications meant that systems could be replaced with whatever was deemed fit. Once all teams finished a scale draft the entire group could put their models together and decide what was missing or off. The final quarter scale model represented our end goal, which was to create a fully functioning cart to deploy in Central Square.

One of the advantages of doing the cart in separate groups is that all of the machinery is inherently separate and thus every piece is mounted, not integrated. The chassis, the prep section and the assembly top section can be removed and replaced easily. No piece is directly linked onto another, only mounted to the delivery top (and the prep section). That means that by using the same mounts, we could switch out the falafel equipment for equipment designed for making other foods. For example, you could turn our cart into a pizza cart or a hamburger cart. You could also make the cart have a deployable seating area by changing the chassis to include storage for tables and chairs. 

The over all goal of this studio was to create a high tech and soulful Falafel Cart. My group mainly worked on the storage and the preparation area. 

Inspirations 

Our first step was to look at what exists and then use that research and our own knowledge to create our own NuVu style Falafel Cart. We also looked at different ways of interacting with the customer while selling the Falafel. Since my group's pecific focus was the prep and storage area. We looked at the kitchens of food trucks. Basically, the first week was spent in research and brain storming. In the second week, we started to put an actual quarter scale Falafel Cart model together.

Iterations

For the first iteration, everyone was required to create a card board version for the part of the Falafel Cart that they were working on. This helped us to get an idea of what the final product would look like.

In the second iteration we started to add details to our designs. This was where we started to face design challenges.

One of the challenges we came across was having to give up a large part of our design to chassis group. In the beginning we didn't want to give this space to chassis group because my group thought that it would make our prep part smaller, and harder for us designers to give the necessary ingredients equal space. By the end, when figured out the exact measurements, we learned that giving up that space would have made no difference.

Knowing the exact measurements in full scale, led to challenge two. Inside of the prep area was too deep. It would have been hard for a cook to reach all the way to grab something. In the second iteration we solved this problem and added some new things to the design.

Iteration Two

In the second iteration we cut the prep area horizontally in equal halves to make it less deep. We added a storage area for the dough. We also added doorways to the front to provide access to the ingredients without pulling the prep area out.

In the very beginning of the process, we did a lot of research and brainstorming. The majority of the research we did was looking for precedents that provided a culinary experience through either performance, technology, or a bit of both. Seeing as our cart was going to be somewhere in between performance and technology, we had to have a firm grasp on how people and robots have made food in the past. Two of the robot precedents we looked at were Let's Pizza and a programmable robot chef. Both of these were a bit creepy for different reasons. The "Let's Pizza" was essentially a vending machine, so the food looked very mathematical and soulless. The robot chef made good food, but the kitchen had to set up in a exact way, and no one wants a robot with a knife swinging around their kitchen. This was when we realized that to make an automated falafel cart, we would have to deal with the uncanny valley. In other words, we would have to make sure the automation was advanced, but not so advanced as to be creepy. The brainstorming we did mainly focused on food carts and food trucks. We asked why would a person go to a food cart, what the problems with food carts were and how we would address these in our cart. We mapped the falafel process out after a live demonstration of how to make a falafel wrap. Then, we split into two groups: a group to focus on experience in favor of mechanics, and a group that used mechanics as the main attraction of their cart. The experience was the Falafoucel, and they had a rotating design, where their were chefs in the middle, and you ordered at one side. The FFFF focused on mechanics and was Their carts was a glass box with a nearly autonomous falafel machine inside. You would order at one side, sit in a chair, and the chair would move around a track, following the falafel. 

We were tasked with creating the tomato dicing mechanism for our falafel cart. The general idea was to create a mechanism that could chop tomatoes (and potentially other vegetables) into cubes with either the push of a button or the pull of a lever. We decided to go the route of using a lever, as a button would have involved more automation of our task than we had the time or resources for. We ended up using a system with two perpendicular grates, one of which moved along the axis of a handle and one of which was fixed to the bottom of the dicer. We also had a press that was parallel to the bottom grate and could press the tomatoes down once they had been sliced in one direction. This mechanism requires only two movements of the handle, one left to right and one down.

We were tasked with creating the tomato dicing mechanism for our falafel cart. The general idea was to create a mechanism that could chop tomatoes (and potentially other vegetables) into cubes with either the push of a button or the pull of a lever. We decided to go the route of using a lever, as a button would have involved more automation of our task than we had the time or resources for. We ended up using a system with two perpendicular grates, one of which moved along the axis of a handle and one of which was fixed to the bottom of the dicer. We also had a press that was parallel to the bottom grate and could press the tomatoes down once they had been sliced in one direction. This mechanism requires only two movements of the handle, one left to right and one down.

Ezra and Will created a design for the overall mechanism at the end of the first week of the studio. They also created a rough cardboard model for the mechanism. Their idea included a manually operated platform crusher that sliced the tomato in two directions. We followed their sketches and continued to perfect our mechanism in rhino. We spent a lot of our time going back and forth from the laser cutter to our work space for the majority of the studio. Early in the second week we created a working prototype of the basics of our mechanism out of cardboard. This gave us a good idea on how to construct the overall design. We also consulted with other groups and determined the exact dimensions in which our design needs to be. We only have 35 cm of space according to the prep kitchen group.

    With the help of Andrew we decided to put tracks in our design. We decided to do this because it would allow the moving platform and slicer to move easier. Our track is an L shaped. This allows us to have the pieces move sideways and vertically. This helps us because we want our pieces to move sideways and vertically, so the object that we want to slice can get sliced in both directions. We built a test track to see if it would work. The track we created seems complex, however, it is actually fairly simple. This is because it is just a fitting piece, fitting into a similar sized slot.

    We had three main iterations with our dicer. One of our first iterations was fairly similar to our dicer but had no tracks. We originally planned our moving platform/slicer to move along the walls without directional tracks. Due to friction this idea failed,for the moving platform/slicer was fairly hard to move. Then Andrew gave us the idea of using directional tracks and he helped us create them. We then put a track on one of the walls. This gave our platform somewhat stable movement throughout our mechanism. However, it was unstable on the other side, slanting the platform. We then added tracks to the other side perfecting our platform mechanism. After a few tests we are confident that our mechanism will be able to dice tomatoes. However, we are still trying to perfect the actual slicing mechanism as we are trying to determine what material to use.

In this project, one of the biggest challenges we faced was ourselves. The sheer number of people working on the project (over 20) is more than any that Nuvu has attempted to coordinate before. The idea of getting 20+ people to work togther in a way that will create a detiled, correct product was intimidating, and the coaches approached this difficult task carefully. We began by working in small groups, on simple "automata" that would serve a single mechanical function, and help us warm up to design work and get us thinking about ideas around mechanical movement, which would be key in the later stages of cart design. After that, we spent a full day talking about the challenges and necessities for a falafel cart, and goals we wanted to achieve with our deisgn. This was a full group discussion, and we concluded by starting to come up with very basic large ideas of a type of cart that we wanted to create.

Onwards, the next few days were a constant recombining of ideas and groups. We had two different ideas, then those groups split into small deicated groups, focused on smaller aspects of their respective carts, then an intervention from the coaches brought us back together and refused the groups, leading to the creation of a singular design that we all agreed upon. The group once again divided into more clear cut roles, and from there on, it was a matter of making sure all the groups could make informed design decisions, and work around and together to create a good model. With minor confusion over measurements, this worked well.

Our goal was to create a conveyor belt that is meant to move large pieces of lavash bread from one side of a falafel cart to the other, while allowing falafel and condiments to be dropped onto the bread. We wanted the belt to be large enough fit the lavash bread, but also small enough that it would fit into the cart. We decided to make a basic conveyor belt that carries light objects while still being small and easy to use.

The first iteration of our conveyor belt was built during the brainstorming process when we were still deciding what our conveyor belt was going to be.  We decided to make a device that would rotate all the way around the cart, similar to an airport luggage carousel.  We quickly designed and cut a conveyor belt out of cardboard with three main pieces connected with pins.  This model turned out well, and it resembled a snake toy.  Since this model worked well, we decided to make a second version out of wood, but with three changes; it was made of wood, it was more stable because we added an extra layer in every connection, and we put electrical tape on the ends to add friction for holding the plates.  This model was great, but after a meeting with the full team, we decided that a design more similar to a treadmill or a conveyor belt in a checkout aisle would work better in our cart, because we no longer needed it to turn.

The second iteration of the conveyor belt was a cardboard model for the twenty-five percent scale model of the falafel cart. It uses paper for the belt, and wood for the rollers. In this iteration, the rollers had to be moved manually since there was no motor, and the belt did not grip onto the rollers well enough to move. There were also five missing rollers and there was no tension so the paper was loose. In the cardboard model the sides were too long, and the system was too tall. Because of this the sides had to be cut down so they were not very clean looking. In the final iteration we hoped to make the sides out of wood, as well as get the belt to move using a motor. We also wanted to make a more durable belt and smaller laser cut rollers to fit in the scaled down model. The third iteration allowed us to see what the system would look like, but it was just a visual representation of what we would make.

The final iteration of our conveyor belt is made of wood. There is a spring mechanism to automatically adjust tension on the belt, and rubber grips that are meant to allow for the rollers to move the belt easily. This iteration is powered by a motor that has a foot pedal to control when the belt moves, allowing it to stop at different intervals so that cooks can add condiments to the falafel. The final iteration is still twenty-five percent scale, but that allows it to easily fit into the full model of the falafel cart, Because of scaling issue certain parts of the conveyor belt are very large and do not leave enough room for the belt.  This means the belt is not wide enough to carry a twenty-five percent scale piece of lavash bread. That then means that the belt is not completely to scale, and it does not move by itself. If we had more time we would have made a full scale version of the conveyor system to get rid of scaling issues as well as gotten the belt moving, but for now this belt is great for showing the concept of our design.