For this project, Jordan Duhe and Miguel Jimenez Gomez would work to create a mechanical model inspired by a flat engine layout. The mechanical model would consist of 2 pistons and a set of gears that would spin a fan, overall replicating the movement of Mechanism #92 from the book 507 Mechanical Movements, the ordinary crank motion.

PHASE 1: Design of the initial model.

To begin the project, we started looking at design possibilities for the model. We both provided some sketching and diagram ideas for the engine. Initially, the concept would have been to have multiple pistons in an enclosed box, so the early designs reflected this draft. Furthermore, a box was created in MakerCase as well as part of the initial concepts. A lot of calculation had to be done to develop the sketches as we had to consider the stroke size of the pistons, which ultimately would have to drive the length of the model. Furthermore, the idea of the fan was still up in the air. Originally, a bevel gear concept would be proposed for Gate 1, but it was commented that bevel gears would be difficult to develop. However, we managed to find a concept only that could provide the movement necessary for the fan, which was two gears spinning a sideways gear acting as a mock-bevel gear system. With that, we used Gear Generator to prepare one large gear that would act on the same axle as the pistons, a middle gear that would be driven by the large gear, and it would also drive the sideways placed gear for the fan.

Link To Sketches Presentation

It was important that we also prepared files to be able to prototype the idea using the laser cutter, so work was done to prepare that. Jordan would develop a model on SolidWorks of a potential layout for the mechanism, and from there I managed to extract DXF files of the relevant parts, which included the pistons, the flywheels and shaft parts, and guides for the piston (the development of the guides was done after discussion on how the flywheels and pistons would be held in a theoretical box). In the model, the parts are shown as press-fitting. As Jordan stated, the concept of press fitting our parts in design would create a simplified assembly without having to utilize adhesives or fasteners. When the files were exported, there was some issue with parts that had circles, as all circles would export as semi-circles. This was fixed by creating a whole circle with the ellipse tool in Adobe Illustrator.

PHASE 2: Low-fidelity prototype

With the DXF files prepared, it was time to develop a low-fidelity prototype for the mechanism. Our chosen medium would be cardboard as it is easy to laser cut relatively accurate parts from the files. As such, parts were cut for the low fidelity prototype at a 1:1 scale as it would give us a better idea of how the mechanism could work. For each piece, 50% power and 50% speed at 10% frequency was used as tipped by another ENGI 210/BIOE 555 student in the GroupMe. After some discussing too, a guide for the driveshaft would be developed the same as one for the pistons was.

Overall, the development of the low-fidelity prototype went well, thought it was noted how the parts where designed for 0.20 in plywood sheets, so the thinner cardboard resulted in parts not being press fit. To compensate for this, some masking tape was wrapped around the press-fitting parts of the pistons. Overall, glue has to be used to assemble the low-fidelity prototype due to this mismatch of widths.

Link to video of low-fidelity prototype theoretical movement

The key takeaways from the low-fidelity model were the general sizing and theoredical movement of parts. We took note of how accurate we need to be with the width of parts so they may be able to press-fit properly, and to account for the proper stroke size of the pistons as to adjust the length of the guides. Movement replication was difficult due to the corrugated edges, so a higher-fidelity prototype material, like wood, would be needed to fully test the concept. Furthermore, we would have to find a way to hold the guides together, so a base concept would have to be developed.

PHASE 3: Medium Fidelity Prototype

For the medium fidelity prototype, we were to utilize plywood to obtain an improvement in fitting parts and in their movement. We also planned to introduce the fan movement, so parts had to be designed and prototyped.

First, a base part was created by Jordan to provide a place for the guides to press-fit into. This also meant that the guide parts would be redesigned to account for the press-fitting depth. This would allow for the guides to properly attach to the base. She would also go to develop the fan platform/guides that would also press-fit onto the base. The features of this part would account for a 0.25″ dowel rod that would allow the gear and fan to spin on the platform. This part would be laser cut. Another redesigned part would be the piston guides. For added straightening of the path, the guides were closed off as a complete housing.

For the other plan, I designed a fan part in Illustrator, and this part would be decided as the metal component for our mechanism. As such, this part would be water-jet cut on an aluminum sheet. Ideally, we would have looked for a much thinner piece of metal to that we could fold the flaps down, but such piece of metal stock was not available in the wet lab for the size of the fan. The fan was designed to account the driveshaft joint (see the plus in the middle).

The final part that was designed in this prototyping stage was the “dowel gear.” We would come to find out that linking two identical gears would not be feasible in creating the movement we want for the fan, similar to the original idea of a bevel gear we had. As such, a custom “gear” would have to be created to engage the fan mechanism to the large gear on the driveshaft. After some discussions in design, we developed a “gear” with three layers, one to attach to the dowel housed by the platform, one to house the dowel “teeth,” and one to attach to the driveshaft holding the fan.

All parts would be re-cut on plywood to account for the proper width they were designed with, allowing us to test the press-fitting concept. Some parts were able to press fit, but it was noted that the guides were still loose, so wood glue was used to attach them to the base. Wood glue was used as well on the pistons, which did not press fit either.

When assembled, the prototype was very stiff in movement, and the driveshaft would lift up and not allow for the proper movement of the pistons. It was also noted that the pistons would twist on the dowel, and within their housing, they would get stuck slightly on the sides. Jordan would identify this as the binding of the pistons to the housing. More notes of improvement for the prototype also included to increase the size of the dowels on the dowel gear, as the would not engage with the gear as planned due to the short dowels and unstable movement of the dowel in the guide.

However, the medium-fidelity prototype provided some positives outside of what needed to be improved, such as proving the structure concept of the mechanism, and some movement of the pistons was still possible.

Movement of medium-fidelity prototype

PHASE 4: Final mechanism

To finalize the design of the mechanism, Jordan would develop a final CAD model assembly in SolidWorks. This helped visualize the assembly of the piece with refined parts.

For the final design, all part files were overhauled to aim for proper press-fitting of the parts, proper movement, and new parts would be designed to help achieve the aimed goals.

For new parts, bushings were created so that they would fit between the flywheels and the pistons to prevent twisting motion of the rod, preventing any unwanted movement by the piston rods.

The remaining parts would be redefined to better accommodate assemblies and to facilitated improved motions. For the dowel gear platform, an extra layer was added to add stability to the dowel that is housed within the platform. This would prevent it from spinning off-axis, which would make the dowel gear and fan unstable. To better house the driveshaft, a redefined guide featuring a closed-off housing and different sized openings would be developed. This redesign intends to completely house the driveshaft guides to prevent vertical and horizontal motion, which caused issues in the medium-fidelity prototype.

A notable fix would be with the piston and guide interactions. As stated, the issue was noted to be the binding of the pistons within the housings, resulting in very stiff motion, and often the piston getting stuck to the guide. A solution would come from  Jordan’s job experience. In her career, she has worked with systems that utilized sliding table designs, and a concept she learned was the ratio of length and diameter of a part. To reduce the binding of a sliding part, it was stated to increase the ration of length and diameter, and so she applied this concept to the pistons by increasing the length of the piston head and of the guides.

The remaining parts were adjusted either by length or to be more press-fitting.

After the redesigns, all parts were laser cut one last time.

After the parts were cut, post processing would begin. Each part was stained, the structural parts in black and American Walnut as the stained looked nice, and the pistons in North Sea to emulate metal. The dowel rod gear was also reduced and stained in North Sea.

Post-processing was also done for our fan piece. Jordan sandblasted the fan piece and helped design a decal for the fan. This decal aimed to resemble fan blades, and it would met our required vinyl cut aspect as part of our mechanism. I went ahead and worked with the vinyl cutter to cut out the decal.

Finally, our name plate was also created and stained.

After the parts were done drying, the final mechanism would be assembled. During assembly, it would be noted that the improvements to the piston and piston guides worked, with the binding being reduced significantly enough so that proper motion could be achieved. The driveshaft guides also kept the driveshaft stable, although it would be a very tight fit, resulting in some rougher motion, although functional. An issue would be noted with the dowel gear, as it Jordan stated it would disengage often with the platform not being stable. Finally, some structural elements could not be press-fit still, including the dowel-rod platform and the driveshaft guides. This assembly would be noted as a preliminary assembly before making some changes for the final submission.

Video of piston motion

Video of the mechanism motion

To improve on the mechanism before final submission, the dowel rod would be recreated with longer dowels, and the remaining structural elements would have to be attached with glue or an adhesive.

During the refining process, wood glue was successful in attaching the dowel gear platform to the base, but not with the driveshaft guides. The movement of the handle and flywheels resulted in the glue breaking down, so a stronger epoxy was used to attach it to the base. Epoxy was also used to attach the fan to the dowel gear. After all the adhesives dried properly, the structure worked much better, apart from the dowel gear getting stuck for being too low engaged to the big gear. As such, a bushing was used to lift the dowel the dowel gear ran on, and that improved the motion.

With that, the mechanism was complete, and the function worked relatively well.

PHASE 6: Final deliverables

The following are videos showcasing the motion of the mechanism, both in slow and a faster pace.

General motion of mechanism

Faster motion of the mechanism

Finally, here is the Google slide deck of the entire project iterations and mechanism explanations:

Presentation of the mechanism

PHASE 7: Reflection and cost breakdown

Working as team through this project was a very rewarding and interesting experience for the both of us. It was very rewarding for us to culminate our skills together to develop a complex mechanical model.

For myself, it was definitely a new challenge to develop a mechanism of this extent, but aspects engineering design process were deployed and a lot of creative thinking needed to be used to develop parts such as the guides and the dowel gear. This is one of the most complex things I have contributed to building, and I say it was a highly rewarding experience.

For Jordan, she stated that it was very helpful to have recalled her experience in the engineering field to engineer through problems, most importantly her fix of the piston movement. A lot of her comments apply to the prototyping aspect of the project, noting the helpfulness of prototyping to find what needs to be improved with each iteration, while also considering that things may function differently due to material choices. An example of this was the cardboard being quick to prototype, but not representative of the final motion of the mechanism due to its physical properties. Speed was also key, and the laser cutter was noted by Jordan as a helpful tool in creating quick iterations.

Overall, we are happy on how our model turned out. For a theoretical V2 of the mechanism, we would look to improve on the motion of the driveshaft, potentially using bearing for smoother motion, and to also make sure all parts are press-fit onto the base.

The following is an estimated cost of the mechanical model:

• 1/4″ 24″ x 48″ plywood panel (Home Depot): $20.99
• 1/4″ 24″ x 24″ plywood panel (Home Depot): $7.28 x 6 = $43.68
• Scrap wood estimate: $30
• 1/8″ 24″ x 24″ cardboard sheets (Amazon): $2.44 x 6 = $14.64
• Overhead/Service cost for Laser Cutter (LaserHints): $1 x 1200 minutes estimate = $1200
• Labor: $20/hr (national average for a laser cutting operator by ZipRecruiter) x 10 hours = $200 x 2 users = $400
• 12″ x 12″ Aluminum Sheet (Home Depot): $11.47
• Overhead/Service cost for water jet cutter (TechniWaterjet): $37.50 x 1 hour = $37.50
• Labor: $18/hr (national average for a water jet cutting operator by ZipRecruiter) x 1 hour = $18
• Labor: $18/hr (national average for a woodworker by the Bureau of Labor Statistics) x 4 hours = $72 x 2 workers = $144
• Wood Glue (Amazon): $2.12 x 4 fl oz = $8.46
• Epoxy: Negligible
• Overhead/Service cost for vinyl cutter: Negligible
• Vinyl roll: Negligible
• Dowels: Negligible
• Stain: Negligible

The estimated cost of the mechanism is about $1,928.74. This project was very labor-intensive and material costly, so a lot of the costs come from labor materials. To potentially reduce the cost of the mechanism, less labor time could be spent on the project, and/or more efficient use of materials could save some cost with that aspect.