Candy Dispenser, 2021

To access the PDF of my Candy Dispenser Photo Essay, please follow this link here.

In my Stanford ME102: Foundations of Product Realization class, we focused a lot on taking our product design abilities to the next level, beyond rapid prototyping. We would design 3D models using Computer-Aided Design (CAD), and use laser-cutters or 3D printers to bring our designs to life. The class is a very thorough introduction to product realization, as the designs were not finished once they were 3D printed or laser-cut. We would continue to assemble our models using hardware provided for us in the Product Realization Lab, and we would even keep a bill of materials.

For my final project, I was tasked with the challenge of sketching, designing, prototyping, and digitally fabricating a candy dispenser, and I was only given the following criteria:

The candy dispenser must:

• store at least 20 artificial candies, which will be provided to you. Candies will be loaded without preference to position relative to one another.
• dispense only 1 candy per user actuation/input and reset for the next dispensing
  • the dispensing action can be “passive” (such as a gumball machine where the gumball falls due to gravity)
  • the dispenser can be “active” with some kind of launching action
  • the dispenser can be “silly” and dispense back into itself (search “useless machine”)
• include a controlled, specific location for the dispensed “candy”; the “candy” cannot just drop onto the table below the dispenser.
• use rotary motion to actuate; this could be turning a knob or moving a lever through an arced path
• incorporate mechanical stock components, including:
  • hardware for rotary motion (bushings, bearings, shafts, etc.)
  • a spring (compression or extension) subject to a preload, hard stops, and a purely linear load between its endpoints.
• be operable with one hand
• fit within an approximately 6” X 6” X 6” volume

Skills/Programs Utilized:

• SolidWorks (Computer-Aided Design)
• Ultimaker CURA
• Laser-Cutter
• Ultimaker 3D Printer
• Adobe Illustrator
• Design Sketching
• Rapid Prototyping

Getting Started

In my initial brainstorming stages, I wanted to focus on these assignment requirements: 

• 

• 

• 

Refining the Design

Rapid Prototyping

Back to the Drawing Board

Functional Prototype

I started the prototype from scratch, applying the changes from my rapid prototype. I fixed the disk underneath the container to hold the candies, used the spring to reset the disk and a static connection to join the disk and the lever.

• 

• 

• 

• 

• 

Digital and Physical Modeling

• 

• 

• 

• 

• 

• 

Critical Part – Rotating Disk

I initially intended to laser-cut this piece, with a hole to press-fit a 10-32 machine screw (instead of the extruded knob as pictured) but decided to change the design to be 3D printed upon realizing that using a machine screw would require a hex nut to secure in place, and the back of disk needed to be flush to the interior mechanism surface. 

• 

• 

Tweaking the Interior Mechanism

The CAD Assembly

• 

• 

• 

• 

Bill of Materials

QTY	Part Number	Description	Cost / unit	Cost
1	90272A242	10-24 x ½” Zinc-Plated Steel Pan Head Phillips Screw	$0.05 ea	$0.05
2	6432K12	Set Screw Shaft Collar, ¼”	$1.25 ea	$2.50
4	90272A148	6-32 x ½” Zinc-Plated Steel Pan Head Phillips Screw	$0.03 ea	$0.12
4	90480A007	6-32 Low-Strength Steel Hex Nut	$0.02 ea	$0.08
1	91259A537	¼” Alloy Steel Shoulder Screw, ½” length	$1.25 ea	$1.25
1	90480A011	10-24 Low-Strength Steel Hex Nut	$0.02 ea	$0.02
1	9654K515	Steel Extension Spring	$1.05 ea	$1.05
2		12 x 12 in Acrylic Plastic, Fluorescent Red	$3.50 ea	$7.00
2	3D Print	Disk, Red Generic PLA (5g)	$0.39	$0.78
1	3D Print	Lever, White Tough PLA (13g)	$0.93	$0.93
1	3D Print	Interior Mechanism, Black Tough PLA (72g)	$5.23	$5.23
			Total	$19.01

The Final Product

• 

• 

• 

• 

Analysis

Overall, my dispenser does function as intended, as it fulfills each of the requirements as outlined in the assignment guidelines. As there were a lot of tweaks and improvements made in each prototyping stage, the final product has evolved greatly since the initial concept sketches: 

My initial design had the rotational disk situated next to the storage compartment and the actuation was intended to align the openings in the disk and the compartment wall, so that the candies would just pass through. However, I realized that this method made it difficult to dispense one candy at a time—if the lever was held down, numerous candies would tumble out. I thus edited my design so that the resting state of the dispenser would separate the to-be-dispensed candy from the compartment, so that the actuation would just release this already separated piece of candy. I specifically designed the disk to achieve this by creating a groove that would perfectly fit one piece of candy, and situating the disk directly beneath the opening of the compartment so that the next candy would be ready to enter the groove of the disk, but only after the dispenser is reset. I also changed the reset mechanism by connecting the disk to the top of the box via an expansion spring, as opposed to using it to connect the disk to the lever (for which I now use a static wire).

I found that this method of approach was extremely effective, as my functional prototype was a just that—functional. However, I was not fully satisfied with my prototype, as there were issues with input (the lever needed to be actuated from inside the dispenser) and spring mounting (the way that I had attached it to the dispenser had slightly bent it). I solved my input problem by creating an arced slot in the wall to which the lever arm was adjacent, and solved my spring mounting problem by using extrusions and screws to secure the ends of the spring. With a few minor adjustments to my CAD models to allow for 3D printing shrinkages and tolerances, my model was ready to assemble. 

In terms of next steps, I found one concern with my reset mechanism. Although the reset functions perfectly, I wish that the lever arm would be brought straight up to a vertical orientation when the dispenser resets. Even though this is a purely aesthetic concern—and does not impact the functionality—I believe that editing the point at which the disk connects to the static wire would solve this issue, as the wire just needs to be more taut. As the dispenser functions perfectly, however, I do not have any unanswered questions.

Reflection

Throughout the process of creating my candy dispenser, I have learned so much about my design process. I realized that, as pictured below, I am a person who likes to have his functional prototype very closely resemble the final product, because I find solace in knowing that every aspect of my design is fully functional before I even start 3D modeling it. Although it was a useful tactic in this context, I wonder if I might fare better if I spend less time on my functional prototype—only creating the mechanism and not worrying so much about the aesthetics in the prototyping process. I also realized that since I am a bit of an indecisive designer, I took a lot of decisive design action. For example, once I had an idea that I believed will work, I chose to play it safer with a simpler design rather than designing something complicated and worrying if it would even function. However, this might have led to me not even considering more creative alternatives. In the future, I hope to take more risks with my designs—i.e. using a launch mechanism instead of a passive dispensing action. Had I taken more risks and spent a little more time brainstorming my ideas and options, my mechanism might have been a little more advanced. Ultimately, although I do believe that I played it a little safer with my design, it made me more confident in 3D printing and laser-cutting my pieces, as I strongly believe that I would be hesitant to print anything if I even had a shred of doubt that it would function properly. However, this did not mean that I did not face any challenges in my process. A large obstacle I ran into was designing my CAD models with the proper tolerances to account for shrinkage from the 3D printer. Numerous times, since I had calculated every aspect of my mechanism perfectly, I ended up having to file down the through-holes in order to fit the proper hardware. Another challenge was definitely the assignment restriction against adhesives, as I had to place tabs and fasteners in creative places, use collars and arced slots, and also account for perfect/press-fits. Moving forward, I have a better sense of who I am as a designer—playing it safe never hurts, as the ultimate goal is to achieve functionality. However, I can sometimes cave into my inner overachiever and take some design risks now and again, even if it might make life a bit more interesting.