My Senior Independent Project is creating the housing for the vibrotactile haptic feedback device replica created by Mr. Callahan and Mr. Gammel. The goal is for it to be durable, ergonomic for long wear periods, accessible to put on and taken off given the constraints of the individual, and easily replicable. For the initial prototype,  Ms. Knowlton will be the subject that I am designing the housing to fit.  An extended goal, if time allows, is that I begin the research to make this entire device wearable, meaning that the user is not tethered to a board or a computer, as the current model requires a constant connection to a driver board and a PC for two-hour increments twice a day. Reimagining this device as a wearable technology makes it more easily replicable for larger groups of people while maintaining a low production cost. 

Week 0

Before the Senior independent project began, I met with Lori Knowlton, and  Phillip Callahan via Zoom and in person to discuss the details and necessities for the project. I was able to obtain their current design, which is featured below with the tactor driver and two sets of tactors. I analyzed their issues: fragile wire connections, bulky design, and difficulty to put on and wear alone. This let me begin background research on similar devices.

Week 1

This week I continued my literary research, ordered materials, began experimenting with materials, and scheduled a time to meet with Lori. I was able to order a molding kit so that I could use a mold reference of her hand during the design process, which would not require her to come into the PIRL as often. Lori is coming to the PIRL Wednesday, April 12th for the molding process and I am planning to discuss her comfort with different materials (i.e. fabrics, zippers, straps), and to take measurements of her hands to begin prototyping. This week I also came up with the first design plan. I am not strong with drawing diagrams, but I did create a form of a mood board with different design plans and ideas featured below. The left image is the hand-casting kit we ordered. The right image is my mood board. Through literary research, I found a stroke treatment glove that uses fluid actuation (airflow) for treatment. I like the use of tubing but instead would like to adapt it for the wire housing. This should ensure the wires are protected and do not break--an issue in the original design.  We went ahead and ordered accordion-style straws I hope to use. I am also hoping to use TPU filament for the skeletal structure of the glove. TPU is a 3D printing filament that is very flexible. When Lori comes in this week, I plan to measure her hands so that I can order the two styles of nylon gloves that fit her hands. Once the two models come in, I am interested to see if the velcro straps or the zippers are easier for her to use by herself.

Week 2

This week I made great progress on my first prototype. On Wednesday the 12th, Lori came into the PIRL and I was able to show her and Philip Calahan (he attended the meeting via Zoom) my initial design plans. We were able to discuss the strengths of this design, and they offered feedback. Philip was able to try on the glove that Stanford designed a few weeks ago and reported to us the strengths and weaknesses of the design. After the meeting concluded, I started measuring and molding Lori's right hand. Using the Hand Casting Kit, I created the mold for her hand and filled it with plaster. After the plaster hardened, I removed the mold. Unfortunately, none of her four fingers were attached after removing the mold. I was left with a creepy fingerless mold of her hand. Unfortunately, I was unable to do much work on Thursday and Friday due to the ISAS Festival, but I began work on my 3D Printing CAD file. This Monday, I came into the PIRL and began reassembling the fingers to the plaster hand. I did not have any luck using plaster (the plaster just began disintegrating) so I used Epoxy. It took a while for the Epoxy to set, but the fingers became reattached and looked great. Tuesday, April 18th, I took more measurements of the casted hand and messed with the materials ideating my design. I am currently designing two CAD files I hope to print by Thursday. One is the skeletal structure that runs along the top of her hand that will hold the wire casings secure. The second is the tactor holder that resembles the one Philip created using hard plastic but is modified to fit Lori's feedback. Both prints will be made using flexible TPU filament. This week was super beneficial for the first prototype, which I hope to have created or almost done by next week.  

Week 3

This week, I started 3D printing and decided to use TPU as my filament, even though I wasn't familiar with it. TPU, thermoplastic polyurethane, is a flexible filament used in 3D printing that can be bent without losing its shape. I was excited to test it out, especially since I wanted to create a flexible structure for the top of the glove to enclose the hardware. However, I soon realized that TPU is difficult to print compared to other filaments I'm familiar with due to its unique properties. It requires special temperatures and speeds, and it clogs easier. I learned this the hard way. My first print finished, but the purge tower was misplaced, and there were a lot of stringy pieces. I have included an image of the prints below. On my second print attempt, the filament jammed only two hours into the print. However, I was able to make adjustments and fix the issue, and I am now waiting for the results of my third test print. The purge tower serves as a spot for the printer to remove any excess or leftover filament from the nozzle between layers of printing the actual model. Despite the position it was set at, the purge tower was in the way again, so I removed the tower for the third attempt. I am excited to see how this works out. 

These prints are for the exoskeleton that will be on top of the glove holding the hardware intact. The flexibility should help with mobility, but retain the rigid structure. For the first design, I wanted to maximize my options. Each finger piece adopts a different design. There is variability in the lengths, thicknesses, distances between the loops, size of the loops, and  end pieces, etc. Once this print becomes successful I hope to test all of the different fingers and decide which one is the most efficient. 

I am also currently working on the electronics aspect of this project. I am meeting with Mr. Lanier this week to receive his feedback. My personal goal for the electronics is to make the system less bulky and more efficient. Currently, the device is connected to a large piece of wood with the drivers and all the electronics, which are connected to a PC. I hope to make this part fully untethered to a PC. I have already researched some Arduino products I hope to use for this design.

After I decided on the base design for all of the fingers, I began designing an addition for the tactor to the finger. In my design below, you can see that I added a second, larger loop that rests on Lori's fingertips perfectly. The straw that will hold the wires will run on top of her finger, and then curve to connect to the tactor on her fingertip. After printing this I confirmed that I liked the base model for all of the fingers, but I did not like the tactor design. 

My third design was printed today and the support materials are currently dissolving in the PIRL. I made the bottom piece larger so that it can hold the tactor snugly and will protect the wires. The circular vibration pad should fit in the notch I created on the design. I only created her pinky finger because it is the shortest print time, and allows for me to move forward with designing.

Mr. Lanier and I began discussing some options for wearable technology and have come up with a few ideas. Our current favorite option is using an Adafruit Audio FX Sound Board. I have connected the board to the tactors and attached a photo of it connected on the far right to a breadboard. I  created a test sound file and recorded the vibrations; the far-right video shows it vibrating (trust me it's vibrating even though you cannot really tell). One of the sets seen in the middle image acts as the whole setup for two fingers and is very small. Four of those sets should replace the large, bulky, and tethered design on the far left. The WAV file developed by Dr. Gammel will be downloaded on it, and I have hooked it up to a rechargeable battery (shiny silver piece) so that it is fully untethered. This way, when Lori is done using it, she can plug it into her computer and it will be charged for the next session. However, there is a 16MB maximum limit for the files. The current files Dr. Gammel created are about 60 MG ,which is huge. This board is a 2 Watt amplifier, but we have a 20 Watt amp version if needed (although it's a larger design). I hope to work with him on both of these things to see what his opinions and thoughts are. On the far left you can see the battery, the soundboard, and the tactor. This design is super small (the board is  1.9" x 0.85"). This week I am hoping to create a few prototypes and designs to show Lori on Friday morning, and plan to meet via Zoom with the whole team to discuss the electronics.  As advice for next year's students, I would recommend ordering parts ahead of time and getting prototypes done quickly so that you have more time to fine-tune small things. I also am really appreciative that I have taken so many photos along the way even of stuff that failed or was not my best work.