A Compact Piezoelectric Tilt Table for Cryo-EM
Summary: Despite the promise of cryogenic electron microscopy, the field has been hampered by the lack of effective mechanisms to correctly orient a specimen at such low temperatures, while being compact enough to fit within a transmission electron microscope (TEM). Current options either sacrifice resolution for mobility, vice versa, or are just too large. This is mainly due to the use of externally coupled tilting stages or cryogenic cartridges, which are subject to mechanical vibrations and thermal drift. The goal of this project is to develop an isolated, internal, specimen tilting stage for cryogenic electron microscopes, which could enable researchers to more accurately orient quantum materials and gain clearer insights into their properties and structure. This project was my independent capstone project within the Rowland Institute at Harvard, as part of my B.S. degree.
The Process: By using piezoelectric chips, I designed compact motors that are compatible with cryogenic temperatures and high magnetic fields, as well as being precise and stable with no backlash. The first stage of this project was design and material selection. To work with cryogenic temperatures and strong magnetic fields as well as machinability with my available tools, I selected stainless steel. The prototypes were machined using a laser cutter and 3D printers , as it is quick and precise, but the final product was machined with a wire EDM mill.
The designs use a slip-stick mechanism in which a sawtooth signal gradually deforms the piezoelectric element on the ramp, rotating the axle using friction. On the ledge of the sawtooth signal, the piezoelectric element will quickly revert to its initial state, but the axle will remain rotated due to its mass inertia. This process will repeat quickly and the many individual small displacements will sum to a large rotation.
Skills/Tools: Piezoelectric actuators, electronics, material selection, laser cutter, wire EDM milling, 3D printing
Click to Read
Automatic TourniQuet: A Rapid-Deployable Pneumatic Tourniquet for Pre-Hospital Emergency Care
Summary: Traumatic hemorrhage accounts for 1.5 million deaths per year worldwide. Emergency tourniquets are widely considered the standard-of-care for controlling extremity hemorrhage in trauma scenarios. However, the commercial tourniquet – a manual windlass system – falls short in consistency and reliability due to a high potential for user error when tightening the strap. The goal of this project was to develop a self-tightening tourniquet that could rapidly deploy as well as more consistently apply the tourniquet with minimal error. To achieve this, we designed an inflatable tourniquet actuated by a one-button puncture of a pressurized CO2 cartridge. This work was done as a group project for the Harvard course ES 227: Medical Device Design. In this class, each group worked on a design project proposed and advised by working doctors and engineers in the medical device field. While all team members contributed to all parts of the project, my focus was on the manufacture of the tourniquet as well as maintaining a consistent air-seal on the device.
The Process: The design process had a few stages: (1) understanding the problem, (2) developing design ideas, and (3) prototyping and testing. For the first stage, we developed our priorities and identified the key challenges. For the tourniquet to be effective, it needed to be able to do two things: (1) apply the correct occlusion pressure every deployment, and (2) maintain that pressure until the patient got to a hospital. However, it also had to be simple and comfortable enough to use such that an untrained person could deploy it, as well as have a lightweight and compact form factor. Considering these priorities, we identified the challenge areas to be maintaining a constant pressure and being easy to apply by untrained users.
For stage two, we identified multiple potential mechanisms for the tourniquet, drawing inspiration from similar devices in use, and arrived at two final designs: a compressed CO2 actuated tourniquet and a ratchet-spiral spring actuated mechanism. Through modeling with Matlab and rough prototypes with laser cutters and 3D printers, we decided on the compressed CO2 design. For stage three, we prototyped tourniquets iterating on the air seal as well as the pressure release and modulation and experimenting with different puncture mechanisms and pressure valves. We also manufactured the device more robustly, using machined metal components rather than plastic 3D-printed ones.
The Result: Through testing the tourniquet on a model arm with blood vessels, the tourniquet successfully occluded the blood vessels and maintained that pressure for an extended period of time. Additionally, we conducted user testing with trained EMTs, inexperienced users who were trained on the tourniquet, as well as inexperienced users who were not trained. Across all three groups, this tourniquet outperformed the standard windlass tourniquet used today in terms of ease of use as well as how intimidating it is to use. Based on this testing, we determined that our tourniquet could be a suitable replacement for currently used tourniquets.
Skills/Tools: Matlab, Solidworks, 3D-printers, laser cutters, manual milling, metal lathes, silicone casting, pneumatics.
Wood Joint Extension and Compression Load Testing & Simulation
Summary: Wood joinery is used everywhere, from houses to furniture, and structural stability is crucial. In this project, we investigated the mechanical properties of different joinery with two common woods, poplar and pine, when subjected to various loading conditions. We tested the materials under compression and tension, and we tested four different types of wood joints: lap joints, mortise and tenon joints, dovetail joints, and bridle joints. Our goal was to determine the strengths and weaknesses of these wood materials and joints, and to provide valuable information for woodworkers and engineers who are designing structures or products that involve wood. This work was done as a group project for the Harvard course ES 192: Materials Selection and Design.
The Process: Using band saws, hand saws, a router table, and wood glue, we built identical wood joints (lap, dovetail, bridle, mortise and tenon) out of both woods (pine and poplar). All joints were tested in compression and tension using an Instron machine. All joints were also computationally simulated using Comsol.
The Result: From material testing with the Instron as well as the Comsol simulations, we found that poplar lap and bridle joints were the strongest among the tested joinery geometries.
Skills/Tools: Band saws, hand saws, router table, material characterization, Comsol
WormBots - Earthworm Inspired Robots
Summary: As natural disasters and building collapses remain an ever present and growing danger, search and rescue operations are critical. However, they present potentially fatal risks to those first responders. The goal of this project was to design a robot that would be able to maneuver through tight and awkward spaces to find trapped survivors. This robot was designed to move like an earthworm. This was an independent project, completed using household tools and materials.
The Process: This research was done at home using inexpensive materials from Home Depot and Amazon as well as standard woodshop tools. The first iteration of the WormBot utilized three parallel inflatable actuators and extendable claws controlled by an Arduino and compressed air to move. The actuators were made using latex surgical tubing and cut up flexible vacuum hose. Using these actuators and claws, the robot was able to crawl like an earthworm.
In the second iteration of the WormBot, I replaced the inflatable actuators with servomotors, flexing elbow joints, and elliptical claws controlled by an Arduino. Using these components, the robot was not only able to crawl like an earthworm, but also to roll horizontally. This second robot had the capability to connect to and communicate with another identical robot to form a longer robot capable of more complex motions or potentially someday, a swarm of robots working together.
The Result: Through testing the robots, I determined that both versions were able to crawl across both smooth and rough surfaces. Furthermore, the second iteration of the robot was able to roll across both smooth and rough surfaces as well as connect to and communicate with other identical robots.
Skills/Tools: Electronics, Arduino electronics, woodshop tools, robotics, pneumatics
Desalinating Water Using Electric Fields
Summary: As many parts of the world struggle with freshwater shortages, effective methods of desalination are crucial. However, many current techniques such as reverse osmosis are expensive, complicated, and energy consuming. In this project, I designed and tested capacitive desalination apparatuses that use electric fields to attract away the positively charged sodium ions and the negatively charged chlorine ions more efficiently than current methods. This was an independent project completed using household tools and materials.
The Process: This research was conducted at home using inexpensive materials from Home Depot and Amazon as well as standard household tools. There were three designs for this method. The first design used a rotating armature with three porous carbon electrodes. The electric fields attracted the ions to the carbon rods which absorb the salt. The rotation of the armature allowed the system to quickly alternate between desalinating the water and flushing clogged electrodes.
The second design used two copper plate electrodes and an alternating on-and-off electric field to attract the ions near the electrodes without them sticking to and coating them. This eliminated the need to flush the electrodes, saving time, money, and energy. The third design used a rotating coil of tubing and an external electric field. As the coil rotated, the external electric field dragged the ions down to one side of the tube leaving the water at the other end desalinated. All three designs effectively lowered the salinity of a saltwater solution.
The Result: From testing the prototypes on a prepared saltwater solution, I found that all designs successfully desalinated the feed water by between 5-20% using less energy than currently used methods.
Skills/Tools: Electronics, soldering, hand tools
Lucy Steam Engine
Summary: This is a one-cylinder oscillating steam engine made using standard metal machine shop tools. It was made using aluminum, brass, steel, and acrylic. This was done as my individual machining project to become fully certified at Harvard’s student machine shops.
The Process: I used a bandsaw to rough cut aluminum, brass, steel, and acrylic stock to machinable pieces. I then started on the milled components. Using a manual mill, I squared the parts and machined them down to the correct size. Following that, I drilled and threaded appropriate holes and contoured the parts according to technical drawings provided. Following that, I moved onto components made on the lathe. Based on the technical drawings, I turned the components down to the correct geometry. All parts of the engine were machined within a tolerance of 0.005” and then fully assembled.
The Result: Using compressed air, I ran the engine and it worked immediately, showing that all components were machined and assembled correctly with tight tolerances.
Skills/Tools: Machining, reading technical drawings, manual milling, metal lathe, bandsaw
Puzzle Box
Summary: This is a set of wooden puzzle boxes with a built-in maze. To open the box, a pin must be moved through the maze. These puzzle boxes were stepping stones to make a Japanese puzzle box. This was done as a personal project.
The Process: The first step was to design the components on Solidworks. The box would be made using laser cut sheets, so it was important to model how 2D parts would interact and connect for a 3D product. To do this, I used a Solidworks assembly to model how the parts would connect and how they would move within the locking mechanism. The components were then laser cut from birch plywood and glued together.
The Result: All laser-cut boxes worked as intended. The maze mechanism moved smoothly and effectively locked the boxes and only unlocked if the maze was solved correctly.
Skills/Tools: Laser cutter, Solidworks
Brass Hammer
Summary: This is a brass machinist hammer that was made during a metal lathe workshop.
The Process: The first step was to make a 2D sketch for the hammer's pattern. Using this sketch, I marked the brass shaft before profiling it. After turning the shaft down based on my sketch, I used a series of sandpapers with increasing grits to polish it. I then parted the head from the handle, tapped a hole in the head, threaded the connecting rod in the handle, and then assembled.
Skills: Manual metal lathe, metal polishing
