Engineering Projects
Democratizing In-Plane Thermal Conductivity - Capstone Design Project
Problem:
There does not yet exist an economic way to measure in-plane thermal conductivity, a key measurement in developing novel materials for thermal management. Our team was tasked with creating a scientific device to measure in-plane thermal conductivity using the Ångström method, using strong design principles such that the design could become standard in labs around the world.
CAD Model + Subsystem Breakdown
Front View of Sample in Device
Device as Presented on Capstone Day
Action:
Over the course of 5 months our team of 5 solved numerous engineering challenges to create our final product. While I, as well as the rest of the team, was involved in discussing nearly every decision made, my primary responsibilities were leading software development, designing the Thermal Feedthrough, running subsystem and full-scale tests, and performing project management tasks. More info on my software work can be found in Programming.
Thermal Feedthrough, Connected and Exploded
Thermal Feedthrough:
The device was required to be placed inside a vacuum chamber to reduce convection, theoretically leading to more accurate results by aligning more with the 1D heat equation. With this feature, there needs to be a way to dissipate the heat built up inside the chamber, leading to requiring a thermal feedthrough. There were several key design considerations at play here:
The block must be able to be machined out of copper for high thermal conductivity
This lead to the design splitting into three parts: the cylinder, and two rectangular blocks, for conducting heat against the clamps and dispersing it once outside. The blocks would be connected by screws, as they would not come apart once placed in the chamber, outside of special maintenance.
The clamps must be able to connect to the passthrough at any height to accommodate for any size thickness sample. Additionally, the frame of the device must be able to detach from the passthrough for easy maintenance.
From this, magnets were used to connect the passthrough to the clamps. This allowed for the clamps to still connect to the passthrough for thermal conduction, but allowed the clamps to connnect to the passthrough at any height, and easy be taken away from the passthrough entirely. The magnet was embedded in the rectangular indent shown on the front plate.
Testing and Validation:
I ran tests that validated critical subsystem components as well as proved the viability of the device as a whole. This included:
Testing the integrity of our vacuum chamber before and after any modifications were made. A 3 day test revealed the pre-modified chamber to not leak at all, while our modifications led to a leak measured at 8.5 kPa/s. Not a critical error, but indicative of more work to be done.
Validating Thermocouples to give accurate readings w.r.t. each other. By placing each TC into different temperature water baths, it was validated that no TC was giving innacurate readings outside of the expected error margins. This test also validated the time-lag between TC readings by the DAQ, which was accounted for in software.
Overall system testing, measuring the in-plane thermal conductivity of aluminum in a 4 Hour Test, and Zinc in an overnight 12 hour test.
Thermocouple Validation Testing, including the DAQ in blue, and the two water baths used
Snippet of the project Gantt Chart at the midway point
Project Management:
Using knowledge gained from Northeastern's Engineering Project Management course, I helped strategize and manage our team to accomplish all of our goals. This involved:
Creating a detailed Gantt Chart and updating it as progress was made
Setting regular meeting times, keeping an agenda, and taking meeting notes
Ensuring each team member had something to work on following each meeting, and that workload was shared equally
Results:
Due to the breakneck pace of capstone, we were only able to test the device twice, and without vacuum pressure. Regardless, our first test measured the thermal conductivity of aluminum at an 8.05% error, and our second test measured zinc at 3.23% error. By using a vacuum, running for a longer timescale, and understanding the best parameters for the Ångström method, higher accuracy with the device is expected. A research paper regarding this is currently being written, with submission for publication expected in May of 2024.
Our team won 1st place by unanimous vote of a panel of experts and alumni judges, beating out 7 other highly competitive teams.
Lessons Learned:
Independently running an engineering project
Setting regular deliverables and milestones, and ensuring both myself and my teammates would make the deadlines
Reviewing design plans with our client and experts in electronics and machining to ensure feasibility
Working with a vacuum chamber, thermocouples, proprietry DAQs, Thermal FEA, and more specific lessons
Press Optimization - Vaxess Technologies
Problem:
In the production line of Vaxess Technologies, backings were pressed onto molds to extract the product using an off-the-shelf Arbor Press. This method was not standardized, required human effort, and was inefficient.
Action:
The Original Method was first measured using a Force Plate and Pressure Sensitive Film to establish a baseline of how force was being applied.
Experiments were then carried out over 100+ molds to determine the minimum and maximum force to press with to achieve successful manufacturing yields. This also took into account the varying printing quality sometimes present on the manufacturing line.
Improved Arbor Press Designs were then designed and created. A force wrench attachment for the arbor press was created to finely control the force applied for experimental use as described above.
Different faces were then produced for the Arbor Press, allowing for multiple molds to be pressed at once. This underwent multiple design iterations and testing to ensure all molds experienced equal force.
A pneumatic cylinder was swapped out for the arbor press, and after designing fixtures to allow it to work and undergoing testing, it was determined to be a superior alternative to the press in providing consistent force efficiently without human effort. Further expansion only concepted before my co-op ended.
Results:
This project successfully increased the production line's efficiency and autonomy. Removing the human element from this step allowed the team to run more consistent experiments on other parameters in the production. And the experiments I ran gave useful insight into what parameters would need to be set for a larger overhaul of the production line to become more fully autonomous.
Lessons Learned:
Running Independent Experiments
Designing for R&D vs Designing for Regular, Efficient Use
Interfacing with Pressure Sensitive Film, Force Plates, Pneumatic Cylinders
Presenting my experiment's findings succinctly to the Engineering Team in rapidly made prototypes
Original Arbor Press Model and In-Use on Production Line
Establishing Baseline using Force Plate and Pressure Sensitive Film, Showing Results and Methodology
Torque Wrench Attachment and Example Results from Force Bounds Testing
Evolution of the Extended Face for the Press, along with their associated force distribution patterns
Temporary Pneumatic Cylinder Housing, And First Concept of Permanent Housing
Inspection Fixture - Vaxess Technologies
Design Progression of Inspection Fixture
Problem:
Inspecting a micro-needle patch at the end of a production run is done under a microscope, and needs to be held at certain angles for extended periods of time to count imperfections on translucent needles.
Action:
As one of my first projects on my co-op at Vaxess, I took it upon myself to improve the existing inspection fixture, shown above on the far left, in as many ways as I could. Through my design process, I used my own technician experience and interviewed my coworkers to see what needed to change. I then rapidly modeled and 3d printed different prototypes, seeing how my solutions fared and what problems I had inadvertently created.
Results:
The newest inspection fixture was machined by a third-party company and is used by all departments at Vaxess whenever patch inspection is required. Having three viewing angles one can easily tilt to with comfortable hand movement is incredibly useful for the imaging required of the translucent micro-needles.
Lessons Learned:
Rapid Prototyping of New Ideas and Solutions
Designing products for human use, understanding the design loop of incorporating feedback
Often times, solving one problem can create new unexpected problems
Rapid 3D Printing techniques, printer maintenance and upkeep