Engineering simulation
At Ansys Canada in winter 2020, I performed crucial validation and testing work on the multiphysics System Coupling engine, which enables the exchange of information between different physics solvers. However, the most comprehensive project I worked on was in a realm of physics I had no experience or classes in - electromagnetics. I developed recommendations for the multiphysics simulation of induction heating by coupling Ansys Maxwell (the flagship electromagnetics solver) with Ansys Mechanical (the flagship finite element solver.) My work term report on this project was awarded an Outstanding rating for clear technical presentation:
With my recommendations, we were able to obtain an accuracy of approximately 15% relative to experimental data obtained at LANL.
I became proficient enough in numerical electromagnetics that just a couple of months later, I was able to pass on what I had learned to my engineering capstone design team.
I became proficient enough in numerical electromagnetics that just a couple of months later, I was able to pass on what I had learned to my engineering capstone design team.
WatFly student design team
The WatFly student design team's goal is to create a personal flying vehicle for the Boeing GoFly competition. The design is propelled by twin ducted propellers driven by electric motors, and carries a single person. I was primarily involved with aerodynamics engineering, where I conducted propulsion testing, analysis, and composites fabrication.
The most significant project I conducted for WatFly was an aircraft propeller aeroacoustic analysis algorithm. The algorithm predicts the aerodynamic noise emitted by the engines. I developed an innovative method that combined the semiempirical model by Brooks, Pope, and Marcolini with ANSYS Fluent numerical analysis. In this manner, computation time was kept at a minimum, while the results provided noise breakdowns by mechanism and spatial distribution. This enabled directed noise mitigation efforts by highlighting areas of interest. The algorithm was verified to be accurate to ±10 dB experimentally.
Post-processing of the results included signal convolution to reflect human perception of the noise and working with frequency spectra.
This project improved my skills in rotor aerodynamics, ANSYS CFD, aeroacoustics, signal processing, and experimental design. Before beginning the project, I had no knowledge of aeroacoustics but was able to learn the requisite knowledge within the space of two months.
In March 2019, I presented my work to judges from ANSYS as part of WatFly's entry into the ANSYS AIM analysis competition, where we placed second and won a prize of $1500.
The most significant project I conducted for WatFly was an aircraft propeller aeroacoustic analysis algorithm. The algorithm predicts the aerodynamic noise emitted by the engines. I developed an innovative method that combined the semiempirical model by Brooks, Pope, and Marcolini with ANSYS Fluent numerical analysis. In this manner, computation time was kept at a minimum, while the results provided noise breakdowns by mechanism and spatial distribution. This enabled directed noise mitigation efforts by highlighting areas of interest. The algorithm was verified to be accurate to ±10 dB experimentally.
Post-processing of the results included signal convolution to reflect human perception of the noise and working with frequency spectra.
This project improved my skills in rotor aerodynamics, ANSYS CFD, aeroacoustics, signal processing, and experimental design. Before beginning the project, I had no knowledge of aeroacoustics but was able to learn the requisite knowledge within the space of two months.
In March 2019, I presented my work to judges from ANSYS as part of WatFly's entry into the ANSYS AIM analysis competition, where we placed second and won a prize of $1500.
Test apparatus built to validate propeller aeroacoustic model. Equipment provided by WatFly design team.
|
Aeroacoustic simulation of aircraft engine conducted in ANSYS Fluent using broadband noise model.
|
Below is the update presentation prepared for the team. It provides a brief introduction to the subject matter and documents the validation of the semiempirical aeroacoustic prediction method.
The culmination of this work is my second work term report, which documents the entire process and methodology of the analysis.
Due to the high rotational speeds of an aircraft propeller (~5000 RPM), even small mass imbalances can cause serious vibration problems. For this reason, I proposed and implemented adding balancing holes to the propeller hub design, allowing rotor balancing. As we did not have access to aircraft balancing weights, this was designed to use car tire balancing weights. This was inspired by the technique used to balance turbofan rotors, which I observed during my co-op term at Pratt and Whitney Canada.
Proposed and implemented balancing holes and stress-relief fillets in propeller rotor hub (see rectangular holes along circumference) to minimize vibration and fatigue propagation.
|
The final product (one of the blades detached). This was used as a mold core for the propeller carbon-fiber pre-preg layup.
|
Robotics and electronics projects
This system allows a user to play checkers by entering the appropriate moves, then the robot uses a magnet to move the checkers pieces around the board. The mechanism was built using LEGO NXT and programmed using RobotC. I was responsible for designing the board, piece-moving mechanisms, and writing the motor movement code.
|
|
|
Demonstration videos for robotic checkers board.
This project was completed at the MakeUofT hackathon, where our team decided to construct a scale model of an automobile-refueling robot. I was in charge of designing and programming the motor control electronics, for which I used a Raspberry Pi and stepper motors. The Raspberry Pi was able to receive requests over the internet from a smartphone to operate. More details on hackster.io here.
Wiring for robotic automobile refueling system, using a Raspberry Pi 3 and stepper motors. Created for MakeUofT 2019.
|
Our MakeUofT 2019 team on presentation day, on two hours of sleep. From left to right: Mohammad Mustafa Sajjad, Shawn Paul Michael, Jin Sing Sia.
|
Software projects
Numerical orbit solver: github.com/jssia/orbits
This code solves numerically for the orbit of a satellite orbiting a celestial body using Newton's law of gravity and creates a CSV file containing all output.
This code solves numerically for the orbit of a satellite orbiting a celestial body using Newton's law of gravity and creates a CSV file containing all output.
An example output, plotted in matplotlib.
WatSub student design team
The WatSub student design team aimed to create a human-powered 'wet submarine' (interior filled with water, pilot breathes using scuba gear) for the International Submarine Races (ISR) in Bethesda, Maryland. I was the team's propulsion lead designer for the 14th submarine races in spring 2017. My drive train design weighed 40% less than the previous year's design, and featured counter-rotating propellers to increase efficiency and to minimize rolling moments. The design was inspired by the drivetrain design for aircraft with counter-rotating propellers.
The drivetrain and propeller blades were made of aluminum, while the propeller hub was 3D-printed nylon.
The drivetrain and propeller blades were made of aluminum, while the propeller hub was 3D-printed nylon.
Drivetrain and propeller hub design, propeller blades and chain to pedals not included.
Drivetrain design.
1. Aft driven bevel gear.
2. Driving bevel gear.
3. Forward driven bevel gear.
4. Top cover.
5. Bearing mount.
6. Bearing cover.
7. Connector shaft to drivechain.
8. Forward gear shaft.
9, 10. Mounting bars.
11. Angle brackets.
12. Mounting plate.
13. Inner shaft.
14. Outer shaft.
Design of propeller hub.
1) Endcap.
2) Propeller blade mount.
3) Midsection.
4) Inner shaft mount.
5) Front section.
6) Outer shaft mount.
7) Hull adapter.
Rapid prototyping projects
The following are projects conducted during first-year, during which I learned rapid prototyping and SolidWorks rendering.
3D-printed phone stand, designed in SolidWorks for aesthetics and minimal support material usage.
Laser-cut plywood parts for a robotic checkers board. Used slots reinforced by glue for maximum joint strength.
|
The final product.
|
Mancor Industries
As part of a Lean initiative at Mancor Industries, I designed and directed assembly of a structure to hold automobile part paint racks. The work term report below demonstrates my skill with engineering analysis and technical report writing.
Blender artwork
The following are space-related renderings created with Blender, an open-source 3D software.
An Orion spacecraft in Jupiter orbit. Project Orion was a concept for a spacecraft propelled by nuclear detonations, which was developed by physicists Ted Taylor and Freeman Dyson from 1957 to 1964. These, in different history, could have taken humanity to Saturn by 1970. Background image credit: NASA.
An Orion spacecraft approaching Titan, a moon of Saturn. Background image credit: NASA.
|
An Orion spacecraft near Enceladus, a moon of Saturn. Background image credit: NASA.
|
The Valkyrie spacecraft, a concept by Charles Pellegrino. Matter and antimatter are annihilated, and the resulting charged particles are deflected backwards by a large, toroidal magnet. The glowing vapor in front of the spacecraft is a cloud of droplets sprayed into space to dissipate heat and to act as a dust shield. Background image credit: NASA.
