MENG 472/474 Projects - 2017

In MENG 472/474, students work on independent projects that cover a wide range of topics, from traditional mechanical engineering topics (e.g., mechanical device design, fluid flow, and materials analysis) to interdisciplinary topics at the interface between mechanical engineering and other branches of engineering such as biomedical, chemical, electrical, or environmental engineering. Under the supervision of faculty advisers, students investigate physical phenomena through experimental measurement and/or numerical simulation, and they design and construct functioning prototypes to solve engineering problems. The majority of the faculty advisers come from within the mechanical engineering department, with the remaining advisers distributed across the University (and occasionally outside the University). Funding for projects is generously provided by the Yale SEAS Dean's Office and, in some cases, through the faculty advisers. The students were asked to write the following short summaries two-thirds of the way through the semester, when they still had a few weeks to go on their projects. All projects are represented here, except for those that cannot be publicized due to information of a proprietary nature.

App Development

Developing an app to combat the normative social network

Lance Chantiles-Wertz
Advisers: Prof. Kyle Jensen, School of Management, and Prof. Corey O’Hern, Mechanical Engineering and Materials Science

Current social media platforms, such as Facebook, Instagram, and Snapchat, create a constant bombardment of what our friends are doing, but do not alert us to these events until after they have passed. This is labeled as “fear of missing out”, or FOMO for short. Waiv turns the tide on this condition by allowing people to have filterable real-time access to everything going on around them. Waiv identifies and organizes events happening around a user through an integrated mobile map application. At its core, Waiv employs patent-pending user tracking and geofencing technology to allow a user to interact with the information surrounding him or her. We are building Waiv on the Ionic platform which uses one technology stack to be exportable to both iOS and Android platforms. We are implementing a system of implied design which tests the limits of how much a user can intuitively understand by symbolic representation and how much interaction can be based on multi-gesture and timed responses. Waiv will be launched the first half of April on campus on both the iOS and Android platforms, with support of other Yale developers. The image shows sample screenshots of the current build of the application.

Basic Research

Three-dimensional dithering and its implications

Taylor Rogers
Advisers: Dr. Larry Wilen and Prof. Corey O’Hern, Mechanical Engineering and Materials Science

Our current research builds on Dr. Larry Wilen’s work on acoustic resonance spectroscopy of soft solids, when questions arose regarding the possibility of creating variable material properties in three-dimensionally printed objects. We found that premium printing techniques are capable of varying properties, but there has been little to no research or development in using more accessible software and printers. We aim to research, analyze, and develop a method for utilizing the current ability to continuously vary material properties anywhere in three-dimensional space with just two basic components. Our original hypothesis was that an effective continuum can be achieved using a method analogous to dithering in traditional printing. Dithering is the reprographic technique that simulates a continuous gradient using only dots. We utilize an acoustic resonance apparatus suitable for probing mechanical properties of materials with elastic moduli less than 5 Gigapascals. The method employs stereo phono-needle transducers to weakly perturb samples until normal modes can be identified by received signals of polarization excitation. We successfully applied two-dimensional error-diffusion dithering in three dimensions and have begun to achieve noticeable gradients in mechanical properties, which could have implications on improving design and testing capabilities for a wider population of engineers.

Biomechanics Research

Bilateral jumping deficit: A learned behavior

Dante Archangeli
Adviser: Prof. Madhusudhan Venkadesan, Mechanical Engineering and Materials Science

Athletic performance is highly valued in modern society. Olympic medalists are idolized and professional athletes make millions of dollars each year. In international jumping events, mere centimeters can be the difference between victory and defeat. Yet, as athletes push themselves to break records, the likelihood of injuries increases. A better understanding of jumping can create the inches needed to win or the medical support to bring athletes back into the game. Our research explores the bilateral jumping deficit, the observation that the force and work per limb in a one-footed jump is greater than the force and work per limb in a two-footed jump. Explanations of this deficit focus on reduced neural drive and the force-velocity relationship during muscle contraction. We take a novel approach and explore psychological and other physiological factors that might affect jumping. We hypothesize that muscles inconsistently produce maximum effort. This inconsistency can cause a variation in force across the body midline, causing a jumper to erratically take off or land. To test this hypothesis, we will use an array of strain gauges to measure variation between intent and actual force production during isometric contractions. We hope our research findings will improve athletic training and recovery.

Determining joint mobility based on articular surface geometry

Lucia Korpas
Adviser: Prof. Madhusudhan Venkadesan, Mechanical Engineering and Materials Science

The articular surfaces and soft tissue structures of joints permit movement in specific directions while providing stability and minimizing mobility in others. The exact mechanisms which prescribe these directions are not fully understood. At the Yale Biomechanics and Control Lab, we are applying analytical, computational and experimental approaches to test theories of joint mobility, with a focus on joints of the leg and foot. This semester, this has involved developing a strategy for obtaining and analyzing 3D surface models of bones and their articular surfaces and collecting preliminary data. This work may have a range of implications in evolutionary biology and human health, such as to facilitate the extraction of information on extinct animals’ walking habits from fossils.

Body orientation stability during maximum-impulse two-legged jumping

Alexander Lee
Adviser: Prof. Madhusudhan Venkadesan, Mechanical Engineering and Materials Science

Whenever the body applies a force, there is a difference between the intended and the actual forces. This is the reason why activities such as shooting a three-pointer in basketball or hitting a serve in tennis are so difficult. In the case of two-legged jumping, unintended forces can cause the jumper to rotate in the air, leading to injury upon landing. Preliminary measurements in the Yale Biomechanics and Control Lab indicate that the magnitude of the difference between intended and realized forces during a jump should cause a human jumper to fall for all but the smallest of jumps. The purpose of this project is to validate a hypothesis for why it is that humans do not fall when they jump.

Foot stiffness in diabetes-induced Charcot foot

Lucinda Peng
Adviser: Prof. Madhusudhan Venkadesan, Mechanical Engineering and Materials Science

The aim of this project is to determine if reinforcing the transverse arch of a diabetic foot can stiffen the foot and quantify the stiffness added to the foot. Through uncontrolled inflammation, diabetes often leads to Charcot foot, a condition where weakened soft tissue collapses the midfoot. It is known that reinforcing the transverse arch, along the width of the foot, in healthy populations stiffens the foot, so we will attempt to verify this relationship in diabetic populations by loading the foot on both sides with springs. There are two sets of ligaments that stiffen the transverse arch, the tarsal-metatarsal ligaments, and the metatarsal-phalangeal ligaments. It is not known which ligaments are weakened for each diabetic patient, so we will individually reinforce the ligaments to test how increasing stiffness of different ligaments affects the stiffness of the foot. Stiffness will be measured through bone curvature data from CT imaging. The patients will wear a boot that immobilizes the ankle and raises the toes, as shown in the diagram. We will compare the change in foot curvature between the reinforced and unreinforced scans.

Developing a novel model of human throwing

Petter Wehlin
Adviser: Prof. Madhusudhan Venkadesan, Mechanical Engineering and Materials Science

Human throwing abilities are yet to be fully explained. Based on the accuracy and speed with which skilled individuals throw, one can show that they consistently let go of the object during a launch window of about one millisecond. This launch window is much shorter than what nerve impulses can reliably control. Hence, in the Yale Biomechanics and Control Lab, we hypothesize that some other biomechanical mechanism exists that allows humans to throw objects as accurately as we do. Our research focuses on testing this hypothesis through a physical model of a human arm, as well as through numerical simulations of the dynamics of throwing. Our results will hopefully advance our understanding of human throwing, and they may have potential applications in high-speed throwing robots.


Modeling and analysis of heat rejection in thermoGreenwalls

David Amanfu and Christian White
Advisers: Prof. Alexander Felson, Schools of Architecture and Forestry and Environmental Studies, and Prof. Corey O’Hern, Mechanical Engineering and Materials Science

thermoGreenwalls are non-infrastructural units composed of pre-vegetated panels that serve as sustainable alternatives to cooling towers and provide heat rejection and water and air quality treatment. Under the supervision of Yale’s Urban Ecology and Design Lab (UED) and based off of previous work from the UED, we are conducting the modeling and analysis of heat rejection in a thermoGreenwall to implement a physical model in a site on Yale’s campus. We started the modeling process by analyzing different functions of the thermoGreenwall including heat rejection, solar radiance, and water quality treatment. We used MATLAB SimuLink to analyze the heat transfer into and out of the wall through solar radiation, convection, conduction, and advection from adjacent panels. So far we have obtained a preliminary model addressing the one-dimensional heat transfer of a wall. Additionally, we have been developing a design for the implementation of a prototype. We coordinated meetings with the Yale Office of Sustainability who provided us with metrics on assessing an ideal test location. Subsequently, we coordinated with the Facilities Department to gather building documents and arrange tours of potential sites. These meetings have yielded preliminary sketches and designs for the building attachments and HVAC (heating, ventilation, and air conditioning) system interfaces. In the coming weeks we are preparing to integrate the models for different functions of the thermoGreenwall and to continue designing the interfaces, attachments, monitoring and maintenance systems for the physical prototype.

Electronic steering systems for high-performance rowing shells

Noah Baily
Advisers: Mr. John Telfeyan, Resolute Racing Shells, and Prof. Corey O’Hern, Mechanical Engineering and Materials Science

Rowing races are often decided by tenths of a second, and although the materials and construction of high performance rowing shells have evolved to reflect this reality, the technology inside has not. Specifically, steering adjustment in rowing shells is the single greatest external interrupter of boat speed. Steering systems are currently actuated by two wires attached to a rudder column, which creates a strength requirement to counter the rudder torque at high speeds and causes imprecise or erratic rudder movement. The aim of our project is two-fold: to measure the effect of rudder movement on boat speed using a purpose-built rudder data collection system, and to use the knowledge gained to design an electronic steering system. By using servo motors rated for the maximum rudder torque, the steering system to be designed in this project ensures that the coxswain or athlete does not need strength for, or divert unnecessary mental focus to, smooth rudder actuation. The aim for this project is to eliminate the extraneous and unnecessary rudder movements or course corrections which interrupt the fluid flow around the shell and can lead to extra distance travelled, decreased boat speed, and ultimately cost the race.

A wearable device PCB for university access control and emergency alerts

Tilman Bartelsmeyer
Advisers: Mr. Kevin Ryan and Prof. Mark Reed, Electrical Engineering

The inconvenience of conventional ID cards and student safety are pressing issues for colleges. With existing options like BlueLight Phones and smartphones, it is infeasible to alert emergency services until after an assailant has left. Additionally, RFID-based identification cards are easily lost, cumbersome, and outdated. However, they remain ubiquitous given the time and cost of system overhaul. We are creating a modular, wearable device to attack these issues. It enables users to subtly report their location to emergency services, without a phone. It also acts as an inexpensive ID that thwarts replication, provides continuous authentication, and reduces loss while avoiding prohibitive door reader replacements. This spring, we are designing and manufacturing the device's Printed Circuit Board (PCB), which is housed in a 3D-printed case. The key design constraints are size, cost, water resistance, and durability. The primary functions are RFID signaling, communication with nearby smartphones via Bluetooth Low Energy (BLE), button input, and user notification through haptic motor and LED. The biggest challenge is maximizing battery life in an ultra-small package without limiting functionality. We seek to sell our product to universities across the country as an inexpensive alternative to traditional ID cards that enhances campus safety and increases utility.

Impedance matching RF circuit for DEP electrodes

Katherine Berry
Adviser: Prof. Mark Reed, Electrical Engineering

We are designing and prototyping an impedance-matching circuit to help streamline the functioning of a dielectrophoresis (DEP) device. Currently, radio frequency signals are being sent through an amplifier across a series of interlaid electrodes in order to isolate bacteria from fluid samples. Due to a mismatch between the output impedance of the amplifier and the impedance of the electrode geometry, the signals are reflecting back down the transmission line, decreasing necessary power across the electrodes, increasing power along the line, and burning out the amplifier. To address this, we characterized the impedance of several electrode geometries, considering each to be analogous to a small capacitor in series with a massive resistance. The image shows the real portion of the impedance of one of the electrode geometries submerged in various fluids to mimic the final use environment, plotted against the frequency of the incoming signals. Fitting this curve, we determined the impedances of all the electrode geometries. From here we used software to design a simple circuit that could match these impedances within our frequency range of interest. However, we found that due to the high radio frequencies used and the very high impedances of the electrodes, stray capacitances in the actual circuit components as well as the types of leads passing the signals into our circuit create effects not predicted by the theoretical simulation. Currently we are selecting more specialized circuit components to eliminate these problems and allow the circuit to function as theorized.

Developing and optimizing a continuous positive airway pressure device for low-resource settings

Sydney Costantini, Jessica Lee, and Ryan Reza
Advisers: Dr. Alyssa Siefert, Center for Biomedical Innovation and Technology, and Prof. Corey O’Hern, Mechanical Engineering and Materials Science

Continuous positive airway pressure (CPAP) devices are used as a therapy for neonatal respiratory distress syndrome (RDS), a leading cause of respiratory failure and mortality in infants in the developing world. Traditional devices in the developed world can cost several thousand dollars, making them inaccessible in low-resource settings. Dr. Ryan Carroll and the Consortium for Affordable Medical Technology at Massachusetts General Hospital have developed a low-cost CPAP prototype for hospitals in Mbarara, Uganda, but the device runs on wall power. Since the region has inconsistent power, with outages nearly every week, we are working to build an affordable CPAP device that is powered by a solar-rechargeable battery. This project began last semester, with a prototype of the CPAP device, structured as shown in the diagram, and a proof-of-concept for the solar system. This semester, we are working to optimize the device by making the motor more efficient, improving the humidifier by adding heating elements, reducing losses in flow and pressure through the device, and implementing a solar recharging system. Last semester’s design used an aquarium pump to generate flow; however, it was significantly more powerful than necessary. By sourcing a better-suited motor, we can save power and run the device more efficiently. Moreover, standard CPAP devices offer heating in the humidifier to improve its performance and make patients more comfortable. We are investigating heating possibilities for our humidifier that balance performance and power draw. Additionally, we are working to characterize the flow of air through the device to reduce losses between the pump and the patient. Finally, the solar system will power our device during outages, and we plan to make our solution scalable for multiple devices in one hospital. Through this project, we hope to make CPAP devices more accessible in the developing world.

EZ ICE™: Reducing component weight using topology optimization

Dylan Gastel
Adviser: Dr. Ahmet Becene, UTC Aerospace Systems

EZ ICE, Inc., a Yale startup, has created The 60 Minute Backyard Rink ™, the only backyard ice skating rink that can be assembled in under an hour, with no tools, and on any surface. EZ ICE’s novel and patent pending technology provides an entire kit that snaps together with ease — saving users dozens to hundreds of hours of backbreaking labor and frostbite compared with alternative construction methods. The aim of this project is to minimize the weight of the rink’s components such that the rink can still support the 280 pounds of force exerted by the water and ice, and such that it can still be manufactured using simple side-action molds without any undercuts. We are using GENESIS with Vanderplaats R&D topology optimization software to minimize material usage based on these constraints. Topology optimization optimizes material distribution within a given design space for a given set of loads and boundary conditions such that the resulting design meets prescribed performance targets. We are using the results of the topology optimization to design a composite board that we expect to be 10–40% lighter than our current design and at least as strong and easy to manufacture.

Revision of a Formula SAE race car: BR14

Taha Ramazanoglu
Advisers: Dr. Joseph Zinter and Prof. Corey O’Hern, Mechanical Engineering and Materials Science

The revision of the BR14 Formula SAE race car aims to preserve the engineering knowhow acquired by Bulldogs Racing through the design and manufacturing of the vehicle in 2014. BR14 remained tabled and partially disassembled during the all-electric race car project (BR16) in 2015 and 2016. Since the last team members who have seen the vehicle in working order belong to Class of 2017, one member decided that it would serve the team well to restore the vehicle back to working order and document the revised engineering subsystems. The first step was assessing which subsystems needed repairs and redesigns. The cooling, ergonomics, and electrical subsystems were determined to be in need of revisions due to a leaking radiator, malfunctioning dashboard, and a worn-out battery as well as a wiring scheme that was prone to faults. The car was also in need of painting due to rusting parts. So far we have replaced the battery, redesigned the wiring harnesses and acquired a new radiator. In the upcoming days we will have put the radiator in the car, installed the wiring, and finished painting. The revised car and the accompanying documentation will enable the team to build upon its previous experiences.

Automated platform for MAGE (Multiplexed Automated Genome Engineering)

Benjamin Rosenbluth
Adviser: Prof. Farren Isaacs, Molecular, Cellular, and Developmental Biology

Multiplex Automated Genome Engineering (MAGE) is a genome engineering technique designed to cultivate genomically diverse E. coli populations with broad applications spanning scientific discovery, drug development, and industry. Despite its conception as an automated procedure, MAGE in its current practice is performed laboriously by hand. Our research aims to build a scalable automated MAGE machine to meet growing academic and industrial demand for genetically engineered organisms. In the fall we built and characterized a flexible turbidostat (pictured), a brand of bioreactor that will form a closed loop with a custom microfluidic genome editor as part of a two-part automated MAGE system. The turbidostat maintains an active cell culture at constant growth conditions by using a feedback control system consisting of laser sensors and 3D-printed syringe pumps to maintain optical density. This semester’s work has centered on developing the microfluidic genome-editing platform, with capacity for temperature-controlled protein expression, tangential-flow cell filtration, and microfluidic electroporation. The completed device will be a fully automated genome editor to provide academic and industrial research facilities with access to automated genome engineering capabilities.

Experimental Testing

Energy savings with permanent magnet synchronous alternating current motors

Jonathan Lai and Tyler Lu
Advisers: Prof. Corey O’Hern, Mechanical Engineering and Materials Science, and Mr. Kevin Ryan, Electrical Engineering

Two forms of refrigeration motors, the industry standard electronically commutated motor (ECM) and a newly developed permanent magnet synchronous motor (PMSM), are to be tested against each other to determine energy savings when switching from an ECM to a PMSM. ECMs are a form of synchronous motor that run on DC power gathered through an integrated inverter. While PMSMs are also a type of synchronous motor, PMSMs theoretically have better energy efficiencies than ECMs because of their use of AC power. As power is often transferred in AC form, the PMSMs’ direct usage of AC means that they constantly save one step in their transformation of electrical energy to mechanical energy. Previous studies between the two motors involved the retrofitting of ECMs with PMSMs in large-scale installations (supermarkets). The difference in input power required to generate equal output power was shown to be about approximately 40%. We aim to confirm these results through laboratory testing of the motors by measuring the electrical power needed to run both types of motors at the same output power. Our experiment involves two PMSMs and one ECM, a smaller-scale comparison of the two motors that can be applicable in demonstrating energy efficiency in both commercial and residential settings. In an industry worth billions of dollars, energy savings of even small margins stand to result in millions of dollars in financial savings as well a significant reduction in our carbon footprint.


Tuning cellular function through novel nanopatterning techniques

Saisneha Koppaka
Advisers: Prof. Jan Schroers, Mechanical Engineering and Materials Science, and Prof. Themis Kyriakides, Biomedical Engineering and Pathology

The longevity of implants can be extended by carefully orchestrating the foreign body response (FBR) through nanopatterning. Nanopatterning is a process for generating surface features, such as rods or pores, on the nanometer length scale. One biomaterial that offers excellent processing capabilities for forming these nanoscale patterns is polymers. With this project, we optimize a protocol for nanopatterning polymeric substrates via a type of compression molding process known as thermoplastic forming. By heating a material to its glass-transition temperature and then using a compressive force to press the material into nano-scale porous molds, we form the nanorods on various polymer substrates. Tuning the tip shape of the polymer nanorod, as shown in the accompanying scanning electron microscopy image (right image — convex tip shape; left image — concave tip shape), is still a work in progress. By the end of the semester, we aim to characterize the optimal processing conditions of a range of polymers to determine which temperature and pressure conditions will yield uniform arrays of nanorods of desired morphology. We hope that the successful fabrication of polymer nanorods can have applications in the field of nanofabrication and biomaterials.

Statistical Analysis

Clean water access in West Sumatra, Indonesia

Amelia Dobronyi and Kartik Srivastava
Advisers: Dr. Sara Hashmi, Chemical Engineering, and Prof. Corey O’Hern, Mechanical Engineering and Materials Science

Access to clean water is lacking for many people in the West Sumatra region of Indonesia. Various levels of government within Indonesia and several international agencies have intervened with programs and schemes to help tackle this problem. However, there has been little concerted effort to document the effects of these policy interventions. Our project attempts to link demographic and infrastructural factors within local communities and ways in which they determine the success of water access programs in this region. We are using data from a primary household survey conducted in 2016, as well as data from local authorities and third parties to develop a model that can determine correlations between factors and increased access. We are using clustered regressions and other statistical techniques to establish these relationships. Drawing on data from PAMSIMAS, a World Bank-sponsored program to provide to access to clean water at the community level, we have found West Sumatra is lagging national health and clean water access statistics. We have also found that increased rural-to-urban migration in this region is hampering clean water access and that ability to pay and funding programs are able to ensure some success. We are augmenting these findings by evaluating other data sources, such as geological data from other researchers at Yale. In publishing the results of our analysis, we hope to draw attention to the issue within the academic community and encourage collaboration to provide a robust, long-term, and potentially adaptive solution to the problem of clean water access.