Student: Nicholas Crispie
Professor/Sponsor: Professor Hayden Taylor
Mentors: Vivek Subramanian and Will Scheideler
Research Project Title: Development of Novel Gravure Printer Design for Improved Printed Circuit Manufacturing
Abstract:
Performance and resolution of printed electronics is highly dependent on the accuracy and quality of the tools used in the fabrication process. Many patterning processes that have been previously developed have a trade-off between speed and accuracy. This project focuses on the development of a new Gravure printer design, making use of roll-on-sheet manufacturing, to produce high quality prints of 2 micron resolution with a high throughput of 1m/s surface speed. This involves implementing precise pressure adjustment, synchronization of the roller and substrate, and developing and repeatable alignment method.
The pressure adjustment is determined by a electro-mechanical system consisting on a roller mounted on a spindle block with an integrated load cell, with additional circuitry to read the values to monitor and adjust the pressure in pneumatic actuators to apply a precise amount of pressure on the substrate to get a clean print without the printing ink running off of the substrate, decreasing the quality of the print. The alignment relies on a combination of mechanical and software systems to provide a basic alignment point, which is then optimized using a combination of mechanical actuators with feedback from mounted cameras with live feed into a custom software interface.
For next steps, the plan is to fully integrate the subsystems that have been developed to test the alignment and repeatability marks to see if it meets the desired specifications, making modifications as necessary to make improvements to the printing process.
Student: Lucas Filshill
Professor/Sponsor: Professor Hayden Taylor
Mentor: Brett Kelly
Research Project Title: Carbon 3D printer Speed and Resolution
Abstract
Student: Rhett Gentile
Professor/Sponsor: Professor David Dornfeld
Mentor: Raunak Bhinge
Research Project Title: Design and Manufacture of Waterproofing Test Chamber and Wired Infinite Uptime Sensor
Abstract:
In the highly automated world of modern manufacturing, monitoring the status of a machine tool is critical to prevent part spoilage or damage to machines due to worn tools. The Infinite Uptime sensor package contains a microphone and accelerometer to monitor machine status in real time. In order to certify that all Infinite Uptime package designs are proof against intrusion from flood coolant, testing was previously performed on the Student Machine Shop’s CNC mills. This report details the design and production of a test chamber to simulate CNC conditions without taking up valuable shop time. The first chamber was a custom-built acrylic cube, but suffered waterproofing issues. A second chamber was successfully constructed from a repurposed chemical glovebox. Flood coolant was provided by a 0.5HP pump. Additionally, an enclosure for a wired Infinite Uptime sensor was designed. The primary focus of this design was to allow for the use of a standard microUSB charging and data cable while maintaining a waterproof enclosure for the sensor package. Several iterations were printed in ABS using the machine shop DimensionSST printers; the final iteration was successfully tested in the previously constructed chamber. Industrial interest has already been expressed in the lower profile and maintenance free wired version of the sensor package.
Student: Alan Harbottle and Hani Esmaeili
Professor/Sponsor: Professor Paul Wright
Mentor: Alic Chen
Research Project Title: Piezoelectric Fan used for Cooling LEDs
Abstract:
This research focuses on cooling high-power LEDs using a combination of a piezoelectric fan and a custom-designed heatsink. A piezoelectric fan contains a piezoelectric material such as PZT, that produces mechanical strain when a voltage is applied, attached to a mylar blade which flexes back and forth with an AC voltage. This fan is ideal because it consumes little power and has low acoustic emission. The research was funded by Siemens AG with the stated goal of cooling a high-power LED, with surface area of 50mm x 60mm, from 100°C down to 50-60°C as LED performance and lifespan is negatively affected by high temperatures. Several parameters were optimized over the course of this research including: fan size and orientation, design of the heatsink channels and fins to maximize airflow, size of the entire system and overall cost. The piezoelectric fan’s “waving” motion presents challenges with irregular airflow characterized by the asymmetric generation and propagation of vortices. To aid in the design of heatsinks we used COMSOL modeling and Particle Image Velocimetry to better visualize the resultant airflow. The piezoelectric fan, paired with a custom-designed, in-house CNC’d T6061 aluminum heatsink cooled the LED down to steady-state temperature of 41°C.
Student: Debera Hsiao
Professor/Sponsor: David Dornfeld
Mentor: Mickey Clemens
Research Project Title: Heat Transfer in 3D Printing
Abstract:
Additive manufacturing, also known as 3D printing, is gaining popularity is a cost-efficient method of rapid prototyping. It can create 3D solid objects from a digital model within hours, using a thermoplastic called acrylonitrile butadiene styrene, or ABS. Since it is a relatively new process, limited modeling has been done on the process and limited resources are available for information about 3D printing materials. ABS Plus is supposedly a 40% stronger version of ABS, but information containing its compositional differences are near impossible to find. This project focuses on modeling the heat transfer in the 3D printing process using MatLab. This simplistic model takes into account the energy necessary to heat up a segment of ABS and then the heat loss when each ABS particle is deposited. The code only accounts for a few particles, but can eventually be used to determine the energy necessary to print an entire model. It can also be used to determine the energy consumption trade-offs between nozzle speed and deposition particle size because the data shows that as the particle size increases, the cooling rate decreases.
Student: Roger Isied
Professor/Sponsor: Professor Hayden Taylor
Mentor: Brett Kelly
Research Project Title: Comparison of the Response of Additive Manufactured Specimens to Monotonic Tensile Loading
Abstract
Student: Roger Isied
Area: Manufacturing, Materials
Professor/Sponsor: Professor Hayden Taylor
Mentor: Brett Kelly
Research Project Title: Curing Dose Impact on Mechanical Properties of PEGDA Specimens created using Computed Axial Lithography
Abstract
Student:Shayan Javaherian
Professor/Sponsor: Professor Reza Alam
Mentor: Dr. Mohsen Saadat
Research Project Title: CalSat
Abstract:
The purpose of this research is to make underwater wireless communication possible by using ROVs and laser tractions. The calsat project is consist of two different version which they call CalSat 1 and CalSat 2. For CalSat 1 the purpose of this project is to modification of controls of two submarine model to carry out the proof of concepts of underwater optical communication using a swarm of autonomous underwater vehicles. For CalSat 2 we made our own ROV that is an Agile and robes underwater platform used for underwater communication by using laser tractions. I widely work on design, prototyping, and manufacturing of CalSat 2. CalSat 2 Has different versions which each one of them developed and improved based on the previous version. Different version of CalSat 2 are as following: CalSat 2A, CalSat 2B, CalSat 2C, CalSat 2D. Following pictures are for CalSat 2C while testing for leakage and performance in O’Brien facility at UC Berkeley.
Student: Abhishek Khemka
Professor/Sponsor: Professor David Dornfeld
Mentor: Dr.Karl Walczak
Sub Area: Energy Science and Technology
Research Project Title: SWOT analysis on the future of the Hydrogen economy pertaining to solar-driven Photoelectrochemical cells.
Abstract:
Primary goal of the project was to analyze the Global Energy sector with respect to solar fuel production using photoelectrochemical cells(PEC) and its future techno-economic viability. A framework for the technology is sought to be built which could evaluate its needs for future development. A Strength-Weakness-Opportunity-Threat (SWOT) analysis is performed so that key parameters of the assignment can be identified while keeping in mind the future goal of the technology.
There exists a bi directional causality between Energy consumption and GDP growth for the US which necessitates government investment in Energy R&D to grow the GDP while meeting the growing energy demand. A current R&D initiative undertaken by the government is investment in solar fuel generation which has the potential to be economically and environmentally sustainable. Solar fuels can be generated using Photoelectrochemical(PEC) cells which would in the future lead to development of a parallel energy economy: the Hydrogen economy, thereby enabling the United States to induce growth in its Energy sector which will correspondingly create economic growth.
In this study, four types of PEC systems were evaluated to analyze the techno-economic feasibility for introduction of this technology on an industrial scale. A levelized cost of electricity(LCOE) comparison was made for each of the 4 PEC systems with conventional sources of energy which were then further evaluated based on parameters such as the hydrogen infrastructure, sustainability, economic viability and scaling up the technology for industrial use.
It was concluded that as economies of scale improves the manufacturing related cost aspects of the 4 PEC systems, further advancement can be attained if the government provides subsidies for private industry players to adopt this technology. Difficulties such as hydrogen storage due to a low energy density per unit volume, relatively long period of development of the technology and a lack of hydrogen infrastructure are key issues which need to be addressed innovatively. With an acute energy shortage in developing regions of the world and fluctuations in oil price and imports which adversely affect major economies, especially the US, it becomes imperative to develop PEC cells as an integral part of the overall energy economy.
Student: Qiaohao (Harry) Liang
Professor/Sponsor: Professor Liwei Lin
Mentor: Emmeline Kao
Research Project Title: Photocatalytic Water Splitting/ Electropolymerized Devices for Photocatalytic Hydrogen Gas Harvesting
Research Areas: Materials, Photocatalytic Materials
Abstract:
Due to the foreseeable shortage of fossils fuels in the future, it remains significant to discover a new source of energy that is both abundant and clean to replace of fossil fuels. One solution researchers have come up with is to acquire hydrogen gas as fuel from electrolysis of water due to the relatively abundancy of water recourses and the sustainability of such clean energy source. Photoelectrochemical (PEC) water splitting devices, widely used in hydrogen gas harvesting, can convert sunlight into electron/hole pairs that oxidize water and reduce resulting H+ ions. However, efficient conversion of solar energy into hydrogen fuel has been difficult to accomplish mainly because ideal water splitting materials have been difficult to fabricate. These materials must have bands gap small enough to absorb energy from solar spectrum effectively and positioned correctly to allow for the reduction and oxidation reactions to take place, necessary for initializing the water splitting process. Therefore, this research project’s focus would be on thiophene-based devices made by electropolymerization for possible usages in solar-powered, photocatalytic hydrogen gas (H2) harvesting. The novelty of this research lies in the first demonstration of electropolymerized photoelectrochemical (PEC) devices for water splitting, and as such, this work opens up a new class of material and device fabrication for cheaper and efficient H2 harvesting systems. At the current stage, stainless steel is used as the substrate for a two-step electropolymerization: (1) 10 second seed layer deposition with Vbias = 2V; and (2) a 60 second film deposition with Vbias = 1.95V. Successfully electroplating thiophene onto conductive substrates, which precludes alignment of hydrophobic hexyl groups and therefore allows electron flow perpendicular to the substrate surface, limits the travel distance of carriers and thus increase conductivity and charge separation of the device. The photocatalytic materials in such device contribute to more efficient conversion of solar energy into H2, shown by the excellent onset voltage during photocurrent testing. The existence of such onset voltage allows lower energy input during water splitting process to activate the reactions, therefore lowering the cost of H2 by some degree. Future experiments will be focused on increasing the onset voltage by improving electroplating recipes so that the electroplated polymers are more smooth and even spread across the substrates’ surface. Different substrates and monomers will also be tested to find the ideal combination that yields largest onset voltage.
Student: Kevin M. Ninomiya
Professor/Sponsor: Professor David Dornfeld
Mentor: Dr. Moneer Helu
Research Project Title: Predictive Surface Roughness Model for a Freeform Geometry: A Material Removal Rate Approach and the Process Energy Requirement
Abstract:
As energy costs and number of environmental regulations are on the rise, it is becoming more important for manufacturers to seek environmentally benign manufacturing practices while maintaining the same part quality. This is especially true for gas turbine compressor vanes/blades employed in the energy and aerospace industries where the former is concerned with energy generation while the latter is associated with consumption. This is also true for many types of gears such as bevel and worm gears applied in various industries. Simply applying sustainable methods in a process may seem less complicated than anticipated; i.e., less process energy consumption. However, manufacturers should take care when deciding the appropriate amount of effort spent on implementing such process parameters in order to maintain the same quality and performance of their products. A poorly planned process improvement may trigger customers to other competitors who have achieved better quality while increasing their sustainable assets. Process optimization enables manufacturers to make swift decisions quantitatively such that the best part quality can be produced while integrating sustainable strategies effectively.
This semester, I researched under Mr. Moneer Helu on the topic of inter-phase correlations regarding the relationship between manufacturing phase process parameter selection and use phase product performance (efficiency). We worked on planning two different projects related to turbine engine blades/vanes and gears. My contributions included a literature review and a potential methodology with the target CAD model for the turbine engine vane project. A literature review was conducted and possible research topics were identified for the gear project.
Student: Brian Salazar
Professor/Sponsor: Professor Hayden Taylor
Research Project Title: ADRIENNE: A Deep Reactive Ion Etch Non-Uniformity Estimator Microsystems and Integrated Circuits
Abstract:
The deep reactive-ion etching process allows for vertical etching of silicon substrates and has many applications in MEMS and IC fabrication. DRIE use has grown in recent years, as it becomes an integral part of 3D wafer-level packaging since it allows for the manufacture of through-silicon vias and serves as a method for plasma dicing. A cycle of the Bosch process consists of silicon etching, followed by polymer deposition, and then removal of the polymer from the bottom of the silicon features. However, this anisotropic etching process has many factors that lead to non-uniform etch rates including at the wafer-scale (the average pattern density across the wafer affects the etch rate), at the pattern-scale (the local pattern density affects competition for reactants), and at the feature-scale (aspect-ratio dependence). Using a synergism model, a control volume analysis of the fluorine, and etch-depth data gathered by Sandia National Labs, this project seeks to determine a model for simulating the etch depths across a wafer when given a design mask. The model performs a least-squares fit on the empirical data to determine the time evolution parameters, spatial maps, and diffusion constants and uses a time-stepping technique to account for fluorine concentration variations. The simulation fits the experimental data with an r.m.s. of 5%. This simulation can be incorporated into CAD software, where it can be used to optimize the location of features on a silicon wafer.
Student: Amrita Srinivasan
Professor/Sponsor: Professor David Dornfeld
Mentor: Dr. Moneer Helu
Research Project Title: In-Situ Characterization of Surface Finish: Part II
Abstract:
The surface finish of a machined part is a highly valuable piece of information that is currently challenging to quantify in-situ. Previous research has demonstrated the existence of some correlation between a machine tool’s energy consumption and the imparted surface finish, though this relationship has not been precisely defined. The objective of this project was to mine the huge amounts of data obtained from previously conducted experiments on this topic to find trends that further elucidate the aforementioned relationship. A graphical user interface was created in MATLAB to facilitate visual and quantitative determination of trends in the data. In particular, the effects of the variance in power demand and of different combinations of spindle speed and feedrate, in conjunction with power consumption, on surface finish were evaluated.
Student: Amrita Srinivasan
Professor/Sponsor: Professor Dornfeld
Mentor: Dr. Moneer Helu
Research Project Title: Measuring Surface Roughness with a MTConnect™ Application
Abstract:
Manufacturing processes are rapidly becoming more advanced and precise, but certain important outcomes of said processes are still not easily measurable. The surface roughness of a finished part is one of those results that currently requires expensive and wasteful techniques to directly measure. If the surface roughness of a machined part could be accurately predicted based on more conveniently measurable manufacturing parameters such as energy consumption, then gauging that part’s surface finish could become the simple, hands-off matter of ob-serving the output of an application that analyzes all the relevant data and performs the necessary computations. In order to form a sufficient base of knowledge and skills to develop such an application that utilizes the MTConnect™ standard for communications between manufacturing devices, such as machines and sensors, and software applications that are driven by the manufacturing data, a test part was designed and prepared for fabrication. The part was modeled in Solidworks and a toolpath to machine it on a Mori Seiki CNC mill was generated in Esprit. Various designs and toolpaths were considered in a process that enhanced understanding of common procedures and challenges faced in precision manufacturing. A MTConnect™ adapter for the Modbus Wattnode was written to gain practice and familiarity with the MTConnect™ standard and with machine monitoring procedures. Ultimately, the valuable lessons and skills gained from developing the test part and studying hardware-software communication will be implemented to write a MTConnect™ application for the Android platform that will take into account the relevant and quantifiable parameters of a CNC milling operation to output a prediction of a part’s surface roughness.
Student: Amrita Srinivasan
Professor/Sponsor: Professor David Dornfeld
Mentor: Raunak Bhinge
Research Project Title: A Generalized Data-driven Energy Prediction Model with Uncertainty For a Milling Machine Tool Using Gaussian Process
Abstract:
Using a machine learning approach, this study investigates the effects of machining param- eters on the energy consumption of a milling machine tool, which would allow selection of optimal operational strategies to machine a part with minimum energy. Data-driven prediction models, built upon a nonlinear regression approach, can be used to gain an un- derstanding of the effects of machining parameters on energy consumption. In this study, we use the Gaussian Process to construct the energy prediction model for a computer numer- ical control (CNC) milling machine tool. Energy prediction models for different machining operations are constructed based on collected data. With the collected data sets, optimum input features for model selection are identified. We demonstrate how the energy prediction models can be used to compare the energy consumption for the different operations and to estimate the total energy usage for machining a generic part. We also present an uncertainty analysis to develop confidence bounds for the prediction model and to provide insight into the vast parameter space and training required to improve the accuracy of the model. Generic parts are machined to test and validate the prediction model constructed using the Gaussian Process and we consistently achieve an accuracy of over 95 % on the total predicted energy.
Additionally, twenty-eight parts of the same design were fabricated on a Mori Seiki NVD 1500 Vertical Milling Center utilizing the same tool path; the controller information and power consumption data were captured simultaneously using MTConnect and a High Speed Power Meter (HSPM). Thirteen of these parts were 1018 mild carbon steel; five were 6061 T6 aluminum; and the remainder were of Grade 2 titanium. Four flute, uncoated 5/16″ diameter carbide end mills were used to fabricate the desired geometry. Spindle speed, feed rate, and depth of cut were varied over the twenty-eight parts so that their effects on power consumption could be investigated.
Student: Amrita Srinivasan
Professor/Sponsor: Professor David Dornfeld
Mentor: Raunak Bhinge
Research Project Title: Integrated Vibration And Acoustic Data Fusion For Chatter And Tool Condition Classification In Milling
Abstract:
Improved data quality and availability, along with lower computation costs, have generated interest in sensor-based tool condition monitoring technologies. In this study, an integrated vibration and acoustic sensor is used for tool condition monitoring, particularly for chatter detection and tool condition classification. Based on feature extraction in the time and frequency domains, chatter and tool condition classification studies are conducted using linear Support Vector Machines. The combination of acoustic and vibration data is found to have a higher classification accuracy as compared to the individual data sources.
Student: Aldo Suseno
Area: Manufacturing, Materials
Professor/Sponsor: Professor Hayden Taylor
Mentor: Brett Kelly
Research Project Title: Investigating the Monotonic Tensile Properties of Dogbone Specimens Created Using Additive Manufacturing on Carbon3D Printer
Abstract
Student: Travis Tindell
Professor/Sponsor: Professor Hayden Taylor
Mentor: Brett Kelly
Research Project Title: Testing Additive Manufacturing Technologies with Unique Geometries
Abstract
Student: Zea Wang
Professor/Sponsor: Professor Tarek Zohdi
Mentor: Maxwell Micali
Research Project Title: Variable 3D printer nozzle
Abstract:
As additive printing is gaining in popularity and increased in its uses, it is important to minimize build time while maintaining resolution throughout the part. A variable nozzle is able to accomplish this by changing the extrusion diameter while printing. A variable nozzle introduces additional flexibility in the 3D printing process. Not only will this make additive manufacturing more efficient it will allow for artists to explore a new feature, further expanding the abilities of the machine.
Our team’s design features the use of a mechanical iris mechanism to vary the diameter of the nozzle. This allows for the cross section of the mechanism to remain relatively circular and continuous variation of the diameter while printing. The 3D Potterbot, a ceramic printer, was chosen in order focus on the mechanical design without interference of heat and phase transitions in the material. In testing, the mechanical iris was successful in changing the size of the extruded material from 6mm to 20mm continuously. Problems came about as the iris reached the smaller diameters due to the bunching of the rubber liner between the clay and the mechanism. High pressure is also applied to the mechanism from the clay during extrusion making the rotation of the iris and therefore the changing of the diameter difficult.
Future works include the integration of a motor in order to control the diameter of the iris automatically and precisely. This will also require proper slicing and G code to control the rate of extrusion of the 3D Potterbot to the nozzle diameter. Further developments will include the integration of a slicing algorithm to maximize the build time and the resolution of each slice.
Student: Kriya Wong
Professor/Sponsor: Professor Grace Gu
Mentor: Zhizhou Zhang, Kahraman Demir
Research Project Title: OwlFoil: Development of Bio-Inspired Multimaterial Composites
Abstract:
The power of silent flight achieved by owls extends further than simple domination of the evolutionary arms race between predator and prey. Successful modeling and printing of wings have the potential to reform turbine and aerodynamic technology in terms of both energy efficiency and noise reduction. The characteristics of owl wings that render them silent are primarily the leading edge feathers and the trailing fringe of the wing, which work jointly to break up oncoming air currents and channel them along an invariant surface, minimizing the sound during flight. The leading edge feathers, which are typically smaller and more circular in shape, are lined with tiny serrations along the feather that are called pennula, whose primary purpose is to create roughness and texture along the wing that will break up the air currents into smaller streams called micro-turbulences, which raise the noise frequency of the air rushing over the wing to a higher frequency that is not detectable by prey and also humans. The trailing fringe further differentiates the owl from other birds in that the substructure of these feathers allow them to mesh into one another when the wings unfold, such that when the feathers spread, the outer fringe of the feathers create almost a single sheet with very little overlap, maximizing area and creating smoother surface which reduces noise and tapers out into larger, less densely packed barbule areas that break the air currents further into smaller streams to reduce noise. This project aims to create a base model for the computer-aided design (CAD) of synthetic, multi-material bird feathers, specifically of the male barn owl for the rapid prototype and development of 3D-printed feathers. Using an online database of primary feathers collected from the barn owl, three models from different regions of the wing were generated taking into account external feather spline, rachis or stem characteristic, curvature and barbule density. The properties of owls’ silent flight deemed to be the most impactful have been determined to be the comb-like pennula on the leading edge feathers and the fluid-like trailing fringe of the lower wing feathers, which work together to break air currents into smaller pockets as well as smooth the underside of the wing. The successful modeling and 3D-printing of these characteristic feathers unique to the owl have the potential to transform airfoil and turbine technology. As a crucial step towards the modeling of an entire wing, this project defines the parameters necessary for the realistic multi-material generation of owl flight feathers.
Student: Xianxin Zhang
Professor/Sponsor: Professor David Dornfeld
Mentor: Raunak Bhinge
Research Project Title: Development Of The Housing For Manufacture Monitoring Sensors
Abstract:
While manufacturers are progressively increasing production rate, or the effectiveness, maximum product quality is also required at the same time. A key thing to ensure product quality is to periodically check the machine tool condition because a worn machine tool not only can reduce production rate, but also can damage the machine itself resulting high maintenance cost. A smart vise jaw device with sensors embedded has been developed to monitor the machine tool condition and minimize manufacturing lead time. Such a device needs safe environmental conditions for sensors to work as well as the easiness to disassemble from the vise for future configurations. A proper housing, thus is needed to isolate coolant or fine particles from entering the sensors. This paper approaches from the original concepts and details the design and fabrication of our housing prototypes to show the development of the device.