Student: Arash Ahmadi
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentor: Dan Murphy
Research Project Title: Scale model building smoke viewing
Abstract:
Scale model building smoke viewing have been completed to address a significant barrier to the technical problem of accounting for wind in naturally ventilated atria or malls for different sizes of buildings. The purpose is to validate this model for different scales of buildings. This exercise is also intended to validate the uncertainty associated with the effects of wind on smoke control systems utilizing passive or natural ventilation methods in atria and malls for different sizes of buildings. The scaling technique can validate or provide an alternative to CFD models. The effect of external wind upon heat and mass transport due to fires in a naturally ventilated atrium is investigated using scaled physical models in an atmospheric boundary layer wind tunnel. Utilizing SolidWorks to design parts and assemblies of the building; Construction of the scaled buildings was done using Polycarbonate plastic. Specific change was made on the program which was intended to analyze the data and pictures taken from the experiments to increase the efficiency of data analysis. Development of an experiment to figure out what fuel should be used to create heat needed in each building based on their size. Scale models were validated against reduced scale physical models. These issues are relevant to design of smoke control systems in naturally ventilated buildings and demonstrate the ability of scaled physical models using modern visualization and experimental techniques to complement and validate numerical fire modeling.
Student:Nicholas Bachand
Professor/Sponsor: Professor Tarek Zohdi
Mentor:Bernard Kim
Research Project Title: Layer-by-Layer Additive Printing of Zn-Mn02 Batteries
Abstract:
Fully printed Zn-Mn02 cells have promising applications in flexible electronics and the IoT. One current limitation of fully printed battery cells is creating a uniform cathode surface. If the surface is nonuniform, the electrolyte may insufficiently cover the cathode, leading to shorting across the battery. Smaller electrolyte layers offer better battery performance, so it is desirable to address this without compromising electrolyte thickness. In this work, we investigated how to create a uniform cathode surface using doctor blade printing. The first variable we modified was the ink viscosity. Shear thinning inks are ideal because they can be applied using doctor blading and hold their form after printing. We found that the ideal ink viscosity was 300 cP using a shear rate of 500 1/s.We also found that printing 175 mm thick cells offered a good compromise between energy density and surface profile. To achieve even better printing, we investigated mechanically altering the Mn02 powder in the inks. After testing six ball milling parameter combinations, we found that ball milling using 10 mm steel balls at 30 Hz for 8 hours led to the most uniform cathode surfaces. We then followed up the ball milling procedure with a variety of sieving parameters. Sieving the particles through 3 sieves of decreasing mesh sizes, we were able to reduce the waviness of the cathode surface by over and order of magnitude. We are continuing to investigate the benefits of this improved printing mechanism on battery performance.
Student: Urvashi Betarbet
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentor: James Urban
Sub Area: Combustion
Research Project Title: Ignition of Fuel Beds by Hot Particles
Abstract:
The increasing rates of unplanned fires in both wild and urban settings, especially in the Western portion of the United States, has sought calls for research concerning its instigators. This research project specifically concerns the effects of spot fire ignition on characteristic combustible materials, and modeling them. In real world scenarios, spot fire ignition is commonly instigated by metal particles or sparks. Thus, to further investigate determining factors, individual tests were performed by dropping heated (ranging from 575 to 1100oC) aluminum particles (ranging from 2 to 8 mm) onto a given fuel bed. In these experiments, three types of fuels were used to mimic main targets of flame propulsion from spot fire ignition: thermal insulation, dry grasses, and duff powders. From any drop, data regarding the potential for no ignition, smolder, and fire ignition was recorded. Before this term, data sets were already completed for cellulose strips (thermal insulation), a barley hay grass blend (dry grass), and both cellulose and barley hay dust (duff powders). Therefore this term was moreso focused on mimicking further combustible materials, reevaluating past data collection, modeling past data collection, and making future data collection more efficient.
First, pinestraw was chosen as a new fuel material to expand upon the dry grass and duff data given from the barley hay grass blend. To prepare for this, pinestraw needed to be collected and cut for dry grass testing, and additionally ground into a dust and then sieved to collect fine dust for duff powder testing. Once these preliminary steps were taken, ignition data was gathered. A brief test was also done using live fuel, but the lack of flame spread even when exposed to a blow-torch heated steel ball, halted further prospects. Then, past cellulose powder data was reevaluated against previous data taken by a different investigator but under similar conditions. Because of a discrepancy between the two sets, further tests were performed by “fluffing” the powder and thus creating a less dense fuel bed than ones test prior. As a result of this change, the discrepancy was eliminated and the packing factor density was taken into consideration for future testing. Additionally, a computer program was used to model the heat dissipation from the heated particles after being dropped onto a fuel bed. In order for the application to have an accurate visual representation, the shape of the particle must have a clearly defined boundary from which heat is transferred. For this reason, a Python function was created to take inputs of radius and pixel size (for sharpness of boundary) as variables to output the image of a representative particle. Finally, at the end of the term, a Raspberry Pi device was included in the experimental apparatus to test for temperature conditions within an oven used for drying fuel bed material. To use the device for outputting current data but also for performing analysis on past data, it was necessary to generate another Python code.
As for research results, the hyperbolic data curve from previous results was maintained, where for smaller particles a higher particle temperature was necessary to instigate ignition (both smoldering and fire). However, these tests also involved constant moderation to keep the particles from melting within the furnace and therefore the real-world imitation for these tests became less accurate than for others. Concerning the types of fuel however, the barley hay powder and cellulose powder were more prone to smoldering whereas the cellulose strips and barley hay were more prone to fire ignition-cellulose strips being the more frequent of the two. As for pinestraw grass and powder, fire ignition seldom occurred even with large particles and smolder occurred just as infrequently.
Student: Bradley Cage
Professor/Sponsor: Professor Robert Dibble
Mentors: Timothy Sennott and Miguel Sierra Aznar
Research Project Title: Acquisition and Control for the Argon Power Cycle
Abstract:
The Argon Power Cycle is a groundbreaking method that aims to increase the thermal efficiency of combustion by means of combusting methane in pure oxygen and argon as opposed to normal air. Implementing the Argon Power Cycle (APC) requires the creation of a robust control and data acquisition software suite capable of analyzing engine parameters in real time. First, the software system of the CNGDSF V8 engine was studied. This system uses two programs in parallel, engine control running in DSpace and data acquisition being done in LabVIEW. After becoming familiarized with this implementation in DSpace and LabVIEW, work began on the creation of an allinclusive engine management software written in LabVIEW.
In order to begin work on the engine management software, the correct versions of LabVIEW and its Xilinx compiler must be installed with all of their respective drivers. To figure out what drivers are needed, a complete cataloging and inventory was performed to understand what sensors would be used, how many of each would be used, and how many had already beThis proves to be nontrivial, as it was discovered that the National Instruments Compact Reconfigurable Input/Output (NI cRIO) chassis has convoluted requirements for each I/O module, with one requiring the repurposement of a separate power supply. The NI cRIO was found to behave very differently from the NI DAQmx chassis that was used on the CNGDSF system, with specifics that are still currently being worked out.
Once the NI cRIO chassis was successfully set up and the workstation ready, work began on the creation of a fuel injector driver system. The goal is to integrate the fuel injection control into the LabVIEW program, however as this requires the engine management software to be in a more completed state, the injector driver system was built as an analog circuit. Texas Instruments LM1949 Injector Drive Controller integrated circuits were used to obtain a ‘peak and hold’ style behaviour, where the current given to the injector peaks and then drops to a lower value for a longer duration of time. This minimizes heat dissipation from the injector and increases the lifetime of the injector. The LM1949 integrated circuits turned out to be inconsistent in performance, so the injector was driven by an Arduino Mega and an external voltage source.
With the injector driver functionality verified, work continues forward on creating and documenting the electrical systems to power the engine and all the data acquisition electronics. Going forward, the next challenge is to fire the engine and understand the encoder that measures crank rotation and begin the work of mapping all measured values, which are measured in time space, to crank angle space. Once in crank angle space, the bulk of the software setup will be done, and work can begin on analysing the measured data and performing experiments.
Student: Bradley Cage
Professor/Sponsor: Professor Robert Dibble
Mentor: Miguel Sierra Aznar
Research Project Title: Acquisition and Control for the Argon Power Cycle
Abstract:
Building off of previously completed work verifying the functionality of key components of the Argon Power Cycle (APC) engine, efforts have been focused on continual progress and development of the data acquisition and control interface for the engine. The entirety of this data acquisition and the majority of the control is done in LabVIEW, with physical I/O being run on a NI cRIO 9074 chassis. This I/O chassis has an onboard Field Programmable Gate Array (FPGA), a piece of rewriteable hardware so to speak. The FPGA enables for highly accurate, high throughput, incredibly fast, and truly parallel data collection and control, by physically changing logic gates to create a unique and custom piece of hardware. This FPGA is the keystone of the software being developed.
Having previously inventoried and detailed all of the sensors and actuators that would be present on the engine, work proceeded on building and wiring all of these devices on to the engine. Several implementations of cable management were attempted, before finally settling upon a modular system that allows for easy swapping of all sensors. All sensors and actuators run to a custom box suspended above the engine, where they meet panel connectors, allowing for easy interchanging of the sensors themselves without having to extensively rewire anything. The cRIO chassis is mounted on the inside of this box, and communicates to the main PC over the network by means of a bridged LAN connection.
With the hardware set up, efforts were then directed towards gathering data in software. This proved to be a big challenge and took up the majority of the semester, and continues to be ongoing. Compile times for code on the FPGA run into several hours, so small changes in code are difficult to test on the spot. Currently, the software can acquire data from resistive thermal devices (RTDs) and thermocouples in parallel, as well as converting their raw values into standard units and notations. SubVIs for other analog and digital inputs, fuel injectors, and analog and digital outputs have been created and are part of the main code, however are still in the process of being tested.
Work will proceed over the coming months to finalize and test all other inputs and outputs to the software, and then unify the I/O portion of the code to the data processing part. This data processing code is taken from the CNG-DSF project, being created over the course of several months by Miguel Aznar. The interface between the FPGA I/O and the data post-processing has been taken into account while all code has been being written, so this union should not present any large problems. Consistent data types and programming practices have been used across the whole project. Over the course of this semester the majority of the creative software work has been done, and testing and integration are the two main steps to come.
Student: Bradley Cage
Professor/Sponsor: Professor JY Chen
Mentor: Daniel Pineda
Research Project Title: Working fluid replacement in gaseous direct-injection internal combustion engines: A fundamental and applied experimental investigation
Abstract:
Replacing air with argon theoretically allows for large thermal efficiency increases in internal combustion engines. Before such cycles can be realized, fundamental research on fuel injection into argon and laboratory-scale engine tests are needed. We investigated non-reacting methane jets into argon and nitrogen atmospheres in a constant volume chamber using high-speed schlieren imaging. We subsequently assessed the feasibility of methane direct-injection in a modified single cylinder research engine with an argon-oxygen mixture as the working fluid. We compared engine performance by measuring fuel flow, in-cylinder pressure, torque, and emissions. Results show that the penetration depth and spread angles of methane jets are notably different but not significantly reduced in argon compared to nitrogen. Additionally, running the modified engine with an argon-oxygen mixture in compression ignition operation leads to improvements in efficiency up to 50 percent relative to spark-ignited air cycles, and NOx emissions are nearly eliminated. The results encourage more studies in which the exhausted argon is recycled into the intake.
Student: Bradley Cage
Professor/Sponsor: Professor Robert Dibble
Mentor: Miguel Sierra Aznar
Research Project Title: Development of an ECU for the Argon Power Cycle
Research Areas: Controls, Energy Science and Technology
Abstract:
The Argon Power Cycle is a novel technique that replaces air as the working fluid in internal combustion with a blend of argon and oxygen in order to increase engine efficiency. A unified Engine Control Unit (ECU) and Data Acquisition system (DAQ) was designed and built in LabVIEW, leveraging the fast processing speed of an integrated Field Programmable Gate Array (FPGA) to perform time-critical tasks such as fuel injection and spark ignition. Wiring design and sensor placement were done alongside code development to ease the work in connecting hardware to software. The ECU and DAQ systems were found to be able to control and acquire data with the engine running with a variety of fuels (Methane, Gasoline, and CNG) and in a variety of conditions and combustion schemes (varying compression ratio, port/direct injection, spark-assisted HCCI, etc). Control of the argon/oxygen dilution ratio has also been achieved, although lacks a robust controller. Overall, the system performed at a level adequate for preliminary data to be collected and submitted for grant reports, however work remains in improving signal noise reduction and data acquisition consistency.
Student: Matthew Cheung
Professor/Sponsor: Professor Dennis Lieu
Mentor: John Madura
Research Project Title: Optimization of the Design of Inductive Charging Systems for Electric Vehicles
Abstract:
The demand for new technologies to quickly, efficiently, and conveniently charge electric vehicles increases based on electric vehicle adoption by consumers. Inductive charging could be a promising alternative, replacing conventional methods of plugging in vehicles to top up on energy. This paper focuses on the fundamentals of how inductive charging systems work and how to increase the speed and efficiency of inductive charging systems.
Student: Johnathan Corvello
Professor/Sponsor: Professor Robert Dibble
Mentor(s): David Vuilleumier and Miguel Sierra Aznar
Research Project Title: Homogeneous Charge Combustion Ignition Engine Research
Abstract
Student: Jonathan Corvello
Professor/Sponsor: Professor Robert Dibble
Mentor: Miguel Sierra Aznar and David Vuilleumier
Research Project Title: Dynamic Skip Fire Engine Research
Abstract:
From working in Professor Dibble’s Combustion Laboratory this past semester, I have put my engine knowledge and capabilities to use through mechanical design and fabrication of custom components for the Cooperative Fuel Research (CFR) engine and VW engine in Hesse Hall.
The CFR engine will be combusting a liquid or gaseous fuel mixed with pure oxygen in closed loop, one cylinder cycle using Argon as the working fluid. The intent of this project is to discover more efficient ways of cleaner energy generation for power plants.
Currently the CFR engine test cell is finishing its re-construction with installation of new intake, exhaust, cooling and data acquisition systems, which is being lead by Miguel Aznar. I have helped with the manufacturing of the port intake system shown in Figures 1 – 3 as well as smaller components such as pressure sensor inserts, shown in Figures 4 and 5, that adapt the sensor to the existing engine housing. The custom VW fuel pump gearline I designed and constructed, shown in Figure 6, was ran successfully this semester, helping an international master thesis student measure an operating baseline of the diesel engine for a research paper.
In summary, the opportunity to join this lab’s research has reinforced what I have learned in my engineering classes. I look forward to continuing these combustion research endeavors.
Student: Johnathan Corvello
Professor/Sponsor: Professor Robert Dibble
Mentor: Miguel Sierra Aznar
Research Project Title: Internal Combustion Research
Abstract
Student: Davis Day
Professor/Sponsor: Professor J.Y. Chen
Mentors: Darren Sholes and Peter Therkelsen
Research Project Title: “Ring” Burner Combustion Research
Energy Science & Technology
Abstract:
The “ring” burner is a novel design for cook-stove burner plates that is aimed towards creating a more stable flame under lean combustion conditions (air-biased fuel mixture), resulting in a decrease of NOx emissions due to combustion. Ideally, this can be achieved with little to no drawback for the end user and minor changes in the manufacturing processes for stoves, burners, and of particular interest in this case, camping stoves. The research objective was to compare the performance of the ring burner with that of existing camp stove technology.
Tests were performed in accordance with ASTM standards for energy input rate and time-to-boil tests, first performing stabilization of the burner for each test and then repeatedly boiling 10lb pots of water. Parameters that were measured included the flow rate of propane into the burner, and the temperature of the water versus time. In order to prevent freezing of the propane cylinders due to the long duration of each test, heating tape was used in order to maintain a constant exterior temperature. Data analysis was performed for both the ring burner and the original camp-stove burner using a python script.
The ring burner showed a variance of about 5% in terms of time-to-boil, and a variance of under 5% in terms of cooking energy efficiency when comparing burner energy input rate and total heating energy absorbed by the water. Further improvements to the ring burner would involve simpler integration into the design of the burner assembly, and proper spacing of holes throughout the plate in order to facilitate even heating without inducing a warping about the boundary between the flame and the plate mounting.
Student: Nathan Duncan
Professor/Sponsor: Professor Dennis Lieu
Research Project Title: Lithium Ion Battery Charge Behavior
Abstract:
Due to the rising cost of oil and increased concern over carbon emissions, automakers have been researching, developing, and selling hybrid and electric vehicles (EVs) as an alternative to IC engine-powered vehicles. Current production hybrids and EVs typically use lithium ion batteries, but the use of lithium ion cells presents a problem. Charging is current limited due to the electrochemistry of the cell. Specifically, when a lithium ion cell is charged at an excessively high current, the resistive heating of the cell breaks down the solid electrolyte interphase. Once this happens, the unstable electrolyte exothermally reacts with the exposed negative electrode, creating a positive feedback loop. This state is known as thermal runaway and leads to exothermal decomposition of cell’s internal chemicals. The increasing heat and pressure will eventually rupture the cell casing, causing a possible vent with fire (VWF) event. If there are nearby cells, they can also be triggered into also undergoing thermal runaway. To avoid this catastrophic failure, manufacturers provide charging current specifications. Charging at these specified levels will not lead to thermal runaway, though gradual cell degradation will still occur. In order to circumvent this failure mode (which in turn limits charging current) there must be a change in cell chemistry. Battery manufacturers have developed a variety of chemicals for use as electrodes and electrolytes, such as lithium titanate electrodes. Other solutions include the use of non-metallic cells and mechanical energy storage devices.
Student: Khanh Trung Do
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentors: Casey Zak and James Urban
Subarea: Combustion
Research Project Title: Fuel Bed Ignition by Heated Particles: An Experimental and Phenomenological Study
Abstract:
According to the National Fire Protection Association of the United States, “outside and other” fires, or wildland and wildland urban interface, has caused more than $500 million dollars in property damage and killed 55 civilians in the year 2010 alone . Many of these fires are allegedly ignited by heated particles generated by power line interactions, welding and other sources of hot particles. However, there has been relatively little research on the ignition of fuel beds by hot particles. Our research focuses on the study of ignition of powdered cellulose fuel beds by hot metal particles. Stainless steel, brass and aluminum spheres, whose diameters ranges from 1.59 mm to 12.7 mm were heated to temperatures between 773 and 1373K in a furnace and immediately dropped onto cellulose fuel beds with moisture contents of 1.5% and 4.5%. The effects of varying particle diameter, temperature and thermal conductivity and fuel bed moisture content on flaming ignition propensity are discussed. Additionally, high-speed videos taken of three ignition events are presented and used in conjunction with phenomenological arguments to develop a description of the ignition process. The results of our works so far suggest that ignition of fuel beds by hot particles is a very rapid surface phenomenon that most strongly depends on particle size and temperature, with a possible dependence on fuel bed moisture content.
Before the ignition process occurs, the biomass must be broken down by pyrolisys. The products of this process are H2, CO, CH4, C2H4, levoglucosan, and a small amount of other hydrocarbon. Depending on the temperature, the ratio of the products will differ. As increasing the temperature, the amount of H2 increases while the percent of methane and levoglucosan decrease.
Student: Galen Elias
Professor/Sponsor: Professor Reza Alam
Mentor: Mohsen Saadat
Research Project Title: Design of a High Performance Ferrofluid Engine
Abstract:
The purpose of this project was to experimentally test the efficacy of increasing the heat transfer area in a Stirling engine by using porous, conductive material in conjunction with ferrofluid. The porous material increases the heat transfer area to the working gas by a factor of ten, and a fluid is necessary to travel through the porous material to displace the working gas. By using ferrofluid, we can control the fluid with a magnetic field. We were interested in seeing how the power density (power per-unit-volume of working gas) of the engine would change. Theoretical results show that it is possible to increase the power density of the engine by 5-10 times[4]. This would be a huge evolutionary step for Stirling engines since although they are relatively efficient engines, they suffer from a low power density, making them unviable options for compact power generation.
To test the theory, we built a setup (figure 1) that allows us to measure the pressure of a fixed volume of gas. Our experimental results show that it is possible to reach gains in power density from 1.5 to 2 times at varying temperatures. We also observed that due to the high heat capacity of ferrofluid, we could not maintain a high temperature on the hot side – the heat from the plates was quickly absorbed by the fluid each cycle, quickly driving down the average temperature of the gas.
This report is intended to document the design and fabrication process to create the test setup, as well as share and discuss experimental results.
Student: Galen Elias
Professor/Sponsor: Professor Reza Alam
Research Project Title: Load Shedding in a Wave Energy Converter
Research Areas: Energy Science and Technology, Ocean Engineering
Abstract:
Wave Energy Converters are devices which convert the renewable energy in ocean waves to electricity. A submerged pressure differential WEC uses a rigid absorber to split a wave’s orbital, creating a pressure gradient which drives a generator. One of the engineering challenges of WECs is to make the device robust enough to handle extreme ocean conditions, during which waves can carry upwards of 30 times more power than usual.1 As such, we looked into ways to reduce the load the device would experience under extreme conditions. Due to the high buoyancy of the device and the high-energy cost of increasing its depth, we focused mainly on the effect of changing the device’s shape. In particular, we analyzed trends in front-to-back hole placement and trends in wall thickness between holes within a constant footprint.
Student: Felix Sebastian Frank
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentor: Daniel Murphy
Subarea: Combustion
Research Project Title: Effect of Wind on Fires Inside Naturally Ventilated Buildings
Abstract:
A team from the Combustion Lab conducted scaled wind tunnel experiments to analyze the effect of wind on fires inside naturally ventilated atria or malls. Motivation of this experiment was to get a clear picture of the different hot gas layers and exhaust vent flows which develop in such a situation. Plenty of measurements were taken with the Schlieren digital photography, however, due to the fact that the analysis of the Schlieren data was only possible one-by-one, time constraints obviated the examination of the data. Consequently, we needed a way to efficiently assess the data. Therefore, we programmed a script, based on PIVIab, which is able to analyze complete packs of data all at once. This major improvement enables us to use the Schlieren digital photography in the future as a very efficient tool for the analysis of general fluid flows.
A second emphasis lies on the validation of the scale model technique. Scale Modeling is an engineering theory that is used to analyze systems where calculations and computer simulations are not reliable or at least very time-consuming. The theoretical approach is to build a downscaled version of the original system while maintaining the same geometry, kinetic and kinematic. Non-dimensional analysis delivers a proof for the similarity between these systems, as long as the governing parameters are kept constant when scaling down the reality. However, in practice, there have to be variations regarding the numerous non-dimensional parameters. As a result the theory of scale modeling needs to be validated to prove that an application is justifiable. Building various sizes of the same geometry and conducting experiments with exactly the same data proves the applicability and secures the competitive ability of scale modeling compared to the in business primarily used computer simulations.
Student: Zachary Hammond
Professor/Sponsor: Professor Robert Dibble
Mentor: Dr. Hunter Mack
Research Project Title: The Addition of Hydrogen Peroxide to Methane Fueled HCCI Engines Through Numerical Simulation
Abstract:
Although Homogeneous Charge Compression Ignition (HCCI) has proven to be an attractive internal combustion technique in attaining low emissions and high fuel efficiency, its ignition process remains dependent on chemical kinetics thus making it difficult to control. One promising approach to controlling combustion timing in HCCI uses two fuels to drive combustion. This study proposes that hydrogen peroxide may be a suitable additive to introduce as a means of control because of its role as a critical oxidizer in the initiation of thermal ignition, and its effect on indicated mean effective pressure and ignition timing. The authors employed hydrogen peroxide as a means of control by studying the effects of directly injecting a hydrogen peroxide solution into the combustion chamber of an HCCI engine through numerical simulation. A single cylinder of a methane fueled HCCI engine was modeled with MATLAB using the Cantera 2.0 flame code toolkit, the GRI-Mech 3.0 chemical reaction mechanism, and a single zone slider-crank engine model. The numerical simulation was used to develop relationships between the amount of hydrogen peroxide injected, the start of injection timing, the intake temperature, and the combustion timing.
It was found that the addition of hydrogen peroxide provided a significant advance in combustion timing. In small concentrations, hydrogen peroxide could induce combustion of methane under conditions that would otherwise result in a misfire. While holding intake temperature and SOI constant, varying the amount of hydrogen peroxide allowed for the control of combustion timing. To expand the range in which the injection of hydrogen peroxide provides a beneficial effect on combustion, injection timing was also utilized as an ignition control parameter. An advance in SOI while holding injected mass and intake temperature constant produced an advance in combustion timing. The use of hydrogen peroxide as a secondary fuel in HCCI can provide control over combustion timing while retaining the factors that make HCCI an attractive technique. The simulations showed that the addition of hydrogen peroxide has a slightly positive effect on the emissions of NOx and other emissions. The simulations also showed that addition of hydrogen peroxide has no discernible effects on peak pressures within the cylinder, which is important considering high peak pressures can damage the engine.
This preliminary experiment shows that hydrogen peroxide may be effectively introduced as a means of control in HCCI combustion. The application of internal injection of hydrogen peroxide has the potential to negate some of the difficulties associated with controlling the combustion event and expand the range of combustion regimes at which HCCI operates. This can be achieved while maintaining low emissions and peak in-cylinder pressures.
Student: Mitchell Heschke
Professor/Sponsor: Professor Carlos Fernández-Pello
Research Project Title: Material Flammability (NASA Space Flight Program)
Abstract:
This research investigates the effect of space “exploration atmospheres” on the flammability of flame retardant materials. Space exploration atmospheres are characterized by elevated oxygen levels, reduced ambient pressure, low flow velocities, and low gravity. In order to reduce pressurization and depressurization times during space walks and cargo transport, it is desirable to have a low-pressure environment. In order to have lower than atmospheric pressure and also remain safe for human inhabitants, the partial pressure or concentration of oxygen must be increased. With increased oxygen concentration, the risk of ignition, and subsequently combustion is greatly increased. By gathering data of flame spread rates and flammability limits under these conditions we gain a greater understanding of the safety of exploration atmospheres. Flame retardant materials are ignited in a vacuum chamber where pressure, oxygen concentration, and external heat flux are varied with a constant flow velocity. Tested flame retardant materials include Nomex and a woven fabric, which comprise the majority of space suit materials and fabrics found on space flight vehicles. In general, higher oxygen concentrations increase flame spread and increase the flammability range, external heat flux has minimal effect when compared to pressure, and reduced pressure exhibits the opposite trend of increasing oxygen.
Student: Jimmy Huang
Professor/Sponsor: Professor Carlos Fernandez-Pello
Research Project Title: The Effect of Fuel Moisture Content on Smoldering Ignition of Sawdust by a Firebrand
Abstract:
Firebrand spotting is an important mechanism by which wildland fires can rapidly spread. When the firebrands from a fire upwind land on a fuel downwind they may initiate flaming or smoldering, which later may transition to a flame in the fuel. The moisture content of fuel bed plays an important role in determining whether or not a firebrand will ignite the fuel. In this work, the effect of fuel moisture content on the smoldering ignition of sawdust by a firebrand is studied experimentally, flaming ignition was not studied or reported. The firebrands were made from wooden cylinders of various sizes from 2 to 12 mm in diameter and length. The cylinders were brought to glowing combustion with a flame and dropped onto the sawdust below. From these experiments we were able to determine a ‘moisture content ignition boundary’ where ignition would not occur at higher moisture contents, for a given firebrand size. The results show that large firebrands are capable of igniting sawdust with a higher moisture content, thus having a higher moisture content ignition boundary. The moisture content ignition boundary then decreased as the particle size decreased. We then found the smallest ember size which would ignite the sawdust with a moisture content of <1%. There was a range of moisture contents over which both smoldering and no ignition events occurred, due largely to the inherent variability in a natural fuel. This semester, progress was made by conducting additional smoldering tests specifically using larger diameter wooden firebrands. In addition, a four component instrumentation panel was designed and manufactured from scratch. This will significantly help organize and streamline implementation of common hardware used during these ignition tests.
Student: Zeqing Jin
Professor/Sponsor: Professor Costas Grigoropoulos
Mentor: Letian Wang
Research Project Title: Resistive Temperature Detector for Pulsed Laser Processing
Abstract
Student: Aman Khan and Alyssa Scheske
Professor/Sponsor: Professor Karl Hedrick
Mentor: Dr. Samveg Saxena
Research Project Title: GPS Hardware for Vehicle to Grid simulation and MyGreenCar system
Abstract:
MyGreenCar is an application in development in the Vehicle to Grid Simulation Lab, as a part of the Environmental Energies Technology Division at Lawrence Berkeley National Labs 1. The purpose of this project was to create a device to determine the accuracy of MyGreenCar by registering raw GPS data and comparing it to the data received from Google Maps on a cell phone. This would ultimately provide a sanity check to verify existing data produced by the Vehicle to Grid Simulation lab and provide a framework for its potential application in car and bus fleets. This report will cover the background of our reasoning behind selecting appropriate hardware and the methods used to construct a device that logs GPS data, which can then be passed to the MyGreenCar server, which produces a trip report. This report will also serve as a user manual for the hardware (GPS 1) in its current state as well as the technical difficulties I encountered while creating the device, and existing bugs. Current usage and future applications to validate data and implement in fleets will also be covered.
Student: Christiaan Khurana
Professor/Sponsor: Professor Robert Dibble
Mentor: David Vuilleumier
Subarea: Combustion
Research Project Title: Homogeneous Charge Compression Ignition
Abstract:
Two semesters ago, I developed an exhaust system for the Combustion Laboratory’s HCCI engine (Homogeneous Charge Compression Ignition). The main design goal of the system was to measure the exhaust gas properties (temperature, pressure, and composition) of a single cylinder of the four-cylinder engine. The exhaust from the measured cylinder was carefully routed such that flow disturbances and other sources of measurement inaccuracy were prevented. Last fall, my contribution to the HCCI engine was to develop a heat exchanger. This exchanger reduced the temperature of exhaust gasses before recirculating them back into the engine’s intake, thereby increasing temperature regulation during the combustion process.
Over the course of a year, I planned how each piece would theoretically fit together. This spring, I was excited to fulfill this commitment and complete my work. Thus, I have finalized the exhaust gas recirculation system (EGR) to unite the exhaust, heat exchanger, and engine intake. This enables more in-depth engine testing by determining which degree of exhaust recirculation is optimal for engine performance. Furthermore, I designed a valve actuator using a stepper motor and a chain drive. Using electronic controls, the valve can be opened to any desired position, allowing efficient access to varying degrees of exhaust recirculation versus exhaust disposal. I hope that these contributions will help promote and refine an environmentally beneficial solution such as HCCI.
Student: Harlan Kuo
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentor: Daniel Murphy
Research Project Title: Advanced Diagnostic of Scaled Compartment Fires
Abstract:
Compartment fires are areas where a flame is present in an enclosed space with one or more outlets. Analysis on such environments is needed to determine whether structures, such as large atriums, are safe enough to allow people to evacuate unharmed in the case of a fire. The computational models such as the Fire Dynamics Simulator can take immensely large resources and computer hours to calculate larger rooms making scaled modeling a much more attractive method of providing the needed analysis. Scaled modeling when combined with Particle Image Velocimetry and Background Oriented Schlieren produce a full set of data needed to verify if the room is safe. The objective of this research is to scale and model an atrium fire in a wind tunnel with similar conditions as to what an actual atrium must be tested in to be considered safe. A scaled down atrium will be placed in a wind tunnel seeded with droplets and the atrium will have a heat source seeded with aluminum oxide powder. A Nd YAG laser is used to illuminate the local area allowing for a camera to take measurements of the velocity field. A special pattern is then placed behind the atrium and imaged to detect for perturbation in the air due to refractions caused by hotter air generating a temperature field of the area as well. These combined measurements will produce all the relevant data to the model. This methods promises to be cheaper and faster than computational models of comparable accuracy.
Student: Gurshaan Madan & Stephen Chu
Professor/Sponsor: Professor Dennis Lieu
Research Project Title: Wireless Charging Applications for Electric Vehicles
Abstract:
In this paper, we will discuss the application of wireless charging to electric vehicles, in particular electric scooters in Asian markets. The increase in population densities in Asian financial centers has caused the emergence of electric scooters to replace their internal combustion counterparts, but the sales are lagging due to a lack of infrastructure. By researching wireless charging technology, we hope to develop a charging system that can be imbedded in the roads of Taipei. We will explain the need for the implementation of this technology and examine current companies and researchers who are developing and improving this technology. The general theory behind high-efficiency wireless charging involves resonant inductive coupling, which is the wireless transfer of electrical power through magnetic fields between two coupled coils from tuned resonant circuits. The key characteristics of such circuits are a transmitter circuit that stores energy in a coil’s magnetic field, and a physically separated receiver circuit that receives the energy in a receiver coil and converts it to an electric current, which can then be used to power a device. We will discuss the theory in greater detail, and outline a procedure and schematics to build a prototype wireless charging device next semester.
Student: Rob McAdam
Professor/Sponsor: Professor Dennis Lieu
Research Project Title: INSTAR HMI Touchscreen Display Integration
Abstract
Student: Tony Ngo
Professor/Sponsor: Professor Dennis K. Lieu
Mentor: Cyndia Cao and John Madura
Research Project Title: Model Development and System Identification of INSTAR’s Test Vehicle
Abstract:
This paper serves to create a basic dynamic model of the current that runs throughout the Inertia Storage and Recovery (INSTAR) vehicle such that data can be acquired and fitted to a transfer function that represents the entire closed loop system. Using the methodology of system identification, and therefore recording the input current and output current of every subsystem, we can tune a PID controller to monitor the current that runs through the battery, and every individual motor controller. Such testing procedure is described further within the paper, where it explains how step inputs are used to receive the transient and steady state behavior of each subsystem. Though, the work within this paper does not fully address the implementation of the closed loop controller within LabView, it can be the framework to replace the open-loop model that exists within the vehicle’s code. Through the implementation of the closed loop model, efforts can be made to improve battery life, while also addressing the current draw issues that limits the performance of the vehicle. Serving as the stepping stones of more advance current controllers as well, the transfer function created can be used to optimize current flow during the different transient phases that exist while the vehicle is running. The creation of such a model can then be scaled and used to implement and optimize the concept of a triple hybrid system within a passenger vehicle.
Student: Eric Olson
Professor/Sponsor: Professor Robert Dibble
Mentor: Dr. Hunter Mack
Sub Area: Combustion
Research Project Title: Investigating the use of Compressed Natural Gas in a Variable Displacement Spark Ignition Engine
Abstract:
With the number of vehicles on the road today the need for improved emission control is readily apparent. The modern spark ignition engine is designed to be most efficient when operated at full load, but normal operating conditions rarely achieve this type of loading. Instead, most engines operate at loads significantly lower than their designed capacity, which increases their contribution of harmful exhaust emissions to the environment. In order to overcome partial-loading conditions, industry has developed “Variable Displacement” engines that dynamically deactivate certain cylinders and effectively lower the displacement of the engine, which allows the remaining functional cylinders to operate at nearly full-load conditions.
This research utilizes Dynamic Skip Fire (DSF), which is the newest in Variable Displacement Technology developed by Tula Technology Inc. DSF technology shuts down individual cylinders for improved performance under partial-load conditions. In addition to utilizing variable displacement, lowered emissions can also be achieved through the use of alternative fuels, such as Compressed Natural Gas (CNG). CNG offers favorable engine-fuel properties, so CNG-fueled vehicles produce lower levels of all pollutant emissions than either conventional or reformulated gasoline and diesel fuels. The goal of this research is to operate a Dynamic Skip Fire equipped GM L94 6.2 Liter V8 on Compressed Natural Gas to determine the overall improvement to emissions achieved through this coupling.
The first phase of the project was the construction of an engine test cell and the breaking in of the engine. Specific tasks assigned to the undergraduate assistants included fabrication of brackets and mounting hardware in the student machine shop, wiring and calibration of sensors between the interior of the test cell and the controlling computer system, and installation of the heat exchanger for the fuel delivery system. Once the engine test cell construction was completed, a start-up procedure was developed and the engine was test fired. A series of runs were then conducted to break in the engine, and data was collected and analyzed to verify that frictional losses would not interfere with future experiments. These preliminary tests have been completed, and the engine will now begin to collect data for examination into the benefits of DSF run on CNG.
Student: Sivam Paramanathan
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentor: James Urban
Research Project Title: The Effect of Fuel Moisture Content on Smoldering Ignition of Sawdust by a Firebrand
Abstract:
Wildfires cause problems, especially in California, where they endanger lives and lay waste to people’s homes. A better understanding of how they spread can help combat them. One mechanism by which they spread is firebrand spotting. When flaming particles, also known as firebrands, land on a fuel downwind they may initiate flaming or smoldering, which later may transition to a flame in the fuel. This allows the fire to travel at the speed of wind. With the rapid air currents, the fire can spread out of control. The moisture content of fuel bed plays an important role in determining whether or not a firebrand will ignite the fuel. The question becomes: how moist does a fuel bed have to be to prevent the ignition from a firebrand of a given size. For this experiment, the effect of fuel moisture content on the smoldering ignition of sawdust by a firebrand is studied experimentally, flaming ignition was not studied or reported. The firebrands in this experiment are wooden cylinders of various sizes from 2 to 12 mm in diameter and length. The firebrands were brought to glowing combustion with a blowtorch and dropped onto sawdust of varying moisture contents. By recording when smoldering occurred vs. when there has no ignition in the fuel bed, we were able to determine a ‘moisture content ignition boundary’ where ignition would not occur at higher moisture contents, for a given firebrand size and wind speed. Intuitively, the results show that large firebrands are capable of igniting sawdust with a higher moisture content, therefore having a higher moisture content ignition boundary. The moisture content ignition boundary decreased as the particle size decreased. To establish the bare minimum required firebrand size, we found the smallest ember size which would ignite the sawdust with a moisture content of <1%. Near the boundary, there was a range of moisture contents over which both smoldering and no ignition events occurred, due largely to the inherent variability in a natural fuel. The results seem straightforward, but the quantitative data gathered could help protect people and their property from wildfires.
Student: Sivam Paramanathan
Professor/Sponsor: Professor Fernandez-Pello
Mentor: James Urban
Research Project Title: Firebrands
Abstract:
The ignition of combustible material by hot metal particles is an important pathway by which wildland and urban spot fires are started. Upon impact with combustible material (e.g. vegetation, cellulosic industrial material or polymer foams), these particles can initiate spot fires. In spite of interest in the subject, there is little work published that addresses the ignition capabilities of hot metal particles landing on natural fuels. This work is an experimental study of how the flaming ignition propensity of fuel beds in contact with hot aluminum particles is affected by the characteristics of the fuel bed. Two fuel beds were tested: pine needles and a fine powder formed by grinding the pine needles which are representative of forest litter and duff respectively. Litter constitutes things like whole leaves, whereas duff constitutes decomposed or trampled organic material. Comparing the ignition characteristics of these fuels will give insight into the effects of fuel macrostructure on the conditions which could initiate spot fires from metal particles. In the experiments, aluminum particles ranging from 2 – 8mm in diameter are heated to various temperatures between 575 – 1100oC and dropped into the different fuel beds. The results show that the pine needle powder fuel was capable of ignition at lower temperatures, but the flame spread faster on the pine needle fuels.
To achieve this, many preparations had to be taken. The particles were hand cut to precise masses, fuel beds were carefully made to be the right consistency, and laboratory conditions were meticulously recorded. After data was gathered, computer algorithms were used to analyze it and determine which tests still needed to be conducted.
Student(s): Alyssa Scheske
Professor/Sponsor: Professor George Johnson
Mentor: Dr. Samveg Saxena
Sub Area: Green and Sustainable Technologies
Research Project Title: GPS Data Comparison and Confirmation
Abstract:
The goal of this research was to analyze and scrutinize location data points along a driven route. These points would be taken via phone through the lab group’s star and shinning app named “My Green Car” and a GPS device simultaneously. Both data sets would be sent to the server hosted at LBNL and run through a program called “V2G Sim”. In return, the user receives an email with their trip displayed as a graph. Additional analytics are attached, including trip duration, state of charge, total distance traveled, ect. The projects “My Green Car” and “V2G Sim” are longstanding undertakings at LBNL by Samveg Saxena, Ph.D. It is hoped to promote electric vehicles with these projects.
The purpose of this research was to create an installable and portable independent GPS device to measure and compare against the following: distance, speed, time step, and change in altitude. A huge challenge for the project included programming altitude data into the calculation using applicable elevation APIs.
The implications of these results lay under two domains. The first domain would include phone (Android and iPhone) result verification. The more tangible and practical application of these results applies to the fleet services transportation sector. The project created a complete procedure, software (written in bulk by Aman Kahn), and hardware device. With a working prototype and data results, the project may enter the exciting next stage of development for users.
Student: Parsa Taleb
Professor/Sponsor: Professor Reza Alam
Research Project Title: An Approach To Analyze Environmental Noise Pollution Via Smartphone Mass Data Collection
Student: Tuong-Vi Tran
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentors: Casey Zak and James Urban
Sub Area: Combustion
Research Project Title: An experimental and phenomenological study of powdered fuel bed ignition by heated particles
Abstract:
Since ancient times, wildland fires have been a threat to both human life and property as well as the environment. Many fires are allegedly ignited when incandescent particles, among them are hot metal particles from electrical arcing of power lines or welding, land in combustible fuels beds, such as saw dust or forest floor. So far, there has been little research done about the impact of hot metallic particles on cellulose fuel bed, thus our research aims to discover a better understanding of this ignition pathway to assist regulatory agencies in predicting causes of wildland fires, as well as preventing it from happening. The experiment was setup so that a metal particles, such as aluminum, stainless steel or brass spheres, get heated inside a furnace until they reach the desired temperature, then are promptly withdrawn from the furnace and drop onto a fuel bed, whose properties were also being measured. The variety of particle mass, diameter, temperature, and material as well as thermal conductivity and fuel bed moisture content were discussed and analyzed. This research is continuing from Fall 2012, using the same methods but also adding the use of thermocouples, implemented on the spoon that holds the particle, to ensure it reaches desired temperatures. High-speed videos captured the moment when the spheres landed into the fuel bed, thus enabling us to determine whether it is flaming, non-flaming or smoldering ignition as well as study their ignition behaviors. Our work so far has indicated the trend that separate the combinations of mass and temperature where a certain metal particle will ignite the fuel bed from those where the fuel bed was not being ignited at all. We concluded that the ignition of a fuel bed is a rapid phenomenon that strongly depends on particle mass and temperature.
In Fall 2013, our research continued using the same methods but focusing on perfection the ignition curve for steel, copper and brass particle. Instead of using the relatively small, less than 3.18 mm, and relative big, larger than 10.5 mm, we conduct tests based heavily on metal ball with diameter ranges from 3.96 mm to 7.92 mm. Our work so far has indicates that the ignition trend we predicted based on relative small and large particles also hold for medium size particles. We also did experiment with aluminum particles using same method but dropping on a barley hay fuel bed instead of cellulose. Our results so far indicated that aluminum particle fully ignite at higher temperature when dropping on barley hay fuel bed due to barley hay’s moisture content is larger than cellulose’s.
Student: Daniel Weeks
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentor: James Urban
Research Project Title: Spot Fire Ignition of Natural Fuels by Hot Metal Particles
Abstract:
Metal particles expelled from metal-on-metal contact, such as railways, clashing powerlines, engines, welding, or other industrial work, often have dangerously high temperature levels. In numerous cases, these particles have been cited as ignition sources for wildfires. In this research, the threshold of size and temperature of steel and aluminum particles for ignition of grass powder fuel beds is studied. Natural fuel beds are simulated by powdered wheat, oat, and barley grasses. Spherical particles with diameters between two and eight millimeters are heated to temperatures ranging from 550 to 1100 degrees Celsius and dropped on the grass powder fuel bed. Although flaming ignition occurs almost immediately, evidence of a smoldering reaction might not be observable for up to five minutes, so the temperature of the fuel bed is monitored with an infrared camera. A camera with infrared imaging overlaid on visible light images takes photographs of the fuel bed at three second intervals to record the path and rate of growth of smoldering reactions. The results show a hyperbolic relationship between the size of the metal particle and the temperature that yields ignition. For each particle diameter, the flaming ignition temperature is greater than the smoldering ignition temperature, but both ignition boundaries follow the same hyperbolic trend. The ignition thresholds for smaller particles are more sensitive to changes in size and temperature than for larger particles; this corresponds to a steeper slope on the temperature-diameter plot. As such, the thresholds for smoldering ignition and flaming ignition occur at closer temperatures for smaller particles than for larger particles. In cases of smoldering ignition, the reaction front initially progresses radially outward from the dropped particle in all directions, but continue outward in only one or two opposing directions. In addition, conditions for flaming ignition appear to extend beyond simply the size and temperature of the dropped particle. Different landing behaviors, such as bouncing or rolling, are observed to affect the likelihood of flaming ignition.
Student: Daniel Weeks
Professor/Sponsor: Professor Carlos Fernandez-Pello
Mentor: James Urban
Research Project Title: The effect of fuel moisture content on smoldering ignition of sawdust by a firebrand
Abstract:
Firebrand spotting is an important mechanism by which wildland fires can rapidly spread. When the firebrands from a fire upwind land on a fuel downwind they may initiate flaming or smoldering, which later may transition to a flame in the fuel. The moisture content of fuel bed plays an important role in determining whether or not a firebrand will ignite the fuel. In this work, the effect of fuel moisture content on the smoldering ignition of sawdust by a firebrand is studied experimentally, flaming ignition was not studied or reported. The firebrands were made from wooden cylinders of various sizes from 2 to 12 mm in diameter and length. The cylinders were brought to glowing combustion with a flame and dropped onto the sawdust below. From these experiments we were able to determine a ‘moisture content ignition boundary’ where ignition would not occur at higher moisture contents, for a given firebrand size. The results show that large firebrands are capable of igniting sawdust with a higher moisture content, thus having a higher moisture content ignition boundary. The moisture content ignition boundary then decreased as the particle size decreased. We then found the smallest ember size which would ignite the sawdust with a moisture content of <1%. There was a range of moisture contents over which both smoldering and no ignition events occurred, due largely to the inherent variability in a natural fuel. My role as an undergraduate researcher consisted mostly of running the experiments and recording the results. After dropping an ember into the sawdust fuel bed, I would monitor the test until the smolder spread to a size where it was self-propagating (indicating smoldering ignition), or the temperature of the cylinder dropped below 100 degrees Celsius (indicating no ignition). I also designed and machined guides for cutting wooden dowels to our set lengths so that we could quickly produce high quantities of cylinders of the correct size.
Student: Michael Wiley
Professor/Sponsor: Professor Robert Dibble
Mentor: Russell Labrie
Sub Area: Combustion
Research Project Title: Benefits of Dynamic Skip Fire (DSF) for Improved Natural Gas Engine Performance
Abstract:
Over the past two semesters I have had the privilege of working in the Combustion Analysis Lab alongside a select group of graduate and undergraduate students on the “Benefits of Dynamic Skip Fire (DSF) for Improved Natural Gas Engine Performance” project. At its heart, this project’s purpose was to demonstrate the possible “fuel economy gain, which results from the combination of Dynamic Skip Fire technology and compressed natural gas”[1]. While the main objective of the project was realized, the data that was collected during the testing phase came at a time prior to the completion of the engine’s break-in. As a result, the conclusions drawn from this data needed to be verified in order to assess their validity. It was this task that myself and the other members of the research team set out to begin this semester.
This semester started out with myself and the other members of the team becoming familiar with the start up procedures and data acquisition system for the engine we would be working with, a refurbished General Motors 6.2L V8. Once familiarized, our task was to run the dynamometer and data acquisition system in order to map the engine’s parameters (focusing especially on the Friction Mean Effective Pressure) at specific engine conditions. The goal was to conduct these tests in order to fully break-in the engine with the intent of using FMEP as a guide to track when full engine break-in had occurred. At this point in time, the data acquisition portion of our project remains incomplete. Hopefully the research team and I will be able to continue our research during the summer.
[1] M. S. Aznar, Dynamic Skip Fire Optimization for a V8 Natural Gas SI Engine. Master Thesis Jan 20, 2014.
Student: Albert Zhou
Professor/Sponsor: Professor J.Y. Chen
Mentor: Vi Rapp
Research Project Title: Performance Characterization of Tankless Water Heaters
Energy Science and Technology
Abstract:
Current tank water heaters are extremely inefficient because of the need to continually heat a large quantity of water. Tankless, or “on demand” water heaters attempt to remedy this problem by turning on only when water begins running and heating according to load. However these water heaters are expensive due to complex burners and their controls. The upsides of Low Swirl Burners (extremely lean flames producing low ultra-NOx, ease of manufacture (3D-Printing), and high turndown) make it a very attractive solution to water heater manufacturers. If successful, the LSB has the potential to greatly reduce wasted energy in millions of households throughout the country. In this project, a burner performance baseline including emissions results, btu output, and water temperature monitoring for three tankless water heaters were established as a benchmark for the design of a Low Swirl Burner (LSB) to replace the current burners and controls in tankless water heaters.