CALCE EPSC Graduate Student Theses and Dissertation Abstracts (2011)

Boteler, Lauren (Ph.D. Mechanical Engineering)
Microfabrication And Analysis Of Manifold Microchannel Coolers For Power Electronics
This research presents the analysis and realization of a single phase high performance manifold microchannel cooler for improving the thermal and hydrodynamic performance of multi-chip power electronic modules. This heat exchanger, microfabricated directly into the substrate, enables higher power density electronic products by more efficiently removing the high levels of heat generated. The improved thermal performance and efficiency of the heat exchanger is demonstrated using both numerical and experimental techniques. The improved heat removal is due to the reduction in the number of packaging layers between the device and the heat exchanger and by improvement in convective heat transfer. In addition, the efficiency of the device is enhanced by minimizing fluid pressure drop through the use of large manifold channels to transport fluid to the cooling area and smaller crossover microchannels in the active cooling area. This combination of channels also improves the uniformity of the temperature distribution across the device. The manifold microchannel coolers were fabricated and tested both with and without electrical isolation between the chip and the coolant. Experimentally, the coolers without electrical isolation demonstrated thermal resistivity values as low as 0.06 K/(W/cm2), which is up to a 50X improvement over a standard power package with significant size and weight reduction. The coolers with an incorporated aluminum nitride electrical isolation layer experimentally demonstrated up to a 15X improvement. In addition to experimental results, the interaction between the manifold channels and multiple microchannels was numerically modeled and compared to simpler, one-dimensional approximations based on the Hagen-Poiseuille equation. The comparison shows that the one-dimensional model, while under-predicting total pressure drops, can provide insight into the effect of varying dimensions on system performance. The numerical models were used to identify the impact of varying dimensions across the entire length of the cooler, and a sensitivity analysis was performed with respect to system pressure drop, thermal resistance and uniformity. Additionally, large microchannel velocity gradients, some larger than 10X, were observed along the length of the device which impacts the chip non-uniformity. The simulations showed that when comparing the manifolded design to a comparable straight microchannel cooler, there is a 38X reduction in system pressure drop for similar thermal performance.

Mark, Paulus (Ph.D. Mechanical Engineering)
Fatigue Damage Accumulation Due to Complex Random Vibration Environments: Application to Single-Axis and Multi-Axis Vibration
A combination of experiments and modeling are used to address the vibration durability of structures subjected to different random vibration environments. Presented in this work are a set of experimental data comparing the rate of change of the first natural frequency and the measured time to failure, of simple structural members under repetitive shock (RS) vibration, single-axis electrodynamic (ED) vibration and multi-axis ED vibration. It was found that multi-axis testing is more severe than single-axis testing at the same level. In addition the RS system low frequency amplitude is often too weak to efficiently propagate the crack. Smoothing of the input power spectral density (PSD) or poor line resolution was also shown to change the time to failure of a test. A poor correlation was shown between the PSD and the rate of natural frequency change (RFC) over a wide frequency shift. The change in natural frequency caused the initial PSD to be ineffective in determining the total time to failure. A predictive, analytic methodology to quantify the RFC was developed to predict the fatigue life of a structure experiencing random vibration excitation. This method allows the estimation of fatigue life using the frequency domain, where only the input power spectral density, damping factor and structural information are required. The methodology uses linear elastic fracture mechanics for fatigue crack propagation and accounts for the frequency shifting that occurs due to fatigue crack evolution. The analytic model has been shown to compare favorably to both finite element analysis (FEA) and experimental results, for mild-steel cantilever beams. Monte Carlo simulations have been conducted to assess the sensitivity of the model predictions to uncertainties in the input parameters. In addition a semi-empirical model was developed whereby the input PSD and damping factor are measured from life tests, and the resulting time to failure and the acceleration factors between different vibration environments can be determined. The improved modeling methodology developed by this work are of value not only to structural designers who wish to design for dynamic environments, but also to test engineers who wish to qualify products through accelerated life testing, and to vibration engineers who wish to compare the relative severity of different random vibration environments, in terms of their potential to cause fatigue damage accumulation.

Cheng, Shunfeng (Ph.D. Mechanical Engineering)
Prognostics and Health Management
During my Ph.D study, I have finished several projects including: 1) the selection of sensor systems for prognostics and health management: a guideline is generated to help users to select sensor systems for prognostics applications; 2) Prognostics for aging systems: the methodology and algorithms are developed to detect the anomaly of aging systems and to identify if the detected anomalies are duo to natural aging; 3) Prognostics for multilayer ceramic capacitors using data-driven methods: data-driven methods are developed and tested to detect the anomalies exhibiting in the monitored data and provide advanced warnings; 4) Failure tracking for medical devices. This project was worked with Food and Drug Administration to use codes to track failure of medical devices. Now I am working on prognostics for polymer positive temperature coefficient (PPTC) resettable fuses. PPTC resettable fuses are made of semi-crystalline polymer and conductive fillers. As a circuit protection component, the failure of the PPTC fuses will damage the circuit or cause to the improper operation of the circuit. In this research, the failure mechanisms of PPTC resettable fuses are been identified and the failure models are in development. The remaining useful life will be predicted based on the combination of the failure models and the data-driven methods.

Crandall, Mike (MS)
High Temperature Lead Free Solder
In 2009 CALCE identified the development of a high temperature lead free solder that minimizes interfacial IMC growth and thereby maximizing reliability as one of the greatest needs in high temperature electronics.  Studies correlating IMC thickness to reliability at elevated temperatures is lacking.  This study assesses the reliability of 2 commercial lead free solders and 3 novel lead free solder ideas by using steady state aging and a combined thermal and vibration cycling.  During cycling samples are removed for shear strength testing and cross sectioning to measure the IMC layer at the joint interface.  All solders use ENIG plated polyimide test boards.

Haddad, Gilbert (PhD)
Decision Support for Systems with Prognostic Capabilities
Developed maintenance options; and methodologies to quantify them. After a prognostic indication is obtained, the decision-maker is faced with multiple maintenance options. Quantifying such options results in a system-level value of PHM. Beyond my PhD dissertation, I worked on data-driven prognostics (machine learning and data mining) and decision support systems.

Jazouli, Taoufik ( PhD)
Design for Availability and PHM Cost Modeling
Development of a new “design for availability” methodology that starts with an availability requirement and uses this requirement to predict the necessary design (e.g., reliability), logistics and operations (e.g., spares inventory and maintenance planning) parameters. This includes the life cycle cost and return on investment prediction for the specific availability requirement. The method is general and can be applied when the inputs to the problem are uncertain. The method has been demonstrated on several examples with and without PHM.

Sharon, Gil (Ph.D. Mechanical Engineering)
Improved Flex Cracking Calculator and Failure Analysis for Multilayer Ceramic Capacitors
Develop improved failure analysis techniques for MLCCs with low insulation resistance, and apply these to capacitors from prior temperature-humidity-bias and storage tests. Use finite element modeling to analyze factors affecting cracking of flexible and standard termination capacitors and extend the CALCE flex cracking calculator.

Sotiris, Vasilis (PhD)
Stochastic Processes, Probability Theory, and Data Driven Prognostics
My research is centered on the area of Prognostics and Health Management (PHM) applied to electronic products and systems. The theoretical background is based on statistics and probability theory as well as computational science. One part of the research focuses on Support vector machine (SVM) and Gaussian process (PG) classification in development of Anomaly detection algorithms. I am interested in the Bayesian machine learning setting where non-parametric SVM function estimators can be used as priors to improve more general Gaussian process estimation models. The other part focuses on nonparametric and semi-parametric statistical estimation theory for survival and reliability problems related to electronics life time and survival data. Here event driven degradation processes are modeled using stochastic counting processes and Cox semi-parametric proportional hazard models. In this context I am interested in a) the competing risks environment of the event type random variables b) the likelihoods for censored, truncated and time dependent events c) the product limit estimator for the transition matrix in non homogeneous Markov Process models and d) inference in low population data sets.

Williard, Nicholas Dane (MS)
Degradation Analysis and Health Monitoring Lithium Ion Batteries
Degradation and health monitoring of lithium-ion batteries is explored through life-cycle testing and failure analysis. Test samples comprised of four different battery types from two different manufactures underwent aging and charge/discharge cycling using a variety of load profiles including constant current discharge, pulsed discharge, and varying depths of discharge. Data from in situ monitoring of several parameters including current, voltage, temperature and internal resistance, was analyzed in order to find the best features that could be used to track and characterize battery performance degradation. Degraded samples were disassembled according to a newly developed disassembly methodology that considered the effects of the environment on post-disassembly failure analysis results. Several different failure analysis methods were used in order to gain an understanding of how degradation mechanisms propagate from a materials stand point. Battery electrodes were investigated to observe changes in their chemical and mechanical structures.

Copyright © 2011 by CALCE and the University of Maryland, All Rights Reserved