Dissertations & Theseshttp://hdl.handle.net/10106/248272024-03-28T19:53:38Z2024-03-28T19:53:38ZInvestigation of Thermal Performance Enhancement and Thermo-mechanical Assessment of ITE using Single-Phase Immersion Cooling Technologyhttp://hdl.handle.net/10106/318002023-11-09T22:34:52Z2023-08-25T00:00:00ZInvestigation of Thermal Performance Enhancement and Thermo-mechanical Assessment of ITE using Single-Phase Immersion Cooling Technology
**Please note that the full text is embargoed until 08/01/2024** The use of air-cooling as a thermal management technique in data centers has consistently maintained its importance, however its efficacy is limited to CPUs with lower power requirements. The expenses and energy use related to the air circulation process for high-power density racks are deemed excessively costly. The limits of air cooling have aroused concerns owing to its low specific heat capacity and poor thermal conductivity, which result in reduced effectiveness. Consequently, a substantial discourse has emerged about alternate and effective cooling methods that provide supplementary advantages, such as the recuperation of waste heat. In light of the limitations associated with conventional air-cooling approaches, several operators of data centers are embracing inventive methodologies to alleviate the consequences caused by escalating power densities. A solution gaining traction for high-power IT equipment is immersion cooling. A comparison between forced convection air cooling and single-phase liquid immersion cooling (Sp-LIC) highlights substantial advantages in the latter. The direct contact of dielectric fluids with components enables higher heat dissipation.. This technique enhances IT equipment reliability by safeguarding against pollutants and harsh conditions. Furthermore, Sp-LIC reduces capital expenditure (CapEx) and energy costs by eliminating the necessity for fans and computer room air handler units.
The objective of this investigation are divided into two parts: thermal efficiency of immersed information technology equipment (ITE) and operational efficiency (overall life cycle of fluid, reliability of components, and serviceability). The thermal part of the study is further divided into 3 section: In first section, an in-depth numerical study is performed which explores multi-parametric optimization for heat sinks in forced convection within an open compute server design when immersed in single phase immersion fluid. Optimization at constant pumping power iterates pressure drop and thermal resistance minimization as objective functions. Varying fin count and fin thickness for a constant base thickness in aluminum heat sinks reveals the correlation between geometric parameters and objective functions. These results contribute to standard methodologies for optimizing heat sinks in immersion cooling. In the second section of the thermal study, a systematic empirical study is undertaken to examine the heat transfer and pressure loss characteristics of aluminum metal foams while immersed in a dielectric synthetic fluid. The research used metal foam specimens with core heights of 0.0127m (0.5 in), varying relative densities between 10.7 and 12.3 percent, and pores per inch (PPI) values of 5, 10, 20, and 40. The metal foams underwent exposure to several flow rates, heat fluxes and inlet fluid temperature. In the final section of the thermal study, the experimental methodology for obtaining the velocity fields using Particle Image Velocimetry (PIV) in a single-phase immersion tank is discussed. PIV is a non-intrusive optical measurement technique used to visualize and measure fluid flow velocity patterns. The velocity field data obtained from these experiments provides valuable insights into the fluid flow behavior and aids in understanding the thermal performance of the immersion cooling system.
The second part of the research aims to investigate the impact of immersion cooling on server reliability. While immersion cooling excels in thermal energy management compared to conventional air-cooling, there have been limited studies on its reliability. The assessment of material characteristics, such as modulus and glass transition temperature, has substantial significance in the mechanical engineering of electronic components. The substrate, which is an essential element of an electronic package, significantly impacts the dependability and failure mechanisms of electronics, both at the package and board levels. The use of established material compatibility tests, such as ASTM 3455, is applied with appropriate adjustments in compliance with the design recommendations for immersion-cooled IT equipment outlined by the Open Compute Project (OCP). The main objective of this research is to investigate the effects of thermal aging on the thermo-mechanical properties of various substrate cores when immsered in single-phase dielectric fluids. This research examines the effects of aging on the substrate core by subjecting it to synthetic hydrocarbon fluid (EC100), Polyalphaolefin 6 (PAO 6), and ambient air for a duration of 720 hours. The aging process is conducted at two different temperatures, namely 85°C and 125°C. The complex modulus and the glass transition temperature of the substrate core are then determined and compared before and after the aging process.
2023-08-25T00:00:00ZANALYZING THE THERMOMECHANICAL PERFORMANCE OF TG400G MATERIAL SUBSTRATE CORE UNDER IMMERSION COOLINGhttp://hdl.handle.net/10106/317902023-11-09T22:37:02Z2023-08-15T00:00:00ZANALYZING THE THERMOMECHANICAL PERFORMANCE OF TG400G MATERIAL SUBSTRATE CORE UNDER IMMERSION COOLING
The relentless surge in demand for seamless information exchange through consumer electronics, driven by the indispensable role of the Internet, has given rise to an unprecedented need for data centres. Yet, the energy consumption of conventional data centres, where a significant one-third of energy usage is attributed solely to cooling, has triggered an urgent quest for energy-efficient solutions. Immersion cooling technology appears as a promising contender due to its exceptional prowess in managing thermal energy. However, its potential impact on the reliability of IT equipment needs a more profound exploration before widespread adoption can be realized.
This study embarks on a focused mission: to unravel the intricate effects of thermal aging on the thermo-mechanical attributes of low loss printed circuit boards (PCBs), specifically homing in on the TerraGreen 400G variant, within ambient air conditions. The investigation subjects these low-loss PCBs to varying temperatures (85°C and 125°C) and durations (720 hours) of thermal aging, both within EC100 and PAO6 environments. By meticulously scrutinizing alterations in complex modulus and Glass Transition Temperature (Tg) before and after aging, the study endeavours to unearth any shifts in the material's fundamental properties.
Anticipated outcomes of this research stand to give invaluable insights into the dependability and adaptability of TerraGreen 400G PCBs within immersion cooling scenarios. Such insights hold profound implications for the relentless pursuit of energy-efficient and environmentally considerate data centres. Moreover, the study's findings promise to cast a luminous beam on the terrain of electronics mechanical design by illuminating material behaviour amidst the rigors of thermal aging. In a world propelled by digital expansion, this investigation serves as a beacon, illuminating pathways to both greener data infrastructure and a more profound comprehension of materials under demanding thermal conditions.
2023-08-15T00:00:00ZEVALUATION AND OPTIMIZATION OF THERMAL SOLUTIONS IN AIR AND LIQUID COOLING SYSTEMS FOR DATACENTERShttp://hdl.handle.net/10106/317832023-11-09T22:38:34Z2023-08-28T00:00:00ZEVALUATION AND OPTIMIZATION OF THERMAL SOLUTIONS IN AIR AND LIQUID COOLING SYSTEMS FOR DATACENTERS
**Please note that the full text is embargoed until 8/1/2024** ABSTRACT: A data center is a specifically constructed establishment designed to accommodate computing, storage, and transmission equipment, thereby facilitating essential business operations spanning diverse sectors of the economy. This facility serves as the designated physical infrastructure for housing servers, networking components, and data storage systems.
In the context of traditional data centers, a substantial portion of energy, specifically around 45% to 55%, is directed towards supporting information technology (IT) operations. Additionally, a considerable allocation, varying from 30% to 40%, is attributed to the facilitation of cooling processes.
The scope of this dissertation study comprises nine distinct chapters, each contributing to addressing the challenges to establish energy-efficient and optimized data centers. The research commences by focusing on energy-efficient control strategies for hyperscale data centers through scaled-down Computational Fluid Dynamics (CFD) modeling. This approach involves validating the model against real data center datasets and conducting simulations tailored to specific needs. Leveraging machine learning techniques, this investigation aims to anticipate control setpoints for optimizing airflow and chilled water control at the data center room level. As the research progresses, the study addresses the limitations of CRAH units in sustaining high-density rack setups. Consequently, the feasibility and energy implications of implementing Rear Door Heat Exchangers (RDHx) are explored. This includes modeling a single rack with RDHx, assessing both passive and active modes. Moving forward, the study expands to assess the energy efficiency of RDHx across data center deployment.
Furthermore, in response to the challenges posed by hybrid cooled data centers, an investigation delves into the heat capture ratio for servers equipped with cold plates. The study quantifies heat capture by different mediums, considering varying temperatures and flow rates. It also delves into the impact of immersing hybrid cooled servers under diverse ambient conditions.
The following research also evaluates strategies to optimize heat sinks for immersion cooling environments, considering a wide range of fluids and design parameters. This includes simulation-driven optimization of cold plates' configurations, enabling improved performance for processors characterized by high Thermal Design Power (TDP). Moreover, the study delves into the enhancement of air-cooled heat sinks for single-phase immersion cooling setups, considering fluid thermo-physical properties for optimal design.
Through these interconnected chapters, the dissertation endeavors to provide a comprehensive understanding of energy-efficient and optimized data centers, encompassing a spectrum of strategies and technologies aimed at addressing the challenges associated with increasing power densities in modern chip architectures.
2023-08-28T00:00:00ZHigh Magnification Surface Topography based Digital Image Correlation (DIC) for Identifying Damage Accumulation and Crack Growth in Polycrystalline Nickelhttp://hdl.handle.net/10106/317242023-11-09T22:50:55Z2023-08-09T00:00:00ZHigh Magnification Surface Topography based Digital Image Correlation (DIC) for Identifying Damage Accumulation and Crack Growth in Polycrystalline Nickel
Damage index parameters that have been formulated to identify damage localization in plastically deformed materials typically require all six stress or strain components. Recently, simulations through Crystal Plasticity Finite Element Method (CPFEM) have been widely used to extract damage initiation parameters. However, these models are rarely verified due to the challenges in measuring the out-of-plane deformation/strain experimentally. A damage index combining the effective plastic strain and surface roughness change is investigated for identifying damage accumulation sites and predicting crack propagation path in polycrystalline pure Nickel. A novel Digital Image Correlation (DIC) technique is developed to measure both in-plane and out-of-plane deformation from surface topography images acquired using optical interferometry. A simple technique that creates random, micro size reflective speckles for sub-grain strain calculation is demonstrated. Damage accumulation sites detected by the effective strain, surface roughness change, and the combined damage index are assessed in terms of damage localization and localization consistency. The combined damage index provides an enhanced damage localization and localization consistency compared to effective plastic strain or surface roughness damage index alone. The detected damage accumulation sites were correlated with grain orientations that favor “sunken” out-of-plane deformations and large misorientations among neighboring grains. The proposed combined damage index is then used to predict the future propagation path of microstructurally small crack (MSC) in a fatigue sample. Effective plastic strain, surface roughness and combined damage index maps were constructed during the crack arresting period. The crack future propagation path was then predicted by two approaches based on ‘highest intensity’ and ‘confidence threshold’. The predicted path by the three damage indices was compared to the actual crack path. The combined damage index provided a more accurate, consistent, and confident prediction of sharp turns in crack tortuous path. Finally, A 2D Finite Element Model (FEM) is generated to verify the strain calculation approach based on theoretical strain-displacement equations. The verified strain-displacement equation can be later used to calculate the out-of-plane strain components based on surface height difference. However, the small height difference in a highly polished sample leads to large errors in out-of-plane strain calculation. Thus, a threshold value for minimum height difference between two points should be defined for error reduction in out-of-plane shear strain calculation. At the end of this study, a methodology for estimating this threshold value through FE analysis is suggested for future studies.
2023-08-09T00:00:00Z