Avram Bar-Cohen

Bio: Avram Bar-Cohen is an academic researcher from University of Minnesota. The author has contributed to research in topics: Chip-scale package & Dielectric. The author has an hindex of 4, co-authored 5 publications receiving 41 citations.

Immersion Cooling of Digital Computers

Abstract: The present work reviews the theory and practice of direct liquid cooling of microelectronic components. A morphological analysis is suggested for the classification of liquid-cooling concepts. While both immersion and microgroove cooling of chips are discussed, the emphasis is on immersion cooling. The performance of individual chips and liquid incapsulated modules, including the submerged condenser, is reviewed in detail, with data presented. Flow-through modules and falling-film techniques are also discussed. Finally, figures-of-merit for coolants are noted.

Determination of the Weighted-Average Case Temperature for a Single Chip Package

Abstract: The RJC thermal characterization technique is commonly used for determining the maximum temperature a given chip will experience when operating in a specified environment. As such, RJC provides a simple, inherent figure-of-merit that is also readily adaptable to the numerical analysis of virtually any packaging design or type. Unfortunately, RJC is strictly valid only for an isothermal package surface and, when significant temperature variations are encountered, the use of the reported value of RJC can lead to substantial errors in chip temperature. The use of the junction-to-case thermal resistance can be extended to non-isothermal packages by defining an appropriately weighted, average surface temperature based on numerically derived “Thermal Influence” coefficients for each package surface (or segment) of interest. The present study applies the expanded RJC methodology to an actual PLCC package, and demonstrates its use.

Fundamentals of Nucleate Pool Boiling of Highly-Wetting Dielectric Liquids

Abstract: Direct cooling with inert, dielectric liquids may well become the technique of choice for the thermal management of future electronic systems. Due to the efficiency of phase-change processes and the simplicity of natural circulation, nucleate pool boiling is of great interest for this application.

Trends in the Packaging of Computer Systems

Abstract: The incessant drive toward higher quality, functional density, and speed, in nearly all categories of electronic equipment, is severely constrained by the available packaging technology. The packaging of high-performance electronic systems is highly interdisciplinary and succeeds best when pursued systematically, with attention to both constraints and technological alternatives, as well as opportunities for component and system optimization. The development of a science-based physical design methodology for electronic systems constitutes the primary packaging challenge for the 1990s.

Thermal Control for Cryogenically Cooled Computer Systems

Abstract: This paper summarizes the work that has been done at the University of Tennessee on the development of thermal control systems for computers designed to operate at cryogenic temperatures. A number of the major thermal design issues are discussed in detail. These include: (1) the selection of a working fluid, (2) the determination of the mode, or modes, of heat transfer to be employed, (3) the selection, or development, of any required heat transfer correlations, and (4) the selection of a refrigeration system. The results of a preliminary analysis of a candidate thermal control system is presented. The results of an extensive optimization study of the thermal control system are also briefly discussed.

Review of cooling technologies for computer products

Abstract: This paper provides a broad review of the cooling technologies for computer products from desktop computers to large servers. For many years cooling technology has played a key role in enabling and facilitating the packaging and performance improvements in each new generation of computers. The role of internal and external thermal resistance in module level cooling is discussed in terms of heat removal from chips and module and examples are cited. The use of air-cooled heat sinks and liquid-cooled cold plates to improve module cooling is addressed. Immersion cooling as a scheme to accommodate high heat flux at the chip level is also discussed. Cooling at the system level is discussed in terms of air, hybrid, liquid, and refrigeration-cooled systems. The growing problem of data center thermal management is also considered. The paper concludes with a discussion of future challenges related to computer cooling technology.

Thermal characterization of electronic devices with boundary condition independent compact models

Abstract: The accurate prediction of operating temperatures of temperature-sensitive electronic parts at the component-, board-, and system-level is seriously hampered by the tack of reliable, standardized input data. The situation which prevails today is that component manufacturers supply to end users experimental data which characterizes the thermal behavior of packages under a set of standardized and idealized conditions. Such characterizations normally involve the junction-to-case thermal resistance or the junction-to-ambient resistance according to MIL or SEMI standards. There are several practical difficulties associated with such an approach, which will be shortly commented upon. Today, the need for more accurate junction temperature prediction becomes increasingly urgent, and the call for a precise definition of the various thermal resistances is heard by a growing number of researchers. An earlier paper discussed the pros and cons of several methods that describe the thermal behavior of electronic parts. It was concluded that none of these methods is capable of meeting the objectives that are proposed. In this paper, a novel approach is introduced, based on the derivation of a simple resistance network starting from a detailed model, using optimization techniques. The proposed method is applied to two cases:a so-called "validation" chip, functioning as a benchmark for the software that is used to generate the detailed model; and a 208-PQFP component. It is demonstrated that it is possible to create a compact model comprising a simple resistance network, representing the detailed model to a high accuracy, which is independent of the boundary conditions.

A novel approach for the thermal characterization of electronic parts

Abstract: The accurate prediction of operating temperatures of temperature sensitive electronic parts at the component, board and system level is seriously hampered by the lack of reliable, standardized input data. The situation which prevails today is that component manufacturers supply to end-users experimental data which characterizes the thermal behaviour of packages under a set of standardized and idealized conditions. Such characterizations normally involve the junction-to-case thermal resistance or the junction-to-ambient resistance according to MIL or SEMI standards. There are several practical difficulties associated with such an approach, which will be briefly commented upon. Today, the need for more accurate junction temperature prediction becomes increasingly urgent, and the call for a precise definition of the various thermal resistances is heard by a growing number of researchers. The paper continues with a survey of the open literature and discusses the pros and cons of several methods that describe the thermal behaviour of electronic parts. It is concluded that none of these methods is capable of meeting the objectives that are proposed. A novel approach is introduced, based on the derivation of a simple resistance network starting from a detailed model using optimization techniques. The proposed method is applied to two cases: a so-called 'validation' chip, functioning as a benchmark for the software that is used to generate the detailed model, and a 208-PQFP component. It is demonstrated that it is possible to create a compact model comprising a simple resistance network, representing the detailed model to a high accuracy which is independent of the boundary conditions.

Compact models for accurate thermal characterization of electronic parts

Abstract: This paper discusses several aspects regarding the derivation, accuracy, applicability, and possible future developments in the field of "compact models." A "compact model" is a simplification of a full or detailed thermal model of an electronic package. As such, it consists of a simple network comprising a limited number of thermal resistances (typically 7), connecting the critical part of the device (usually the junction) to the outer parts of the device. Furthermore, the "compact model" is independent of the applied boundary conditions, and is an accurate representation of the full model. It is found that the compact models values typically approach the full model values within 6%. Compact models are suited for embedding in design environments in use by the electronics industries, because they can be incorporated into the component libraries linked to board and system level thermal analysis software packages.