chapter one
Electronic cooling and the need for experimentation
Kaveh Azar
1.1 Introduction
1.2 The objective of thermal management
1.3 Thermal phenomenon in electronic enclosures
1.3.1 Component
1.3.2 Board/shelf
1.3.3 Frame (enclosure)
1.3.4 Environment
1.3.5 System level approach to thermal management
1.4 Experimental characterization of system, boards, and components
1.4.1 System level
1.4.2 Board level
1.4.3 Component level
1.5 Thermal transport model in circuit boards
1.5.1 Fluid flow in circuit boards
1.5.2 Heat transfer in circuit boards and its effect on components
1.5.3 Heat transfer in a channel
1.5.4 Heat transfer from a component
1.5.5 Parameters to measure
1.5.5.1 Geometry
1.5.5.2 Properties
1.5.5.3 Temperature
1.5.5.4 Fluid velocity
1.6 Analysis tools and their limitations
1.6.1 Overview of design analysis tools
1.6.2 Fundamental analysis tools
1.6.3 Analysis procedure
1.6.4 Analytical modeling: integral method
1.6.4.1 Example 1: governing equations for a single component residing in a circuit board channel
1.6.4.1.1 Component
1.6.4.1.2 Board
1.6.4.1.3 Fluid
1.6.4.1.4 Component interior
1.6.5 Computer-based tools: numerical method
1.6.5.1 Example 2: air temperature distribution for a natural and forced convection component residing on a horizontal board
1.6.6 Experimentation: why, when, and how
1.6.6.1 Example 3: component thermal characterization
1.6.7 How-to lists for experimentation
1.6.8 Steps for successful measurement
1.7 Summary
1.8 Terms defined
Nomenclature
Subscripts
References
âExperimentation is required when the problem is not phenomenologically understood.â
Thermal transport in electronic systems is a prime example of a class of phenomenologically not understood problems. This difficulty can be attributed to the presence of composite materials with different thermal conductivities, multiple heat sources with different intensities, and multi-regime fluid flows on a typical PCB. This is a unique set of problems that are typically difficult to address by classical theory. Since analytical or computational solutions may not be adequate to resolve these intricate differences, often experimentation is required and, in many cases, is the only solution approach.
As we embark on experimental simulation, there are many questions that confront the experimenter. These range from the parameters that are needed to be tracked to the measurement techniques and their accuracy. This is further aggravated by the typical question that one faces: âhow close is the experimental simulation (and the results) to the real problem?â A bit of contemplation on this question reveals that a successful experimental study mandates understanding of the problem and all parameters contributing to it. This includes the physics that governs the problem, the capability or adequacy of sensors measuring, and finally, the interpretation of data and its subsequent relation to the real problem.
Experimentation is another solution technique, like analytical or computational approaches, that strives to answer a specific question. Hence, there is a need to be focused and clear on the objective(s) as we attempt to set up an experiment. In electronics cooling, the objective is to ensure that component junction temperature meets the design specifications, and the cooling system is sufficient in providing error-free operations for the worst environment. Therefore, a successful investigation mandates the understanding of all parameters affecting junction temperature and incorporating them into the analysis. This lends itself to the discussion of thermal coupling in electronic enclosures, and how the heat generated at the component level is distributed throughout the system.
To familiarize the reader with the basic characterization schemes, the ensuing discussion will show the required experimental characterization at different levels of electronics packaging â component, board, and system. The discussion will show the process of experimentation and what is typically done if we decide to embark on an experimental study and what the expected outcome is.
Experimental characterization requires the knowledge of what parameters need to be measured and what their impact is in satisfying the final objective. Measurement in electronics cooling is no exception; there is a need for scaling, planning, error analysis, and quantification of the data measured. In this chapter, parameters that should be me...