Über die Autorin bzw. den Autor

Allan D. Kraus, PhD, is Professor of Mechanical Engineering at the University of Akron, Ohio, and is principal associate at Allan D. Kraus Associates. He is the author of many works on thermal systems.

Abdul Aziz, PhD, is Professor of Mechanical Engineering at Gonzaga University in Spokane, Washington.

James Welty, PhD, Professor of Mechanical Engineering at Oregon State University in Corvallis, is co-author of Fundamentals of Momentum, Heat, and Mass Transfer (Wiley).

Auszug. © Genehmigter Nachdruck. Alle Rechte vorbehalten.

Extended Surface Heat Transfer

John Wiley & Sons

Chapter One

1.1 INTRODUCTION

Three quarters of a century ago, a paper by Harper and Brown (1922) appeared as an NACA report. It was an elegant piece of work and appears to be the first really significant attempt to provide a mathematical analysis of the interesting interplay between convection and conduction in and upon a single extended surface. Harper and Brown called this a cooling fin, which later became known merely as a fin. It is most probable that Harper and Brown were the pioneers even though Jakob (1949) pointed out that published mathematical analyses of extended surfaces can be traced all the way back to 1789. At that time, Ingenhouss demonstrated the differences in thermal conductivity of several metals by fabricating rods, coating them with wax, and then observing the melting pattern when the bases of the rods were heated. Jakob also pointed out that Fourier (1822) and Despretz (1822, 1828a,b) published mathematical analyses of the temperature variation of the metal bars or rods.

Although these ancient endeavors may have been quite significant at the time they were written, it appears that the Harper-Brown work should be considered as the forerunner of what has become a burgeoning literature that pertains to a very significant subject area in the general field of heat transfer.

The NACA report of Harper and Brown was inspired by a request from the Engineering Division of the U.S. Army and the U.S. Bureau of Standards in connection with the heat-dissipating features of air-cooled aircraft engines. It is interesting to note that this request came less than halfway through the time period between the first flight of the Wright Brothers at Kitty Hawk and the actual establishment of the U.S. Air Force. The work considered longitudinal fins of rectangular profile and trapezoidal profile (which Harper and Brown called wedge-shaped fins) and radial fins of rectangular profile (which Harper and Brown called circumferential fins). It also introduced the concept of fin efficiency, although the expression employed by Harper and Brown was called the fin effectiveness. From this modest, yet masterful beginning, the analysis and evaluation of the performance of individual components of extended surface and arrays of extended surface where individual components are assembled into complicated configurations has become an art.

Harper and Brown (1922) provided thorough analytical solutions for the two-dimensional model for both rectangular and wedge-shaped longitudinal fins and the circumferential fin of uniform thickness. The solutions culminated in expressions for the fin efficiency (called the effectiveness) or in correction factors that adjusted the efficiency of the rectangular profile longitudinal fin. They concluded that the use of a one-dimensional model was sufficent and they proposed that the tip heat loss could be accounted for through the use of a corrected fin height, which increases the fin height by a value equal to half of the fin thickness. Lost in the shuffle, however, was the interesting observation that with dx as the differential element of fin height, the differential face surface area of the element is dx/ cos [kappa], where [kappa] is the taper angle, which is 90 for rectangular profile straight and circumferential fins as well as for spines of constant cross section.

Schmidt (1926) covered the three profiles considered by Harper and Brown from the standpoint of material economy. He stated that the least material is required for given conditions if the fin temperature gradient (from base to tip) is linear, and he showed how the fin thickness of each type of fin must vary to produce this result. Finding, in general, that the calculated shapes were impractical to manufacture, he proceeded to show the optimum dimensions for longitudinal and radial fins of constant thickness (rectangular profile) and the longitudinal fin of trapezoidal profile. He also considered the longitudinal fin of triangular profile as the case of the longitudinal fin of trapezoidal profile with zero tip thickness.

The case of integral pin fins of different profiles was considered by Bueche and Schau (1936). They determined for conical fins that the heat dissipation was a function of the Biot modulus based on the base radius and aspect ratio of fin height to base radius. They also showed that a weight optimization could be effected.

Murray (1938) considered the problem of the radial fin of uniform thickness (the radial fin of rectangular profile) presenting equations for the temperature gradient and effectiveness under conditions of a symmetrical temperature distribution around the base of the fin. He also proposed that the analysis of extended surface should be based on a set of assumptions that have been known since 1945 as the Murray-Gardner assumptions. These assumptions are deemed to be of considerable importance because their elimination, either one at a time or in combination, provided a series of paths for subsequent investigators to follow.

A stepwise procedure for calculating the temperature gradient and efficiency of fins whose thickness varies in any manner was presented by Hausen (1940). The temperature gradient in conical and cylindrical spines was determined by Focke (1942) who, like Schmidt, showed how the spine thickness must vary to keep the material required to a minimum. He too, found the resulting shape impractical and went on to determine the optimum cylindrical and conical spine dimensions.

Avrami and Little (1942) derived equations for the temperature gradient in thick bar fins and showed under what conditions fins might act as insulators on the base or prime surface. Carrier and Anderson (1944) discussed straight fins of constant thickness, radial fins of constant thickness, and radial fins of constant cross-sectional area, presenting equations for the fin efficiency of each. However, in the latter two cases, the efficiencies are given in the form of an infinite series.

Gardner (1945), in a giant leap forward, derived general equations for the temperature excess profile and fin efficiency for any form of extended surface for which the Murray-Gardner assumptions are applicable and whose thickness varies as some power of the distance measured along an axis normal to the base or prime surface (the fin height). He proposed the profile function (his nomenclature)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

for the straight or longitudinal fins,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

for spines and

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

for radial or circumferential (radial) fins.

These equations depend on the assignment of some number to n; for example, the straight fin of rectangular profiles results when n = 0 in the first of these. This also serves to show that in Gardner's profile functions, the positive sense of the height coordinate x is in a direction from fin tip to fin base.

With the foregoing profile functions in hand and working with a general differential equation that he derived, Gardner was able to provide solutions for the temperature excess profile in terms of modified Bessel functions. For n equal to zero or an integer,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

and for n equal to a fraction,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where u depends on the type of fin or spine and where

= [I.sub.n-1]([u.sub.a])/[K.sub.n-1]([u.sub.a])

if n is equal to zero or an integer and

= - [I.sub.n-1]([u.sub.a])/[I.sub.1-n] ([u.sub.a])

if n is equal to a fraction.

For straight (longitudinal) fins

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

for spines

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

and for radial fins

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The fin efficiency, defined as the ratio of the heat transferred from the fin to the heat that would be transferred by the fin if its thermal conductivity were infinite (if the entire fin were to operate at the base temperature excess), was provided by Gardner for all the fins that he considered. Gardner designated the fin efficiency [eta], and for n equal to zero or an integer,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

and for n equal to a fraction,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Graphs were provided that plotted the efficiency as a function of a parameter that embraced the fin dimensions and thermal properties. Two of the graphs (for straight fins and spines) are reproduced here as Figs. 1.1 and 1.2.

Gardner also pointed out that the terms fin efficiency and fin effectiveness had not been used consistently in the English literature. He redefined the fin effectiveness as the ratio of the heat transferred through the base of the fin to the heat transferred through the same prime or base surface area if the fin were not present. He also provided a relationship to permit the conversion from fin efficiency to fin effectiveness.

It is felt that the Gardner paper is remarkable for several reasons. First and probably foremost is the fact that he reemphasized the concept of the fin efficiency, thereby creating an itch that literally thousands of equipment designers have been scratching ever since. Moreover, it appears that Gardner was one of the first to demonstrate the use of applied mathematics to yield concepts that engineers could use to build equipment that worked. He may not have been the first to show the modified Bessel functions to the working mechanical engineer, but he certainly provided an intense reexposition of these interesting functions.

One may observe that as 1945, the year of Gardner's paper, drew to a close, the extended surface technology was on a firm foundation. What began with Harper and Brown and what had concluded with the Gardner paper had established useful design equations for the construction of working heat transfer hardware containing finned surfaces. It is also interesting to note that the 1945 volume of ASME Transactions contained, in addition to Gardner's pioneering effort, correlations for the heat transfer coefficient between fin and fluid. DeLorenzo and Anderson (1945) provided a correlation for the heat transfer coefficient and friction factor for the longitudinal fin-axial flow exchanger. Jameson (1945) provided a heat transfer correlation, and Gunter and Shaw (1945) presented flow friction data in what were then called transverse fins (now radial fins).

Attention now turns to a formal study of extended surface heat transfer and begins with fin analyses based on a consideration of all the Murray-Gardner assumptions.

1.2 EXTENDED SURFACE HEAT TRANSFER

A growing number of engineering disciplines are concerned with energy transitions requiring the rapid movement of heat. They produce an expanding demand for high-performance heat transfer components with progressively smaller weights, volumes, costs, or accommodating shapes. Extended surface heat transfer is the study of these high-performance heat transfer components with respect to these parameters and of their behavior in a variety of thermal environments. Typical components are found in such diverse applications as air-land-space vehicles and their power sources, in chemical, refrigeration, and cryogenic processes, in electrical and electronic equipment, in convential furnaces and gas turbines, in process heat dissipators and waste heat boilers, and in nuclear-fuel modules.

In the design and construction of various types of heat transfer equipment, simple shapes such as cylinders, bars, and plates are used to implement the flow of heat between a source and a sink. They provide heat-absorbing or heat-rejecting surfaces, and each is known as a prime surface. When a prime surface is extended by appendages intimately connected with it, such as the metal tapes and spines on the tubes in Fig. 1.3, the additional surface is known as extended surface. In some disciplines, prime surfaces and their extended surfaces are known collectively as extended surfaces to distinguish them from prime surfaces used alone. The latter definition prevails throughout this book. The elements used to extend the prime surfaces are known as fins. When the fin elements are conical or cylindrical, they may be referred to as spines or pegs.

The demands for aircraft, aerospace, gas turbine, air conditioning, and cryogenic auxilliaries have placed particular emphasis on the compactness of the heat exchanger surface, particularly on those surfaces that induce small pressure gradients in the fluids circulated through them. Several are shown in Fig. 1.4. Compactness refers to the ratio of heat transfer surface per unit of exchanger volume.

An early definition by Kays and London (1950) established a compact exchanger element as one containing in excess of 245 [m.sup.2] per cubic meter of exchanger. Compact exchanger elements have been available with over 4100 [m.sup.2] per cubic meter compared with 65 to 130 [m.sup.2] per cubic meter for conventional heat exhangers with 5/8- to 1-in. tubes. Many compact heat exchanger elements consist of prime surface plates or tubes separated by plates, bars, or spines, which also act as fins. As shown in Fig. 1.4d, each of the fins may be treated as a single fin with fin height equal to half of the separation plate spacing and with the separation plate acting as the prime surface. Thus, the compact heat exhanger is considered as another form of extended surface.

1.2.1 Fin Efficiency

It can be shown quite readily that when a fin and its prime surface are exposed to a uniform thermal environment, a unit of fin surface will be less effective than a unit of prime surface. Consider the plate with a longitudinal fin of rectangular cross section shown in Fig. 1.5. Let the inner plate surface remove heat from a source with a uniform heat transfer coefficient and temperature [T.sub.1], and let the outer plate and fin surfaces reject it to colder surroundings with a uniform heat transfer coefficient and temperature [T.sub.s]. The colder surface of the plate is at some intermediate temperature [T.sub.p], and the heat from the source leaves the plate because of the temperature potential, [T.sub.p] - [T.sub.s]. Similarly, the fin surface is at some temperature, T, and the heat leaves the fin because of the temperature potential, T - [T.sub.s]. The heat enters the fin through its base, where it joins the plate and moves continuously through it by conduction. In most cases, the temperature at the base of the fin will be very nearly the same as [T.sub.p]. Heat absorbed by the fin through its base can flow toward its tip only if there is a temperature gradient within the fin such that [T.sub.p] is greater than T. For this condition, because the temperature T varies from the base to the tip of the fin, the temperature potential T - [T.sub.s] will be smaller than [T.sub.p] - [T.sub.s] and a unit of fin surface will be less effective than a unit of plate or prime surface.

This inescapable loss of performance of a unit of fin surface compared to a unit of prime surface is the inefficiency of the fin. The fin efficiency is defined consistently throughout this book as the ratio of the actual heat dissipation of a fin to its ideal dissipation if the entire fin were at the same temperature as its base. Other indexes of perfomance are also employed, such as the fin effectiveness, weighted fin efficiency, overall passage efficiency, fin resistance, and fin input admittance. Most of these are discussed in later chapters. Fins of given size, shape, and material possess different fin efficiencies, and the efficiency of any fin will vary with its thermal conductivity and the mode of heat transfer with respect to its environment.

(Continues...)

Excerpted from Extended Surface Heat Transferby Allan D. Kraus Abdul Aziz James Welty Copyright © 2001 by Allan D. Kraus. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.