![7218: “intro” — 2007/8/28 — 18:10 — page 9 — #9
Introduction to Microwaves and RF I-9
In addition to these extensions of “traditional” microwave applications, other fields of electronics
are increasing encroaching into the microwave-frequency range. Examples include wired data net-
works based on coaxial cable or twisted-pair transmission lines with bit rates of over 1 Gb/s,
fiber-optic communication systems with data rates well in excess of 10 Gb/s, and inexpensive per-
sonal computers and other digital systems with clock rates of well over 1 GHz. The continuing
advances in the speed and capability of conventional microelectronics is pushing traditional circuit
design ever further into the microwave-frequency regime. These advances have continued to push
digital circuits into regimes where distributed circuit effects must be considered. While system- and
board-level digital designers transitioned to the use of high-speed serial links requiring the use of
distributed transmission lines in their designs some time ago, on-chip transmission lines for distribu-
tion of clock signals and the serialization of data signals for transmission over extremely high-speed
serial buses are now an established feature of high-end designs within a single integrated circuit.
These trends promise to both invigorate and reshape the field of microwave engineering in new and
exciting ways.
I.3 Frequency Band Definitions
The field of microwave and RF engineering is driven by applications, originally for military purposes
such as radar and more recently increasingly for commercial, scientific, and consumer applications.
As a consequence of this increasingly diverse applications base, microwave terminology and frequency
band designations are not entirely standardized, with various standards bodies, corporations, and other
interested parties all contributing to the collective terminology of microwave engineering. Figure I.5 shows
graphically the frequency ranges of some of the most common band designations. As can be seen from
the complexity of Figure I.5, some care must be exercised in the use of the “standard” letter designations;
substantial differences in the definitions of these bands exist in the literature and in practice. While the
IEEE standard for radar bands [8] expressly deprecates the use of radar band designations for nonradar
applications, the convenience of the band designations as technical shorthand has led to the use of these
band designations in practice for a wide range of systems and technologies. This appropriation of radar
band designations for other applications, as well as the definition of other letter-designated bands for other
applications (e.g., electronic countermeasures) that have different frequency ranges is in part responsible
for the complexity of Figure I.5. Furthermore, as progress in device and system performance opens up
new system possibilities and makes ever-higher frequencies useful for new systems, the terminology of
microwave engineering is continually evolving.
Figure I.5 illustrates in approximate order of increasing frequency the range of frequencies encompassed
by commonly-used letter-designated bands. In Figure I.5, the dark shaded regions within the bars indicate
the IEEE radar band designations, and the light cross-hatching indicates variations in the definitions by
different groups and authors. The double-ended arrows appearing above some of the bands indicate other
non-IEEE definitions for these letter designations that appear in the literature. For example, multiple
distinct definitions of L, S, C, X, and K band are in use. The IEEE defines K band as the range from 18 to
27 GHz, while some authors define K band to span the range from 10.9 to 36 GHz, encompassing most of
the IEEE’s Ku, K, and Ka bands within a single band. Both of these definitions are illustrated in Figure I.5.
Similarly, L band has two substantially different, overlapping definitions, with the IEEE definition of
L band including frequencies from 1 to 2 GHz, with an older alternative definition of 390 MHz–1.55 GHz
being found occasionally in the literature. Many other bands exhibit similar, though perhaps less extreme,
variations in their definitions by various authors and standards committees. A further caution must also
be taken with these letter designations, as different standards bodies and agencies do not always ensure
that their letter designations are not used by others. As an example, the IEEE and U.S. military both
define C, L, and K bands, but with very different frequencies; the IEEE L band resides at the low end of the
microwave spectrum, while the military definition of L band is from 40 to 60 GHz. The designations (L–Y)
in Figure I.5a are presently used widely in practice and the technical literature, with the newer U.S. military](https://image.slidesharecdn.com/566857-250525211033-a753caa5/85/Rf-And-Microwave-Handbook-Rf-And-Microwave-Circuits-Measurements-And-Modeling-2nd-Mike-Golio-34-320.jpg)
![7218: “intro” — 2007/8/28 — 18:10 — page 10 — #10
I-10 RF and Microwave Circuits, Measurements, and Modeling
L 0.39 GHz
1.55 GHz
3.9 GHz
2 GHz
12 GHz
5.2 GHz
6.2 GHz
8 GHz
10.9 GHz
12.4 GHz
17.25 GHz
18 GHz
26 GHz
27 GHz
36 GHz
15.35 GHz
33 GHz
24.5 GHz
40 GHz
46 GHz
50 GHz
60 GHz
75 GHz
90 GHz
110 GHz
170 GHz
325 GHz
220 GHz
140 GHz
0.1 GHz
0.5 GHz
2 GHz
4 GHz
8 GHz
40 GHz
100 GHz
140 GHz
60 GHz
20 GHz
10 GHz
6 GHz
3 GHz
1 GHz
0.25 GHz
56 GHz
40 GHz
1 GHz
4 GHz
S
C
X
Ku
K
Band
designation
K1
Ka
Q
U
V
E
W
D
G
Y
A
B
C
D
E
F
Band
designation
G
H
I
J
K
L
M
N
0.1 1 10
Frequency (GHz)
100
0.1 1 10
Frequency (GHz)
100
K
10.9 GHz
FIGURE I.5 Microwave and RF frequency band designations [1–7]. (a) Industrial and IEEE designations. Diagonal
hashing indicates variation in the definitions found in literature; dark regions in the bars indicate the IEEE radar band
definitions [8]. Double-ended arrows appearing above bands indicate alternative band definitions appearing in the
literature, and K† denotes an alternative definition for K band found in Reference [7]. (b) U.S. military frequency band
designations [2–5].](https://image.slidesharecdn.com/566857-250525211033-a753caa5/85/Rf-And-Microwave-Handbook-Rf-And-Microwave-Circuits-Measurements-And-Modeling-2nd-Mike-Golio-35-320.jpg)
![used for cider, but some are more especially adapted for house
purposes:—
I.—LIST OF APPLES.
Those marked with (A) are good for hoarding, and those with † are good for
boiling.
Skyrme’s Kernel—Tart; good for cider.
Royal Wilding—Bitter sweet; good for cider.
Black Foxwhelp—Moderately tart; good for cider.
† Red Foxwhelp (A)—Moderately tart; good for cider.
Cowan Red—Sweet; good for cider.
† Dymock Red (A)—Very sweet; good for cider.
White Norman—Bitter sweet; good for cider.
Red Norman—Bitter sweet; good for cider.
Hagloe Crab—Tart; good for cider.
Pawson—Tart; good for cider.
† Redstreak—Sweet; good for cider.
Yellow Styre—Sweet; good for cider.
† Hooper’s Kernel (A)—Moderately sweet; good for cider.
† Hill Barn Kernel (A)—Sweet; good for cider.
† Ribston Pippin (A)—Sweet; good for table and keeping.
Golden Harvey (A)—Sweet; good for table and for cider.
Siberian Harvey—Sweet; good for cider.
Farewell Blossom—Tart and bitter; large bearer.
Upright French—Bitter sweet; large bearer.
Black or Red French—Bitter sweet.
Knotted Kernel—Tart.
Leather Apple—Hardly any taste.
Ironsides (A)—Hardly any taste; good for keeping.
† Cats’-heads (A)—Sweet; good for cider.
Pigs’-eyes—Sweet.
Downton Pippin (A)—Sweet; table and eating.
† [335]Codlings (A)—Sweet; good as boilers and for cider.
† May Blooms (A)—Sweet; good for cider, boiling, and keeping.
Rough Coat (A)—Dry and sweet; good keepers.
Brandy Apple (A)—Very sweet; makes strong cider.
† Cowarne Quinin (A) Sweet; good for cider.
† Blenheim Orange (A)—Very sweet; good for table.](https://image.slidesharecdn.com/566857-250525211033-a753caa5/85/Rf-And-Microwave-Handbook-Rf-And-Microwave-Circuits-Measurements-And-Modeling-2nd-Mike-Golio-61-320.jpg)
![† Golden Pippin (A)—Very sweet; good for table.
Old Pearmain (A)—Very sweet; good for table.
Brown Crests—Very sweet.
Under Leaves—Sweet; large bearer.
Red Kernel—Sweet; good for cider.
† Reynolds’s Kernel (A)—Sweet; large pot-fruit.
Newland Kernel—Bitter sweet; good for cider.
Jackson’s Kernel—Tart.
† Sam’s Crab—Tart.
† Bridgewater Pippin (A)—Sweet.
† Spice Apple (A)—Sweet.
White Beach—Bitter sweet; good for cider.
Handsome Mandy—Bitter sweet; good for cider.
Golden Rennet (A)—Sweet.
Pine Apple—Moderately tart; wood cankers.
Stoke Pippin (A)—Sweet; good bearers; pot-fruit and for cider; and numerous
others.
From Prize Essay on Orchards, by Clement Cadle, from the Journal of the Royal
Society.
The next list is taken from Scott’s Descriptive Catalogue, by way of
contrast and comparison with the above, as it is more particularly
adapted to Devon, Somerset, and Dorset.
LIST II. CIDER APPLES.
The following is a list of some of the best Cider fruit,
cultivated in the best Cider counties throughout England.
167. Best Bache, spec. grav. 1073. A Herefordshire fruit of great excellence.
168. Bringewood, a good cider fruit.
169. Bovey Redstreak.
170. Cadbury, supposed to be the same as Royal Somerset.[336]
171. Coccagee, a splendid cider fruit of first-rate excellence.
172. Cowrane, red, spec. grav. 1069; an excellent sort.
173. Devonshire Redstreak.
37. Devonshire Quarrenden, a valuable hardy fruit; well known.
35. Downton Pippin, a most prolific and valuable cider fruit.
174. Forest Styre, spec. grav. 1076 to 1081, esteemed fruit.](https://image.slidesharecdn.com/566857-250525211033-a753caa5/85/Rf-And-Microwave-Handbook-Rf-And-Microwave-Circuits-Measurements-And-Modeling-2nd-Mike-Golio-62-320.jpg)
![then the wind will only partially dry it, and the result will be a
general heating of the mass, which results, if not in quick decay
amounting to absolute rottenness, yet in that state, technically called
“moisey,”[31] or dead, in which the juices are nearly dried up and the
fruit flavourless.
[31] Apple moise, or apple moce, was an old dish made of pressed apples.
In cider counties apples are called moisey when they are juiceless, dry,
and without flavour—dead. (See Archaic Dictionaries.)
We have seen heaps of apples, consisting of many waggon-loads, in
the orchard at Christmas, when wet and frost had so preyed upon
them that none of their proper juices remained. This is certain to
make a cider which will be of inferior quality; and though some of
our friends boast of the good quality of their cider which has been
made in the roughest manner, yet one cannot help thinking how
much better it might have been with the fruit carefully collected, and
kept until it could be ground. Still, with all our care in this matter,
disappointment is sometimes the result; for it is with cider as with
wine, the season will have a great deal to do with it, though with
both, the manner of making and storing will be all-important
matters, to which we shall advert in the next chapter.
We much object to the gathering of fruit for any purpose in the wet.
Were it not for the expense, it would be better to take advantage of
dry weather, and to collect even cider-fruit by hand-picking before it
has become dead ripe, and so let the ripening process be completed
in some dry storing-place. In our own experience of cider-making,
the two or three casks made for home consumption from carefully
picked and well-kept fruit are usually of the best quality, and so
made we believe cider to be a most agreeable and very wholesome
beverage,—to paraphrase Isaac Walton, only fit for farmers or very
honest men. As long, however, as rough people are about who never
know when they have had enough, the rougher cider made by a
ruder process is quite good enough.
It must be obvious to all that if a man can drink as much as four
gallons of good cider in a day’s mowing, he would be better off with](https://image.slidesharecdn.com/566857-250525211033-a753caa5/85/Rf-And-Microwave-Handbook-Rf-And-Microwave-Circuits-Measurements-And-Modeling-2nd-Mike-Golio-68-320.jpg)
![Portable mills of this kind are now very general, but we so fully
agree with the remarks of Mr. Cadle, that we here quote his
description of some portable cider-mills, with his comments upon
their action.
About twenty-six years ago, Mr. Coleman, of Chaxhill, Westbury-on-
Severn, commenced making an improved cider-mill and press, which could
act either as a fixture or a portable mill. It was found that the cider thus
made fined better, and the process was also more expeditious. These
advantages, together with the cost of keeping the old kind of mills in
repair, which landlords were unwilling to undertake, led to their being
superseded, as they wore out, by Coleman’s, or a similar mill.
Coleman’s mill consists of two pairs of rollers fixed in a strong wooden
frame; it is fed from a hopper, the apples passing through the first pair of
rollers, which are made of hard wood, with iron teeth, so as to break the
apples, which fall next between a pair of stone rollers set close enough to
break the kernels, and from these the pulp drops into a trough placed
beneath to receive it.
Mr. Latchem, of Hereford, has also paid considerable attention to the
construction of these mills, and has taken out a patent for doing away
with the iron in the feed-rollers, and substituting steel teeth fitted into one
roller, and working through other steel teeth on a fixed plate, partly on the
same principle as a curd-mill. The fruit, after passing this “chewer,” is
ground between a pair of stone rollers, as before described.
Until the portable apple-mills became general, we had a mill to almost
every farm, and even to many of the cottages; but in Devonshire one mill
or pound-house serves for a number of makers, and sometimes for a
parish, each person paying so much per hogshead for the making.
[347]Most of the travelling portable machines in Herefordshire have two
presses with each mill, and are worked by two horses, making 1,000 to
1,500 gallons in a day; sometimes they are worked by a small portable
steam-engine. They are very expeditious, and do very well for a second-
class cider, but if you would have the best, they are very objectionable,
because the different sorts of fruit very rarely get ripe at once in sufficient
quantities to enable you to make much at a time. Much cider is therefore
spoiled, the fruit being ground when too green, by those who are
impatient to finish the process. I think that each farm or holding should
have a mill of its own, even if it be only a small hand-mill.](https://image.slidesharecdn.com/566857-250525211033-a753caa5/85/Rf-And-Microwave-Handbook-Rf-And-Microwave-Circuits-Measurements-And-Modeling-2nd-Mike-Golio-70-320.jpg)
Rf And Microwave Handbook Rf And Microwave Circuits Measurements And Modeling 2nd Mike Golio
- 1. Rf And Microwave Handbook Rf And Microwave Circuits Measurements And Modeling 2nd Mike Golio download https://ebookbell.com/product/rf-and-microwave-handbook-rf-and- microwave-circuits-measurements-and-modeling-2nd-mike- golio-1133714 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Rf And Microwave Circuits Measurements And Modeling The Electrical Engineering Handbook Series Rf And Microwave Handbook 2nd Edition 2nd Mike Golio https://ebookbell.com/product/rf-and-microwave-circuits-measurements- and-modeling-the-electrical-engineering-handbook-series-rf-and- microwave-handbook-2nd-edition-2nd-mike-golio-2135042 Rf And Microwave Handbook Rf And Microwave Applications And Systems 1st Mike Golio https://ebookbell.com/product/rf-and-microwave-handbook-rf-and- microwave-applications-and-systems-1st-mike-golio-1133712 The Rf And Microwave Handbook Golio John Michael https://ebookbell.com/product/the-rf-and-microwave-handbook-golio- john-michael-5311554 The Rf And Microwave Handbook 1st Edition Mike Golio https://ebookbell.com/product/the-rf-and-microwave-handbook-1st- edition-mike-golio-1053620
- 3. The Rf And Microwave Handbook 3 Volume Set 2nd Edition Mike Golio Author https://ebookbell.com/product/the-rf-and-microwave-handbook-3-volume- set-2nd-edition-mike-golio-author-12055158 Handbook Of Rf And Microwave Power Amplifiers The Cambridge Rf And Microwave Engineering Series John L B Walker https://ebookbell.com/product/handbook-of-rf-and-microwave-power- amplifiers-the-cambridge-rf-and-microwave-engineering-series-john-l-b- walker-2517326 Handbook Of Rf And Microwave Power Amplifiers John L B Walker https://ebookbell.com/product/handbook-of-rf-and-microwave-power- amplifiers-john-l-b-walker-4720412 Rf And Microwave Passive And Active Technologies Electrical Engineering Handbook 2nd Edition Mike Golio https://ebookbell.com/product/rf-and-microwave-passive-and-active- technologies-electrical-engineering-handbook-2nd-edition-mike- golio-2136708 Handbook Of Rfmicrowave Components And Engineering 1st Edition Kai Chang Ed https://ebookbell.com/product/handbook-of-rfmicrowave-components-and- engineering-1st-edition-kai-chang-ed-52032956
- 6. 7218: “7218_c000” — 2007/11/16 — 10:28 — page i — #1
- 7. 7218: “7218_c000” — 2007/11/16 — 10:28 — page ii — #2
- 8. 7218: “7218_c000” — 2007/11/16 — 10:28 — page iii — #3
- 9. 7218: “7218_c000” — 2007/11/16 — 10:28 — page iv — #4
- 10. 7218: “7218_c000” — 2007/11/16 — 10:28 — page v — #5 Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Advisory Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix Introduction to Microwaves and RF Patrick Fay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I-1 1 Overview of Microwave Engineering Mike Golio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1 SECTION I Microwave Measurements 2 Linear Measurements Ronald E. Ham . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1 3 Network Analyzer Calibration Joseph Staudinger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 4 Absolute Magnitude and Phase Calibrations Kate A. Remley, Paul D. Hale, and Dylan F. Williams . . . . . . . . . . . . . . 4-1 5 Noise Measurements Alfy Riddle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1 v
- 11. 7218: “7218_c000” — 2007/11/16 — 10:28 — page vi — #6 vi Contents 6 Nonlinear Microwave Measurement and Characterization J. Stevenson Kenney . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1 7 Theory of High-Power Load-Pull Characterization for RF and Microwave Transistors John F. Sevic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1 8 Pulsed Measurements Anthony E. Parker, James G. Rathmell, and Jonathan B. Scott . . . . . . . . . . 8-1 9 Microwave On-Wafer Test Jean-Pierre Lanteri, Christopher Jones, and John R. Mahon . . . . . . . . . . . 9-1 10 High Volume Microwave Test Jean-Pierre Lanteri, Christopher Jones, and John R. Mahon . . . . . . . . . . . 10-1 11 Large Signal Network Analysis/Waveform Measurements Joseph M. Gering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-1 SECTION II Circuits 12 Receivers Warren L. Seely . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-1 13 Transmitters Warren L. Seely . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-1 14 Low Noise Amplifier Design Jakub Kucera and Urs Lott . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-1 15 Microwave Mixer Design Anthony M. Pavio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-1 16 Modulation and Demodulation Circuitry Charles Nelson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-1 17 Power Amplifier Fundamentals Douglas A. Teeter and Edward T. Spears . . . . . . . . . . . . . . . . . . . . . 17-1 18 Handset Power Amplifier Design Douglas A. Teeter and Edward T. Spears . . . . . . . . . . . . . . . . . . . . . 18-1 19 Class A Amplifiers Warren L. Seely . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-1 20 High Power Amplifiers Brent Irvine, Todd Heckleman, and Aaron Radomski . . . . . . . . . . . . . . 20-1
- 12. 7218: “7218_c000” — 2007/11/16 — 10:28 — page vii — #7 Contents vii 21 Oscillator Circuits Alfy Riddle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21-1 22 Phase Locked Loop Design Robert Newgard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22-1 23 Filters and Multiplexers Richard V. Snyder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23-1 24 RF Switches Robert J. Trew . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24-1 25 Monolithic Microwave IC Technology Lawrence P. Dunleavy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25-1 26 Bringing RFICs to the Market John C. Cowles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26-1 SECTION III CAD, Simulation and Modeling 27 System Simulation Joseph Staudinger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27-1 28 Numerical Techniques for the Analysis and Design of RF/Microwave Structures Manos M. Tentzeris . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28-1 29 Computer Aided Design of Passive Components Daniel G. Swanson, Jr. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29-1 30 Nonlinear RF and Microwave Circuit Analysis Michael B. Steer and John F. Sevic . . . . . . . . . . . . . . . . . . . . . . . . 30-1 31 Computer Aided Design (CAD) of Microwave Circuitry Ron Kielmeyer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31-1 32 Nonlinear Transistor Modeling for Circuit Simulation Walter R. Curtice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32-1 33 Behavioral Modeling of Microwave Power Amplifiers Troels S. Nielsen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33-1 34 Technology CAD Peter A. Blakey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34-1 Appendix A: Mathematics, Symbols, and Physical Constants . . . . . . . . . . . . A-1
- 13. 7218: “7218_c000” — 2007/11/16 — 10:28 — page viii — #8 viii Contents Appendix B: Microwave Engineering Appendix John P. Wendler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I-1
- 14. 7218: “7218_c000” — 2007/11/16 — 10:28 — page ix — #9 Preface The first edition of the RF and Microwave Handbook was published in 2000. The project got off to an inauspicious start when 24 inches of snow fell in Denver the evening before the advisory board planned to hold their kick-off meeting. Two members of the board were trapped for days in the Denver airport since planes were not arriving or leaving. Because of road closures, one member was stranded only miles away from the meeting in Boulder. And the remainder of the board was stranded in a Denver hotel 10 miles from the airport. Despite this ominous beginning, a plan was formed, expert authors recruited, and the book was developed and published. The planning and development of this second edition have been very smooth and uneventful in comparison to our first efforts. Since publication in 2000, the value of the RF and Microwave Handbook has been recognized by thousands of engineers throughout the world. Three derivative handbooks have also been published and embraced by the microwave industry. The advisory board believes that this edition will be found to be of even greater value than the first edition. Prior to the 1990s, microwave engineering was employed almost exclusively to address military, satellite, and avionics applications. In 1985, there were a few limited applications of RF and microwave systems that laymen might be familiar with such as satellite TV and the use of satellite communications for overseas phone calls. Pagers were also available but not common. In contrast, by 1990 the wireless revolution had begun. Cell phones were becoming common and new applications of wireless technology were emerging every day. Companies involved in wireless markets seemed to have a license to print money. At the time of the introduction of the first edition of the RF and Microwave Handbook, wireless electronic products were pervasive, but relatively simple, early generations of the advanced wireless products available today. At present, the number of people using wireless voice and data systems continues to grow. New systems such as 3G phones, 4G phones, and WiMAX represent emerging new wireless markets with significant growth potential. All of these wireless products are dependent on the RF and microwave component and system engineering, which is the subject of this book. During this time the military, satellite, and avionics systems have also become increasingly complex. The research and development that drives these applications continues to serve as the foundation for most of the commercial wireless products available to consumers. This edition of the handbook covers issues of interest to engineers involved in RF/microwave system and component development. The second edition includes significantly expanded topic coverage as well as updated or new articles for most of the topics included in the first edition. The expansion of material has prompted the division of the handbook into three independent volumes of material. The chapters are aimed at working engineers, managers, and academics who have a need to understand microwave topics outside their area of expertise. Although the book is not written as a textbook, researchers and students will find it useful. Most of the chapters provide extensive references so that they will not only explain fundamentals of each field, but also serve as a starting point for further in-depth research. ix
- 15. 7218: “7218_c000” — 2007/11/16 — 10:28 — page x — #10 x Preface This book, RF and Microwave Circuits, Measurements, and Modeling, examines three areas of critical importance to the RF and microwave circuit designer. Characterization and measurement of components, circuits, and systems at high frequencies are unique and challenging tasks. Standard, low frequency equipment fails to provide meaningful information for the RF and microwave engineer. Small-signal, large-signal, phase, pulsed, waveform, and noise measurements are discussed in detail. Calibration procedures are extremely important for these measurements and are also described. RFandmicrowavecircuitdesignsareexploredintermsofperformanceandcriticaldesignspecifications. Transmitters and receivers are first discussed in terms of functional circuit blocks. The blocks are then examined individually. Separate chapters consider fundamental amplifier issues, low noise amplifiers, power amplifiers for handset applications, and high power amplifiers. Other circuit functions including oscillators, mixers, modulators, phase locked loops, filters, and multiplexers are each considered in individual chapters. The unique behavior and requirements associated with RF and microwave systems establish a need for unique and complex models and simulation tools. The required toolset for a microwave circuit designer includes unique device models, both 2D and 3D electromagnetic simulators, as well as frequency domain based small-signal and large-signal circuit and system simulators. This unique suite of tools requires a design procedure that is also distinctive. Individual chapters examine not only the distinct design tools of the microwave circuit designer, but also the design procedures that must be followed to use them effectively.
- 16. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xi — #11 Acknowledgments Although the topics and authors for this book were identified by the editor-in-chief and the advisory board, they do not represent the bulk of the work for a project like this. A great deal of the work involves tracking down those hundreds of technical experts, gaining their commitment, keeping track of their progress, collecting their manuscripts, getting appropriate reviews/revisions, and finally transferring the documents to be published. While juggling this massive job, author inquiries ranging from, “What is the required page length?”, to, “What are the acceptable formats for text and figures?”, have to be answered and re-answered. Schedules are very fluid. This is the work of the Managing Editor, Janet Golio. Without her efforts there would be no second edition of this handbook. The advisory board has facilitated the book’s completion in many ways. Board members contributed to the outline of topics, identified expert authors, reviewed manuscripts, and authored several of the chapters for the book. Hundreds of RF and microwave technology experts have produced the chapters that comprise this second edition. They have all devoted many hours of their time sharing their expertise on a wide range of topics. I would like to sincerely thank all of those listed above. Also, it has been a great pleasure to work with Jessica Vakili, Helena Redshaw, Nora Konopka, and the publishing professionals at CRC Press. xi
- 17. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xii — #12
- 18. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xiii — #13 Editors Editor-in-Chief Mike Golio, since receiving his PhD from North Carolina State University in 1983, has held a variety of positions in both the microwave and semiconductor industries, and within academia. As Corporate Director of Engineering at Rockwell Collins, Dr. Golio managed and directed a large research and devel- opment organization, coordinated corporate IP policy, and led committees to achieve successful corporate spin-offs. As an individual contributor at Motorola, he was responsible for pioneering work in the area of large signal microwave device characterization and modeling. This work resulted in over 50 publications including one book and a commercially available software package. The IEEE recognized this work by making Dr. Golio a Fellow of the Institute in 1996. He is currently RF System Technologist with HVVi Semiconductor, a start-up semiconductor company. He has contributed to all aspects of the company’s funding, strategies, and technical execution. Dr. Golio has served in a variety of professional volunteer roles for the IEEE MTT Society including: Chair of Membership Services Committee, founding Co-editor of IEEE Microwave Magazine, and MTT- Society distinguished lecturer. He currently serves as Editor-in-chief of IEEE Microwave Magazine. In 2002 he was awarded the N. Walter Cox Award for exemplary service in a spirit of selfless dedication and cooperation. He is author of hundreds of papers, book chapters, presentations and editor of six technical books. He is inventor or co-inventor on 15 patents. In addition to his technical contributions, Dr. Golio recently published a book on retirement planning for engineers and technology professionals. Managing Editor Janet R. Golio is Administrative Editor of IEEE Microwave Magazine and webmaster of www.golio.net. Prior to that she did government, GPS, and aviation engineering at Motorola in Arizona, Rockwell Collins in Iowa, and General Dynamics in Arizona. She also helped with the early development of the personal computer at IBM in North Carolina. Golio holds one patent and has written six technical papers. She received a BS in Electrical Engineering Summa Cum Laude from North Carolina State University in 1984. When not working, Golio actively pursues her interests in archaeology, trick roping, and country western dancing. She is the author of young adult books, A Present from the Past and A Puzzle from the Past. xiii
- 19. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xiv — #14
- 20. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xv — #15 Advisory Board Peter A. Blakey Peter A. Blakey obtained a BA in applied physics from the University of Oxford in 1972, a PhD in electronic engineering from the University of London in 1976, and an MBA from the University of Michigan in 1989. He has held several different positions in industry and academia and has worked on a wide range of RF, microwave, and Si VLSI applications. Between 1991 and 1995 he was the director of TCAD Engineering at Silvaco International. He joined the Department of Electrical Engineering at Northern Arizona University in 2002 and is presently an emeritus professor at that institution. Nick Buris Nick Buris received his Diploma in Electrical Engineering in 1982 from the National Technical University of Athens, Greece, and a PhD in electrical engineering from the North Carolina State University, Raleigh, North Carolina, in 1986. In 1986 he was a visiting professor at NCSU working on space reflector antennas for NASA. In 1987 he joined the faculty of the Department of Electrical and Computer Engineering at the University of Massachusetts at Amherst. His research work there spanned the areas of microwave magnetics, phased arrays printed on ferrite substrates, and broadband antennas. In the summer of 1990 he was a faculty fellow at the NASA Langley Research Center working on calibration techniques for dielectric measurements (space shuttle tiles at very high temperatures) and an ionization (plasma) sensor for an experimental reentry spacecraft. In 1992 he joined the Applied Technology organization of Motorola’s Paging Product Group and in 1995 he moved to Corporate Research to start an advanced modeling effort. While at Motorola he has worked on several projects from product design to measurement systems and the development of proprietary software tools for electromagnetic design. He currently manages the Microwave Technologies Research Lab within Motorola Labs in Schaumburg, Illinois. Recent and current activities of the group include V-band communications systems design, modeling and measurements of complex electromagnetic problems, RF propagation, Smart Antennas/MIMO, RFID systems, communications systems simulation and modeling, spectrum engineering, as well as TIA standards work on RF propagation and RF exposure. Nick is a senior member of the IEEE, and serves on an MTT Technical Program Committee. He recently served as chair of a TIA committee on RF exposure and is currently a member of its Research Division Committee. Lawrence P. Dunleavy Dr. Larry Dunleavy, along with four faculty colleagues established University of South Florida’s innovative Center for Wireless and Microwave Information Systems (WAMI Center—http://ee.eng.usf.edu/WAMI). xv
- 21. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xvi — #16 xvi Advisory Board In 2001, Dr. Dunleavy co-founded Modelithics, Inc., a USF spin-off company, to provide a prac- tical commercial outlet for developed modeling solutions and microwave measurement services (www.modelithics.com), where he is currently serving as its president. Dr. Dunleavy received his BSEE degree from Michigan Technological University in 1982 and his MSEE and PhD in 1984 and 1988, respectively, from the University of Michigan. He has worked in industry for E-Systems (1982–1983) and Hughes Aircraft Company (1984–1990) and was a Howard Hughes Doctoral Fellow (1984–1988). In 1990 he joined the Electrical Engineering Department at the University of South Florida. He maintains a position as professor in the Department of Electrical Engineering. His research interests are related to microwave and millimeter-wave device, circuit, and system design, characterization, and modeling. In 1997–1998, Dr. Dunleavy spent a sabbatical year in the noise metrology laboratory at the National Institute of Standards and Technology (NIST) in Boulder, Colorado. Dr. Dunleavy is a senior member of IEEE and is very active in the IEEE MTT Society and the Automatic RF Techniques Group (ARFTG). He has authored or co-authored over 80 technical articles. Jack East Jack East received his BSE, MS, and PhD from the University of Michigan. He is presently with the Solid State Electronics Laboratory at the University of Michigan conducting research in the areas of high- speed microwave device design, fabrication, and experimental characterization of solid-state microwave devices, nonlinear and circuit modeling for communications circuits and low-energy electronics, and THz technology. Patrick Fay Patrick Fay is an associate professor in the Department of Electrical Engineering at the University of Notre Dame, Notre Dame, Indiana. He received his PhD in Electrical Engineering from the University of Illinois at Urbana-Champaign in 1996 after receiving a BS in Electrical Engineering from Notre Dame in 1991. Dr. Fay served as a visiting assistant professor in the Department of Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign in 1996 and 1997, and joined the faculty at the University of Notre Dame in 1997. His educational initiatives include the development of an advanced undergraduate laboratory course in microwave circuit design and characterization, and graduate courses in optoelectronic devices and electronic device characterization. He was awarded the Department of Electrical Engineering’s IEEE Outstanding Teacher Award in 1998–1999. His research interests include the design, fabrication, and characterization of microwave and millimeter-wave electronic devices and circuits, as well as high-speed optoelectronic devices and optoelectronic integrated circuits for fiber optic telecommunications. His research also includes the development and use of micromachining techniques for the fabrication of microwave components and packaging. Professor Fay is a senior member of the IEEE, and has published 7 book chapters and more than 60 articles in refereed scientific journals. David Halchin David Halchin has worked in RF/microwaves and GaAs for over 20 years. During this period, he has worn many hats including engineering and engineering management, and he has worked in both academia and private industry. Along the way, he received his PhD in Electrical Engineering from North Carolina State University. Dave currently works for RFMD, as he has done since 1998. After a stint as a PA designer, he was moved into his current position managing a modeling organization within RFMD’s Corporate Research and Development organization. His group’s responsibilities include providing compact models for circuit simulation for both GaAs active devices and passives on GaAs. The group also provides compact models for a handful of Si devices, behavioral models for power amplifier assemblies, and physics-based simulation for GaAs device development. Before joining RFMD, Dave spent time at Motorola and Rockwell working
- 22. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xvii — #17 Advisory Board xvii in the GaAs device development and modeling areas. He is a member of the IEEE-MTT and EDS. Dave is currently a member of the executive committee for the Compound IC Symposium. Alfy Riddle Alfy Riddle is vice president of Engineering at Finesse. Before Finesse, Dr. Riddle was the principal at Macallan Consulting, a company he founded in 1989. He has contributed to the design and development of a wide range of products using high-speed, low-noise, and RF techniques. Dr. Riddle developed and marketed the Nodal circuit design software package that featured symbolic analysis and object-oriented techniques. He has co-authored two books and contributed chapters to several more. He is a member of the IEEE MTT Society, the Audio Engineering Society, and the Acoustical Society of America. Dr. Riddle received his PhD in Electrical Engineering in 1986 from North Carolina State University. When he is not working, he can be found on the tennis courts, hiking in the Sierras, taking pictures with an old Leica M3, or making and playing Irish flutes. Robert J. Trew Robert J. Trew received his PhD from the University of Michigan in 1975. He is currently the Alton and Mildred Lancaster Distinguished Professor of Electrical and Computer Engineering and Head of the ECE Department at North Carolina State University, Raleigh. He previously served as the Worcester Professor of Electrical and Computer Engineering and Head of the ECE Department of Virginia Tech, Blacksburg, Virginia, and the Dively Distinguished Professor of Engineering and Chair of the Department of Electrical Engineering and Applied Physics at Case Western Reserve University, Cleveland, Ohio. From 1997 to 2001 Dr. Trew was director of research for the U.S. Department of Defense with management responsibility for the $1.3 billion annual basic research program of the DOD. Dr. Trew was vice-chair of the U.S. government interagency group that planned and implemented the U.S. National Nanotechnology Initiative. Dr. Trew is a fellow of the IEEE, and was the 2004 president of the Microwave Theory and Techniques Society. He was editor-in-chief of the IEEE Transactions on Microwave Theory and Techniques from 1995 to 1997, and from 1999 to 2002 was founding co-editor-in-chief of the IEEE Microwave Magazine. He is currently the editor-in-chief of the IEEE Proceedings. Dr. Trew has twice been named an IEEE MTT Society Distinguished Microwave Lecturer. He has earned numerous honors, including a 2003 Engineering Alumni Society Merit Award in Electrical Engineering from the University of Michigan, the 2001 IEEE-USA Harry Diamond Memorial Award, the 1998 IEEE MTT Society Distinguished Educator Award, a 1992 Distinguished Scholarly Achievement Award from NCSU, and an IEEE Third Millennium Medal. Dr. Trew has authored or co-authored over 160 publications, 19 book chapters, 9 patents, and has given over 360 presentations
- 23. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xviii — #18
- 24. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xix — #19 Contributors Peter A. Blakey Northern Arizona University Flagstaff, Arizona John C. Cowles Analog Devices–Northwest Labs Beaverton, Oregon Walter R. Curtice W. R. Curtice Consulting Washington Crossing, Pennsylvania Lawrence P. Dunleavy Modelithics, Inc. Tampa, Florida Patrick Fay University of Notre Dame Notre Dame, Indiana Joseph M. Gering RF Micro Devices Greensboro, North Carolina Mike Golio HVVi Semiconductor Phoenix, Arizona Paul D. Hale National Institute of Standards and Technology Boulder, Colorado Ronald E. Ham MW and RF Consulting Austin, Texas and Kitzbuhel, Austria H. Mike Harris Georgia Tech Research Institute Atlanta, Georgia Todd Heckleman MKS Instruments, Inc. Rochester, New York Brent Irvine MKS Instruments, Inc. Rochester, New York Christopher Jones M/A-COM Tyco Electronics Lowell, Massachusetts J. Stevenson Kenney Georgia Institute of Technology Atlanta, Georgia Ron Kielmeyer RF Micro Devices Chandler, Arizona Jakub Kucera AnaPico AG Zürich, Switzerland Jean-Pierre Lanteri M/A-COM Tyco Electronics Lowell, Massachusetts Urs Lott AnaPico AG Zürich, Switzerland John R. Mahon M/A-COM Tyco Electronics Lowell, Massachusetts Charles Nelson California State University Sacramento, California Robert Newgard Rockwell Collins Cedar Rapids, Iowa Troels S. Nielsen RF Micro Devices Greensboro, North Carolina Anthony E. Parker Macquarie University Sydney, Australia Anthony M. Pavio Microwave Specialties Paradise Valley, Arizona Aaron Radomski Harris RF Communications Rochester, New York James G. Rathmell University of Sydney Sydney, Australia Kate A. Remley National Institute of Standards and Technology Boulder, Colorado Alfy Riddle Finesse, LLC Santa Clara, California xix
- 25. 7218: “7218_c000” — 2007/11/16 — 10:28 — page xx — #20 xx Contributors Jonathan B. Scott University of Waikato Hamilton, New Zealand Warren L. Seely Ubidyne, Inc. Tempe, Arizona John F. Sevic Maury Microwave Corporation Ontario, California Richard V. Snyder RS Microwave Butler, New Jersey Edward T. Spears RF Micro Devices Chandler, Arizona Joseph Staudinger Freescale Semiconductor, Inc. Tempe, Arizona Michael B. Steer North Carolina State University Raleigh, North Carolina Daniel G. Swanson, Jr. Tyco Electronics Lowell, Massachusetts Douglas A. Teeter RF Micro Devices Billerica, Massachusetts Manos M. Tentzeris Georgia Institute of Technology Atlanta, Georgia Robert J. Trew North Carolina State University Raleigh, North Carolina John P. Wendler Tyco Electronics Wireless Network Solutions Lowell, Massachusetts Dylan F. Williams National Institute of Standards and Technology Boulder, Colorado
- 26. 7218: “intro” — 2007/8/28 — 18:10 — page 1 — #1 Introduction to Microwaves and RF Patrick Fay University of Notre Dame I.1 Introduction to Microwave and RF Engineering ...... I-1 I.2 General Applications .................................... I-8 I.3 Frequency Band Definitions ............................ I-9 I.4 Overview of The RF and Microwave Handbook ...... I-11 References ....................................................... I-12 I.1 Introduction to Microwave and RF Engineering Modern microwave and radio frequency (RF) engineering is an exciting and dynamic field, due in large part to the symbiosis between recent advances in modern electronic device technology and the explosion in demand for voice, data, and video communication capacity that started in the 1990s and continues through the present. Prior to this revolution in communications, microwave technology was the nearly exclusive domain of the defense industry; the recent and dramatic increase in demand for communication systems for such applications as wireless paging, mobile telephony, broadcast video, and tethered as well as untethered computer networks has revolutionized the industry. These com- munication systems are employed across a broad range of environments, including corporate offices, industrial and manufacturing facilities, infrastructure for municipalities, as well as private homes. The diversity of applications and operational environments has led, through the accompanying high pro- duction volumes, to tremendous advances in cost-efficient manufacturing capabilities of microwave and RF products. This in turn has lowered the implementation cost of a host of new and cost-effective wireless as well as wired RF and microwave services. Inexpensive handheld GPS navigational aids, automotive collision-avoidance radar, and widely available broadband digital service access are among these. Microwave technology is naturally suited for these emerging applications in communications and sensing, since the high operational frequencies permit both large numbers of independent channels for the wide variety of uses envisioned as well as significant available bandwidth per channel for high-speed communication. Loosely speaking, the fields of microwave and RF engineering together encompass the design and imple- mentation of electronic systems utilizing frequencies in the electromagnetic spectrum from approximately 300 kHz to over 100 GHz. The term “RF” engineering is typically used to refer to circuits and systems hav- ing frequencies in the range from approximately 300 kHz at the low end to between 300 MHz and 1 GHz at the upper end. The term “microwave engineering,” meanwhile, is used rather loosely to refer to design and implementation of electronic systems with operating frequencies in the range from 300 MHz to 1 GHz on the low end to upwards of 100 GHz. Figure I.1 illustrates schematically the electromagnetic spectrum from audio frequencies through cosmic rays. The RF frequency spectrum covers the medium frequency (MF), high frequency (HF), and very high frequency (VHF) bands, while the microwave portion of the I-1
- 27. 7218: “intro” — 2007/8/28 — 18:10 — page 2 — #2 I-2 RF and Microwave Circuits, Measurements, and Modeling 3×101 ELF (extremely low frequency) SLF/VF (super low/voice frequency) VLF (very low frequency) LF (low frequency) MF (medium frequency) HF (high frequency) VHF (very high frequency) UHF (ultra high frequency) SHF (super high frequency) EHF (extremely high frequency) THz radiation Infrared Visible light Ultraviolet light X-rays, gamma rays, cosmic rays 107 106 Audio frequencies RF: AM/FM radio, VHF television Microwaves; millimeter, submillimeterwaves 105 104 103 102 10 1 10−1 10−2 Wavelength (m) 10−3 3×10−5 10−6 4×10−7 10−8 3×103 3×105 3×107 3×109 3×1011 Frequency (Hz) 3×1014 3×1016 >3×1024 <10−16 FIGURE I.1 Electromagnetic frequency spectrum and associated wavelengths. electromagnetic spectrum extends from the upper edge of the VHF frequency range to just below the THz radiation and far-infrared optical frequencies (approximately 0.3 THz and above). The wavelength of free-space radiation for frequencies in the RF frequency range is from approximately 1 m (at 300 MHz) to 1 km (at 300 kHz), while those of the microwave range extend from 1 m to the vicinity of 1 mm (corresponding to 300 GHz) and below. The boundary between “RF” and “microwave” design is both somewhat indistinct as well as one that is continually shifting as device technologies and design methodologies advance. This is due to implicit connotations that have come to be associated with the terms “RF” and “microwave” as the field has developed. In addition to the distinction based on the frequency ranges discussed previously, the fields of RF and microwave engineering are also often distinguished by other system features as well. For example, the particular active and passive devices used, the system applications pursued, and the design techniques and overall mindset employed all play a role in defining the fields of microwave and RF engineering. These connotations within the popular meaning of microwave and RF engineering arise fundamentally from the frequencies employed, but often not in a direct or absolute sense. For example, because advances in technology often considerably improve the high frequency performance of electronic devices, the correlation between particular types of electronic devices and particular frequency ranges is a fluid one. Similarly, new system concepts and designs are reshaping the applications landscape, with mass market designs utilizing ever higher frequencies rapidly breaking down conventional notions of microwave-frequency systems as serving “niche” markets. The most fundamental characteristic that distinguishes RF engineering from microwave engineering is directly related to the frequency (and thus the wavelength, λ) of the electronic signals being processed. This distinction arises fundamentally from the finite speed of propagation of electromagnetic waves (and thus, by extension, currents and voltages). In free space, λ = c/f , where f is the frequency of the sig- nal and c is the speed of light. For low-frequency and RF circuits (with a few special exceptions such as antennae), the signal wavelength is much larger than the size of the electronic system and circuit components. In contrast, for a microwave system the sizes of typical electronic components are often comparable to (i.e., within approximately 1 order of magnitude of) the signal wavelength. A schematic diagram illustrating this concept is shown in Figure I.2. As illustrated in Figure I.2, for components much smaller than the wavelength (i.e., λ/10), the finite velocity of the electromagnetic signal as it propagates through the component leads to a modest difference in phase at opposite ends of the com- ponent. For components comparable to or larger than the wavelength, however, this end-to-end phase difference becomes increasingly significant. This gives rise to a reasonable working definition of the two design areas based on the underlying approximations used in design. Since in conventional RF design, the circuit components and interconnections are generally small compared to a wavelength, they can be
- 28. 7218: “intro” — 2007/8/28 — 18:10 — page 3 — #3 Introduction to Microwaves and RF I-3 Component Signal waveform Component Signal waveform (a) (b) FIGURE I.2 Schematic representation of component dimensions relative to signal wavelengths. Conventional lumped-element analysis techniques are typically applicable for components for which λ/10 (a) since the phase change due to electromagnetic propagation across the component is small, while for components with λ/10 (b) the phase change is significant and a distributed circuit description is more appropriate. modeled as lumped elements for which Kirchoff’s voltage and current laws apply at every instant in time. Parasitic inductances and capacitances are incorporated to accurately model the frequency dependencies and the phase shifts, but these quantities can, to good approximation, be treated with an appropriate lumped-element equivalent circuit. In practice, a rule of thumb for the applicability of a lumped-element equivalent circuit is that the component size should be less than λ/10 at the frequency of operation. For microwave frequencies for which component size exceeds approximately λ/10, the finite propagation velocity of electromagnetic waves can no longer be as easily absorbed into simple lumped-element equi- valent circuits. For these frequencies, the time delay associated with signal propagation from one end of a component to the other is an appreciable fraction of the signal period, and thus lumped-element descrip- tions are no longer adequate to describe the electrical behavior. A distributed-element model is required to accurately capture the electrical behavior. The time delay associated with finite wave propagation velocity that gives rise to the distributed circuit effects is a distinguishing feature of the mindset of microwave engineering. An alternative viewpoint is based on the observation that microwave engineering lies in a “middle ground” between traditional low-frequency electronics and optics, as shown in Figure I.1. As a con- sequence of RF, microwaves, and optics simply being different regimes of the same electromagnetic phenomena, there is a gradual transition between these regimes. The continuity of these regimes results in constant re-evaluation of the appropriate design strategies and trade-offs as device and circuit technology advances. For example, miniaturization of active and passive components often increases the frequen- cies at which lumped-element circuit models are sufficiently accurate, since by reducing component dimensions the time delay for propagation through a component is proportionally reduced. As a con- sequence, lumped-element components at “microwave” frequencies are becoming increasingly common in systems previously based on distributed elements due to significant advances in miniaturization, even though the operational frequencies remain unchanged. Component and circuit miniaturization also leads to tighter packing of interconnects and components, potentially introducing new parasitic coup- ling and distributed-element effects into circuits that could previously be treated using lumped-element RF models.
- 29. 7218: “intro” — 2007/8/28 — 18:10 — page 4 — #4 I-4 RF and Microwave Circuits, Measurements, and Modeling The comparable scales of components and signal wavelengths has other implications for the designer as well, since neither the ray-tracing approach from optics nor the lumped-element approach from RF circuit design are valid in this middle ground. In this regard, microwave engineering can also be considered to be “applied electromagnetic engineering,” as the design of guided-wave structures such as waveguides and transmission lines, transitions between different types of transmission lines, and antennae all require analysis and control of the underlying electromagnetic fields. Guided wave structures are particularly important in microwave circuits and systems. There are many different approaches to the implementation of guided-wave structures; a sampling of the more common options are shown in Figure I.3. Figure I.3a shows a section of coaxial cable. In this common cable type, the grounded outer conductor shields the dielectric and inner conductor from external signals and also prevents the signals within the cable from radiating. The propagation in this structure is controlled by the dielectric properties, the cross-sectional geometry, and the metal conductivity. Figure I.3b shows a rectangular waveguide. In this structure, the signal propagates in the free space within the structure, while the rectangular metal structure is grounded. Despite the lack of an analog to the center conductor in the coaxial line, the structure supports traveling-wave solutions to Maxwell’s equations, and thus can be used totransmitpoweralongitslength. Thelackofacenterconductordoespreventthestructurefromproviding any path for dc along its length. The solution to Maxwell’s equations in the rectangular waveguide also leads to multiple eigenmodes, each with its own propagation characteristics (e.g., characteristic impedance and propagation constant), and corresponding cutoff frequency. For frequencies above the cutoff frequency, the mode propagates down the waveguide with little loss, but below the cutoff frequency the mode is Outer conductor Inner conductor Upper conductor Dielectric, r Dielectric, r Dielectric, r Center conductor Center conductor Outer conductors Lower conductor Lower conductor Upper conductor Outer conductor (b) (a) (d) (c) (e) FIGURE I.3 Several common guided-wave structures. (a) coaxial cable, (b) rectangular waveguide, (c) stripline, (d) microstrip, and (e) coplanar waveguide.
- 30. 7218: “intro” — 2007/8/28 — 18:10 — page 5 — #5 Introduction to Microwaves and RF I-5 evanescent and the amplitude falls off exponentially with distance. Since the characteristic impedance and propagation characteristics of each mode are quite different, in many systems the waveguides are sized to support only one propagating mode at the frequency of operation. While metallic waveguides of this type are mechanically inflexible and can be costly to manufacture, they offer extremely low loss and have excellent high-power performance. At W-band and above in particular, these structures currently offer much lower loss than coaxial cable alternatives. Figure I.3c through I.3e show several planar structures that support guided waves. Figure I.3c illustrates the stripline configuration. This structure is in some ways similar to the coaxial cable, with the center conductor of the coaxial line corresponding to the center conductor in the stripline, and the outer shield on the coaxial line corresponding to the upper and lower ground planes in the stripline. Figures I.3d and I.3e show two planar guided-wave structures often encountered in circuit-board and integrated circuit designs. Figure I.3d shows a microstrip configuration, while Figure I.3e shows a coplanar waveguide. Both of these configurations are easily realizable using conventional semiconductor and printed-circuit fabrication techniques. In the case of microstrip lines, the key design variables are the dielectric properties of the substrate, the dielectric thickness, and the width of the top conductor. For the coplanar waveguide case, the dielectric properties of the substrate, the width of the center conductor, the gap between the center and outer ground conductors, and whether or not the bottom surface of the substrate is grounded control the propagation characteristics of the lines. For all of these guided-wave structures, an equivalent circuit consisting of the series concatenation of many stages of the form shown in Figure I.4 can be used to model the transmission line. In this equivalent circuit, the key parameters are the resistance per unit length of the line (R), the inductance per unit length (L), the parallel conductance per unit length of the dielectric (G), and the capacitance per unit length (C). Each of these parameters can be derived from the geometry and material properties of the line. Circuits of this form give rise to traveling-wave solutions of the form V (z) = V + 0 e−γ z + V − 0 eγ z I(z) = V + 0 Z0 e−γ z − V − 0 Z0 eγ z In these equations, the characteristic impedance of the line, which is the constant of proportionality between the current and voltage associated with a particular traveling-wave mode on the line, is given by Z0 = (R + jωL)/(G + jωC). For lossless lines, R = 0 and G = 0, so that Z0 is real; even in many practical cases the loss of the lines is small enough that the characteristic impedance can be treated as real. Similarly, the propagation constant of the line can be expressed as γ = α +jβ = (R + jωL)(G + jωC). In this expression, α characterizes the loss of the line, and β captures the wave propagation. For lossless lines, γ is pure imaginary, and thus α is zero. The design and analysis of these guided-wave structures is treated in more detail in Chapter 30 of the companion volume RF and Microwave Applications and Systems in this handbook series. The distinction between RF and microwave engineering is further blurred by the trend of increasing commercialization and consumerization of systems using what have been traditionally considered to be microwave frequencies. Traditional microwave engineering, with its historically military applications, R L C G I(z + ∆z,t) I(z,t) V(z,t) V(z + ∆z,t) + − − + FIGURE I.4 Equivalent circuit for an incremental length of transmission line. A finite length of transmission line can be modeled as a series concatenation of sections of this form.
- 31. 7218: “intro” — 2007/8/28 — 18:10 — page 6 — #6 I-6 RF and Microwave Circuits, Measurements, and Modeling has long been focused on delivering performance at any cost. As a consequence, special-purpose devices intended solely for use in high performance microwave systems and often with somewhat narrow ranges of applicability were developed to achieve the required performance. With continuing advances in silicon microelectronics, including Si bipolar junction transistors, SiGe heterojunction bipolar transistors (HBTs) and conventional scaled CMOS, microwave-frequency systems can now be reasonably implemented using the same devices as conventional low-frequency baseband electronics. These advanced silicon-based act- ive devices are discussed in more detail in the companion volume RF and Microwave Passive and Active Technologies, Chapters 16–19. In addition, the commercialization of low-cost III–V compound semi- conductor electronics, including ion-implanted metal semiconductor field-effect transistors (MESFETs), pseudomorphic and lattice-matched high electron mobility transistors (HEMTs), and III–V HBTs, has dramatically decreased the cost of including these elements in high-volume consumer systems. These compound-semiconductor devices are described in Chapters 17 and 20–22 in the RF and Microwave Passive and Active Technologies volume of this handbook series. This convergence, with silicon microelec- tronics moving ever higher in frequency into the microwave spectrum from the low-frequency side and compound semiconductors declining in price for the middle of the frequency range, blurs the distinc- tion between “microwave” and “RF” engineering, since “microwave” functions can now be realized with “mainstream” low-cost electronics. This is accompanied by a shift from physically large, low-integration- level hybrid implementations to highly-integrated solutions based on monolithic microwave integrated circuits (MMICs) (see Chapters 25–26 of this volume and Chapters 24–25 in the companion volume RF and Microwave Passive and Active Technologies). This shift has a dramatic effect not only on the design of systems and components, but also on the manufacturing technology and economics of production and implementation as well. A more complete discussion of the active device and integration technologies that make this progression possible is included in Section II of the companion volume RF and Microwave Passive and Active Technologies while modeling of these devices is described in Section III of this volume. Aside from these defining characteristics of RF and microwave systems, the behavior of materials is also often different at microwave frequencies than at low frequencies. In metals, the effective resistance at microwave frequencies can differ significantly from that at dc. This frequency-dependent resistance is a consequence of the skin effect, which is caused by the finite penetration depth of an electromagnetic field into conducting material. This effect is a function of frequency; the depth of penetration is given by δs = (1/ πf µσ), where µ is the permeability, f is the frequency, and σ is the conductivity of the material. As the expression indicates, δs decreases with increasing frequency, and so the electromagnetic fields are confined to regions increasingly near the surface as the frequency increases. This results in the microwave currents flowing exclusively along the surface of the conductor, significantly increasing the effective resistance (and thus the loss) of metallic interconnects. Further discussion of this topic can be found in Chapter 28 of the companion volume RF and Microwave Applications and Systems and Chapter 26 of the RF and Microwave Passive and Active Technologies volume in this handbook series. Dielectric materials also exhibit frequency-dependent characteristics that can be important. The permeability and loss of dielectrics arises from the internal polarization and dissipation of the material. Since the polarization within a dielectric is governed by the response of the material’s internal charge distribution, the frequency dependence is governed by the speed at which these charges can redistribute in response to the applied fields. For ideal materials, this dielectric relaxation leads to a frequency-dependent permittivity of the form ε(ω) = ε∞ + (εdc − ε∞)/(1 + jωτ), where εdc is the low-frequency permittivity, ε∞ is the high- frequency(optical)permittivity, andτ isthedielectricrelaxationtime. Lossinthedielectricisincorporated in this expression through the imaginary part of ε. For many materials the dielectric relaxation time is sufficiently small that the performance of the dielectric at microwave frequencies is very similar to that at low frequencies. However, this is not universal and some care is required since some materials and devices exhibit dispersive behavior at quite low frequencies. Furthermore, this description of dielectrics is highly idealized; the frequency response of many real-world materials is much more complex than this idealized model would suggest. High-value capacitors and semiconductor devices are among the classes of devices that are particularly likely to exhibit complex dielectric responses. In addition to material properties, some physical effects are significant at microwave frequencies that are typically negligible at lower frequencies. For example, radiation losses become increasingly important
- 32. 7218: “intro” — 2007/8/28 — 18:10 — page 7 — #7 Introduction to Microwaves and RF I-7 as the signal wavelengths approach the component and interconnect dimensions. For conductors and other components of comparable size to the signal wavelengths, standing waves caused by reflection of the electromagnetic waves from the boundaries of the component can greatly enhance the radiation of electromagnetic energy. These standing waves can be easily established either intentionally (in the case of antennae and resonant structures) or unintentionally (in the case of abrupt transitions, poor circuit layout, or other imperfections). Careful attention to transmission line geometry, placement relative to other components, transmission lines, and ground planes, as well as circuit packaging is essential for avoiding excessive signal attenuation and unintended coupling due to radiative effects. A further distinction in the practice of RF and microwave engineering from conventional electronics is the methodology of testing and characterization. Due to the high frequencies involved, the imped- ance and standing-wave effects associated with test cables and the parasitic capacitance of conventional test probes make the use of conventional low-frequency circuit characterization techniques impractical. Although advanced measurement techniques such as electro-optic sampling can sometimes be employed to circumvent these difficulties, in general the loading effect of measurement equipment poses significant measurement challenges for debugging and analyzing circuit performance, especially for nodes at the interior of the circuit under test. In addition, for circuits employing dielectric or hollow guided-wave structures, voltage and current often cannot be uniquely defined. Even for structures in which voltage and current are well-defined, practical difficulties associated with accurately measuring such high-frequency signals make this difficult. Furthermore, since a dc-coupled time-domain measurement of a microwave signal would have an extremely wide noise bandwidth, the sensitivity of the measurement would be inadequate. For these reasons, components and low-level subsystems are characterized using specialized techniques. One of the most common techniques for characterizing the linear behavior of microwave components is the use of s-parameters. While z-, y-, and h-parameter representations are commonly used at lower frequencies, these approaches can be problematic to implement at microwave frequencies. The use of s-parameters essentially captures the same information as these other parameter sets, but instead of directly measuring terminal voltages and currents, the forward and reverse traveling waves at the input and output ports are measured instead. While perhaps not intuitive at first, this approach enables accurate characterization of components at very high frequencies to be performed with comparative ease. For a two-port network, the s-parameters are defined by: V − 1 V − 2 = s11 s12 s21 s22 V + 1 V + 2 where the V − terms are the wave components traveling away from the two-port, and the V + terms are the incident terms. These traveling waves can be thought of as existing on “virtual” transmission lines attached to the device ports. From this definition, s11 = V − 1 V + 1 V + 2 =0 s12 = V − 1 V + 2 V + 1 =0 s21 = V − 2 V + 1 V + 2 =0 s22 = V − 2 V + 2 V + 1 =0
- 33. 7218: “intro” — 2007/8/28 — 18:10 — page 8 — #8 I-8 RF and Microwave Circuits, Measurements, and Modeling To measure the s-parameters, the ratio of the forward and reverse traveling waves on the virtual input and output transmission lines is measured. To achieve the V + 1 = 0 and V + 2 = 0 conditions in these expressions, the ports are terminated in the characteristic impedance, Z0, of the virtual transmission lines. Although in principle these measurements can be made using directional couplers to separate the forward and reverse traveling waves and phase-sensitive detectors, in practice modern network analyzers augment the measurement hardware with sophisticated calibration routines to remove the effects of hardware imperfections to achieve accurate s-parameter measurements. A more detailed discussion of s-parameters, as well as other approaches to device and circuit characterization, is provided in Section I of this volume. I.2 General Applications The field of microwave engineering is currently experiencing a radical transformation. Historically, the field has been driven by applications requiring the utmost in performance with little concern for cost or manufacturability. These systems have been primarily for military applications, where performance at nearly any cost could be justified. The current transformation of the field involves a dramatic shift from defense applications to those driven by the commercial and consumer sector, with an attendant shift in focus from design for performance to design for manufacturability. This transformation also entails a shift from small production volumes to mass production for the commercial market, and from a focus on performance without regard to cost to a focus on minimum cost while maintaining acceptable performance. For wireless applications, an additional shift from broadband systems to systems having very tightly-regulated spectral characteristics also accompanies this transformation. For many years the driving application of microwave technology was military radar. The small wavelength of microwaves permits the realization of narrowly-focused beams to be achieved with antennae small enough to be practically steered, resulting in adequate resolution of target location. Long-distance terrestrial communications for telephony as well as satellite uplink and downlink for voice and video were among the first commercially viable applications of microwave technology. These commercial commu- nications applications were successful because microwave-frequency carriers (fc) offer the possibility of very wide absolute signal bandwidths (f ) while still maintaining relatively narrow fractional bandwidths (i.e., f /fc). This allows many more voice and data channels to be accommodated than would be possible with lower-frequency carriers or baseband transmission. Amongthecurrenthostofemergingapplications, manyarebasedlargelyonthissameprinciple, namely, the need to transmit more and more data at high speed, and thus the need for many communication channels with wide bandwidths. Wireless communication of voice and data, both to and from individual users as well as from users and central offices in aggregate, wired communication including coaxial cable systems for video distribution and broadband digital access, fiber-optic communication systems for long- and short-haul telecommunication, and hybrid systems such as hybrid fiber-coax systems are all designed to take advantage of the wide bandwidths and consequently high data carrying capacity of microwave-frequency electronic systems. The widespread proliferation of wireless Bluetooth personal- area networks and WiFi local-area networks for transmission of voice, data, messaging and online services operating in the unlicensed ISM bands is an example of the commoditization of microwave technology for cost-sensitive consumer applications. In addition to the explosion in both diversity and capability of microwave-frequency communication systems, radar systems continue to be of importance with the emergence of nonmilitary and nonnavigational applications such as radar systems for automotive collision avoidance and weather and atmospheric sensing. Radar based noncontact fluid-level sensors are also increasingly being used in industrial process control applications. Traditional applications of microwaves in industrial material processing (primarily via nonradiative heating effects) and cooking have recently been augmented with medical uses for microwave-induced localized hyperthermia for oncological and other medical treatments.
- 34. 7218: “intro” — 2007/8/28 — 18:10 — page 9 — #9 Introduction to Microwaves and RF I-9 In addition to these extensions of “traditional” microwave applications, other fields of electronics are increasing encroaching into the microwave-frequency range. Examples include wired data net- works based on coaxial cable or twisted-pair transmission lines with bit rates of over 1 Gb/s, fiber-optic communication systems with data rates well in excess of 10 Gb/s, and inexpensive per- sonal computers and other digital systems with clock rates of well over 1 GHz. The continuing advances in the speed and capability of conventional microelectronics is pushing traditional circuit design ever further into the microwave-frequency regime. These advances have continued to push digital circuits into regimes where distributed circuit effects must be considered. While system- and board-level digital designers transitioned to the use of high-speed serial links requiring the use of distributed transmission lines in their designs some time ago, on-chip transmission lines for distribu- tion of clock signals and the serialization of data signals for transmission over extremely high-speed serial buses are now an established feature of high-end designs within a single integrated circuit. These trends promise to both invigorate and reshape the field of microwave engineering in new and exciting ways. I.3 Frequency Band Definitions The field of microwave and RF engineering is driven by applications, originally for military purposes such as radar and more recently increasingly for commercial, scientific, and consumer applications. As a consequence of this increasingly diverse applications base, microwave terminology and frequency band designations are not entirely standardized, with various standards bodies, corporations, and other interested parties all contributing to the collective terminology of microwave engineering. Figure I.5 shows graphically the frequency ranges of some of the most common band designations. As can be seen from the complexity of Figure I.5, some care must be exercised in the use of the “standard” letter designations; substantial differences in the definitions of these bands exist in the literature and in practice. While the IEEE standard for radar bands [8] expressly deprecates the use of radar band designations for nonradar applications, the convenience of the band designations as technical shorthand has led to the use of these band designations in practice for a wide range of systems and technologies. This appropriation of radar band designations for other applications, as well as the definition of other letter-designated bands for other applications (e.g., electronic countermeasures) that have different frequency ranges is in part responsible for the complexity of Figure I.5. Furthermore, as progress in device and system performance opens up new system possibilities and makes ever-higher frequencies useful for new systems, the terminology of microwave engineering is continually evolving. Figure I.5 illustrates in approximate order of increasing frequency the range of frequencies encompassed by commonly-used letter-designated bands. In Figure I.5, the dark shaded regions within the bars indicate the IEEE radar band designations, and the light cross-hatching indicates variations in the definitions by different groups and authors. The double-ended arrows appearing above some of the bands indicate other non-IEEE definitions for these letter designations that appear in the literature. For example, multiple distinct definitions of L, S, C, X, and K band are in use. The IEEE defines K band as the range from 18 to 27 GHz, while some authors define K band to span the range from 10.9 to 36 GHz, encompassing most of the IEEE’s Ku, K, and Ka bands within a single band. Both of these definitions are illustrated in Figure I.5. Similarly, L band has two substantially different, overlapping definitions, with the IEEE definition of L band including frequencies from 1 to 2 GHz, with an older alternative definition of 390 MHz–1.55 GHz being found occasionally in the literature. Many other bands exhibit similar, though perhaps less extreme, variations in their definitions by various authors and standards committees. A further caution must also be taken with these letter designations, as different standards bodies and agencies do not always ensure that their letter designations are not used by others. As an example, the IEEE and U.S. military both define C, L, and K bands, but with very different frequencies; the IEEE L band resides at the low end of the microwave spectrum, while the military definition of L band is from 40 to 60 GHz. The designations (L–Y) in Figure I.5a are presently used widely in practice and the technical literature, with the newer U.S. military
- 35. 7218: “intro” — 2007/8/28 — 18:10 — page 10 — #10 I-10 RF and Microwave Circuits, Measurements, and Modeling L 0.39 GHz 1.55 GHz 3.9 GHz 2 GHz 12 GHz 5.2 GHz 6.2 GHz 8 GHz 10.9 GHz 12.4 GHz 17.25 GHz 18 GHz 26 GHz 27 GHz 36 GHz 15.35 GHz 33 GHz 24.5 GHz 40 GHz 46 GHz 50 GHz 60 GHz 75 GHz 90 GHz 110 GHz 170 GHz 325 GHz 220 GHz 140 GHz 0.1 GHz 0.5 GHz 2 GHz 4 GHz 8 GHz 40 GHz 100 GHz 140 GHz 60 GHz 20 GHz 10 GHz 6 GHz 3 GHz 1 GHz 0.25 GHz 56 GHz 40 GHz 1 GHz 4 GHz S C X Ku K Band designation K1 Ka Q U V E W D G Y A B C D E F Band designation G H I J K L M N 0.1 1 10 Frequency (GHz) 100 0.1 1 10 Frequency (GHz) 100 K 10.9 GHz FIGURE I.5 Microwave and RF frequency band designations [1–7]. (a) Industrial and IEEE designations. Diagonal hashing indicates variation in the definitions found in literature; dark regions in the bars indicate the IEEE radar band definitions [8]. Double-ended arrows appearing above bands indicate alternative band definitions appearing in the literature, and K† denotes an alternative definition for K band found in Reference [7]. (b) U.S. military frequency band designations [2–5].
- 36. 7218: “intro” — 2007/8/28 — 18:10 — page 11 — #11 Introduction to Microwaves and RF I-11 designations (A–N) shown in Figure I.5b having not gained widespread popularity outside of the military community. I.4 Overview of The RF and Microwave Handbook The field of microwave and RF engineering is inherently interdisciplinary, spanning the fields of system architecture, design, modeling, and validation; circuit design, characterization, and verification; active and passive device design, modeling, and fabrication, including technologies as varied as semiconductor devices, solid-state passives, and vacuum electronics; electromagnetic field theory, atmospheric wave propagation, electromagnetic compatibility and interference; and manufacturing, reliability and system integration. Additional factors, including biological effects of high-frequency radiation, system cost, and market factors also play key roles in the practice of microwave and RF engineering. This extremely broad scope is further amplified by the large number of technological and market-driven design choices faced by the practitioner on a regular basis. The full sweep of microwave and RF engineering is addressed in this three-volume handbook series. Section I of this volume features coverage of the unique difficulties and challenges encountered in accurately measuring microwave and RF devices and components, including linear and non-linear char- acterization approaches, load-pull and large-signal network analysis techniques, noise measurements, fixturing and high-volume testing issues, and testing of digital systems. Consideration of key circuits for functional blocks in a wide array of system applications is addressed in Section II, including low-level circuits such as low-noise amplifiers, mixers, oscillators, power amplifiers, switches, and filters, as well as higher-level functionalities such as receivers, transmitters, and phase-locked loops. Section III of this volume discusses technology computer-aided design (TCAD) and nonlinear modeling of devices and circuits, along with analysis tools for systems, electromagnetics, and circuits. A companion volume in this handbook series, RF and Microwave Applications and Systems, features detailed discussion of system-level considerations for high-frequency systems. Section I of this companion volume focuses on system-level considerations with an application-specific focus. Typical applications, ranging from nomadic communications and cellular systems, wireless local-area networks, analog fiber- optic links, satellite communication networks, navigational aids and avionics, to radar, medical therapies, and electronic warfare applications are examined in detail. System-level considerations from the viewpoint of system integration and with focus on issues such as thermal management, cost modeling, manufactur- ing, and reliability are addressed in Section II of this volume in the handbook series, while the fundamental physical principles that govern the operation of devices and microwave and RF systems generally are dis- cussed in Section III. Particular emphasis is placed on electromagnetic field theory through Maxwell’s equations, free-space and guided-wave propagation, fading and multipath effects in wireless channels, and electromagnetic interference effects. Comprehensive coverage of passive and active device technologies for microwave and RF systems is provided in a third companion volume in the handbook series, RF and Microwave Passive and Active Tech- nologies. Passive devices are discussed in Section I of this volume, which includes coverage of radiating elements, cables and connectors, and packaging technology, as well as in-circuit passive elements includ- ing resonators, filters, and other components. The fundamentals of active device technologies, including semiconductor diodes, transistors and integrated circuits as well as vacuum electron devices, are discussed in Section II. Key device technologies including varactor and Schottky diodes, as well as bipolar junc- tion transistors and heterojunction bipolar transistors in both the SiGe and III-V material systems are described, as are Si MOSFETs and III-V MESFETs and HEMTs. A discussion of the fundamental phys- ical properties at high frequencies of common materials, including metals, dielectrics, ferroelectric and piezoelectric materials, and semiconductors, is provided in Section III of this volume in the handbook series.
- 37. 7218: “intro” — 2007/8/28 — 18:10 — page 12 — #12 I-12 RF and Microwave Circuits, Measurements, and Modeling References 1. Chang, K., Bahl, I., and Nair, V., RF and Microwave Circuit and Component Design for Wireless Systems, John Wiley Sons, New York, 2002. 2. Collin, R. E., Foundations for Microwave Engineering, McGraw-Hill, New York, 1992, 2. 3. Harsany, S. C., Principles of Microwave Technology, Prentice Hall, Upper Saddle River, 1997, 5. 4. Laverghetta, T. S., Modern Microwave Measurements and Techniques, Artech House, Norwood, 1988, 479. 5. Misra, D. K., Radio-Frequency and Microwave Communication Circuits: Analysis and Design, John Wiley Sons, New York, 2001. 6. Rizzi, P. A., Microwave Engineering, Prentice-Hall, Englewood Cliffs, 1988, 1. 7. Reference Data for Radio Engineers, ITT Corp., New York, 1975. 8. IEEE Std. 521-2002.
- 38. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 1 — #1 1 Overview of Microwave Engineering Mike Golio HVVi Semiconductor 1.1 Semiconductor Materials for RF and Microwave Applications .............................................. 1-1 1.2 Propagation and Attenuation in the Atmosphere ..... 1-3 1.3 Systems Applications .................................... 1-5 Communications • Navigation • Sensors (Radar) • Heating 1.4 Measurements............................................ 1-7 Small Signal • Large Signal • Noise • Pulsed I –V 1.5 Circuits and Circuit Technologies ...................... 1-16 Low Noise Amplifier • Power Amplifier • Mixer • RF Switch • Filter • Oscillator 1.6 CAD, Simulation, and Modeling ....................... 1-19 References ....................................................... 1-20 1.1 Semiconductor Materials for RF and Microwave Applications In addition to consideration of unique properties of metal and dielectric materials, the radio frequency (RF) and microwave engineer must also make semiconductor choices based on how existing semicon- ductorpropertiesaddresstheuniquerequirementsofRFandmicrowavesystems. Althoughsemiconductor materials are exploited in virtually all electronics applications today, the unique characteristics of RF and microwave signals requires that special attention be paid to specific properties of semiconductors which are often neglected or of second-order importance for other applications. Two critical issues to RF applic- ations are (a) the speed of electrons in the semiconductor material and (b) the breakdown field of the semiconductor material. The first issue, speed of electrons, is clearly important because the semiconductor device must respond to high frequency changes in polarity of the signal. Improvements in efficiency and reductions in parasitic losses are realized when semiconductor materials are used which exhibit high electron mobility and velocity. Figure 1.1 presents the electron velocity of several important semiconductor materials as a 1-1
- 39. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 2 — #2 1-2 RF and Microwave Circuits, Measurements, and Modeling 102 103 104 105 106 105 106 107 108 Electric field (V/cm) Electron drift velocity (cm/s) G a . 4 7 I n . 5 3 A s G aAs Si InP FIGURE 1.1 The electron velocity as a function of applied electric field for several semiconductor materials which are important for RF and microwave applications. TABLE 1.1 Mobility and Breakdown Electric Field Values for Several Semiconductors Important for RF and Microwave Transmitter Applications Property Si SiC InP GaAs GaN Electron mobility (cm2/Vs) 1900 40–1000 4600 8800 1000 Breakdown field (V/cm) 3 × 105 20 × 104 to 30 × 105 5 × 105 6 × 105 10 × 105 function of applied electric field. The carrier mobility is given by µc = ν e for small values of E (1.1) where ν is the carrier velocity in the material and E is the electric field. Although Silicon is the dominant semiconductor material for electronics applications today, Figure 1.1 illustrates that III–V semiconductor materials such as GaAs, GaInAs, and InP exhibit superior electron velocity and mobility characteristics relative to Silicon. Bulk mobility values for several important semi- conductors are also listed in Table 1.1. As a result of the superior transport properties, transistors fabricated using III–V semiconductor materials such as GaAs, InP, and GaInAs exhibit higher efficiency and lower parasitic resistance at microwave frequencies. From a purely technical performance perspective, the above discussion argues primarily for the use of III–V semiconductor devices in RF and microwave applications. These arguments are not complete, however. Most commercial wireless products also have requirements for high yield, high volume, low cost, and rapid product development cycles. These requirements can overwhelm the material selection process and favor mature processes and high volume experience. The silicon high volume manufacturing experience base is far greater than that of any III–V semiconductor facility. The frequency of the application becomes a critical performance characteristic in the selection of device technology. Because of the fundamental material characteristics illustrated in Figure 1.1, Sil- icon device structures will always have lower theoretical maximum operation frequencies than identical III–V device structures. The higher the frequency of the application, the more likely the optimum device choice will be a III–V transistor over a Silicon transistor. Above some frequency, fIII−V, compound semiconductor devices dominate the application space, with Silicon playing no significant role in the
- 40. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 3 — #3 Overview of Microwave Engineering 1-3 microwave portion of the product. In contrast, below some frequency, fSi, the cost and maturity advantage of Silicon provide little opportunity for III–V devices to compete. In the transition spectrum between these two frequencies Silicon and III–V devices coexist. Although Silicon devices are capable of operating above frequency fSi, this operation is often gained at the expense of DC current drain. As frequency is increased above fSi in the transition spectrum, efficiency advantages of GaAs and other III–V devices provide com- petitive opportunities for these parts. The critical frequencies, fSi and fIII−V are not static frequency values. Rather, they are continually being moved upward by the advances of Silicon technologies—primarily by decreasing critical device dimensions. The speed of carriers in a semiconductor transistor can also be affected by deep levels (traps) located physically either at the surface or in the bulk material. Deep levels can trap charge for times that are long compared to the signal period and thereby reduce the total RF power carrying capability of the transistor. Trapping effects result in frequency dispersion of important transistor characteristics such as transconductance and output resistance. Pulsed measurements as described in Section 1.4.4 (especially when taken over temperature extremes) can be a valuable tool to characterize deep level effects in semi- conductor devices. Trapping effects are more important in compound semiconductor devices than in silicon technologies. The second critical semiconductor issue listed in Table 1.1 is breakdown voltage. The constraints placed on the RF portion of radio electronics are fundamentally different from the constraints placed on digital circuits in the same radio. For digital applications, the presence or absence of a single electron can theoret- ically define a bit. Although noise floor and leakage issues make the practical limit for bit signals larger than this, the minimum amount of charge required to define a bit is very small. The bit charge minimum is also independent of the radio system architecture, the radio transmission path or the external environment. If the amount of charge utilized to define a bit within the digital chip can be reduced, then operating voltage, operating current, or both can also be reduced with no adverse consequences for the radio. In contrast, the required propagation distance and signal environment are the primary determinants for RF signal strength. If 1 W of transmission power is required for the remote receiver to receive the signal, then reductions in RF transmitter power below this level will cause the radio to fail. Modern radio requirements often require tens, hundreds, or even thousands of Watts of transmitted power in order for the radio system to function properly. Unlike the digital situation where any discernable bit is as good as any other bit, the minimum RF transmission power must be maintained. A Watt of RF power is the product of signal current, signal voltage and efficiency, so requirements for high power result in requirements for high voltage, high current and high efficiency. The maximum electric field before the onset of avalanche breakdown, breakdown field, is the fun- damental semiconductor property that often limits power operation in a transistor. Table 1.1 presents breakdown voltages for several semiconductors that are commonly used in transmitter applications. In addition to Silicon, GaAs and InP, two emerging widebandgap semiconductors, SiC and GaN are included in the table. Interest from microwave engineers in these less mature semiconductors is driven almost exclusively by their attractive breakdown field capabilities. Figure 1.2 summarizes the semiconductor material application situation in terms of the power–frequency space for RF and microwave systems. 1.2 Propagation and Attenuation in the Atmosphere Many modern RF and microwave systems are wireless. Their operation depends on transmission of signals through the atmosphere. Electromagnetic signals are attenuated by the atmosphere as they propagate from source to target. Consideration of the attenuation characteristics of the atmosphere can be critical in the design of these systems. In general, atmospheric attenuation increases with increasing frequency. As shown in Figure 1.3, however, there is significant structure in the atmospheric attenuation versus frequency plot. If only attenuation is considered, it is clear that low frequencies would be preferred for long range communications, sensor, or navigation systems in order to take advantage of the low attenuation of the atmosphere. If high data rates or large information content is required, however, higher frequencies
- 41. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 4 — #4 1-4 RF and Microwave Circuits, Measurements, and Modeling 0.1 1 10 100 10.0 1.0 100.0 S i: B J T Power (W) Frequency (GHz) S iG e : H B T I I I – V : H B T I I I – V : H E M T SiC: MESFET GaN: HEMT 1000 FIGURE 1.2 Semiconductor choices for RF applications are a strong function of the power and frequency required for the wireless application. 20 50 10 100 200 Frequency (GHz) 100 10 1 0.1 Attenuation (dB/km) FIGURE 1.3 Attenuation of electromagnetic signals in the atmosphere as a function of frequency. are needed. In addition to the atmospheric attenuation, the wavelengths of microwave systems are small enough to become effected by water vapor and rain. Above 10 GHz these effects become important. Above 25 GHz, the effect of individual gas molecules becomes important. Water and oxygen are the most important gases. These have resonant absorption lines at ∼23, ∼69, and ∼120 GHz. In addition to absorption lines, the atmosphere also exhibits “windows” that may be used for communication, notably at ∼38 and ∼98 GHz. RF and microwave signal propagation is also affected by objects such as trees, buildings, towers, and vehicles in the path of the wave. Indoor systems are affected by walls, doors, furniture, and people. As a result of the interaction of electromagnetic signals with objects, the propagation channel for wireless communication systems consists of multiple paths between the transmitter and receiver. Each path will experience different attenuation and delay. Some transmitted signals may experience a deep fade (large attenuation) due to destructive multipath cancellation. Similarly, constructive multipath addition can produce signals of large amplitude. Shadowing can occur when buildings or other objects obstruct the line-of-site path between transmitter and receiver. The design of wireless systems must consider the interaction of specific frequencies of RF and microwave signals with the atmosphere and with objects in the signal channel that can cause multipath effects.
- 42. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 5 — #5 Overview of Microwave Engineering 1-5 1.3 Systems Applications There are four important classes of applications for microwave and RF systems: communications, naviga- tion, sensors, and heating. Each of these classes of applications benefits from some of the unique properties of high-frequency electromagnetic fields. 1.3.1 Communications Wireless communications applications have exploded in popularity over the past decade. Pagers, cellular phones, radio navigation, and wireless data networks are among the RF products that consumers are likely to be familiar with. Prior to the growth of commercial wireless communications, RF and microwave radios were in common usage for communications satellites, commercial avionics communications, and many government and military radios. All of these systems benefit from the high frequencies that offer greater bandwidth than low frequency systems, while still propagating with relatively low atmospheric losses compared to higher frequency systems. Cellular phones are among the most common consumer radios in use today. Analog cellular (first generation or 1G cellular) operates at 900 MHz bands and was first introduced in 1983. Second generation (2G) cellular using TDMA, GSM TDMA, and CDMA digital modulation schemes came into use more than 10 years later. The 2G systems were designed to get greater use of the 1.9 GHz frequency bands than their analog predecessors. Emergence of 2.5G and 3G systems operating in broader bands as high as 2.1 GHz is occurring today. These systems make use of digital modulation schemes adapted from 2G GSM and CDMA systems. With each advance in cellular phones, requirements on the microwave circuitry have increased. Requirements for broader bandwidths, higher efficiency and greater linearity have been coupled with demands for lower cost, lighter, smaller products, and increasing functionality. The microwave receivers and transmitters designed for portable cellular phones represent one of the highest volume manufacturing requirements of any microwave radio. Fabrication of popular cell phones has placed an emphasis on manufacturability and yield for microwave radios that was unheard of prior to the growth in popularity of these products. Other microwave-based consumer products that are growing dramatically in popularity are the wireless local area network (WLAN) or Wi-Fi and the longer range WiMAX systems. These systems offer data rates more than five times higher than cellular-based products using bandwidth at 2.4, 3.5, and 5 GHz. Although the volume demands for Wi-Fi and WiMAX components are not as high as for cellular phones, the emphasis on cost and manufacturability is still critical to these products. Commercial communications satellite systems represent a microwave communications product that is less conspicuous to the consumer, but continues to experience increasing demand. Although the percent- age of voice traffic carried via satellite systems is rapidly declining with the advent of undersea fiber-optic cables, new video and data services are being added over existing voice services. Today satellites provide worldwide TV channels, global messaging services, positioning information, communications from ships and aircraft, communications to remote areas, and high-speed data services including internet access. Allocated satellite communication frequency bands include spectrum from as low as 2.5 GHz to almost 50 GHz. These allocations cover extremely broad bandwidths compared to many other communica- tions systems. Future allocation will include even higher frequency bands. In addition to the bandwidth and frequency challenges, microwave components for satellite communications are faced with reliability requirements that are far more severe than any earth-based systems. Avionics applications include subsystems that perform communications, navigation, and sensor applic- ations. Avionics products typically require functional integrity and reliability that are orders of magnitude more stringent than most commercial wireless applications. The rigor of these requirements is matched or exceeded only by the requirements for space and/or certain military applications. Avionics must function in environments that are more severe than most other wireless applications as well. Quantities of products required for this market are typically very low when compared to commercial wireless applications, for example, the number of cell phones manufactured every single working day far exceeds the number of
- 43. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 6 — #6 1-6 RF and Microwave Circuits, Measurements, and Modeling aircraft that are manufactured in the world in a year. Wireless systems for avionics applications cover an extremely wide range of frequencies, function, modulation type, bandwidth, and power. Due to the number of systems aboard a typical aircraft, Electromagnetic Interference (EMI) and Electromagnetic Compatibility (EMC) between systems is a major concern, and EMI/EMC design and testing is a major factor in the flight certification testing of these systems. RF and microwave communications systems for avionics applications include several distinct bands between 2 and 400 MHz and output power requirements as high as 100 Watts. In addition to commercial communications systems, military communication is an extremely import- ant application of microwave technology. Technical specifications for military radios are often extremely demanding. Much of the technology developed and exploited by existing commercial communications systems today was first demonstrated for military applications. The requirements for military radio applic- ations are varied but will cover broader bandwidths, higher power, more linearity, and greater levels of integration than most of their commercial counterparts. In addition, reliability requirements for these systems are stringent. Volume manufacturing levels, of course, tend to be much lower than commercial systems. 1.3.2 Navigation Electronic navigation systems represent a unique application of microwave systems. In this application, data transfer takes place between a satellite (or fixed basestation) and a portable radio on earth. The consumer portable product consists of only a receiver portion of a radio. No data or voice signal is trans- mitted by the portable navigation unit. In this respect, electronic navigation systems resemble a portable paging system more closely than they resemble a cellular phone system. The most widespread electronic navigation system is GPS. The nominal GPS constellation is composed of 24 satellites in six orbital planes, (four satellites in each plane). The satellites operate in circular 20,200 km altitude (26,570 km radius) orbits at an inclination angle of 55◦. Each satellite transmits a navigation message containing its orbital elements, clock behavior, system time, and status messages. The data transmitted by the satellite are sent in two frequency bands at 1.2 and 1.6 GHz. The portable terrestrial units receive these messages from multiple satellites and calculate the location of the unit on the earth. In addition to GPS, other navigation systems in common usage include NAVSTAR, GLONASS, and LORAN. 1.3.3 Sensors (Radar) Microwave sensor applications are addressed primarily with various forms of radar. Radar is used by police forces to establish the speed of passing automobiles, by automobiles to establish vehicle speed and danger of collision, by air traffic control systems to establish the locations of approaching aircraft, by aircraft to establish ground speed, altitude, other aircraft and turbulent weather, and by the military to establish a multitude of different types of targets. The receiving portion of a radar unit is similar to other radios. It is designed to receive a specific signal and analyze it to obtain desired information. The radar unit differs from other radios, however, in that the signal that is received is typically transmitted by the same unit. By understanding the form of the transmitted signal, the propagation characteristics of the propagation medium, and the form of the received (reflected) signal, various characteristics of the radar target can be determined including size, speed, and distance from the radar unit. As in the case of communications systems, radar applications benefit from the propagation characteristics of RF and microwave frequencies in the atmosphere. The best frequency to use for a radar unit depends upon its application. Like most other radio design decisions, the choice of frequency usually involves trade-offs among several factors including physical size, transmitted power, and atmospheric attenuation. The dimensions of radio components used to generate RF power and the size of the antenna required to direct the transmitted signal are, in general, proportional to wavelength. At lower frequencies where
- 44. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 7 — #7 Overview of Microwave Engineering 1-7 wavelengths are longer, the antennae and radio components tend to be large and heavy. At the higher frequencies where the wavelengths are shorter, radar units can be smaller and lighter. Frequency selection can indirectly influence the radar power level because of its impact on radio size. Design of high power transmitters requires that significant attention be paid to the management of electric field levels and thermal dissipation. Such management tasks are made more complex when space is limited. Since radio component size tends to be inversely proportional to frequency, manageable power levels are reduced as frequency is increased. As in the case of all wireless systems, atmospheric attenuation can reduce the total range of the system. Radar systems designed to work above about 10 GHz must consider the atmospheric loss at the specific frequency being used in the design. Automotive radar represents a large class of radars that are used within an automobile. Applications include speed measurement, adaptive cruise control, obstacle detection, and collision avoidance. Various radar systems have been developed for forward-, rear-, and side-looking applications. V-band frequencies are exploited for forward looking radars. Within V-band, different frequencies have been used in the past decade, including 77 GHz for U.S. and European systems, and 60 GHz in some Japanese systems. The choice of V-band for this application is dictated by the resolution requirement, antenna size requirement and the desire for atmospheric attenuation to insure the radar is short range. The frequency requirement of this application has contributed to a slow emergence of this product into mainstream use, but the potential of this product to have a significant impact on highway safety continues to keep automotive radar efforts active. As in the case of communications systems, avionics and military users also have significant radar applications. Radar is used to detect aircraft both from the earth and from other aircraft. It is also used to determine ground speed, establish altitude, and detect weather turbulence. 1.3.4 Heating The most common heating application for microwave signals is the microwave oven. These consumer products operate at a frequency that corresponds to a resonant frequency of water. When exposed to electromagnetic energy at this frequency, all water molecules begin to spin or oscillate at that frequency. Since all foods contain high percentages of water, electromagnetic energy at this resonant frequency interacts with all foods. The energy absorbed by these rotating molecules is transferred to the food in the form of heat. RF heating can also be important for medical applications. Certain kinds of tumors can be detected by the lack of electromagnetic activity associated with them and some kinds of tumors can be treated by heating them using electromagnetic stimulation. The use of RF/microwaves in medicine has increased dramatically in recent years. RF and microwave therapies for cancer in humans are presently used in many cancer centers. RF treatments for heartbeat irregularities are currently employed by major hospitals. RF/microwaves are also used in human subjects for the treatment of certain types of benign prostrate conditions. Several centers in the United States have been utilizing RF to treat upper airway obstruction and alleviate sleep apnea. New treatments such as microwave aided liposuction, tissue joining in conjunction with microwave irradiation in future endo- scopic surgery, enhancement of drug absorption, and microwave septic wound treatment are continually being researched. 1.4 Measurements The RF/microwave engineer faces unique measurement challenges. At high frequencies, voltages and currents vary too rapidly for conventional electronic measurement equipment to gauge. Conventional curve tracers and oscilloscopes are of limited value when microwave component measurements are needed. In addition, calibration of conventional characterization equipment typically requires the use
- 45. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 8 — #8 1-8 RF and Microwave Circuits, Measurements, and Modeling of open and short circuit standards that are not useful to the microwave engineer. For these reasons, most commonly exploited microwave measurements focus on the measurement of power and phase in the frequency domain as opposed to voltages and currents in the time domain. 1.4.1 Small Signal Characterization of the linear performance of microwave devices, components and boards is critical to the development of models used in the design of the next higher level of microwave subsystem. At lower frequencies, direct measurement of y-, z-, or h-parameters is useful to accomplish linear characterization. As discussed in Chapter 1, however, RF and microwave design utilizes s-parameters for this application. Other small signal characteristics of interest in microwave design include impedance, VSWR, gain, and attenuation. Each of these quantities can be computed from two-port s-parameter data. The s-parameters defined in Chapter 1 are complex quantities normally expressed as magnitude and phase. Notice that S11 and S22 can be thought of as complex reflection ratios since they represent the magnitude and phase of waves reflected from port 1 (input) and 2 (output), respectively. It is common to measure the quality of the match between components using the reflection coefficient defined as = |S11| (1.2) for the input reflection coefficient of a two-port network, or = |S22| (1.3) for the output reflection coefficient. Reflection coefficient measurements are often expressed in dB and referred to as return loss evaluated as Lreturn = −20 log(). (1.4) Analogous to the reflection coefficient, both a forward and reverse transmission coefficient can be measured. The forward transmission coefficient is given as T = |S21| (1.5) while the reverse transmission coefficient is expressed T = |S12| . (1.6) As in the case of reflection coefficient, transmission coefficients are often expressed in dB and referred to as gain given by G = 20 log(T). (1.7) Another commonly measured and calculated parameter is the standing wave ratio or the voltage standing wave ratio (VSWR). This quantity is the ratio of maximum to minimum voltage at a given port. It is commonly expressed in terms of reflection coefficient as VSWR = 1 + 1 − . (1.8) The vector network analyzer (VNA) is the instrument of choice for small signal characterization of high-frequency components. Figure 1.4 illustrates a one-port VNA measurement. These measurements
- 46. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 9 — #9 Overview of Microwave Engineering 1-9 Incident signal Reflected signal Signal path RF RF Signal path Digitizer a1 b1 Frequency conversion Device under test (DUT) Directional coupler Signal generator FIGURE 1.4 Vector network analyzer measurement configuration to determine s-parameters of a high-frequency device, component, or subsystem. use a source with well-defined impedance equal to the system impedance and all ports of the device under test (DUT) are terminated with the same impedance. This termination eliminates unwanted signal reflections during the measurement. The port being measured is terminated in the test channel of the network analyzer that has input impedance equal to the system characteristic impedance. Measurement of system parameters with all ports terminated minimizes the problems caused by short-, open-, and test-circuit parasitics that cause considerable difficulty in the measurement of y- and h-parameters at very high frequencies. If desired, s-parameters can be converted to y- and h-parameters using analytical mathematical expressions. The directional coupler shown in Figure 1.4 is a device for measuring the forward and reflected waves on a transmission line. During the network analyzer measurement, a signal is driven through the directional coupler to one port of the DUT. Part of the incident signal is sampled by the directional coupler. On arrival at the DUT port being measured, some of the incident signal will be reflected. This reflection is again sampled by the directional coupler. The sampled incident and reflected signals are then downconverted in frequency and digitized. The measurement configuration of Figure 1.4 shows only one-half of the equipment required to make full two-port s-parameter measurements. The s-parameters as defined in Chapter 1 are determined by analyzing the ratios of the digitized signal data. For many applications, knowledge of the magnitude of the incident and reflected signals is sufficient (i.e., is all that is needed). In these cases, the scalar network analyzer can be utilized in place of the VNA. The cost of the scalar network analyzer equipment is much less than VNA equipment and the calibration required for making accurate measurements is easier when phase information is not required. The scalar network analyzer measures reflection coefficient as defined in Equations 2.1 and 2.2. 1.4.2 Large Signal Virtually all physical systems exhibit some form of nonlinear behavior and microwave systems are no exception. Although powerful techniques and elaborate tools have been developed to characterize and analyze linear RF and microwave circuits, it is often the nonlinear characteristics that dominate microwave engineering efforts. Nonlinear effects are not all undesirable. Frequency conversion circuitry, for example, exploits nonlinearities in order to translate signals from one frequency to another. Nonlinear performance characteristics of interest in microwave design include harmonic distortion, gain compression, intermod- ulation distortion (IMD), phase distortion, and adjacent channel power. Numerous other nonlinear phenomena and nonlinear figures-of-merit are less commonly addressed, but can be important for some microwave systems.
- 47. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 10 — #10 1-10 RF and Microwave Circuits, Measurements, and Modeling Saturated output power P out (dBm) Pin (dBm) Linear region, slope=1 Region of gain compression Ideal linear output 1 dB compression point 1 dB FIGURE 1.5 Output power versus input power at the fundamental frequency for a nonlinear circuit. DUT Directional coupler Signal generator Power meter Spectrum analyzer Attenuator FIGURE 1.6 Measurement configuration to characterize gain compression and harmonic distortion. By replacing the signal generator with two combined signals at slightly offset frequencies, the configuration can also be used to measure intermodulation distortion. 1.4.2.1 Gain Compression Figure 1.5 illustrates gain compression characteristics of a typical microwave amplifier with a plot of output power as a function of input power. At low power levels, a single frequency signal is increased in power level by the small signal gain of the amplifier (Pout = G ∗ Pin). At lower power levels, this produces a linear Pout versus Pin plot with slope = 1 when the powers are plotted in dB units as shown in Figure 1.5. At higher power levels, nonlinearities in the amplifier begin to generate some power in the harmonics of the single frequency input signal and to compress the output signal. The result is decreased gain at higher power levels. This reduction in gain is referred to as gain compression. Gain compression is often characterized in terms of the power level when the large signal gain is 1 dB less than the small signal gain. The power level when this occurs is termed the 1dB compression point and is also illustrated in Figure 1.5. The microwave spectrum analyzer is the workhorse instrument of nonlinear microwave measurements. The instrument measures and displays power as a function of swept frequency. Combined with a variable power level signal source (or multiple combined or modulated sources), many nonlinear characteristics can be measured using the spectrum analyzer in the configuration illustrated in Figure 1.6. 1.4.2.2 Harmonic Distortion A fundamental result of nonlinear distortion in microwave devices is that power levels are produced at frequencies which are integral multiples of the applied signal frequency. These other frequency
- 48. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 11 — #11 Overview of Microwave Engineering 1-11 components are termed harmonics of the fundamental signal. Harmonic signal levels are usually specified and measured relative to the fundamental signal level. The harmonic level is expressed in dBc, which desig- nates dB relative to the fundamental power level. Microwave system requirements often place a maximum acceptable level for individual harmonics. Typically third and second harmonic levels are critical, but higher-order harmonics can also be important for many applications. The measurement configuration illustrated in Figure 1.6 can be used to directly measure harmonic distortion of a microwave device. 1.4.2.3 Intermodulation Distortion When a microwave signal is composed of power at multiple frequencies, a nonlinear circuit will produce IMD. The IMD characteristics of a microwave device are important because they can create unwanted interference in adjacent channels of a radio or radar system. The intermodulation products of two signals produce distortion signals not only at the harmonic frequencies of the two signals, but also at the sum and difference frequencies of all of the signal’s harmonics. If the two signal frequencies are closely spaced at frequencies fc and fm, then the IMD products located at frequencies 2fc − fm and 2fm − fc will be located very close to the desired signals. This situation is illustrated in the signal spectrum of Figure 1.7. The IMD products at 2fc − fm and 2fm − fc are third-order products of the desired signals, but are located so closely to fc and fm that filtering them out of the overall signal is difficult. The spectrum of Figure 1.7 represents the nonlinear characteristics at a single power level. As power is increased and the device enters gain compression, however, harmonic power levels will grow more quickly than fundamental power levels. In general, the nth-order harmonic power level will increase at n times the fundamental. This is illustrated in the Pout versus Pin plot of Figure 1.8 where both the fundamental and the third-order products are plotted. As in the case of the fundamental power, third-order IMD levels will compress at higher power levels. IMD is often characterized and specified in terms of the third-order intercept point, IP3. This point is the power level where the slope of the small signal gain and the slope of the low power level third-order product characteristics cross as shown in Figure 1.8. 1.4.2.4 Phase Distortion Reactive elements in a microwave system give rise to time delays that are nonlinear. Such delays are referred to as memory effects and result in AM–PM distortion in a modulated signal. AM–PM distortion creates Frequency 2fc− fm fm 2fm − fc fc P out (dBm) FIGURE 1.7 An illustration of signal spectrum due to intermodulation distortion from two signals at frequencies fc and fm.
- 49. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 12 — #12 1-12 RF and Microwave Circuits, Measurements, and Modeling Saturated output power P out (dBm) Pin (dBm) Linear region, slope = 1 Ideal linear output Third order, slope = 3 Third-order intercept point, IP3 FIGURE 1.8 Relationship between signal output power and intermodulation distortion product levels. sidebands at harmonics of a modulating signal. These sidebands are similar to the IMD sidebands, but are repeated for multiple harmonics. AM–PM distortion can dominate the out-of-band interference in a radio. At lower power levels, the phase deviation of the signal is approximately linear and the slope of the deviation, referred to as the modulation index, is often used as a figure-of-merit for the characterization of this nonlinearity. The modulation index is measured in degrees per volt using a VNA. The phase deviation is typically measured at the 1 dB compression point in order to determine modulation index. Because the VNA measures power, the computation of modulation index, kφ, uses the formula kφ = (P1dB) 2Z0 √ P1dB (1.9) where (P1dB) is the phase deviation from small signal at the 1 dB compression point, Z0 is the characteristic impedance of the system and P1dB is the 1 dB output compression point. 1.4.2.5 Adjacent Channel Power Ratio Amplitude and phase distortion affect digitally modulated signals resulting in gain compression and phase deviation. The resulting signal, however, is far more complex than the simple one or two carrier results presented in Sections 1.4.2.2 through 1.4.2.4. Instead of IMD, adjacent channel power ratio (ACPR) is often specified for digitally modulated signals. ACPR is a measure of how much power leaks into adjacent channels of a radio due to the nonlinearities of the digitally modulated signal in a central channel. Measurement of ACPR is similar to measurement of IMD, but utilizes an appropriately modulated digital test signal in place of a single tone signal generator. Test signals for digitally modulated signals are synthesized using an arbitrary waveform generator. The output spectrum of the DUT in the channels adjacent to the tested channel are then monitored and power levels are measured. 1.4.2.6 Error Vector Magnitude Adjacent channel power specifications are not adequate for certain types of modern digitally modulated systems. Error vector magnitude (EVM) is used in addition to, or instead of adjacent channel power for these systems. EVM specifications have already been written into system standards for GSM, NADC, and PHS, and they are poised to appear in many important emerging standards.
- 50. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 13 — #13 Overview of Microwave Engineering 1-13 I Q Ideal reference symbol location Measured symbol location Magnitude error Error vector P h a s e e r r o r FIGURE 1.9 I–Q diagram indicating the error vector for EVM measurements. The EVM measurement quantifies the performance of a radio transmitter against an ideal reference. A signal sent by an ideal transmitter would have all points in the I–Q constellation fall precisely at the ideal locations (i.e., magnitude and phase would be exact). Nonideal behavior of the transmitter, however, causes the actual constellation points to fall in a slightly scattered pattern that only approximates the ideal I–Q location. EVM is a way to quantify how far the actual points are from the ideal locations. This is indicated in Figure 1.9. Measurement of EVM is accomplished using a vector signal analyzer (VSA). The equipment demodu- lates the received signal in a similar way to the actual radio demodulator. The actual I–Q constellation can then be measured and compared to the ideal constellation. EVM is calculated as the ratio of the root mean square power of the error vector to the RMS power of the reference. 1.4.3 Noise Noise is a random process that can have many different sources such as thermally generated resistive noise, charge crossing a potential barrier, and generation–recombination (G–R) noise. Understanding noise is important in microwave systems because background noise levels limit the sensitivity, dynamic range and accuracy of a radio or radar receiver. 1.4.3.1 Noise Figure At microwave frequencies noise characterization involves the measurement of noise power. The noise power of a linear device can be considered as concentrated at its input as shown in Figure 1.10. The figure considers an amplifier, but the analysis is easily generalized to other linear devices. All of the amplifier noise generators can be lumped into an equivalent noise temperature with an equivalent input noise power per Hertz of Ne = kTe, where k is Boltzmann’s constant and Te is the equivalent noise temperature. The noise power per Hertz available from the noise source is NS = kTS as shown in Figure 1.10. Since noise limits the system sensitivity and dynamic range, it is useful to examine noise as it is related to signal strength using a signal-to-noise ratio (SNR). A figure-of-merit for an amplifier, noise factor (F), describes the reduction in SNR of a signal as it passes through the linear device illustrated in Figure 1.10. The noise factor for an amplifier is derived from the figure to be F = SNRIN SNROUT = 1 + Te TS (1.10)
- 51. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 14 — #14 1-14 RF and Microwave Circuits, Measurements, and Modeling + Noisy amplifier Noiseless amplifier NS = kTS Ne = kTe Equivalent amplifier noise Ga FIGURE 1.10 System view of amplifier noise. DUT Power meter TS 50 Ω N0 Noise factor measurement instrument Noise source Bandpass filter FIGURE 1.11 Measurement configuration for noise factor measurement. Device noise factor can be measured as shown in Figure 1.11. To make the measurement, the source temperature is varied resulting in variation in the device noise output, N0. The device noise contribution, however, remains constant. As TS changes the noise power measured at the power meter changes providing a method to compute noise output. In practice, the noise factor is usually given in decibels and called the noise figure, NF = 10 log F (1.11) 1.4.3.2 Phase Noise When noise is referenced to a carrier frequency it modulates that carrier and causes amplitude and phase variations known as phase noise. Oscillator phase modulation (PM) noise is much larger than amplitude modulation (AM) noise. The phase variations caused by this noise result in jitter which is critical in the design and analysis of digital communication systems. Phase noise is most easily measured using a spectrum analyzer. Figure 1.12 shows a typical oscillator source spectrum as measured directly on a spectrum analyzer. Characterization and analysis of phase noise is often described in terms of the power ratio of the noise at specific distances from the carrier frequency. This is illustrated in Figure 1.12. 1.4.4 Pulsed I –V Although most of the measurements commonly utilized in RF and microwave engineering are fre- quency domain measurements, pulsed measurements are an important exception used to characterize
- 52. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 15 — #15 Overview of Microwave Engineering 1-15 Frequency dB fc Distance from the carrier dBc FIGURE 1.12 Typical phase noise spectrum observed on a spectrum analyzer. Drain voltage Drain current FIGURE 1.13 Pulsed I–V characteristics of a microwave FET. Solid lines are DC characteristics while dashed lines are pulsed. high-frequency transistors. At RF and microwave frequencies, mechanisms known as dispersion effects become important to transistor operation. These effects reveal themselves as a difference in I–V char- acteristics obtained using a slow sweep as opposed to I–V characteristics obtained using a rapid pulse. The primary physical causes of I–V dispersion are thermal effects and carrier traps in the semiconductor. Figure 1.13 illustrates the characteristics of a microwave transistor under DC (solid lines) and pulsed (dashed lines) stimulation. In order to characterize dispersion effects, pulse rates must be shorter than the thermal and trapping time constants that are being monitored. Typically, for microwave transistors, that requires a pulse on the order of 100 ns or less. Similarly, the quiescent period between pulses should be long compared to the measured effects. Typical quiescent periods are on the order of 100 ms or more. The discrepancy between DC and pulsed characteristics is an indication of how severely the semiconductor traps and thermal effects will impact device performance.
- 53. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 16 — #16 1-16 RF and Microwave Circuits, Measurements, and Modeling Another use for pulsed I–V measurement is the characterization of high power transistors. Many high power transistors (greater than a few dozen Watts) are only operated in a pulsed mode or at a bias level far belowtheirmaximumcurrents. Ifthesedevicesarebiasedathighercurrentlevelsforafewmilliseconds, the thermal dissipation through the transistor will cause catastrophic failure. This is a problem for transistor model development, since a large range of I–V curves—including high current settings—is needed to extract an accurate model. Pulsed I–V data can provide input for model development while avoiding unnecessary stress on the part being characterized. 1.5 Circuits and Circuit Technologies Figure 1.14 illustrates a generalized radio architecture that is typical of the systems used in many wireless applications today. The generalized diagram can apply to either communications or radar applications. In a wired application, the antenna of Figure 1.14 can be replaced with a transmission line. The duplexer of Figure 1.14 will route signals at the transmission frequency from the PA to the antenna while isolating that signal from the low noise amplifier (LNA). It will also route signals at the receive frequency from the antenna to the LNA. For some systems, input and output signals are separated in time instead of frequency. In these systems, an RF switch is used instead of a duplexer. Matching elements and other passive frequency selective circuit elements are used internally to all of the components shown in the figure. In addition, radio specifications typically require the use of filters at the ports of some of the components illustrated in Figure 1.14. A signal received by the antenna is routed via the duplexer to the receive path of the radio. An LNA amplifies the signal before a mixer downconverts it to a lower frequency. The downconversion is accom- plished by mixing the received signal with an internally generated local oscillator (LO) signal. The ideal receiver rejects all unwanted noise and signals. It adds no noise or interference and converts the signal to a lower frequency that can be efficiently processed without adding distortion. On the transmitter side, a modulated signal is first upconverted and then amplified by the PA before being routed to the antenna. The ideal transmitter boosts the power and frequency of a modulated signal to that required for the radio to achieve communication with the desired receiver. Ideally, this LNA LO PA Duplexer Backend electronics Upconverting mixer Downconverting mixer Bandpass filter Antenna FIGURE 1.14 Generalized microwave radio architecture illustrating the microwave components in both the receiver and transmitter path.
- 54. 7218: “7218_c001” — 2007/8/13 — 19:43 — page 17 — #17 Overview of Microwave Engineering 1-17 process is accomplished efficiently (minimum DC power requirements) and without distortion. It is especially important that the signal broadcast from the antenna include no undesirable frequency components. To accomplish the required transmitter and receiver functions, RF and microwave components must be developed either individually or as part of an integrated circuit. The remainder of this section will examine issues related to individual components that comprise the radio. 1.5.1 Low Noise Amplifier The LNA is often most critical in determining the overall performance of the receiver chain of a wireless radio. The noise figure of the LNA has the greatest impact of any component on the overall receiver noise figure and receiver sensitivity. The LNA should minimize the system noise figure, provide sufficient gain, minimize nonlinearities, and assure stable 50 impedance with low power consumption. The two performance specifications of primary importance to determine LNA quality are gain and noise figure. In many radios, the LNA is part of a single chip design that includes a mixer and other receiver functions as well as the LNA. In these applications, the LNA may be realized using Silicon, SiGe, GaAs or another semiconductor technology. Si BJTs and SiGe HBTs dominate the LNA business at frequencies below a couple of GHz because of their tremendous cost and integration advantages over compound semiconductor devices. Compound semiconductors are favored as frequency increases and noise figure requirements decrease. For applications that require extremely low noise figures, cooled compound semiconductor HEMTs are the favored device. 1.5.2 Power Amplifier A PA is required at the output of a transmitter to boost the signal to the power levels necessary for the radio to achieve a successful link with the desired receiver. PA components are almost always the most difficult and expensive part of microwave radio design. At high power levels, semiconductor nonlinearities such as breakdown voltage become critical design concerns. Thermal management issues related to dissipating heat from the RF transistor can dominate the design effort. Efficiency of the amplifier is critical, especially in the case of portable radio products. PA efficiency is essential to obtain long battery lifetime in a portable product. Critical primary design specifications for PAs include output power, gain, linearity, and efficiency. For many applications, PA components tend to be discrete devices with minimal levels of on-chip semiconductor integration. The unique semiconductor and thermal requirements of PAs dictate the use of unique fabrication and manufacturing techniques in order to obtain required performance. The power and frequency requirements of the application typically dictate what device technology is required for the PA. At frequencies as low as 800 MHz and power levels of 1 Watt, compound semiconductor devices often compete with Silicon and SiGe for PA devices. As power and frequency increase from these levels, compound semiconductor HBTs and HEMTs dominate in this application. Vacuum tube technology is still required to achieve performance for some extremely high-power or high-frequency applications. 1.5.3 Mixer A mixer is essentially a multiplier and can be realized with any nonlinear device. If at least two signals are present in a nonlinear device, their products will be produced at the device output. The mixer is a frequency translating device. Its purpose is to translate the incoming signal at frequency, fRF, to a different outgoing frequency, fIF. The LO port of the mixer is an input port and is used to pump the RF signal and create the IF signal. Mixer characterization normally includes the following parameters: • Input match (at the RF port) • Output match (at the IF port)
- 55. Other documents randomly have different content
- 56. 1. The Graft. 2. The Stock. In grafting, the first thing to be done is to secure good shoots from a healthy tree of the sort you wish to grow—these are called the “grafts.” The stem to receive the graft is called the stock. Now a stock may be single, in which case one graft will be sufficient, as in the accompanying diagram, or if an old tree has to be grafted, a graft may be inserted on as many branches as may seem desirable. Our diagram represents the common practice of side grafting, but different plans are adopted according to the difference in size of the stock on the one hand, and the graft on the other, the principle to be aimed at in the process being to get as complete an apposition of as much of the wood and bark of the graft, with that of the stock, as is possible by careful cutting and fitting, and the tact and delicacy in manipulating this matter make that successful result which marks the good grafter. In this as in other matters, practice and experience
- 57. ensure success; and hence it is usually found expedient to employ a person who makes it his profession, and such are always to be obtained in cider countries. Graft protected by a Wicker basket. When the grafts have been fitted, they must be kept in place by some plastic material, and that most commonly used is a compost of cow-dung and clay, well kneaded together, or merely chopped hay and clay; this is pressed round the united parts in the form of a ball, and in cases where every care is taken the graft may be further protected by a wicker basket, as in the diagram. Cutting.—The ease with which apple trees can be multiplied by cuttings was forcibly impressed upon our attention at a very early
- 58. age. When a boy, having seen a most promising branch cut from a favourite apple tree in the process of pruning, the thought struck us that we might get a tree of our own, and so, seizing the branch in question, we planted it in another part of the garden, only—sad to relate—to have it pulled up the first time the gardener passed that way. With a boy’s perseverance or obstinacy—which the reader pleases—again and again did we replant this same branch with a like result, until finding a quiet corner, we once more planted our cutting, and this time, no evil chance overtaking it, it took root; and in two years from that time we enjoyed the taste of apples from what, we hope not undeservedly, was allowed to be considered our own tree. This was a matter for frequent reflection in after-life, for, besides viewing the result as a reward for perseverance, it is just possible that our first disappointment may have tended after all to our success, for doubtless the unexposed sheltered corner was just the place for ensuring this in rooting cuttings. Here, however, the cutting was a large branch, but for general purposes we should recommend cuttings to be made of small unbranched shoots; these may be planted in rows in a somewhat shaded situation, and when they have become rooted and fit for independent trees, they may be removed to their permanent places, and so be either pruned for tall orchard trees, or, as they are well adapted to the purpose, be trained for dwarf orchards. Pruning, in the cultivation and due keeping of an orchard, is one of the most important operations connected with the subject. Its objects are:— 1st. To circumscribe the growth in any given direction, to train the tree on the one hand, and to let in light and air by thinning on the other. 2nd. By pruning fruit trees we operate so as to check undue growth of wood and leaf, and thus, by what the botanist calls the “arrestation of development,” cause flower and fruit to be formed instead of leaves. In the western counties, if a tree or plant of any
- 59. kind grows leaves too freely, it is said to be too “frum,” probably derived from the Saxon from, strong, stout. Pruning, then, hastens the fruiting season in fruit trees, but at the same time it brings on premature age, and hence the operation should be performed with judgment, or else premature decay will be the consequence. In pruning of large trees care should be taken to cut out, as smoothly as possible, all awkward or crossing branches, so as to expose the whole of the fruiting limbs to light, warmth, and air. This again is an operation requiring an experienced hand, and when such an one is known, it is far better to employ him than to trust the matter to those who know little or nothing of the subject. Much has been said and written upon the subject of rearing fruit trees, and when matter of this kind is addressed to the nurseryman, it is to be welcomed if based upon sound botanical principles, but we cannot recommend the farmer to grow his own fruit trees, as he rarely pays sufficient attention to their youthful training, and we therefore recommend the purchase of fruit trees from the best growers, to get the best sorts, and to get well-grown and healthy examples. These should be carefully lifted and planted as soon as possible after leaving the nursery, always avoiding trees that have hawked the market week after week, even if procurable for nothing. Some people insist upon the propriety of planting poor trees grown in poor soil, but our experience has shown that nothing could be a greater mistake. It is true that these often fruit soon; but getting crops of fruit from trees only a quarter grown, though sometimes welcome to a tenant with no sure holding, is a matter which should always be looked to by the landlord, who, indeed, should pay greater attention to his orchards than is usually the case, if his desire be to hand them down to his successors in anything like a good bearing condition. That fruit trees must in time get old is quite true; at the same time it may be stated as an important fact, that poor stunted trees on the one hand, or those too prodigal of their youth on the other, will too surely result in decrepitude ere half the span of a healthy tree be attained.
- 60. Feeling so strongly as we do the importance of healthy young trees from a good soil and climate to plant even in an unfavourable district, instead of, as is generally sought after, trees from a poor soil, we are glad to have our opinion fortified by a successful practical grower of fruit trees, whose samples of young stock in apple trees, as we have seen them exhibited in Yeovil market, are patterns of healthiness in bark and models of form. The cultivator to whom we refer is Mr. J. Scott, whose name and place we have before mentioned. He says, in his Descriptive Catalogue of Fruit Trees:— There remains one thing the writer would especially guard intending planters against; that is, be careful never to purchase trees off a poor soil. I know this is heterodox; but many years’ experience has taught me the fallacy of the popular dogma, i.e., “Get your trees off poor soils, as they will be hardier, and endure the storms better.” I could show examples, in numbers, in my nursery, where the trees came from one of the so-called poor soils, that never will be anything like healthy trees. They were hide- bound and checked in their natures when I received them, and I believe will ever remain so, less or more. A genial, moderately rich, and naturally good soil is the soil I would choose my trees from. Experience and observation, both in the garden and the orchard, fully confirm us in this view of the case, and we would therefore only add to the direction, “Get your trees from moderately rich soil,” that of, “Plant them in a soil of the like kind;” for if trees be brought from a poor soil, not fit for them, to a poorer, they will certainly not succeed, and indeed the choice of poor land for orchard growth will be seen to end in disappointment. In planting apples we should choose a mixture of several of the best sorts, and it is recommended that some should be sour; but we prefer to have those that produce a juice of high specific gravity, though with all cider and perry fruit there will be great diversities in this respect, depending upon soil, climate, and season. The following list of apples contains such as are met with principally in the counties of Worcester, Hereford, and Gloucester; all may be
- 61. used for cider, but some are more especially adapted for house purposes:— I.—LIST OF APPLES. Those marked with (A) are good for hoarding, and those with † are good for boiling. Skyrme’s Kernel—Tart; good for cider. Royal Wilding—Bitter sweet; good for cider. Black Foxwhelp—Moderately tart; good for cider. † Red Foxwhelp (A)—Moderately tart; good for cider. Cowan Red—Sweet; good for cider. † Dymock Red (A)—Very sweet; good for cider. White Norman—Bitter sweet; good for cider. Red Norman—Bitter sweet; good for cider. Hagloe Crab—Tart; good for cider. Pawson—Tart; good for cider. † Redstreak—Sweet; good for cider. Yellow Styre—Sweet; good for cider. † Hooper’s Kernel (A)—Moderately sweet; good for cider. † Hill Barn Kernel (A)—Sweet; good for cider. † Ribston Pippin (A)—Sweet; good for table and keeping. Golden Harvey (A)—Sweet; good for table and for cider. Siberian Harvey—Sweet; good for cider. Farewell Blossom—Tart and bitter; large bearer. Upright French—Bitter sweet; large bearer. Black or Red French—Bitter sweet. Knotted Kernel—Tart. Leather Apple—Hardly any taste. Ironsides (A)—Hardly any taste; good for keeping. † Cats’-heads (A)—Sweet; good for cider. Pigs’-eyes—Sweet. Downton Pippin (A)—Sweet; table and eating. † [335]Codlings (A)—Sweet; good as boilers and for cider. † May Blooms (A)—Sweet; good for cider, boiling, and keeping. Rough Coat (A)—Dry and sweet; good keepers. Brandy Apple (A)—Very sweet; makes strong cider. † Cowarne Quinin (A) Sweet; good for cider. † Blenheim Orange (A)—Very sweet; good for table.
- 62. † Golden Pippin (A)—Very sweet; good for table. Old Pearmain (A)—Very sweet; good for table. Brown Crests—Very sweet. Under Leaves—Sweet; large bearer. Red Kernel—Sweet; good for cider. † Reynolds’s Kernel (A)—Sweet; large pot-fruit. Newland Kernel—Bitter sweet; good for cider. Jackson’s Kernel—Tart. † Sam’s Crab—Tart. † Bridgewater Pippin (A)—Sweet. † Spice Apple (A)—Sweet. White Beach—Bitter sweet; good for cider. Handsome Mandy—Bitter sweet; good for cider. Golden Rennet (A)—Sweet. Pine Apple—Moderately tart; wood cankers. Stoke Pippin (A)—Sweet; good bearers; pot-fruit and for cider; and numerous others. From Prize Essay on Orchards, by Clement Cadle, from the Journal of the Royal Society. The next list is taken from Scott’s Descriptive Catalogue, by way of contrast and comparison with the above, as it is more particularly adapted to Devon, Somerset, and Dorset. LIST II. CIDER APPLES. The following is a list of some of the best Cider fruit, cultivated in the best Cider counties throughout England. 167. Best Bache, spec. grav. 1073. A Herefordshire fruit of great excellence. 168. Bringewood, a good cider fruit. 169. Bovey Redstreak. 170. Cadbury, supposed to be the same as Royal Somerset.[336] 171. Coccagee, a splendid cider fruit of first-rate excellence. 172. Cowrane, red, spec. grav. 1069; an excellent sort. 173. Devonshire Redstreak. 37. Devonshire Quarrenden, a valuable hardy fruit; well known. 35. Downton Pippin, a most prolific and valuable cider fruit. 174. Forest Styre, spec. grav. 1076 to 1081, esteemed fruit.
- 63. 175. Foxley, spec. grav. 1080, hardy and a great bearer, excellent cider fruit. 176. Fox Whelp, spec. grav. 1076 to 1080, a celebrated cider fruit of the richest kind. 54. Golden Harvey, spec. grav. 1085, a first-rate cider fruit. No orchard should be without this. 177. Haglo Crab, spec. grav. 1081. 178. Jersey, early, very fine cider fruit. 179. Jersey, late, a great bearer, and excellent; one of the best. 77. Isle of Wight Pippin, spec. gray. 1074, a fine cider fruit of great excellence. 180. Kingston Black, first-rate cider fruit of first-rate excellence. 97. Minchal Crab, a very fine fruit. 181. Red Must, very large, yielding a fine cider from heavy soils. 182. Red Streak, spec. grav. 1079, one of the best cider apples. 183. Siberian Bitter Sweet, spec. grav. 1091. 184. Sops in Wine. 185. Tom Potter or Tom Put, a fine fruit. Besides the above, many other choice sorts make splendid Cider. Pears for perry differ in one respect from apples, in that, though the best and purest perry is made from only one sort of fruit, and that generally from fruit utterly unfit for any other purpose. Pears, as has been stated, delight in a lighter soil than that which is suitable for apples, and the trees have the advantage of growing so tall that even cereal cultivation is possible under them. It is, therefore, curious to note how scarcely any perry pears are grown in the west of England, unless we view Gloucester as a western county. Though Somerset and Dorset are particularly adapted for the pear, there are many places where its culture is never attempted; we would mention the district of sandy loam around Sherborne, Dorset, as one well adapted for the growth of perry, but where it is nevertheless almost unknown. It may be noted that although good cider—even the best—can be made from dessert and culinary fruit, yet dessert pears are not well adapted for perry, as their produce is usually watery, and does not fine well.
- 64. CHAPTER XLIX. ON FRUIT-GATHERING, ETC. In making cider and perry there are several important matters to be taken into consideration, as upon the due observance of these success will mainly depend. These are— The selection, gathering, and storing of the fruit. The grinding of the fruit, and storage of the drink. The after-management, keeping, fining, c. c. Orchard fruit is economized chiefly in the three following methods:— 1. Cooking Apples—used for culinary purposes. 2. Dessert Apples—some of the fine-flavoured varieties. 3. Cider Fruit—which includes all the others. 1. Cooking apples may be hand-picked as they become ripe, and those that will not keep long, as the various codlins, may be disposed of in the lump to the fruiterer, or sent to market in smaller quantities. The good keeping apples may be sold in the lot when ripe, or kept in store to be retailed at market. Both these sets of apples require to be gathered with some care; in short, to be what are called “hand-picked,” as, when bruised, they not only are injured for present use, but their keeping qualities are greatly affected. For store apples the fruit should be gathered before being what is called “dead ripe,” that is, when they are quite crisp and juicy; one of the best indications of fitness being a bright light-brown kernel as opposed to a dull dark-brown. The fruit should be kept in a dry room, from which frost is entirely excluded, and where air can freely ventilate whenever required. The
- 65. best plan is to fit up such a room with shelves made up of laths three inches wide, and placed an inch and a half or two inches apart. PLAN OF SHELF FOR KEEPING FRUIT. In this way a represents the laths, of which there may be many or few to each shelf according to the breadth required; b, the interspaces. Here, then, the fruit is placed in lines over the interspaces, the object being thus to secure a free passage for the air all around the fruit; if placed in a single layer, faulty ones can be seen at a glance, and these should be removed as soon as detected. If this plan be found too onerous, and fruit must be put together in larger quantity, we would advise that they be so placed as that air can get to them from below. Keeping fruit in heaps in corners, or even spreading them between layers of straw, tends to their destruction rather than preservation. If, then, it be borne in mind that the end to aim at, in order to keep fruit, is that of exposing sound examples to the free access of the air, it will be seen that the nearer we can secure this the better will be our result. We say sound fruit, for it is useless to put spotted and worm-eaten apples or pears in the keeping-room. These had better be put by and used as soon as possible for whatever purpose they may be fit, for whenever the air can get into the interior of fruit by reason of abrasions, borings, c., decay soon sets in; and now, while we are writing, we have a quantity of apples with the plague-spot of rottenness proceeding from their being “worm-eaten.” 2. In storing dessert apples these directions are even more important. If, then, the farm should produce one or several sorts in
- 66. quantity, if they are to be disposed of, we would advise their sale to the fruiterer with the onus of gathering and managing them. Small farmers sometimes make no bad addition to their income by thus disposing of fine fruits, and we always advise that such should be planted to a greater extent than is usually done about farm homesteads. It is not a heavy matter for the landlord to find a few sorts of choice fruit-trees for his smaller or even larger holdings, and, by thus adding to the comfort or even luxuries of his tenants, he will be benefiting not only himself but the country at large. We believe it to be a duty incumbent upon the landed proprietor thus to foster a love of fruits, and we honour the names of Knight, of Downton, and Williams, of Pitmaston, in that they loved to propagate new fruits, and to encourage their dissemination. It is said by Mr. Benjamin Maund, the author of “The Fruitist”:— A propagator of apple and pear trees from seeds may be supposed to possess not only patience, but a desire to benefit posterity. Twelve or fourteen years cast a long shadow before them; and when, after waiting this length of time, the uncertain value of the substance is considered, it must be confessed that men deserve more than praise, who originate new fruits. Apple trees rarely show the real quality of their fruit in less than fourteen years. All, however, who have the convenience of doing so, should raise seedling trees; for it is to these only that we can look with any degree of confidence for permanently furnishing our orchards, and not to old or cankering varieties. It is true that it is not within the province of all, even of the permanent owners of the soil, thus to add to the number of Pomona’s gifts, but all can inquire for and purchase esteemed sorts; and no tenant that is worth having will grudge them care and attention, be his tenure ever so precarious. We would assign to the lords of the soil the duty of improving fruit- trees, while the gentleman who resides in the country, it may be for only a short season, should make the best use of it to encourage a love for the garden, and to increase its various attractions to charm the eye, and to increase and vary the vegetable food of the people.
- 67. 3. Fruit for cider-making will consist of “wind-falls,” that is, such as has fallen prematurely ripe, or been shaken off by the wind; and gathered fruit. As regards wind-falls, it is only necessary to state that, although these can only be employed for an inferior kind of drink, yet even this may be improved by care, as thus:—Instead of picking up the apples while they are still wet with dew, they should be gathered in as dry a state as possible, and then not, as is too often the case, huddled together in a heap in the orchard, exposed to alternations of frost, and wet, and dry. Such fruit will often require to be kept for some time waiting temperate weather, which is best for cider-making. It should be kept then under cover, and in such a manner that the air can get beneath it; and for this purpose we have found a few wattled hurdles well adapted for keeping fruit on that is waiting to be ground. In gathering cider-fruit we should consider it ripe at that period when a not rude shake of a limb would cause most of it to fall pretty well at one and the same time. We dislike beating off fruit with sticks, as it damages the bearing shoots. In fine, in gathering fruit all undue violence should be carefully avoided, as it is unwise to use that amount of hurry, which will only secure a large present crop, unless it can be done in such a manner as not to injure our hopes of the future. It is a curious circumstance that in the garden there is usually something like a crop, even in a bad season; but in the orchard we seldom meet with anything like a crop the year following what is called a “hit of fruit,” and only the finer sorts of apples which are hand-gathered with care are often found to be most constant bearers, while the rougher cider-fruits seldom afford a good crop oftener than once in from three to five years. Surely, then, much of this must be the result of the rougher treatment to which cider-fruit is so carelessly subjected. When the fruit is collected, it should be put in a dry airy place, to await the process of grinding. For this we adopt the plan of spreading it in sheds or outhouses on wattled hurdles. This keeps it from the rain, by which it becomes sodden when in exposed heaps:
- 68. then the wind will only partially dry it, and the result will be a general heating of the mass, which results, if not in quick decay amounting to absolute rottenness, yet in that state, technically called “moisey,”[31] or dead, in which the juices are nearly dried up and the fruit flavourless. [31] Apple moise, or apple moce, was an old dish made of pressed apples. In cider counties apples are called moisey when they are juiceless, dry, and without flavour—dead. (See Archaic Dictionaries.) We have seen heaps of apples, consisting of many waggon-loads, in the orchard at Christmas, when wet and frost had so preyed upon them that none of their proper juices remained. This is certain to make a cider which will be of inferior quality; and though some of our friends boast of the good quality of their cider which has been made in the roughest manner, yet one cannot help thinking how much better it might have been with the fruit carefully collected, and kept until it could be ground. Still, with all our care in this matter, disappointment is sometimes the result; for it is with cider as with wine, the season will have a great deal to do with it, though with both, the manner of making and storing will be all-important matters, to which we shall advert in the next chapter. We much object to the gathering of fruit for any purpose in the wet. Were it not for the expense, it would be better to take advantage of dry weather, and to collect even cider-fruit by hand-picking before it has become dead ripe, and so let the ripening process be completed in some dry storing-place. In our own experience of cider-making, the two or three casks made for home consumption from carefully picked and well-kept fruit are usually of the best quality, and so made we believe cider to be a most agreeable and very wholesome beverage,—to paraphrase Isaac Walton, only fit for farmers or very honest men. As long, however, as rough people are about who never know when they have had enough, the rougher cider made by a ruder process is quite good enough. It must be obvious to all that if a man can drink as much as four gallons of good cider in a day’s mowing, he would be better off with
- 69. a less quantity of an inferior sort, supplemented with tea or coffee. CHAPTER L. ON CIDER-MAKING AND ITS MANAGEMENT. In making cider or perry it is well not to begin unless the weather be moderately cool, as in hot weather the changes in the fluid become too rapid, and it consequently does not keep well. The first process will be to grind the fruit into as perfect a state of pulp as possible. This will be effected when the kernels are decidedly crushed. Such a state of pulp usually ensures the best results, not only from the fact that the whole juice of the fruit is not only set free, but it is all exposed to the action of the air, by which both the colour and quality are greatly improved; and, besides this, every good quality is decidedly increased by having the principles and flavour of the kernels mixed with the other juices. The method by which this is best effected is by grinding in the usual circular stone horse-mill. This is confessedly a slow process, but notwithstanding the newer methods, to be presently described, we still prefer it to all others, and that from the great completeness with which the grinding is effected. Of late years cider-mills have been brought out which essentially consist of a combination of gribbling teeth, by which the fruit is first torn to pieces, and two cylindrical rollers, between which it is afterwards crushed with greater or less completeness. In some cases the rollers are of iron, in others of hard stone: the latter is preferable, as contact with iron, even where but slight, causes the drink to assume a degree of blackness, especially on exposure.
- 70. Portable mills of this kind are now very general, but we so fully agree with the remarks of Mr. Cadle, that we here quote his description of some portable cider-mills, with his comments upon their action. About twenty-six years ago, Mr. Coleman, of Chaxhill, Westbury-on- Severn, commenced making an improved cider-mill and press, which could act either as a fixture or a portable mill. It was found that the cider thus made fined better, and the process was also more expeditious. These advantages, together with the cost of keeping the old kind of mills in repair, which landlords were unwilling to undertake, led to their being superseded, as they wore out, by Coleman’s, or a similar mill. Coleman’s mill consists of two pairs of rollers fixed in a strong wooden frame; it is fed from a hopper, the apples passing through the first pair of rollers, which are made of hard wood, with iron teeth, so as to break the apples, which fall next between a pair of stone rollers set close enough to break the kernels, and from these the pulp drops into a trough placed beneath to receive it. Mr. Latchem, of Hereford, has also paid considerable attention to the construction of these mills, and has taken out a patent for doing away with the iron in the feed-rollers, and substituting steel teeth fitted into one roller, and working through other steel teeth on a fixed plate, partly on the same principle as a curd-mill. The fruit, after passing this “chewer,” is ground between a pair of stone rollers, as before described. Until the portable apple-mills became general, we had a mill to almost every farm, and even to many of the cottages; but in Devonshire one mill or pound-house serves for a number of makers, and sometimes for a parish, each person paying so much per hogshead for the making. [347]Most of the travelling portable machines in Herefordshire have two presses with each mill, and are worked by two horses, making 1,000 to 1,500 gallons in a day; sometimes they are worked by a small portable steam-engine. They are very expeditious, and do very well for a second- class cider, but if you would have the best, they are very objectionable, because the different sorts of fruit very rarely get ripe at once in sufficient quantities to enable you to make much at a time. Much cider is therefore spoiled, the fruit being ground when too green, by those who are impatient to finish the process. I think that each farm or holding should have a mill of its own, even if it be only a small hand-mill.
- 71. There are several other rude plans of grinding, such as nut-mills, graters, scratchers, c., but they are so objectionable that they hardly deserve notice. All metallic substances should be kept from contact with the pulp, as chemical combinations immediately take place on contact; for instance, if you take a clean knife and cut an apple through, the knife quickly becomes black, as well as the apple. For this reason I think the iron teeth and cast-iron in the rollers are objectionable; as also the steel ones, although perhaps not to the same extent. I should recommend that this iron be removed, and fluted rollers of larger diameter be made of some hard wood, such as yew-tree, or American iron-wood. No doubt more power would then be required to work the mills, but this would be of little consequence if the produce was first-class cider. When this new mode of grinding was first tried, there was great complaint amongst the labourers that the cider did not agree with them, and this was generally attributed to the iron; but in my opinion, the green state of the fruit when ground made the juice harsh, and caused irritation in the system.—Journal R. A. S., vol. XXV. page 1. The next point for consideration is the pressing out of the juice. This has been done with screw-presses of various kinds, either wood or iron, with single or double screws. Hydraulic presses are now coming into fashion, and one advantage which they possess is, that of easily and expeditiously getting all the juice from the pulp. In Dorsetshire the ground pulp or “pummy” is usually put upon a flat stage between layers of straw, which are deftly turned up at the edges so as to keep the “cheese” together. Upon the top of the cheese is placed another flat board, which is acted upon by the press. In Worcestershire and Hereford the pulp is pressed in hair cloths, which plan is much more perfect than with straw. In pressing it is well to observe that the pulp be ground on one day and pressed the next, as not only colour but general richness in quality results from exposure. The dark colour which an apple
- 72. assumes on being cut is due to this cause, not as supposed to the steel knife, for the change mentioned is equally certain with a silver one. In the now almost exploded plan of scooping apples, the pulp of even sour apples becomes sweet by the process. As the juice is exuded from the press it falls into a trough beneath, which is divided into two parts by a grating with small holes, by which the particles of pulp are separated, and from this the clearer fluid is conveyed to the cask. As regards straining, we have seen some of the finer sorts of perry made by a more complete straining than the above; in fact, a rough kind of filtering in flannel bags. This would take too long a time for general purposes. It is, however, a good way of making drink for bottling. The after-management of cider and perry is a subject upon which much has been both said and written. We, however, join in the country opinion, that “if it be made well the less it is messed with the better.” We prefer putting cider in large casks in a cool cellar—say of from one to two hundred gallons or more,—to each of which should be two tap-holes, one in the middle and one towards the bottom; the first tapping from the middle hole insures a clear fluid without disturbing the lower part, which thus goes on “settling down.” If cider from good fruit be made well, it will have an agreeable sub- acid flavour, derived from the malic acid, which is the principle which gives the refreshing juice of most fruits. Fermentation is necessary to make good cider, as by it the sugar of the fruit is converted into alcohol or spirit; and if, when this process is complete, the fermentation ceases, we shall have a refreshing, exciting, and generous fluid; if, however, it passes from vinous to acetous fermentation, we get acetic acid, and the product is sour.
- 73. Cider made from good and well-ordered fruit in temperate weather, and put in casks in a cool cellar, will be likely to ferment equably, and to stop at the right time; if so, the product will be of the best; if, however, these conditions have not been complied with, the cider will be more or less harsh or “hard,” and no means will avail to improve it. Sulphur may be burnt in the casks to check fermentation; but we would after all prefer acetic to sulphurous acid. Chalk and lime will decompose the acid, but to little purpose. The London method of adding sugar or sugar-candy and water to sour cider— and to them all mature cider is sour—is in itself innocent enough. There is, then, this consolation: if the cider be harsh, farm labourers will drink it; and as they will not, as a rule, drink half so much of the inferior as of the best, they will after all be the gainers. CHAPTER LI. ON THE USES AND ECONOMY OF CIDER AND PERRY. If we canvass the opinions of the mass of the people in cider- producing and non-cider-producing counties as to the relative merits of cider and beer, we shall find opinions wider apart than even the counties themselves. The “Beer-drinking Briton” cannot at all understand how the lover of cider can skin his throat with such sour stuff as cider, whilst the agricultural labourer in cider districts infinitely prefers harsh cider to the finest ale. We recollect, in one of our geological trips in to Herefordshire, in company with an esteemed clerical friend, that a quarryman, working in Wenlock limestone, tendered us a few shells, on which we offered him sixpence, remarking, “Here’s a quart of beer for your trouble.” This same man then gave our companion a couple of trilobites, who presented him with a coin of like value to our own, but with the remark, “Here, my friend, is a gallon of cider for you.” The effect upon the man’s whole being will never be forgotten. He was the
- 74. slave of the Church for the whole day, and ever thereafter for all we can tell. In cider districts the farmer, his family and friends, all relish cider, and with all, its proper use seems to agree in a most remarkable manner; but it would be fun to a country cousin who could cease to look at the matter in a serious light to see what a face his London relative would make at a draught of his “own peculiar;” and yet he of the town professes to like sweet cider; but as his knowledge of sweet cider is obtained from the summer drink of the London houses, called “Prime Devonshire Cider,” the following recipe will explain it:— Take of Vinegar (or sweeter still, cider) 1 pint. Brown sugar (or treacle) 1 pound. Water 7 quarts. The following will be found in Cooley’s “Cyclopædia of Practical Receipts:”— Cider, Made.—An article under this name is made in Devonshire for the supply of the London market, it having been found that the ordinary cider will not stand a voyage to the metropolis without some preparation. The finest quality of made cider is only ordinary cider racked into a clean cask, and well sulphured; but the mass of that which is sent to London is mixed with water, treacle, and alum, and then fined down, after which it is racked into well-matched casks (i.e., a burnt-sulphur match). The larger portion of the cider sold in London, professing to be Devonshire cider, would be rejected even by the farmers’ servants in that county. No wonder, then, that cider is not a favourite beverage when it is only used as a summer drink in some sophisticated form; but, when understood and obtained at all good, we believe it to be wholesome and palatable, and, indeed, we know it to be preferred before even the best ales in cider districts. There is a common error amongst town-folk who prefer the above mixture that cider is not intoxicating, that it has no strength in it; but
- 75. we regret to say that it is not only intoxicating, but we believe more exciting than beer: it is true that its effects pass off sooner. Drunkenness with cider would seem to be so far different than in the case of beer, in that while the latter makes its victim heavy and stupid, the former incites to motion, and leads to quarrelling, fighting, and foolhardiness. Hence, then, cider so exhilarates the farm labourer that he will do any amount of work if he is constantly plied with it, and all the while that it is but stimulating him, he fancies he is getting strength and vigour from it; but, alas! he is only thus drawing upon his capital; exhaustion follows a hard day’s work got over amid hard drinking, which requires the following day to be spent on the same high- pressure system, or else little will be done. Hence one of our own labourers, during barley mowing at so much per acre, was fain to confess that he “wanted a pint of cider at four o’clock in the morning worse nor any other time of day.” It happens, then, that as harvest work is wanted to be done expeditiously, it is let out by the piece, by which the labourer gets more money and more cider. But consider, my masters, that, when not under these stimulants, you can only expect from the workman a languid day’s work when the excitement is over; and too often, indeed, the poor man gets a long illness as the result of his forced, that is, stimulated labour, and, if not, such a system of drawing upon his capital—strength—is certain to end in premature old age. Seeing, however, that the labourer has got to believe that drink keeps up his strength, it too often follows that he concludes that the more he gets of it the better; and hence, as a rule, there is no satisfying him upon this head, and the result is, that the labourer too often keeps himself in that state of thirst and muzziness during his work that almost compels him to seek the public-house when work is done. Here quarrels ensue, and it is a wonder that manslaughter is not more frequently the result. Expelled from the scene of his
- 76. debauch, he finds his way home, unless, as is not unfrequent, he is “found drowned” in the river by which he may have to pass. This is no fancied sketch, as it is derived from the sad experience of the author and the result of events in his own parish. On one melancholy occasion it was indeed sad to hear the Coroner, among other remarks, observe that full four-fifths of the inquests in a cider county were the result of drink. Is there not, then, a heavy responsibility resting upon the farmer in especial connection with cider, while his men are partially paid in this fluid? It is different in the beer-drinking counties, as beer costs more money, and is never allowed in such quantity as cider. Put it down as true that the farmer at times gets more work out of his men by plying them with cider, yet we feel sure he thereby hastens the time when such men can no longer work, and they have then to be chargeable to the parish, if in the mean time nothing worse should happen. Mechanics are not paid in drink; they purchase what they require out of regular wages, and thus they have the option, which many of them take advantage of, of leaving off strong drink altogether; and though they too are sometimes hard pressed to get a piece of work done, yet, by over-hours, for which they are rightly paid, not, as in the country, wholly by cider, but in money, the business is managed, and the workman can afford extra meat and bread, by which his worn muscles are truly renovated, and not merely stimulated to frantic action as by drink. The great rise in the price of meat, even before cattle disease became rife, is due to the cause that so much more meat has, within the last five years, been eaten by the British workman. In this advance, however, the farm labourer has had no part; he rarely gets meat twice a week, while all this time his wages have advanced so much as 25 per cent., which rise, in nine cases out of ten, is only looked upon as a boon, inasmuch as it enables the recipient to “enjoy himself,” which simply means he has more to spend at the public-house.
- 77. We conclude, as the result of experience, that each sack of corn that finds its way to market from a cider county costs 1s. (or 3d. per bushel) in drink, which, though it is produced on the farm, might yet have been sold to produce that amount. Would it then not be better to sell such farm produce, and, by giving extra money instead of drink to the labourers, and so, by allowing him the option of taking less drink but more meat, gradually to withdraw him from the temptations to get drunk, which beset him under the present system? For, while we feel quite sure that the morbid craving for the public-house has commenced with drinking on the farm, we may be certain that if by any means we can check this system, it will ultimately be a great gain to both master and man. Where are farm labourers best off? We say in the non-cider counties. In these he has learnt the use of skim-milk and the value of meat. In cider counties the farm labourer despises skim-milk as “poor weak tack, only fit for pigs.” He cannot get meat, as he takes part of his wage in a stimulant which excites him to spend some of his money in falsely “keeping up his strength.” Now what are the results? We unhesitatingly assert, muscle, longevity, more robust, honest, well-to-do families, healthier bodies and minds, beyond the cider limits. If, then, these things be so, some change in the use and economy of this wholesome drink is an object worthy of the deepest and most earnest consideration. One man alone can do no good. Beneficial results can only follow upon calm discussion and combined action by the masters, upon well ascertained facts. We would not stint the labourer of that which is to do him good; and if we find that he is really willing and capable of taking the whole responsibility connected with his drinking requirements upon his own shoulders, we cannot help thinking that it would be for the good of all parties to pay increased wages in full rather than any portion in kind, and more especially of the kind we have thus animadverted upon.
- 79. POSTSCRIPT. In bringing these Papers to a conclusion, we would, among other matters, make a few remarks upon the title under which they have been issued, namely, Science and Practice of Farm Cultivation. Now it will be seen that our object has not been to enter into the minutiæ of practical farming, but rather to point out some of the more important scientific principles by which much of practice is regulated. Hence, then, we would beg the reader to amend the title as follows:—“Science of Practice in Farm Cultivation.” This will more fully explain the aim and object we have had in view in the series of Papers now concluded. It is now time to tender our best acknowledgements for the aid we have received in the many drawings with which this small work has been so liberally illustrated. We owe especial thanks to Mr. Hardwicke for several fine plates of interesting agricultural as well as botanical specimens; to the Royal Agricultural Society of England for the loan of the woodcuts of roots; and to our friend Mr. Wheeler, of Gloucester, for the use of the woodcut illustrations of grasses; and as both the drawings of roots and grasses were made by us direct on the wood, rough though they may be, we yet hope they may be deemed more faithful than any second-hand copy. Our labours being ended, it only remains to add that we hope our little work may have the effect of inducing some of our agricultural friends to look into the principles connected with the various operations which they daily superintend, as by so doing agriculture will be really elevated to a science; whereas, by merely copying what
- 80. has been done before, we shall only be empirics, practising rational empiricism it is true, but still coming short of that light and knowledge which is the life,—the science of our profession. J. B. Bradford Abbas, Dorset, Sept. 25, 1865. COX AND WYMAN, PRINTERS, GREAT QUEEN STREET, LONDON, W.C.
- 81. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com