o8e ote ae a8 ae yeyTable of Conients Section 1 1 BASIC RADIO FREQUENCY RELATIONSHIPS Cateulation of wavelength — Characteristic impedance of coaxial and waveguide lines — Voltage Standing Wave Ratio — Reflection Coefficient — Percent power reflected — Attenuation — Transmission line equations, Section 2 19 TRANSMISSION LINE CHARTS — 4 ‘Smith Charts — Coordinate grid system — Plotting impedances/admittances —SWR and f° — Peripheral Scales — Carter Chart — Overlay operations, Section 3 33 TRANSMISSION LINE CHARTS — 2 Impedance trans ine travel — “Lossless” line ~ — Non-uniform line ~ Line segm: shunt network. Section 4 IMPEDANCE MEASUREMENT METHODS Impedance reference plane ~ VHF bridge method — Standing wave detector method — Slotted line method — Equipment calibration — Data reduction, Section 5 NARROW BAND MATCHING Definition of bandwidth — Series and shunt networks — Matching analysis aids — Single element networks — Two element networks — Quarterwave transformers ~ Laboratory tuning devices — Stub tuners — Slide serew tuners — Hatch tuner. Section 6 BROADBAND MATCHING — LINE TRANSFORMER DESIGN CHARTS Line transformer electrical parameters — Design Charts — Chart description — Analysis factors in design selection — Example of chart usage — Additional considerations. Section 7 BROADBAND MATCHING — FOUR ELEMENT NETWORKS. Theory of broadband matching ~ Network configurations — Network application to a high “Q” antenna, Section 8 SELECTED PRACTICAL EXAMPLES Broadband Dipole ~ VOR Slot — UH Blade. Appendix A DESIGN CHARTS FOR SINGLE SECTION LINE TRANSFORMERS 51 77 105 123 141 1614 BASIC RADIO | FREQUENCY RELATIONSHIPS 1.1 INTRODUCTION One of the more interesting aspects of an engineer's task, in the de- sign and development of components for radio frequency (1-f.) s¥s- tems, is impedance compensation or more commonly termed “match- ing.” Matching simply has the meaning that consideration has-been given in the initial design of the rf. component to assure that the impedance properties of that particular device are compatible with the associated transmission system into which it is to be incorporated. A component that is properly matched will function for efficient j Tansmission of power whereas a badly “mis-matched” component re- sults in an unnecessary loss of energy. In antenna systems, for exam- ple, loss of energy as a result of a “mis-matched” antenna may have the meaning that the communication range for that system has been degraded to the extent to be not acceptable. Impedance matching may for one case be relatively simple to achieve whereas for another quite technically complex. As an illustration, single frequency or narrow band matching requires a minimum of development effort and is generally obtainable with either a single or, at the most, a two-element fixed-tuned network. Broadband match- | ing, on the other hand, is often difficult, particularly in those cases, | where broad bandwidths and high “Q” structures are involved,2@ IMPEDANCE cine The intent of this work is to serve as a practical introduction to im- pedance matching and is offered to those who are involved with either the design, testing, or the specification of performance charac- teristics for r.f. devices. No attempt will be made to delve deeply into theory; only the fundamental aspects of the subject will be consid- ered. This first section is a brief review of the relationships that are considered basic to all radio frequency work. 1.2 WAVELENGTH The free space wavelength (q) ofa radio frequency wave is given by: ¥ (300) (39.4) —t where: Xo = free space wavelength (inches) f = frequency (megahertz) (aay ‘The wavelength (4m) in a transmission line filled with a dielectric medium is related 10 Ao as: Ams te = wavelength in the medium = dielectric constant Foraire=1 Consider, for example, a teflon (¢ = 2.1) filled coaxial transmission line whoselengii is 926 inches and at'a frequency of 100 MHz. The wavelength in the line is: do _ (800) (39.4) — = ——— > = 81.5 inches Ye (100) (1) Am = and the electrical length of the line is said to be four wavelengths long. TRANSMISSION un® 3 The product of frequency (f) and (2) is the velocity of electromag- netic energy and is the speed of light, thus: Cm det as) and in various units is: C = 3x10" cmisec 1.18 x 10" in/sec 9184.x 10° te/sec 1.3 TRANSMISSION LINES ‘The primary function of transmission lines is to convey radio fre- quency energy from one point on the line to another. Some typical transmission lines are shown in Figure 1-1. The wave that travels along the line is an electromagnetic wave formed by electric and magnetic fields that are mutually perpendicular. The electric field is always perpendicular to the surface of a conductor and hence can- not have a component parallel to the surface. The magnetic field, on the other hand, is always parallel to a conductor surface and cannot, have a component perpendicular to the surface. An infinite number of waves or modes can exist on a transmission line. Each mode has its own distinctive configuration of electric and magnetic fields that. satisfy Maxwell's equations and fit the boundary conditions imposed by the line. ‘Transmission line modes may be separated into three classes such as: © Principal Mode — This mode can occur only for lines containing not more than two separated conductors and consists only of electric and magnetic fields that are everywhere transverse to theIMPEDANCE Mrcune outer Inner Conductors (2) _Coaxfat Line Ground Plane < Creer conductor (0) strip Line (c) High © Microstrip ®@® @ (@)_Parattel Mires Tens) FIGURE 1-1 i, ® | F (e)_ Rectangular (f) Cireutar Crd (a) Single - Ridgea I} be (h) Doubte = Ridged Peace (WAVE De CTW] eae VARIOUS TYPES OF TRANSMISSION LINES TRANSMISSION une? 5 direction of energy flow. The mode is referred to as TEM signity- ing a Transverse Electromagnetic Wave. The coaxial line is the most common example of this mode of transmission Higher order modes (called waveguide modes) can exist in a coax- ial line if, in some manner, the line is excited at a frequency above cut-off for that mode. Higher order modes may be excited below mode cutoff by a discontinuity in the line but such waves will at- tenuate very rapidly with distance as they travel along the line. Coaxial lines are generally designed to operate only in the princi- pal mode — the approximate limiting frequency, or cut-off fre- queney for the first higher order TE mode (TE, ), given as: te 7.51 x 10° Ve+(a+D) Where: f = limiting frequency (megahertz) 4d, D = as defined in Figure 1-1 (inches) ‘* ‘Transverse Electric (TE) modes of transmission have magnetic field components in the direction of energy flow whereas the electric field is everywhere transverse. ‘© Transverse Magnetic (TM) modes have electric field components in the direction of energy flow and the magnetic field is every- ‘where transverse. ‘TE and TM waveguide modes are termed “higher order” modes and have low frequency cut-off limits below which energy will not be propagated along the line. The higher order modes are further dis- tinguished by two subscripts. The first subscript indicates the num- ber of half-wave variations of the electric field across the wide dimen- sion of the guide. The second subscript describes the number across the narrow dimension of the guide. For example, the dominant mode (lowest frequency waveguide mode) in rectangular waveguide is the TE;,. mode. Figure 1-2 shows the field configurations for the principal (coaxial) and dominant (simplest waveguide) modes of transmission.& IMPEDANCE MATCHING TRANSMISSION LINES e cs e —H-= magnetic Field ——S Hectric Field ae —--— Magnetic Field > Hectric Field (b)_Coaxtal Principal Node ~ TEM (c) Parallel open-Circutted Line FIGURE 1-2 | FIELD CONFIGURATIONS FOR TEM AND TEx,o MODES FIGURE 1-3 | we TYPES OF MATCHING NETWORKS REALIZABLE IN COAXIAL TRANSMISSION LINES9 IMPEDANCE Prcune Other applications of transmission lines include utilizing segments of line as matching networks. It is this application that will be given primary treatment in this work. Figure 1-3 shows various types of matching elements that are realizable in practice in coaxial systems. Each network is characterized by two principal electrical parameters =the characteristic impedance (Zo') and the electrical length (0.2). Zo is constant for a given network whereas 02 varies with frequency. ‘The networks shown in Figures 1-3(b) and (c) add reactance either in series or shunt to the existing impedance at the plane of network insertion, The input impedance (Zs¢) of a short-circuited line seg- ment is given by: Zee = +4 Zo tan 82 (ohms) (1-5) Where: Zo" characteristic impedance of the line segment (ohms) 360 £/A = electrical length of the line seg- ‘ment (degrees) and the input impedance (Zo<) of an open-circuited line segment is: Zoe = -j %o' cot 42 (ohms) (1-6) ‘The network configuration shown in Figure 1-3(a) provides imped- ance matching by the process of transformation. The characteristic impedance of the section (Zo) differs from that of the primary transmission line. Selection of the electrical parameter (Zo' and 02) for optimum performance is accomplished by the application of the theory of transformation circles. 1.4 IMPEDANCE Of the various types of transmission lines shown in Figure 1-1, the coaxial line and the simple rectangular waveguide will be the types that will be given principal consideration here. on A voltage (E) epplied across two conductors of an infinite line causes IMPEDANCE 9 e ‘current (I) to flow. ‘The term characteristic (Zp ) impedance (or surge impedance) is derived from this relationship as: EB \z \[R+ijot 22 EYE YEE scams an Where: Z= R+jaL= series impedance of a unit section of the infinite line Y= G+jwC= shunt admittance of a unit section of the infinite line and for a “lossless” line — that is one in which the attenuation may be considered negligible: aE hi 1-8) 0 “Vg (ohms) (-8) Where: L= inductance per unit length (henries) C= capacitance per unit length (farads) ‘The characteristic impedance of the coaxial line (Figure 1-1) is de- rived from the dielectric constant (c) and line geometry as: 138 D Zo = — logo — (ohm: 1-9) JE lei Z (ohms) (1-9) ‘The characteristic impedance of a segment of transmission line of unknown value may be determined by measurement of two equal segments of the line where one segment is terminated in a short cir- cuit and the other segment terminated in an open circuit. The Zy is calculated as: Zo = VEoe* Zs (ohms) (1-10) Where: Zoe = open circuit impedance Zse = short circuit impedance The characteristic impedance of a uniform loss less coaxial line is always real and does not differ with frequency.1@ IMPEDANCE MRICHING Waveguide characteristic impedance differs with frequency and for mpl rectangular gude, operatingin the TEx mode, 1 vadey cross-sectional dimensions (see Figure 1-1) nut-off frequency frequency at which Zp is desired and the cut-off wavelength (Ae) is: (ohms) (1-41) de= 2a (112) ‘The guide wavelength (Ng) also differs with frequency as: 1 a a (1-13) eo FRR a Where: = wavelength in the unbounded mediam of d- electric constant e. Figure 1-4 shows various matching devices that are useable in wave- guide type structures. IMPEDANCE "4 Ae 1 a a \— ai wt (1-22) and b= 10 logio = = 10 logo 55 1.7 TRANSMISSION LINE EQUATIONS ‘The input impedance (Zin) “looking into” a “lossless” transmission line terminated in a load (Z,) is given by: (1-23) Zy+j Zo tan 9e Tin 0 Fi ttn 2 (24) Where: Zp = line characteristic impedance (ohms) Zz, = load impedance (ohms) 68’ = electrical length of line (degrees) TRANSMISSION LINE eMPations 15 ‘The input impedance is generally a complex quantity. The process of impedance matching requires numerous manipulations of impedance quantities in the derivation of optimum networks. In such operations, application of equation (1-24) is quite tedious and time consuming, Fortunately there are other methods less tedious and more readily applied. Such methods include the utilization of graphical charts (Smith and Carter)*? that not only provide ready solutions to trans- mission line problems but also ‘‘visual” interpretation of parameter variation along a given line. The work herein will treat the solution of network problems by graphical techniques only. Equation (1-24) is given for technical interest.16 IMPEDANCE MATCHING REFERENCES | J.A, Nelson and G. Stavis, “Impedance Matching, Transformers and Baluns,” Very High Frequency Techniques, Chapter 3, Radio Research Laboratory, MieGrav-Hill Book Company, New York and London, 1947. 2, PAH, Smith, “Transmission Line Calculator,” Electronics Mogzzine, January 1939. 3, PS, Carter, “Charts for Transmission Line Measurements and Computa- tions, R.C.A, Review, No. 3, p. 355, January 1939. BIBLIOGRAPHY 1. Theodore Moreno, Microwave Transmission Design Data, Dover Publica tions, Ine., New York, New York, 1958. 2. Gershon J. Wheeler, Introduction to Microwaves, Prentice-Hall, Inc., New Jersey, 1963. 3, ‘Theodore S. Saad, Microwave Engineers Handbook, Volumes I and Ti, “Artech House, Inc., Dedham, Massachusetts, 1971. 4, -M, Slurzberg and W. Osterheld, Essentials of Radio, MeGraw-Hill Book Company, Ine., New York — Toronto — London, 1948. 5. “Transmission Lines,” Reference Data for Radio Engineers, Chapter 20, Titernational Telephone and Telegraph Corporation, fourth edition, 1957.| 2 TRANSMISSION LINE CHARTS—1e 2 © 20 IMPEDANCE MATCHING SMITH CHART e ‘Sections II and ITI will be concerned with the above two types of 241 teguseission line charts. Together the two sections present a basic Gescription of the charts together with a discussion of the fundamen- tal impedance operations that may be performed. The reader thus Gbtains the knowledge of a design tool that is most useful in the graphical derivation of solutions to impedance matching network problems. 2.2 SMITH CHART 2.2.1 Coordinate Grid System ‘The Smith Chart has a coordinate grid system that is arranged in such dinanner to permit the plotting, in rectangular form, of all values of fopedances (R + j X) and admittances (G + j B). The conventional foun of the chart, Figure 2-1, is generally used for impedance opera tions where SWR’s of greater than 2:1 are involved. An expanded Version of the chart, not shown, enlarges the center portion and has application for operations where SWR’s are less than 2.6:1. ‘The grid system is said to be “normalized” as the prime center of the chart, on the vertical resistance/conductance axis, is given @ titue of 1.0. This represents the 100% value of the characteristic Tmpedance (Zo of admittance (Yo ) of the associated transmission Te The scale of the resistance/conductance axis ranges from zero to afinity and normalized value appearing on this axis represent per- ‘centages of the value assigned to the prime center. Reference to Figure 2-1 shows that there are two families of normal- eed Orthogonal circles — the resistance/conductance circles and the Factance/susceptance circles. The resistance/conductance circles are "entered upon the vertical axis and represent the real component of {mpedance/admittance. The reactance/susceptance circles appear on the right and left sides of the chart and represent the imaginary com- ponent of impedance/admittance. Sufficient normalized circles are Erown on the Smith Chart to permit a practical interpolation accu racy when a plotted value falls between two given circles. SMITH CHART FIGURE 2-1 SMITH CHART FORMAT22 IMPEDANCE mat@ine SMITH CHART e 23 e@ 2.2.2 Plotting Impedances/Admittances IMPEDANCE OR ADMITTANCE COORDINATES Since the coordinate grid system of the Smith Chart is normalized, all values of impedances/admittances to be plotted upon the chart ‘must be converted to normalized form. An impedance or admittance, in complex rectangular form, is normalized when their actual values IMPEDANCE are divided by the associated transmission line characteristic imped- CURVE (2) ance or admittance as: ZR YX TZ) Zo 2) ¥_G,.B Ye Yo’ Yo Where: Z', Y'_ = normalized values of impedance/admittance ZY ~actual value of impedance/admittance (obms/ (2-2) mhos) R____ * resistance (ohms) #jX = reactance (ohms) G ‘conductance (mhos) $B = susceptance (mhos) ‘The reverse of the normalization process provides actual values of impedances or admittances — that is actual values are obtained when the normalized values on the chart are multiplied by the characteris tic value (Zp or Yo). Figure 2-2 shows a typical Smith Chart Plot of an impedance/admit- tance curve, A natural phenomena of such a curve is that it tends to proceed in a clockwise manner around the chart — starting at the Jowest frequency of measurement and proceeding clockwise to the higher frequency. Normalized admittances appear diametrically op- posite their respective impedances and are therefore quite easily de- termined by graphical construction. -ADATTTANCE CURVE (¥) SHITH CHART FIGURE 2-2 SHITH CHART PLOT OF IMPEDANCE / ADMITTANCE DATADefeats eae the following chart information need be noted: ” © The characteristic impedance (Zo )/admittance (Yo) of the associ. | ated transmission line. i * Identification of the curve as either impedance or admittance. © Frequency identification of plotted values and the notation of the unit of frequency. © Impedance Reference Plane. The margins of the chart may be utilized to include the date of mea- surement, network schematic, and a simple sketch of the measure. ment setup, Such information is relevant as Smith Chart plots quite | often become enclosures to design documents. e ; e 4 IMPEDANCE MATCHING 2.23 SWRandT Specifications that define the desired performance characteristics for radio frequency devices require that such devices operate over a par. ticular frequency band with an SWR less than some specified value. ‘An SWR circle, constructed upon the Smith Chart, is used to show whether or not the device is performing to specification requirements, Impedance points lying anywhere within this SWR circle are said to be within specification limits whereas those outside the circle are not. } Figure 2-2 shows an SWR = 3:1 circle. The SWR of any particular | impedance point may be determined by utilizing a compass to con- struct an arc of a circle, whose center is at the prime center, through | the impedance point to intersect the resistance/conductance maxic mum axis. The point of intersection shows the desired SWR value. ‘This process is illustrated in Figure 2-2 where the SWR of the 80 MHz impedance point is found to be 4.7: SMITH CHART 25 ‘The voltage reflection coefficient (°) is a complex quantity as noted in Equation (1-15) — having both magnitude and phase. The absolute magnitude, Il, varies from zero at the chart center to unity at the chart rim. If! of any plotted impedance point may be derived by calculation using Equation (1-18) or may be determined by graphical construction from the chart as: distance from chart center to impedance point - (28) distance from chart center to chart rim 2.2.4 Peripheral Scales ‘The peripheral scales that are shown on the rim of the Smith Chart are the Wavelength Scales and the Angle of Reflection Coefficient Scale. ‘The Wavelength Scales are labled “Towards Generator” and “Towards Load.” These identify the direction of travel when it is desired to transform impedance values from one position on a transmission line to another. A travel of one complete circle on the chart represents a distance of one-half wavelength (\/2) along the transmission line. Impedance/admittance relationships repeat each half wavelength along a uniform lossless (negligible attenuation) line. The zero point on the Wavelengths Scale may, therefore, be considered as any multi- ple of a half wavelength as well as zero. In performing impedance admittance transformations, the electrical length of travel must be specified in terms of wavelengths. Equations (1-1) and (1-2) describe calculation of wavelength in both free space and in a dielectric filled line. ‘The Angle of Reflection Coefficient Scale provides the phase angle (9) of the voltage reflection coefficient. 6 is found by.constructing a straight line from the chart center through the impedance point to intersect the Angle of Reflection Coefficient Scale on the chart rim. ‘The Angle of Reflection Coefficient Scale can only be used with the impedance coordinates.IMPEDANCE M@rcHING | CARTERCHART @ 27 e 2 2.3 CARTER CHART 2.3.1 Coordinate Grid System ‘The Carter Chart, more commonly known as the Z-@ chart, permits the plotting of impedances in the polar form. The format that is shown in Figure 2-3 is a 50 ohm chart and is probably the more Popular form in present day useage. The chart is not normalized and is used for plotting impedances where the output data provided from the measurement equipment utilized is in the form of magni- tude (ohms) and phase (degrees). The values shown on the vertical axis are in ohms. Positive phase angles are found on the right side of the chart whereas negative phase angles are on the left side, ‘There are some Carter Charts that are normalized. ‘The chart shown in Figure 2-3 may be normalized by simply dividing all values on the vertical axis by 50 ohms such that the prime center has a value of 1.0. Normalized admittances, as in the case of the Smith Chart, would then appear diametrically opposite their respective impedance points. Determination of SWR and I’ of any plotted impedance point may be found by graphical construction in a like manner as described for the Smith Chart, CARTER CHART 2.3.2 Peripheral Seale ‘There is only one peripheral scale on the Carter Chart shown in Fi ure 2-3 and that is the Wavelengths Toward Load Scale. The process of impedance matching starts with a knowledge of the terminal im- pedance properties of the device to be compensated. Generally the impedance reference plane is at the bridge terminals of the measure- ment equipment with which the Carter Chart is used. Impedance transformation is, therefore, in only one direction and that is towards the load — hence the lack of need for a Towards Generator Wave- length Scale, FIGURE 2-3 CARTER CHART FORMAT28@ IMPEDANCE MA@EHING 2.4 CHART OVERLAY OPERATIONS ‘Smith and Carter Charts are considered standard stockroom items at facilities engaged in the development and testing of radio frequency devices. It is the general case that the diameters of both charts are the same. This feature permits the charts to be used as overlays in several interesting operations. Overlay operations require that the two charts be aligned one on top of the other such that their vertical axis and prime centers coin- cide. When the vertical axis zero points of both charts coincide, the following operations may be performed: ‘© If both charts are normalized, impedances appearing in polar form ‘on the Carter Chart may be transposed to impedances in rectangu lar form on the Smith Chart and vice versa. © When the Carter Chart is a 60 ohm chart and the Smith Chart prime j center is assigned to be 50 ohms, actual values of impedances appearing on the Carter Chart may be transposed to normalized values on the Smith Chart. Figure 2-4 illustrates this process, ‘The other case to consider is the alignment of the vertical axis of the charts such that the zero point on one chart coincides with the in- finity point on the other chart. In this case, impedance values ap- pearing on one chart are transposed to admittance values on the other chart for the conditions described above. CHART OVERLAY OPEM@TIONS 29 FIGURE 2-4 SMITH CHART (RECTANGULAR CooRDINATES)] 0.3 +5 0.8 (15 + 5 20) 25 953.1" CARTER CHART (PoLaR cooROINATES) RECTANGULAR TO POLAR CONVERSION USING TRANSMISSION LINE CHARTS30 IMPEDANCE wBoune REFERENCES Kraus, John D., Antennas, McGraw-Hill Book Company, Inc. Page 506, 1950. 2, Smith Charts may be purchased from Analog Instruments Company, P.O, Box 808, New Providence, New Jersey, 07974 BIBLIOGRAPHY, Southworth, George C., “Using the Smith Diagram — I,” The Microwave Journal, Vol. 2, No. 1, pps 25-81, January 1959, 2, Southworth, George C., “Using the Smith Diagram — II,” The Microwave Journal, Vol. 2, No, 2, pps 24-30, February 1959, 3, Gelman, David S,, “A Fortran IV Subroutine for the Computer Output of Smith Charts,” Report No. MDC J5918, Douglas Aircraft Co., MeDon- nell Douglas Corporation, Long Beach, California, March 1973.3 TRANSMISSION LINE CHARTS—2 1 INTRODUCTIONrd #, IMPEDANCE MATCHING 3.2 IMPEDANCE TRANSFORMATION Impedance transformation, as discussed herein, is the chart operation of transforming a given set of impedance data from one position on a transmission line to another. Such operations are frequently used in matching network analysis where it is desired to position the im- pedance curve to obtain the maximum effectiveness of a particular compensating network. 3.2.1 Line Travel (02) ‘Transformation of impedances or admittances along a transmission line involves primarily the use of the peripheral Wavelengths Scales in the direction either towards the source or towards the load. Line travel must be specified in terms of wavelengths. A given physical length of transmission line has one electrical length at one frequency and a different electrical length at another frequency. If the electri- ‘eal length (02) is known at one frequency, then the electrical lengths at other frequencies may be readily determined by ratio for TEM lines from: 2 . on, (4) 68, = electrical length at frequency f; (wavelengths) 02, = electrical length at frequency f, (wavelengths) Where: ‘As an example consider a rexolite (¢ = 2.5) filled coaxial line whose physical length is 100 inches and it is desired to determine the elec- trical length in the 30 MHz to 75 MHz frequency band. The wave- length in the dielectric medium (Xm), at 30 MHz may be found when Equations (1-1) and (1-2) as: Ao _ (800) (39.4) eo . (800) 894) 049.6 inches rm" Je ~ (80) WEB) IMPEDANCE TRANSFORMATION 35 and the electrical length (9) of the coaxial line at this frequency is: 100" O80" aygr 7 0.4020 ‘The electrical lengths, at other frequencies within the band of inter- est, may be found by applying Equation (3-1) and are as follows: frequency (MHz) 8 (wavelengths —d) 30.0 0.402 40.0 0.536 50.0 0.670 60.0 0.804 70.0 0.936 78.0 1.001 Since impedance relationships repeat each half-wavelength (2/2) along uniform “lossless” line, the line travel in the above example would “4 frequency (MHz) 00 (wavelengths — X) 30.0 0.402 40.0 0.036 50.0 0.170 60.0 0.304 70.0 0.436 75.0 0.001 3.2.2 “Lossless” Line Case A “lossless” transmission line is one that is characterized by negligible attenuation. Equation (1-8) defines the characteristic impedance of such a line in terms of inductance and capacitance per unit length. Figure (3-1) shows a load whose impedance at 30 MHz and 40 MHz 4s Zag = 18 +} 109 and Zao = 10 + j 252. Let it be desired to trans- form this impedance data to Reference Plane B-B which represents a ine travel of .085) at 30 MHz. The process is described in a step-by- step manner as: ‘U/Normalize Z3o and Zyo to Zo = 502 as Z'y9 = 0.3 + j 0.2 and Zug = 0.2 + j 0.5. These values are shown plotted in the figure aud identified asthe impedance curve at Reference Pane A-A AA):36 IMPEDANCE MATCHING SMITH CHART ao + 15+) 108 dg = 1045250 Figure 3-1 IMPEDANCE TRANSFORMATION (LOSS LESS LINE) ‘iio Libera Sie ade e e IMPEDANCE TRANSFORMATION 37 2 [For a “lossless” line, impedances transform on constant SWR circles. Construct these SWR circles to include Z'39 and Z's on the Z'aa impedance curve. 3/Construct radial lines from the chart center through the imped- ance points to intersect the Wavelengths Scales. C (.034A) and D (0764) are the points of intersection and define the initial starting positions for line travel. 4 [Line travel at 30 MHz is .085) and at 40 MHz [from Equation (8-1)] is 0.1132. End positions, C’ and D’, of line travel are found by adding the line travel at each frequency to the starting position values as: 034, + 085A = 0.1190 ‘D' = .076) + .113A = 0.189 5 [Construct radial ines from the chart center to C’ and D’. The points of intersection of these lines and the SWR circles define the impedance values at Reference Plane B-B. These values are: Ziq = 0.5 +5 0.8 Zuo = 25 +5 400 Zao = 1.04521 + Zag = 50 +5 1052 ‘Thus the impedance curve Z'gp represents the impedance charac- teristics of the load as transformed to Reference Plane B-B. 3.2.3 “Lossy” Line Case Figure (3-2) illustrates transformation along a line containing attenu- ation. Load conditions and line travel are the same as that given in Figure (3-1). Transformation on a “lossy” line is accomplished in the same manner as described in 3.2.2 for the “lossless” line case ex- cept that the values obtained at the end of line travel must be cor- rected for attenuation. The transformed impedance points marked38 IMPEDANCE MATCHING thus (*) in Figure (3-2) are for the “lossless” line case as previously described. Correction for attenuation at the end of line travel in- volves the following steps: 1 Calculate the reflection coefficients of the 30 MHz and 40 MHz ‘impedances at Reference Plane A-A utilizing Equation (1-16) SWR-1 1 ose Washo “ SWRo1 351" SWR-1_ 62-1 Paale = Soo = 0.723 aalo * SyRei 6241 2. [Convert the line attenuation that is given in db (a = 2db) to nepers utilizing Equation (1-21) as: 2 = —— = 0.231 nepers ad 8.68 ce 2a = 0.462 nepers 3/ Determine the reflections coefficients at Reference Plane B-B from Equation (1-19) as: Weel = Waals + €% = (0.566) + 4? = 0.857 Ipplo = Waal * © 0.455 (0.723) + e046 4/'The input SWR of the 80 MHz and 40 MHz impedance points at, Reference Plane B-B, using Equation (1-16), is: 1+ IPpplo SWRs = T= Talo asian IMPEDANCE TRANS SBMATION 39 e INBEDANCE OR ADMITTANCE cOOROINATES o 3 + S00 Fos mz SMITH CHART 2 TRANSFORNED IHPEDANCE POINTS FOR LOSSLESS LINE CASE. FIGURE 3-2 IMPEDANCE TRANSFORMATION (LOSSY LINE)@ IMPEDANCE MASEHING. 1+ 0.357 “T 357 _ 14 IPpple “Ts Weel _1+0.455 ©” T0455 oad = 2.121 SWReo 5/Construct the SWRao circle to intersect radial line OC’ and the SWRgo circle to intersect radial line OD’. The points of intersec- tion are the impedance values at Reference Plane B-B and are: Ziq = 0.12 + 50.6, Zo = 36 +5309 Zig = 1A +5118, Za = T0+j 57.52 In the above example, several mathematical operations are involved in the determination of the network input SWR. Charts are avail- able in literature! , however, that provide a graphical solution and are probably more convenient for usage in the process of correc: tion for line attenuation. 2.4 Non-uniform Line Anon-uniform transmission line is here defined as one in which the lectrical characteristics change with distance along the direction of propagation. Figure (3-3) shows one such line containing segments of different characteristic impedance. At Reference A-A, a load whose _ impedance Zr, = 20 + j 302 is given at a frequency of 30 MHz. The following is involved in the process of impedance transformation to Reference Plane D-D: 1/21, is normalized to 2y = 502 and entered upon the chart of Fig- ure (3-3) as Z'aa (so) = 0.4 + j 0.6. This impedance is next trans- formed along the 502 line to Reference Plane B-B, a line travel of 0.12, and identified a5 Z'gp(so) = 1.5 + j 1.6. 2/Prior to line travel on the Zy = 352 section, it is necessary to mul- tiply Z’pp (:o) by the characteristic impedance ratio */35 in order to be normalized to the Z) = 359 line. This is accomplished and IMPEDANCE TRANSA@RMATION a e IWPEDANCE OR ADMITTANCE cOOROINATES FuoURE 3-3 | sur cua = +sme IMPEDANCE TRANSFORMATION. OW A WON-UNIFORM LINE. a 42 IMPEDANCE MATCHING identified as 2'pp (ss) = 2.1 + j 2.8. 2'pg (ss) is transformed to Reference Plane C-C, a line travel of 0.154, and identified as Zice (xs) = 0.48 -j 1.1. 3/2'cc (xe) is multiplied by the next characteristic impedance ratio, which is °"/y5, to give Z'cc (os) = 0.22 -j 0.5. Zico (95) is transformed a line travel of 0.094 to Reference Plane D-D and identified as 2'pp (1) = 0.17 + j 0.08. 4/'The last step, in the process, is the determination of the actual impedance value at Reference Plane D-D which is accomplished by multiplying Z’pp ("*) by the characteristic impedance value of the last line section which is Zo = 7522. ‘The input impedance of this non-uniform line is thus found to be: Zpp = (0.17 + j 0.08) - 75 = 12,75 +5 6.02 Examples of the transformation of impedances along the trans- mission lines described in paragraphs 3.2.2, 3.2.3, and 3.2.4 cover the majority of the cases involved in matching network analysis. 3.3LINE SEGMENTS - ‘Segments of transmission lines are utilized in impedance matching. Figures (1-3) and (1-4) of Section I shows various types of match ing networks that are realizable in coaxial and waveguide transmis sion lines. In scale model antenna radiation pattern studies, it is stan- dard practice to first match the model antenna to a SWR of less than 8:1 prior to performing pattern measurements. The primary reason for this effort is that the transmission line feed cable is more often a better antenna than the model antenna under study. A high SWR on the feed cable may cause the cable to radiate thus giving erroneous pattern measurement results. Line segments, used in shunt configura- tions, are quite convenient as matching elements in this particular application. e e LINE SEGMENTS 43 Bquations (1-5) and (1-6) provide the input impedance of open cir- cuited and short circuited line segments. The Smith Chart provides a ready graphical solution to the above equations. An example is shown in Figure (3-4) where a segment of open and shorted line is given. Since the input impedance of such segments is always reactive (¢ jx), the impedance curves are shown on the rim of the chart. ‘The input impedance, at any position on the line segment, may be determined by multiplying the normalized value on the chart rim by the characteristic impedance assigned to the chart center. For example, let it be assumed that the chart center Zp = 7502 and itis desired to examine the impedance at various positions along the 0.176 shorted segment. Referring to Figure (3-4) the following is compiled: Reference Plane Normalized Z Actual Z AA Za=+) 09 BB Zen = +) 300 cc +) 67.59 va 441502 Examination of the figure also shows that the input ows e input impedance of an /4 shorted cable is infinity ohms and of an open circuited cable is zero ohms. Also a shorted 4/8 cable has a normalized +) 1.0 input impedance whereas an \/8 open circuited cable has a normalized 1.0 input impedance.44 e IMPEDANCE MARHING ‘WEEDANCE 08 ADMITTANCE. COORDINATES al tt an c pet a pom FIGURE 3-4 | INPUT IMPEDANCE OF OPEX age) vO SHORTED. (Zse) LINE. SEGMENTS ‘SERIES/SHUNT newex 45 3.4 SERIES/SHUNT NETWORK Graphical solutions to simple network problems are accomplished quite easily using the Smith Chart. In Figure (3-5), a lumped con- stant T-network is shown that is selected to mateh a given load whose impedance at 30 MHz is 21, = 5 + j 02. The load is pure resistance and a perfect match is sought. At higher frequencies, a quarter wave transmission line transformer section such as that shown in Figure (1-3a) could be considered for the matching network since the physical length of the section would probably not be ob- jectionable. At 30 MHz, however, a X/4 section in a coaxial rexolite line represents a physical length of 62 1/4 inches. The network shown in the figure is an equivalent quarter wave transformer utiliz- ing lumped constant matching elements. This example is selected to describe the method of handling series and shunt impedances in the process of network solution. A good rule to follow is to work with impedances when combining series elements and admittances when combining shunt elements. The solu- tion to the network shown in Figure (3-5) to derive the input imped- ance Zpp follows: 1Zroad (ZAA)s Zis » Zi2, and Zc, are normalized and plotted as: Zan = 0.1450 Zu = Zi, = 0 +5 0.32 Ze, = 0 - 50.32 Since C1 is a shunt element, the admittance Y'¢, (diametrically opposite Z'c, ) = 0 + j 8.1 is also plotted.46 IMPEDANCE MATCHING FIGURE 3-5 GRAPHICAL SOLUTION OF A T-NETHORK ADDITIONAL REMAR® 47 e 2/The impedance at Reference Plane B-B is composed of series ele- ments Z'aq and Zz, which is: Zpp= Zant Zr = (01 +} 0) + (0+ 5 0.92) Oi +5032 3/Y'gp is diametrically opposite Z’gp and is plotted as Y'py = 0.92 -}2.8. Y'pp is added to Y'o, to give Y'oc = 0.92 +} 0.3. Zicc = 1 -j 0.32 is found diametrically opposite Y'cc. 4/'The desired input impedance at Reference Plane D-D is next found bby adding Z'cc and Z'r.2 in series to give: 2pp = Lec + Zs = (1 +} 0.82) + (0 +j 0.32) = 1+) 0= perfect match Zpp =(14)0)- 50=50+j0n 3.5 ADDITIONAL REMARKS Sections II and III have provided the reader with a fundamental knowledge of the two most popular forms of transmission line charts. ‘These charts have become quite important to all practitioners in the field of radio frequency work and particularly to those engaged in im- edance matching. An insight as to the degree of importance of the Smith Chart, for example, may be found in the following remark of Dr. George C. Southworth:* “In this writer’s opinion, the Smith diagram has been one of the more important device contributions made to microwave technique uring the last couple of decades.”48 IMPEDANCE MATCHING REFERENCES 1, Intemational Telephone and Telegraph Corporation, Reference Data for Radio Engineers, Fourth Edition, pages 570 and 871, 1957. 2, Southworth, George C., “More About Phil Smith and His Diagram,” The ‘Microwave Journal, Vol. 1, No. 2, pgs, 26-28, September 1958. BIBLIOGRAPHY 1, Mathis, HLF,, “Logarithmic Transmission Line Charts,” Electronics, September 1960 2, Blanchard, W.C., “A Do-It-Yourself Transmission Line Impedance Chart,” Microwaves, pages 46-52, April 1969. 3, Markin, Joseph, “Smith Chart Applications,” Tele-Tech & Electronic Indus- taies, pp 85-88, May 1953. 4, Cholewski, L.S.,"'Some Amateur Applications of the Smith Chart,” QST., pp 28-31, January 1960.4 IMPEDANCE MEASUREMENT METHODS 4.1 INTRODUCTION Impedance matching analysis must first start with a knowledge of the terminal impedance properties of the r.f. device to be compen- sated. It has been proven by theory and established in practice that, greater impedance bandwidths may be achieved when the matching network is incorporated as near as practical to the device terminals, An inherent characteristic of any impedance curve, as may be shown on the Smith Chart, is that the impedance points tend to “spread” as the electrical length from the device terminals is increased. The spreading of the impedance points has the meaning that the band- width potential has been reduced. It is, therefore, of importance that one entire section of this book be devoted to a summary dis- cussion of various measurement methods in present day usage that describe the terminal impedance properties of r.f. devices. ‘The more frequently used impedance measurement methods are (a) the VHF Bridge, (b) the Standing Wave Detector, and (c) the Slotted Line. The HP 803A VHF Bridge, for example, may be used at fre- quencies from 50 MHz to 500 MHz, whereas the Slotted Line finds applications from 500 MHz throughout the microwave frequency spectrum. The PRD 219 Standing Wave Detector is an instrument that permits impedance measurements within the 20 MHz to 2000 MHz frequency range. These methods are employed to measure de-52 e IMPEDANCE MATEHING vices that radiate r.f. energy such as antennas and also devices that do not radiate such as filters, hybrids, terminations, attenuators, and 0 forth. There are other methods of measuring impedance using such instrumentation as “Q” Meters, Vector Impedance Meters, and swept frequency equipment; however, the reader is referred to the selected bibliography given at the end of this section for a description of these methods, ‘This section is more primarily concerned with the type of data ob- tained from the above noted measurement methods and the tech- nique of data reduction rather than in operational procedures. Spe- cific equipment operational procedures may be found in literature.’ 4,2 IMPEDANCE REFERENCE PLANE In Section II, it was noted that impedance/admittance data plotted upon the Smith Chart is meaningless if such data is not properly identified, Part of the proper identification requires that the Im- pedance Reference Plane be defined. The Impedance Reference Plane simply identifies that position, either on the transmission line or at the device terminals, for which the data plotted upon the ‘Smith Chart is valid. Itis general practice in performing impedance measurements to e3- tablish the Impedance Reference Plane, as close as practical to the device terminals, Broader impedance bandwidths may be achieved when matching networks are incorporated at or near this location. Figure 4-1 shows the preferred location of the Impedance Reference Plane for differing measurement test setups. Establishment of this fey see IMPEDANCE REFERENG? PLANE ee 53 DIPOLE ANTENNA SLOTTED - LINE 7z Na ANTENNA c SS TERMINALS Hl (a) oURECT connection |< IvPevance (ANTENWA TERMINALS ACCESSIBLE) REFERENCE PLANE TNPEDANCE REFERENCE PLANE | SLOTTED - LINE TJ GROUND PLANE Na FLUSH STUB u ANTENNA (&)otREcT cowection (ANTENWA TERNINALS NOT ACCESSIBLE) IMPEDANCE. REFERENCE PLANE SLOTTED - LINE 7 y— NOSE CONE INSTALLATION ___-| ya CAVITY ~ BACKED (e) REMOTE conNecTION j HELIX. ANTENNA (ANTENNA TERMINALS NOT ACCESSIBLE) | FIoURE 4-1 IMPEDANCE REFERENCE PLANE FOR DIFFERING MEASUREMENT SETUPS Cee 54 IMPEDANCE MA@HING VHF BRIDGE METHOD @ 55 e reference plane requires a measurement with a short circuit placed at the desired location. In (a) of the Figure, the device terminals are ac- cessible to be shorted and the Impedance Reference Plane is estab- lished at that location where as in (b) and (c) the device terminals are not readily accessible. In the latter two cases, the Impedance Refer- ence Plane is established at the nearest antenna location which is at the 1, connector located on the antenna cavity. Direct connection of the measurement equipment, as illustrated in Figure 4-1 (a) and (b), is desirable in that transmission line losses are reduced to a minimum and, as such, have little or no affect upon the measured impedance data, In situations requiring remote connection of equipment, as that shown in Figure 4-1(c), line losses may be such to significantly affect measured data. In these cases, the measured data must be corrected for line loss as previously described in para- graph 3.2.3 (“Lossy” Line Case) of Section IIT. 4.3 VHF BRIDGE rence ne METHOD aaa — 4.3.1 General cto HP. 417A Figure 4-2 shows a VHF Bridge Test Setup used to measure imped- ! ance in the frequency range of 50-500 megahertz. The impedance Gata obtained is in polar form (|Z 4.8) and as such the bridge is commonly termed a Z-0 Bridge. The bridge measures absolute mag- nitude of impedance in the range of 2-2000 ohms and is calibrated aieaae erangle (2) at 100 megahertz. Al frequencies other than Seana arent the corrected phase angle (B.) is given by the ex- | fi-—1H pression! ‘Test Frequency — MHz) gg = Crest reaueney = MD) 4) (4) | 00 i wep Where: = corrected phase angle between -90 and +90 degrees. |} Signal Generator RF. Bridge 3. 04 = phase angle as road from the bridge dial — + degrees. { HAP. 6086 HP. 803A T Phones FIGURE 4-2 VHF BRIDGE MEASUREMENT TEST SETUP56 e IMPEDANCE MATCHING TYPICAL CALIBRATION DATA OBTAINED FOR AN H.P. 803A BRIDGE USED TO MEASURE VOR /LOCALIZER ANTENNAS ‘A VHF BRIDGE OPERATING IN THE 108-122 MHZ FREQUENCY BAND: ‘FREQUENCY ques Bee | oa | oc | asc | & | be |Vic - ase voa.o | 381 | -87 | -93.8] 6.2 [+81 |+87.5 | 42.6 /-3.1° to.6 | 386 | -35 | 93.1] 6.35 ]+80 |+97.7 | 43.9 /-2.7° 2.0 | 365 | -0¢ | -94.0] 6.5 [+7a_|+a7.5 | 48.7 (-3.2° 1.6 | 350 | -91 | -98.5] 6.6 [+75 497.5 | 48.1. [-3.0° uia.o | 345 | -20 | -94-5] 6.8 [+75 |+80.5 | 48.4 3.0" yeo.0 | 335 | -79 | -96.8| 7.0 [+79 _|494.8 | 48.4 vee.0 | 328 | -77.5| -98-5| 7-1. |+72.5 |sea.5 | 48.3. (-3.0° Zoc = OPEN CIRCUIT BRIDGE TERMINAL IMPEDANCE - OHS SHORT CIRCUIT BRIOGE TERMINAL IMPEDANCE - OWNS 0g = PHASE ANGLE AS READ ON BRIDGE DIAL - DEGREES Se = PHASE ANGLE CORRECTED FOR FREQUENCY - DEGREES Zo = foc ™ Bsc = BRIDGE TERMINAL CHARACTERISTIC INPEDANCE FIGURE 4-3 | TYPICAL CALIBRATION DATA FOR e e VHF BRIDGE METHOD 57 ‘The uncorrected accuracy of the measurement is 5% to 6% in magni- tude and 3° to 4° in phase angle. These accuracies are usually accept- able for the majority of measurement tasks encountered in general impedance development work. Correction curves are provided with the equipment, however, that may be used to increase the measure- ment accuracy to 2% in magnitude and to 1.2” in phase for those cases requiring greater accuracy. 4.3.2 Equipment Calibration ‘The validity of test data is generally supported by measurements that establish that the test equipment was within proper calibration, Such measurements form a natural part of the test documentation, Figure 4-3 shows typical calibration data obtained for a VHF Bridge used to measure the impedance of VOR /Localizer antennas operat- ing in the frequency band of 108-122 MHz. The data was obtained for open and short circuited conditions at the bridge terminals. The principal results are given in the last column of the figure. The char- acteristic impedance (Zy) of the bridge terminals was determined from the measured data using Equation (1-10). For example, the Za of the bridge terminal at 108 MHz was found to be Zq(108) = Zoe > Zs VGBI B58) (62 [487.8) = 2360 76.3" 48.6 /-3.1° Examination of the results shows good agreement with the published bridge accuracies of 5% in magnitude and 3° in phase. 4.3.3 Data Keduction Bridge measurement data for a slot antenna operating in the 210-240 MHz band is given in Figure 4-4. The particular antenna shown is a cavity backed single folded slot in which partial impedance compen- sation is achieved by the selection of appropriate characteristics for58 IMPEDANCE iicuine VHF BRIDGE etHoo® 59 e the folded and driven slots. The set of measurements is selected to i: lustrate the method of data reduction and the plotting of results on the Carter Chart, The data reduction procedure follows: 1/Impedance Reference Plane (Za) — Zaa is established by ob- taining a set of cable short measurements [Figure 4-4(d)} with a connector short circuit placed at the antenna terminals, The pur- pose of these measurements is to determine the electrical length (@2) of the transmission line from the bridge measurement plane to Zaa in terms of wavelengths. 2/Rinding 6% — Cable short measurements [121A Oc} of Figure 4-4(d) are plotted upon the Carter Chart of Figure 4-5. Through each plotted point, a radial line is constructed to intersect the peripheral wavelengths scales. For example, the radial line through the 210 MHz point intersects the wavelengths scale at 0.211 A. 60 is found by subtracting this value from 0.5 2 as: (2) stor anew (9) FEED TeRWMMAL DETAILS R210) = 0.5X-0.211 = 0.2892 fe oe ‘The electrical lengths of the transmission line at the remaining i reference I ashen frequencies are found in a like re covweeror | PE te 3 [The slot antenna measurements (12! Abe] of Figure 4-4(e) are or “Bye tt - next plotted in Figure 4-6 and corrected for the above values ! ee of 82 to provide the impedance curve Za. For simplicity, con- A struction is shown only for the 210 MHz impedance data in the (e) TEST cra anemone : figure. The graphical construction start with a radial line through the measured Z, 10 impedance point to intersect the wavelengths rest cane #2 | (2) CABLE _sioaT_MEASWRONEN'S —_—_(e)_SLOT_ANTENML_eASuRENENTS seale at an initial st&rting position (0) for line transformation. To iis added 60 to give the transformed position (0¢) on the wave- couner] ia] © | ee | ve | (rvecumy] ii |e | x | engths scale. In this care, 8t ne (oa) | to) | oes | cess | or | ower | tor | en) | (oc | or ot = oivee zo. | avo. | -to.s|-0s.s| 20 || eve. | co. | a | v7.7] tes = 0.142 +0.2804=0.4292 zo. | o.| -so.0[-20.0| 250 || exo. | o.2[ 22 | -s0.s| os |} § A radial line is next constructed from the chart center to intersect 6t. The antenna terminal impedance, at 210 MHz, lies on this 230, 30. | -98-0| -67.4] 414 TS $33} 78.9) ah radial line, at the same distance from the chart center as for the nl | ee Se es initial measured plotted impedance point. The construction as- sumes the “lossless line” case, FIGURE 4-6 BRIDGE MEASUREMENT. ORTA FOR A SLOT. RNTENMA ‘Transformed impedances at 220, 230, and 240 MHz are found in Alike manner and together form the antenna terminal impedance curve Za.60 IMPEDANCE MBcuine VHF BRIDGE metHop® 61 e ayy AS MEASURED CALCULATION FoR 9 om) $09 rent SB. A loa loa fei Isolate ea FIGURE 4-8 PLOTTING. RIDGE CABLE SHORT. HESLREKENTS TO. FINO 08 SLOT ANTORU (FIORE 4-4)oo IMPEDANCE Bounce 4. 4 STANDING WAVE DETECTOR METHOD — Figure 4-7 shows an impedance measurement test setup using the PRD 219 Rotary Standing Wave Detector. The detector consists of a coaxial tee junction, a pickup probe assembly that is manually driven, and a calibrated susceptance. There are three separate cali- brated susceptances that permit the instrument to measure SWR and impedances in the frequency range of 20 MHz to 2 GHz. The instru- ment is compact and particularly useful for the measurement of rf components installed in environments that are not relatively accessi- ble for the setup of test equipment — such as aircraft electronic com- partments, for example. In operation, the calibrated susceptance drum is set to the desired frequency, the input jack is connected to a modulated rf. source, the output jack to a Standing Wave Indicator and the device to be measured is connected to the rf. port of the instrument. The SWR is read from the Standing Wave Indicator and the Angle Of Reflec- tion Coefficient (¢), in the range of -180° to +180", is read directly from the calibrated dial at the top of the PRD 219 Detector. ‘The accuracy of measurement depends upon (a) the residual SWR. of the coaxial tee junction and (b) the precision with which the calibrated variable susceptance is set to the operating frequency. ‘The published maximum residual SWR of the tee junction is 1.04:1 whereas the Reflection Coefficient Angle accuracy is stated to be 26°. 2 STANDING WAVE vefror METHOD: 63 Signal Generator HP. B12, Figure — Standing Wave Indicator Ha P8158 Standing Naye Detector Circular Ground Plane a “7 D219 containing Stub Antennas| bao 219 STANDING WAVE DETECTOR MEASUREMENT TEST SETUPIMPEDANCE MATCHING WPEDANCE 08 ADMITTANCE COORDINATES Flame 6-8 | Finow using sate cna ING LORD TWPEDINCE (24) PRO 219 MEASURED ATR STANDING WAVE DETECTOR METHOD 65 4.4.2 Equipment Calibration ‘The rf. port of the PRD 219 Detector is a coaxial line designed to have a characteristic impedance of 50 ohms. The calibration of the ‘equipment may be verified by obtaining calibrated open and short- circuited readings of this port and calculating the Z. using Equation (1-10). The SWR accuracy may be verified by the measurement of standard mis-matches. In addition, the calibration of the phase angle dial may be verified to within an accuracy of 2° by the short circuit measurement, A perfect short will reflect an infinite SWR and a phase angle reading of #180", ‘The maximum error in SWR will occur for a pure resistive load and vill be about equal to the percent error in the setting of the fre- quency of operation, The maximum error in Angle of Reflection Coefficient occurs for the maximum reactive value possible for a given load SWR and will be about one degree for each percent of error in frequency setting. 4.4.3. Data Reduction ‘The data obtained from the PRD 219 test setup is in the form of SWR. and Angle of Reflection Coefficient (9) and may, therefore, be plotted upon the Smith Chart. The plotted values are normalized to 50 ohms. The measured data consists of (a) Angle of Reflection Co- efficient (6,) of a short circuit placed at the desired Impedance Ref- erence Plane, (b) the load SWR and (c) the Load Angle of Reflection Coefficient (61). ‘The method of data reduction is best described by example. Let it be ascumed that itis desired to plot the following measured data: SWR ‘This data is plotted upon the Smith Chart of Figure 4-8 in the fol- lowing steps: 1/The short measurement, ¢, = -75°, is plotted upon the chart as Point A. A radial line is constructed from the chart center throughe e 66 IMPEDANCE MATCHING: Point A to intersect the Wavelengths Towards Generator Scale. ‘The point of intersection defines the electrical length (82) of the transmission line and, in this case, is equal to 0.354 2. 2/A SWR = 5:1 circle is next identified in the figure. A radial is then constructed from the chart center to intersect 41, = +150" and the Wavelengths Towards Load Scale. The point of intersection of the radial line and the SWR = 5:1 Circle defines the load impedance as measured at the plane of the PRD 219 Detector. The point of in- tersection of the radial line and the Wavelengths Towards Load Scale defines the initial starting position (i) for line transforma- tion which in this case is 0.458 A. 8/Line transformation is accomplished by adding 0i and 6& to give the transformed position (9+) on the Wavelengths Towards Load Scale as: ot = gitee = 0.458 2 + 0.354 01812 d or 0.3124 ‘A radial line is constructed from the chart center through 0t = 0.312 a. The point of intersection of this line and the SWR circle establishes the normalized load impedance (Z't,) to be 1.1 + 41.9. The actual value of the load impedance (Z1,) is: %, = 2+ 50 BB+) 952 ‘The above procedure assumes the “lossless line” case. Had there been line loss, the load impedance (Z'1.) would require further correction using the method described in paragraph 3.2.2 of Section IIL. SLOTTED LINE METHOD 67 4.5 SLOTTED LINE METHOD 4.5.1 General ‘The slotted line is another instrument that measures impedance utilizing the technique of measuring the Standing Wave Ratio (SWR) that exists on a transmission line as a result of a load mis-match. Slotted lines are available in both coaxial and waveguide types. Co- axial lines generally cover the frequency range from 500 MHz to 18 GHz, whereas waveguide types are utilized in the frequency range of 2 GHz to 40 GHz. Figure 4-9 shows a typical test setup. Both coaxial and waveguide lines are shown to illustrate the two principal types. In practice, only one slotted line per test setup is generally required. The Slotted Line contains an adjustable probe that may be extended into the slotted section to sample the r-f. voltage. This voltage is rectified by a detector contained within the probe mount, amplified, and displayed ona SWR Indicator. In operation the probe is moved along the line to measure the maximum and minimum of the voltage Standing Wave pattern. The probe is “loosely” coupled to the line for this measure- ‘ment, in order to cause minimum distortion of the r.f. field within the guide. ‘The accuracy of slotted line measurements is limited primarily by the residual SWR of the line. The residual SWR is that SWR mea- sured when the line is terminated in a perfect match. Residual SWR_ for coaxial lines varies from 1.02:1 to 1.06:1 whereas waveguide lines reflect a residual SWR. of 1.01:1. Other considerations, such as probe penetration, harmonics, and detector “‘square law” response affect the measurement accuracy, however, these factors are con- ‘trollable by proper operational procedures, 4.5.2 Equipment Calibration Calibration of Slotted Line test setups include (a) shorted line mea- surements, (b) detector “square law” response, and (c) calibrated mis-match measurements.68 Frequency Counter systron Gonner 6ST6A | Standing Mave Inéteators HP. 8158 ® qu de ‘ ; 230 08 Dtreetional Narda 3003-30 Signal Generator HP, BIER Frome 4-9 | coupler Slotted Lines coaxial Waveguide HP BO5R WPS 108 SLOTTED LINE MEASUREMENT Test sew? Waveguide tp Coaxiat ‘adapter coaxtat Unknown SLOTTED LINE METHOD 69 ‘The shorted line measurement consists of placing a short at the Im- pedance Reference Plane and measuring the distance between two Successive minima. This measurement multiplied by two defines the guide wavelength (\,) and hence provides a means of verification of the frequency of operation and probe tuning. In an air-filled coaxial Slotted Line, \g for all practical purposes is equal to the free space wavelength (\o) as defined in Equation (1-1) whereas for the air- filled rectangular waveguide, 2, is given by Equation (1-18). If there is an excessive degree of r.f. harmonics emitted from the signal source, the probe may very easily be improperly tuned to the second harmonic instead of the fundamental. Comparison of calculated and ‘measured values of Xg thus provides a means of frequency calibration. ‘The short measurement also describes the presence of FM, Sharp deep nulls indicate FM not excessive whereas broad nulls is a clear indication of excessive FM from the signal source. Excessive FM re- sults in significant errors in SWR measurements. ‘The “square law” response of the probe detector may be verified by decreasing the source power output in discrete increments (usually 3 dB increments) and noting the response on the SWR In- dicator. If the SWR Indicator response closely follows that of the ‘signal source incremental power decrease, the detector is thus op- ‘erating in the “square law” region. ‘The measurement accuracy may be verified by the measurement of standard mis-matches that are calibrated for specific SWR values. The standard mismatches should have SWR values comparable to that of the rf. component to be measured. 4.8.3 Data Reduction ‘The measurement of impedance, using a Slotted Line, requires (a) a measurement of two successive voltage minima for the shorted line case and (b) a measurement of the SWR of the load. The mea- surement data, at a single frequency, shown in Figure 4-10 is selected to illustrate data reduction. The technique is as follows: 1/The antenna terminals are shorted to produce the Standing Wave pattern shown in the figure. Two successive voltage nulls aree e 70 IMPEDANCE MATCHING noted — one towards the load (N1,) and the other towards the generator (Ng). The guide wavelength (Ag) is found as: Ag = (NL-Na)- 2 (4-2) = (27.94 - 5.23) +2 = 45.42 em = 17.88 inches ‘The calculated free space wavelength (Xo) from Equation (1-1) for a coaxial air-filled line is = (800) (39.4) = 17.92 inches Close correlation of the above measured wavelength values pro- vides a verification of frequency setting and probe tuning. 2/The SWR of the antenna is next measured and an antenna null position (Na) that lies between Ng and Ny, is determined. In this case these values are: SWR = 3.5:1 Na =14.82cm 3/Ni, No, Na, and SWR comprises the measured data needed to form a Smith Chart plot. Figure 4-11 is a plot of the measure- ments given in Figure 4-10. In Figure 4-11 a SWR circle of 3.5:1 is first constructed. The impedance to be determined must lie somewhere upon this circle. The angular position (9), in wave- lengths, needed to establish the exact location of impedance is derived from the Nz,,Ng, and Na data. 4/02 determination — 9 may be plotted either towards the gener- ator or towards the load on the wavelengths scales. Either plot e SLOTTED LINE METHOD 71 STANDING WAVE PATTERN WITH ANTENNA TERMINALS SHORTED STANDING WAVE PATTERN OF ANTENNA DIscoNE al [\— srouno % N PLANE “ * . ‘ALSO IMPEDANCE REFERENCE. PLANE MEASURED DATA: FREQUENCY NG 7% uM (wiz) ec cM) (oH 660 as | 52a | tee | 27.94 were: Nig = MULL TOWARDS GENERATOR NL = NULL TOWARDS LOAD Ny = ANTENNA NULL SLOTTED~ LINE MEASURENENT DATA FOR A DISCONE ANTENNA FIGURE 4-107® IMPEDANCE MIRCHING produces the same results. Let it be assumed that 02 is to be plot- ted towards the load. #2 (towards the load) is found as: Nu -Na re. 27,94 - 14.82, 45.42 ag (43) = 0.289 A radial line is constructed from the chart center through the 8 = .289 position on the Wavelengths Towards Load Scale. The in- tersection of this radial and the SWR = 8:6:1 circle defines the antenna normalized impedance (Z'). In this case, Z'q is found to be 2.1 + 1.6. The actual value of antenna impedance (Z'a) is: Za= Za+ 60 = 105+j802 SLOTTED LINE mens 73 IMPEDANCE OR ACHTTANCE COORDINATES Flowe 4-11 - - oe = 0.209 SHIT wer FINOING LOAD IWPEDAICE (24") USING SLoTTeD LINE. DATA74 IMPEDANCE MATCHING REFERENCES 1, Hewlett-Packard Company, “Microwave Measurements for Calibration Laboratories,” Application Note 38, dated June 1960. 2, Pat Tuceiarone, “Making Microwave Measurements,” Electronic Industries, dated June 1962. 3. W.A, Weissman, “Measuring Microwave Impedance,” Electronic Products Magazine, dated May 1966, BIBLIOGRAPHY . LK. Irving, Editor, Microwave Theory and Measurements, Engineeying Staft of the Microwave Division, Hewlett-Packard Company, Prentice-Hall, Inc. Book Company, 1962. 2, Hewlett-Packard Company, “Swept Frequeney Techniques,” Application Note 65, dated August 1965. 3, Hewlett-Packard Company, “Network Analysis at Microwave Frequencies,” Application Note 92, no date. 4, General Radio Company, “Type 1602-B Admittance Mete Operating Instructions, dated May 1965. * Handbook of 5. R.W, Beatty, “Microwave Impedance Measurements and Standards,” National Bureau of Standards, Monograph 2, dated 12 August 1965.ee eo OD) zt 5 NARROW BAND MATCHING 5.1 INTRODUCTION Single frequency or narrow band impedance matching is quite easily accomplished with a minimum of development effort. Moderate “Q” rf. structures may often be matched with a single fixed-tuned network whereas high “Q” structures would require, at the most, a two-element network. A two-element network would provide a perfect or near perfect match at a single frequency. Networks applicable to narrow band matching are discussed in this, section. Such networks include (a) single-element series and shunt, types, (b) two-element types, and (c) simple quarterwave transform: ers that use both “lumped” and “distributed” parameters. Smith Chart illustrations are shown that describe chart areas within which impedances are susceptible to compensation using series or shunt networks. Such illustrations as these facilitate network selection. Practical application of narrow band matching methods are related to stub tuning devices that are used in scale model antenna radiation pattern measurements,e e@ IMPEDANCE MATCHING 5.2 DEFINITION OF BANDWITH The bandwidth of an r.f. component is that frequency range over which the component operates within a specified performance re- quirement. The performance requirement for impedance bandwidth is specified in terms of SWR. Percent bandwidth (%BW) is expressed as: fy BW = 2( ) +100 (5-1) fut fi Where: fy = upper frequency limit, fy = lower frequency limit A VHF aircraft antenna, for example, that operates within the 118 MHz to 136 MHz frequency band is said to have a 14.2% bandwidth in accordance with the above definition of bandwidth. ‘The term “octave” is another expression that is commonly used in 1. work to describe bandwidth. To illustrate, a component that op- erates from 2000 MHz to 4000 MHz is said to operate an octave in bandwidth. Had the upper frequency been 8000 MHz, the same com- ponent could be described as a two-octave device. Other frequently used terms relating to bandwidth and the classifi- cation of rf. components are (a) narrow band, (b) moderate band, and (c)broad band. Component classifications on the basis of these terms are somewhat vague; however, typical classification of band- width limits that are generally acceptable throughout the industry re: Bandwidth Term General Limits Less than 10% Between 10% and 50% Greater than 50% Narrow Band Moderate Band Broad Band SERIES AND SHUNT NETWORKS 79 5.3 SERIES AND SHUNT NETWORKS 5.3.1 Matching Analysis Aids Figures 5-1 and 5-2 are graphical design aids that may be used to de- termine network configurations for series and shunt elements to com- pensate a given load impedance. The 2:1 SWR circle that is shown in the figures is arbitrarily selected for illustration only. Similar con- structions are quite easily defined for other SWR value cireles. The shaded chart areas in the figures define those areas in which imped- ances 50 located are susceptible to matching using series and shunt elements. The application of these graphical aids will be described for single and two-element networks such as those that are shown in Figure 5-3. 5.3.2 Single-Element Networks A single-element, either in series or shunt with a load impedance, forms the most simple type of matching network. The network adds a reactance (+ jX) or susceptance (+ jB) component that tends to can- cel that respective component of the complex load impedance. The compensation effectiveness of a single-element network is dependent upon the chart location of the initial impedance curve. Consider the load impedance curve, Z'a a, that is shown in Figure 5-4. The load impedances normalized to 60 ohms are: F (MHz) Load Impedance (2" 44) 30.0 0.5 +51.0 31.0 0.8 +516 32.0 2.04} 2580 e IMPEDANCE MATCHING iouRe s-1 WOEOANCE OR ADMITTANCE COORDINATES etuoar FOR ‘mnmh oHaRT AREAS IN WHICH DAPEDANCES ARE SUSCEPTIBLE TO MATOHING TO MITHIN A 2:1) SMR CIRCLE USING. SERIES NETWORKS e@ SERIES AND SHUNT NETWORKS 81 [VPEOMCE OR ADUITTNGE cxORDINATES sort ona FlGRE 5-2 | AREAS Im HICH tHPEDMNCES ARE suscEPTIBLE To MATCHING TO MITHIN A 2:1) Sum CIRCLE USING swuNT NETMoREs82 IMPEDANCE buns ‘SERIES AND SHUNT NETWORKS 83 DANCE OR ADMITTANCE COORDINATES A A : | zx | I 4 i \ L Q ‘ I bee i 3 i r 3 | I ERE a Saar ) sneue_euchent A 4 Sr SEES SERIES - SANT © vec cn coees fae cert 7 vara sine A SERIES coon (e) LINE TRANSFORMATION. PLUS SINGLE ELEVENT FIGURE 5-4 | IMPEDANCE HATCATNG. USING FIGURE 5-3 SERIES AND. SHUNT. NeTWDAKS A SINGLE G.DUNT NETORKad IMPEDANCE Rrcune ‘The process involved in the selection of a single-element matching network to compensate impedance curve Z’aq to within an SWR circle of 8:1 is as follows: 1/Construction dash lines are drawn, similar to that of Figure 5-2, to establish the chart area in which the impedances must lie in order to be susceptible to matching using a series network. As one achieves practice in Smith Chart operations, the matching suscep- tible chart areas are readily visualized, and the above construction is not necessary. 2/Examination of the impedance curve Z', shows that the entire ccurve lies within the chart area susceptible to matching by a series capacitance network. In order to determine the capacitance need- ed, a trial impedance value (Z'.) is assigned to one frequency and the compensation effects throughout the band are analyzed. 9/A trial value for Z'c of -j1.5 is assigned at 30 MHz. This imped- ance in series with the initial load impedance (Z'aaio =0.5++j 1.0) results in a compensated impedance at Reference Plane BSB of Z’paso = 0.5 - j0.5. 4/The impedance values (2'.) of the capacitor and its compensation effects throughout the band that results in the matched curve Zipp are: £ (MH) aA = Zap 80.0 05+j10 -j15 0.5- 50.5 31.0 08 +j16 145 08+} 015 32.0 2.04525 141 2.0 +} 1.09 5/The capacitance value (pf) of the capacitor may be calculated from: 10? (af) %) (picofarads) (5-2) Where: C = capacitance (picofarads) Xe = capacitance reactance (ohms) f= frequency (hertz) e SERIES AND SHUNT NETWORKS. 85 ‘The capacitance reactance (X-) at 30 MHz is: Xeso = Zea + 50 +4 1.5 + 50 = 75 ohms and from Equation (5-2): a 1012 (628) (80x 10°) (7a) ‘Thus the chart location of the initial impedance curve, Z’a,, was favorable such that compensation over the frequency band was ac- complished by a single-element capacitor network. = 70.5 pt 5.3.3 Two-Element Networks ‘Two-element matching networks are less dependent upon the chart location of the initial impedance curve and are more flexible in ap- lication as compared to the single-clement network. Also, two -ele- ‘ment networks provide greater bandwidths when properly selected. In Section VII, a theory of broadband impedance matching! is pre- sented that describes the method of selecting the first two elements, ina four-element configuration, that will provide the maximum im- pedance bandwidth. It will suffice here in the present section, how- ever, to describe only the manner in which a two-element network compensates a given impedance curve. In Figure 5-5 is shown the impedance curve, Z’ aa, of a high “Q” structure whose normalized values over the operating frequency band are: £ (MHz) 95.0 975 100.0 Let it be desired to compensate impedance curve Z'a4 to within an ‘SWR = 2:1 circle using a two-element network. For the purpose of this example, let the network elements be arranged in a series-shunt configuration. The process of impedance matching follows:e e 86 IMPEDANCE MATCHING 1/4 seriesshunt network configuration, in this example, requires that the first element be a series inductance and the second ele- ment a shunt inductance. The matching effect desired of the first element is to move the initial impedance curve to within the chart area where impedances are susceptible to matching using a shunt network. Construction dash lines are therefore established in Figure 5-5, similar to that of Figures 5-1 and 5-2 to establish the matching susceptibility chart areas @and 0. 2/A trial impedance value for the first element of Z'., = + j 3.1 is assigned at 95 MHz, This value in series with the load impedance (Z'aaos = 0.6 - j 4.0) provides an impedance at Reference Plane BB of Z'ppys = 0.6 -j 0.9. Zppos falls just within the matching susceptibility chart area 1 . ‘The compensating effect of 2',,, on impedance curve 2’, throughout the frequency band is: £ (MHz) Zax + Zi = Zipw 95.0 06-340 © +j8.1 06 -j0.9 975 05-]3.7 © +4.3.18 0.5 =} 0.52 100.0 04-35 +43.25 0.4 -j 0.25 ‘The trial value selected for 2't, is satisfactory since the compen- sated curve Z'gp falls within the desired chart area 1 that is sus- ceptible to matching using a shunt network. 3/ Admittance curve Y'p is established diametrically opposite Z'pp as the second element will provide shunt matching. Values of gp, as read from Figure 5-5 are: £ (MHz) Ysa 95.0 0.52 + 50.78 975 0.95 +510 100.0 1:80 +5115 A trial value of Y'i2 = -j 0.88 is assigned for the admittance of the second element at 95 MHz. This admittance in shunt with Y’pass (0.52 + j 0.78) gives Y'ccos = 0.62 -j 0.10. ¥'cces falls within SERIES AND SHUNT NETWORKS 87 IN@EDINCE OR ADMITTANCE COORDINATES ‘nme Onn FIOURE 5-5 | IMPEDRNCE MATCHING. USING A Two-CLENENT NETWORKe e 88 IMPEDANCE MATCHING and near the boundary of the SWR = 2:1 circle. The compensat- ing effect of Y'L2 on admittance curve Y’pg over the frequency band is: £ (MHz) Yes + Yns = Yoo 0.52+j0.78 -j0.88 0.52 -j 0.1 095+j10 -jo.86 095 +5014 1.80+j115 -jo.84 1:80 +5031 The trial value selected for Y'r,, proved satisfactory since all points of admittance curve Y'¢c fall within the specified SWR = 2:1 circle. 5.4 QUARTER-WAVE TRANSFORMERS 5.4.1 Line Transformers (distributed parameters) A segment of transmission line inserted in series with a load imped- ance for purposes of matching is termed a “line transformer.” Prin. cipal electrical parameters of a line transformer are the characteristic impedance (2,') and the electrical length (92). In a given application, 2! remains constant whereas 9¢ varies with frequency. ‘The impedance (Zin) “looking ” into the input of a line transformer 21 +) Ze! tanoe 2 + 5%, Yan de -omplex input impedance (ohms) Zin 20 Where: Zin (53) Zz = complex load impedance (ohms) QUARTER-WAVE TRANSFORMERS 89 Zq' = transformer characteristic impedance (ohms) a8. ~ electrical length of transformer (degrees) For the frequency at which #£ = \/4 = 90°, the transformer is termed a “quarter-wave” transformer and the input impedance is given by: _ a? 2 AA schematic of the transmission line (distributed parameters) quar- ter-wave transformer is shown in Figure 5-6(a). Quarter-wave trans- formers are used as matching devices in the higher frequency bands where the physical length of the transformer is not objectionable. At a single frequency, a resistive load (Ry) may be perfectly matched to the associated transmission line (Z) by a quarter-wave transform er whose characteristic impedance (Z,’) is: Lin (5-4) 2) = VR (55) Where: Zs! = transformer characteristic impedance (ohms) 2» = transmission line characteristic impedance (ohms) Ry, = purely resistive load (ohms) Line transformer usage is not necessarily restricted to narrow band ‘matching applications. Section VI shows their application as broad- band matching devices whose Z' and 9¢ are selected by graphical methods. 5.4.2. “PI” and “T” Transformers (Lumped parameters) Equivalent quarter-wave “PI” and “T” transformers are shown in Figure 5-6(b) and (c). These transformers utilize lumped parameters and are particularly useful in the lower frequency bands where the physical length of a line transformer would be objectionable. An interesting relationship to the line transformer exists that permits ready determination of the parameter values (Zz, and Zc) of the “PI” and “T” transformers. This relationship is expressed as:2 impepance SearcHING ; 3 ; a x ' ze i be— ws at to —ol i : % +r ke eS I T A 8 (©) PE Section (Lurped = constants) f 4 LP 1 i ! t Py a 2 | ! I o——_o—_4—__ 4 __] A 6 (c)__T sectfon (Lumped ~ constants) FIGURE 5-6 ~ EQUIVALENT. QUARTERUAVE TRANSFORMERS QUARTER-WAVE TRANSFORMERS aa Ly = +420 (5-6) 2 =-jZd (67) Where: Zy, = inductor impedance (ohms) Zc_ = capacitor impedance (ohms) Zq' = as determined by Equation (5-5) for the design frequency (fo) ‘To illustrate the compensation effects of the equivalent quarter-wave transformers that are shown in Figure 5-6, the following load imped- ances are selected to be compensated: £ (Relative) Z Load Z! Load 0.8 fy 18-510 0.3- 50.2 1.0 f, 15+j0 0.3+j0 12, 15 +j10 03 +502 At the design frequency (fg), the Zy' of a line transformer for this case is: 20! = 50> 15 = 27.4 ohms and for the equivalent “PI” and “T” transformers the impedances %, = +} 27.4 ohms Ze = - 527.4 ohms Figures 5-7, 5-8, and 5-9 show the compensation characteristics of the three equivalent quarter-wave transformers as applied to the same load impedances. Examination of the figures show that, at the design frequency (fp), the compensation effects are the same and a perfect match, SWR = 1:1, is achieved. At frequencies other than the design frequency, the line transformer (Figure 5-7) has some- ‘what greater bandwidth potential as a result of the uniform transfor- mation properties that are characteristic of distributed parameter elements,e IMPEDANCE MATCHING QUARTER-WAVE TRANSFORMERS 93 +302 ‘SMITH CHART PLOT OF IMPEDANCE AT REFERENCE PLANE AA: FIGURE 5-7 J INPEDANCE MATCHING USING A QUARTERHAVE LINE TRANSFORMER +5240 ' Zio fo = alee e 2 Loan LOAD (normalized to Zo = S00} 0.8 fo 18-4109 1.0 fo 1543 on V2 fo B+} 109 SMITM CHART PLOT OF IMPEDANCE AT REFERENCE PLANE A-A: 0.8 fo 1.2 fo 1.0 fo = 500 FIGURE 5-8 | IMPEDANCE MATCHING USING AN EQUIVALENT QUARTERNAVE _PY-NETHORK94 IMPEDANCE MATCHING LABORATORY TUNING DEVICES 5.5 LABORATORY TUNING ! DEVICES : : In paragraph 3.3, open-circuited and short-circuited line segments t zvoap LOAD (normalized to Zo_= 509) were described as impedance matching devices. They behave as pure onares seve reactances whose input impedance values are readily determined by 1.0 fo ej lon graphical construction on the Smith Chart. In the r.f. laboratory, 12 fo 1543 108 line segments are more commonly termed “stub tuners” and are used as “‘in-line” matching devices both in the coaxial and wave- SMITH CHART PLOT OF IMPEDANCE AT REFERENCE PLANE A-A: guide forms. The majority of stub tuners are shorted stubs. Open- circuited stubs are rarely used as a true open circuit is difficult to achieve. Also an open line radiates energy which is not desirable in most laboratory test setups, In Figure 5-10 are shown the schematics of both single and double stub tuners. Since the stubs are shunted across the transmission line, it is more convenient to use admittances to describe the compensa- tion effects. A single stub must be located at the proper position in the transmission line in order to be effective, For example, consider the load Y1, = Gr, + j By, located at Reference Plane A-A of Figure 5-10(a). If Vz, is rotated along the transmission line to a point on the Smith Chart where Yy,= 1 +j By, then a single stub inserted at this point will have the maximum compensation effectiveness. The shorting plunger of the stub may be adjusted to cancel out the sus- ceptance (+ j Br,) of Yy, to give Yz, = 1 j 0. Thus a perfect match is achieved and there will be no reflections on the transmission line. In practice it is usually inconvenient to locate a single stub at its proper position in the line. Double stub tuners, however, may be generally located at most any position in the transmission line and still cancel out most reflections. The distance between the two fixed stubs is usually 8/8 \. To describe the compensation “action” FIGURE 5-9 | INPEOANCE MATCHING USING AN EQUiVALe) NETHORK, QUARTERKAVE® IMPEDANCE Prrcuine (a) simece stue raver ‘s3—> oe Y \~ 98 ‘oc —> FIGURE 5-10 SCHEMATICS. OF MATCHING STUB TUNERS LABORATORY TUNING DEVICES 7 ‘of a double stub tuner, consider the construction shown in Figure 5-11. In the figure, the admittance, Y, , looking towards the load at Reference Plane B-B is 0.5 + j 1.5. The stubs are adjusted as follows: 1/The shorting plunger of the first stub is adusted to make 8, 087 A. This gives Y, = -j 1.63, Ypp is then: Yes = ¥:+¥; = 0.5 +) 1.5 + (41.63) = 05-j0.13 In the above, Ypp is positioned on the chart such that the dis- tance between the stubs (3/8 2) places the admittance, 3, on the G = 1.0 axis. At this point Y; = 1.00.7. 2/The electrical length, #4, of the second stub is adjusted such that Yq = +] 0.7. The final input admittance, Ycc, becomes: Yoo =¥: +¥s =1.0=j0.7+j0.7 .0 + j 0 (a perfect match) The above example shows the theoretical aspects of matching with a double stub tuner. In the laboratory, this is not the case. The stubs are adjusted altemately, without resorting to calculations, until the SWR on the transmission line is reduced to unity. 5.5.2 Slide Screw Tuners Slide-screw tuners are particularly useful for reducing reflections in waveguide transmission line test setups. This type of tuner consists of slotted section of waveguide on which is mounted a precision car- riage containing an adjustable probe. The length of the slotted sec- tion is generally several wavelengths. The penetration of the probe into the guide is controlled by a micrometer drive whereas the probe position, along the guide, is varied by a thumb-operated wheel ‘Thus the depth and position of the probe may be varied to setup a reflection to cancel an existing reflection that is existing on the line,98 IMPEDANCE MATCHING LABORATORY TUNING DEVICES 99 5.5.3 Hatch Tuner? Another interesting transmission line tuning device is the Hatch ‘Tuner whose coaxial representation is shown in Figure 5-12. The’ transmission line may be bent in a circle, Reference Planes A-A and D-D near coinciding, to provide a device consisting of series and shunt sections of transmission lines whose lengths are adjustable by two moving shorting sections. The Hatch Tuner was initially de- veloped to be used in scale model aircraft, such as that shown in Figure 5-13, to provide impedance compensation for model antennas operating in the 300 MHz to 1000 MHz frequency bands. FIGURE 5-11 | IMPEDANCE. MATOHING. USING A oovBLe STUB TUNERay WK 1o® IMPEDANCE vbuine LABORATORY TUNIN@DEVICES 101 RE ween CIRCULAR FORM OF TUNER “re = Mode! antennas Tatteone Frowe 5-12 y= OPEN crRcuTTED LINE fp + SORT CIRCUITED LINE 1) ~ MOVEABLE SHORTING LINE ‘A COAXIAL REPRESENTATION OF THE HATCH TUNER A 1/20th Seale Model Aircraft for Radiation Pattern Studies Hatch Tuner: = (300-1000 ts) FIGURE 5-13 | HATCH TUNER — A VARIABLE MATCHING DEVICE USED IN MODEL PATTERN STUDIES.102 IMPEDANCE Brow ING REFERENCES 1. RLL. Thomas, “Broadband Impedance Matching in High-@ Networks,” EDN Magazine, December 1973, 2. RL, Thomas, “Design Charts for Single Section Line Transformers’ ‘Technical Report No. DAC 66662, Douglas Aircraft Company, Ine, Long Beach, California. 1967. 8, RM, Hatch, “Improvements in Instrumentation for the Investigation of Aircraft Anfenna Radiation Patterns by means of Seale Models,” Technical Report No. 26, Stanford Research Institute, dated August 1952. BIBLIOGRAPHY |. G.d. Wheeler, Introduction to Microwaves, Prentice-Hall Book Co., Ine. New Jersey, 1963, 2. J.A, Nelson and G. Stavis, “Impedance Matching, Transformers and Baluns,” Very High Frequency Techniques, Chapter 3, Radio Research Laboratory, McGraw-Hill Book Company, New York and London, 1947, HLF, Mathis, “L-Network Desig,” Electronics Magazine, pp 186188, dated 1 February 1957. soaatle6 BROADBAND MATCHING— LINE TRANSFORMER DESIGN CHARTS 6.1 INTRODUCTION The term “broadband matching”, as defined in Section V, implies impedance compensation where bandwidths are greater than 50%. ‘This type of matching is more technically complex when compared to single frequency or narrow band matching, The previous section has treated simple networks such as (a) single element, (b) two-ele- ‘ment, and (c) quarterwave transformers where the impedance band- widths of concem were narrow to moderate, In these cases, very little technical effort need be expended to achieve a satisfactory match. Broadband matching, however, may require considerable technical effort and the degree of success is directly related to the ingenuity of the designor in his eelection of appropriate networks. ‘Two sections are here devoted to the discussion of broadband match- ing techniques. The present section describes the use of design charts, as graphical aids, in the derivation of design parameters for broadband line transformers. Such aids permit a rapid accessment of the com- Pensation effects of a given transformer design. Section VII will dis- cuss the use of four-element optimum networks for maximum band- width. Such networks are particularly applicable for compensating high “Q” structuree 106 IMPEDANCE MATCHING 6.2 LINE TRANSFORMER ELECTRICAL PARAMETERS ‘A transformer that is inserted in series with a load and the associated +. transmission line is termed a “line transformer.” Principal para- meters of such a transformer is its electrical length (@) and charac- teristic impedance (Z,'). In a given design Za’ remains constant where- as 02 varies with frequency. Each specific transformer has an associated family of transformation circles or 8 circles. The significance of a particular 8 circle is that the circle is transformed to the definition circle (SWR) when the trans- former electrical length equals 9. 9 circles, plotted upon design charts, provide a method for performing a graphical analysis of the ‘compensation characteristics of a given transformer design. A descrip- tion of such design charts and their application follows. DESIGN CHARTS e 107 6.3 DESIGN CHARTS 6.3.1 General ‘Tabulated data to permit the formulation of design charts for line transformers for various SWR values is reported in Reference 1. De- sign charts have been compiled from this data for the case where the definition circle (SWR) equals 2:1 and are included in Appendix A. These charts permit a rapid analysis of line transformer charac- teristics. igure 6-1 shows a typical design chart. The chart coordinate system (resistance-reactance axis) and transformer characteristic impedance (Zq!) have been normalized to an associated transmission line charac- teristic impedance (Zo) of 1.0. The charts are thus universal in appli- cation and not limited to the specific case of Zy = 50 ohms. In Ap- pendix A, charts are provided with transformer characteristic im- pedance values in 0.1 Z, increments from 0.4 Zy to 3.0 Zy. 6.3.2 Chart Description * Title Block — The SWR ratio and normalized value of Zo and Zp’ are given here. Should the characteristic impedance (Zo ) of the associated line be 50 ohms, this particular chart would reflect the design parameters of a 35 ohm (0.7 Zo) transformer (Figure 6-1). ‘* Coordinate System — The resistance-reactance axis are normalized to 2, = 1.0. Since the impedances plotted on the Smith Chart are in normalized form, they may be directly transferred to the design chart, * Definition Circle (SWR) — The definition circle, or SWR circle, is established by performance specifications and all impedance values contained within the boundaries of this circle are consid- ered to be matched. The SWR 2:1 circle crosses the normalized resistance axis at 0.5 and 2.0.IMPEDANCE MAING OUTER BOUNDARY CIRCLE DEFINITION CIRCLE (VSWR) INNER BOUNDARY CIRCLE (INVERTED IN THIS CASE) TRANSFORMATION. CIRCLES (9-CIRCLES ) RESISTANCE ANO REACTANCE AXES (R-X) (NORMALIZED TO Zo=10) FIGURE 6-1 TYPICAL DESIGN CHART FOR SINGLE SECTION LINE TRANSFORMER ANALYSIS FACTORS Rossin SELECTION 109 * Outer Boundary Circle ~ This circle encloses the SWR circle and all points on the impedance curve that is to be matched. The circle is tangent to the SWR circle, on the resistance axis at 2.0 when Zq' < than Z and 0.5 when Za > Zo. Inner Boundary Circle — The inner boundary circle is tangent to both the d = 90° circle and the SWR circle. This circle is said to be inverted when it is contained within the SWR circle. ‘© Transformation Circles (@ Circles) — Each transformer has an as- sociated family of transformation circles. All @ circles are differ- ent in size and progress in a counterclockwise manner on the charts. All 0 circles are tangent to (a) the outer boundary circle and (b) the inner boundary circle. It is the characteristics of the 9 circles that provides the method of performing a graphical analy- sis of line transformer matching susceptibility. 6.4 ANALYSIS FACTORS IN DESIGN SELECTION Simplicity of design chart usage is based upon an understanding of the following fundamental analysis factors: * A line transformer has different electrical lengths (0) at each fre- quency considered. Each transformer has a specific family of @ circles. The transformer will transform to the SWR circle one 9 circle at one frequency and a different @ circle at another frequency.e oe NO IMPEDANCE MATCHING. ‘* When an impedance point falls within a @ circle, the point is trans- formed to within the SWR circle at the frequency where the elec- trical length of the transformer equals to @. The degree of com- pensation afforded by a4 circle is illustrated in Figure 6-2. Should the impedance point lie near the boundary of the 8 circle, it will be transformed to a position near the boundary of the SWR circle, The impedance point will be transformed to near the center of the SWR circle, if it is initially near the center of the @ circle. © Ifsuccessive points on the impedance curve fall within the respec- tive @ circles for those frequencies, the entire impedance curve may be transformed to within the SWR circle by that particular transformer, This implies that increasing frequency impedance points should “track” increasing 0 circles for effective matching. © The portion of the impedance curve that lies in the region be- tween the inner and outer boundary circles is susceptible to matching. Points outside the outer boundary circle or within the inner boundary circle (except when this circle is inverted) are not susceptible to matching by that particular transformer. © Impedance points within an inverted inner boundary circle remain in that circle (and hence, within the SWR circle) regardless of the electrical length of the transformer. Such points are, therefore, of no concern in the analysis. © In the selection of a trial electrical length (8) for the transformer for matching analysis, consideration should be given to the criti- cal points of the impedance curve. Critical points are those that (a) lie near the outer boundary circle or (b) on a portion of the impedance curve that moves in a clockwise manner (direction of decreasing 6 circles) with increasing frequency. There are two basic electrical characteristics that define a particular line transformer; the characteristic impedance (Zs') and electrical length (0). Zo’ is derived by simple construction methods using the Smith Chart, whereas 92 is selected by trial by judicious consid- eration of the above analysis factors. Both processes are described by the selected example of chart usage that follows. ANALYSIS FACTORS IN DESIGN SELECTION 411 OUTER enNDARY cELE f ewe arcte 7 fs es 7 te) wrenavee Poet HEAR CENTER OF @ ROLE wep [ ouTeR souusy cnet oa 2 8 aa 2 | 5 8 ost (B) IMPEDANCE POINT NEAR BOUNDARY OF @ CIRCLE FicuRe 6-2 COMPENSATION EFFECTS OF A 9 CIRCLE Gv TWO ARBITRARILY SELECTED Po IMPEDANCE POINTS12 IMPEDANCE MATCHING EXAMPLE OF CHART USAGE 113 6.5 EXAMPLE OF CHART USAGE 6.5.1 General ‘The processes involved in the selection of the characteristic imped- ance (hence a particular design chart) and consideration of analysis factors to determine an optimum electrical length will be described using the uncompensated impedance curve that is shown in Figure 6-3. It is desired to compensate this curve to within a SWR of 2:1. | 6.5.2 Design Chart Selection %! of a possible line transformer is determined by construction of an outer boundary circle, Figure 6-3, by inspection to enclose all points on the impedance curve for which compensation is desired. By definition, the outer boundary circle, in Figure 6-3, is tangent, to the SWR circle (0.5 on the resistance axis). Za’ is given by: Product of Outer Boundary Zq' = \{ Circle Resistance Axis Intercepts (6-1) and for this case is: 20! = JOBx46 = 15225 In Appendix A, the design charts are given in 0.1 Zp increments and the chart Z,' ='1.5 Zo is selected as nearest to the above calculated value. The impedance curve of Figure 6-3 is transferred to the Zo’ = 1.5 Z» chart shown in Figure 6-4 for matching analysis. Determina- tion of the transformer electrical length (0) and whether or not all points on the impedance curve may be matched follows from an analysis of this figure. Pome 63 eusmurion yo Fewor seen oF carer WPEDANCE (2p) OF A POSSIBLE LINE TRAKSFORNER