
	 






	                    AO 6.5 ANTENNA OPTIMIZER




	             Copyright 1995 by Brian Beezley, K6STI
	                       All Rights Reserved




		AO 6.5 Antenna Optimizer.......................1
		Running AO.....................................1
		Antenna Files..................................2
		Coordinate System..............................3
		Wires, Segments, and Pulses....................3
		Automatic Wire Segmentation....................5
		Automatic Segment-Length Tapering..............5
		Pulse Identification...........................6
		Sources........................................7
		Loads..........................................8
		Free-Space Symmetric Antennas.................10
		Ground........................................11
		Wire-Conductivity Losses......................14
		Standing-Wave Ratio...........................15
		Symbolic Dimensions...........................15
		Symbolic Expressions..........................17
		Shift and Rotate..............................17
		Polarization Components.......................19
		Near Field....................................20
		Far Field.....................................21
		3-D Geometry/Current Display..................21
		3-D Pattern Display...........................22
		2-D Pattern Generation........................23
		2-D Pattern Plots.............................23
		Screen Colors and Default Menu Settings.......26
		Printing The Screen...........................26
		Graphics Image Files..........................27
		Aborting Calculations.........................27
		Automatic Frequency Sweep.....................27
		Batch Mode....................................28
		Algorithm Limitations and Corrections.........29
		Memory........................................31
		SET Commands..................................33
		Additional Information........................39
		Index.........................................40


	---- AO 6.5 ANTENNA OPTIMIZER ----------------------------------

	        AO is an enhanced version of the MININEC antenna-
	analysis program combined with an automatic optimizer.  AO is
	more powerful than MININEC, more accurate, easier to use,
	faster, and has many additional features.  AO can analyze and
	optimize almost any antenna made of wire or tubing.  However,
	the program is not suitable for designs employing dielectrics,
	ferrites, or surfaces.

	        AO is based on the MININEC 3.13 antenna-modeling
	software developed by J.C. Logan and J.W. Rockway at the U.S.
	Naval Ocean Systems Center.  AO uses a modified algorithm for
	improved accuracy.  AO runs much faster than MININEC.  The speed
	increase is due to code optimization, special analysis modes,
	and extensive use of assembly language.

	        This document covers general aspects of AO.  See OPT.DOC
	for information about its optimizing capabilities.


	---- RUNNING AO ------------------------------------------------

	        The quickest way to start AO is to give the name of an
	antenna file on the command line, for example, AO DIPOLE.  If
	you don't know which file you want, type AO and antenna files in
	the current directory will be listed.

	        Select a file by moving the lightbar with the arrow
	keys, PgUp, PgDn, Home, or End.  Press Enter to select the
	highlighted file.  Alternatively, you can type a filename.  As
	you type, the lightbar moves to the first filename that matches
	the characters entered.  Press Enter whenever the desired file
	is highlighted.

	        Select the last item ("Other") to enter a file or
	directory name not listed.  If you enter a directory containing
	antenna files, AO lists them and you can select one.  See the
	SET Commands section for information on organizing antenna files
	into subdirectories.

	        Instead of analyzing an antenna file, you can display a
	2-D plot file generated previously.  To do this, start AO by
	typing AO PLOT.  Plot files in the current directory will be
	listed and you can select one.  Alternatively, you can specify
	the name of a plot file on the command line, for example, AO
	PLOT DIPOLE, to display the plot immediately.

	        After loading an antenna file, you control AO by making
	menu selections with single-keystroke command keys.  Command
	keys are highlighted by color in the main menu.  Use lower-case
	characters; upper case has a special effect for P and Z.

	        To return to the Main Menu from a submenu, press Esc.
	You also can return by pressing the command key a second time.
	You can terminate data entry and exit a submenu with one

				       1

	keystroke by pressing Esc instead of Enter.  You can bypass the
	Main Menu and go directly from one submenu to another by
	pressing a command key.  Use Esc to hide the Main Menu to view
	the whole screen.

	        The D and N commands output results to a file named RUN.
	RUN is an output file created by AO each time it is executed.
	RUN holds lengthy results and contains a complete record of the
	analysis session.  RUN also contains an image of the antenna
	file.  You can examine RUN from within AO with the R command.
	You can also use the file later, after quitting AO.  To save a
	particular RUN file, you must rename it.  RUN is always
	overwritten with new information whenever you execute AO.

	        To create a summary printout of analysis results, use
	PrtSc to print the main screen or the RUN screen.  You can also
	quit AO, use your text editor to edit RUN, and print selected
	results.

	        Error messages are highlighted in bright yellow at the
	bottom of the screen.  Not all commands are available if an
	error is detected in the antenna file (you must correct the
	error before proceeding).


	---- ANTENNA FILES ---------------------------------------------

	        Each antenna system is described in a file with the
	extension .ANT.  You can create an antenna file with a text
	editor before starting AO or you can use the E command to invoke
	your editor from within AO to edit a file already loaded.  Once
	AO is running, you can load a new antenna file with the A
	command.

	        Antenna files consist of ordinary text characters.  The
	format of an antenna file is illustrated by this sample file for
	a 3-element Yagi.  Anything following a semicolon is a comment.
	Comments can be added freely at the end of a file without using
	semicolons.


	3-Element Yagi                                  ; Antenna title
	Free Space                                 ; Ground description
	14.2 MHz                     ; Frequency.  Use kHz, MHz, or GHz
	3 wires, feet                ; Number of wires, dimension units
	10   -8 -17   0   -8 17   0   1"   ; For each wire: # segments,
	10    0 -16.5 0    0 16.5 0   1"     ; XYZ coordinates of ends,
	10    7 -16   0    7 16   0   1"                 ; and diameter
	1 source                                 ; Number of feedpoints
	Wire 2, center  1  0          ; Source location, voltage, phase
	0 loads                                              ; Optional


	        Details of antenna-file entries are explained in
	subsequent sections.

				       2

	        Antenna-file lines are free-form.  You can position
	items anywhere on a line as long as you maintain item order.
	Use spaces, commas, or tabs to separate items.  Upper or lower
	case can be used.  Most keywords, like "wires", can be singular
	or plural.  Blank lines or comment lines can be inserted
	anywhere to improve readability.

	        Dimensions can be specified in feet, inches, meters,
	centimeters, or millimeters.  AO looks for just the first two
	characters of these keywords, so both American and British
	spellings can be used.  You can specify the length-unit
	abbreviations given next as well.

	        Any dimension in the wire list can use its own unit of
	length.  Simply append one of the following dimension overrides
	to the number (don't leave a space):

	                      ' " ft in m cm mm

	        You can specify wire diameter by entering a wire gauge
	prefixed with #, for example, #12.  AO uses American Wire Gauges
	for annealed bare copper wire.


	---- COORDINATE SYSTEM -----------------------------------------

	        AO uses the Cartesian coordinate system to refer to
	points in space.  X and Y are in the horizontal plane and Z is
	height.  If you think of +X as north, then +Y is west.  AO uses
	the same azimuth-angle convention as ordinary compass bearings:
	0 degrees is along the +X direction (north), and 90 degrees is
	along the -Y direction (east).  +Z is up.  The forward horizon
	is 0 degrees elevation angle, overhead is 90 degrees, the rear
	horizon is 180 degrees, and directly below is -90 degrees.  If
	you aim unidirectional antennas in the +X direction, AO
	automatically computes forward gain, F/B, and beamwidth.

	        To get a feeling for the coordinate-system geometry,
	imagine facing a 3-element Yagi 5 feet off the ground aimed at
	you.  Assume that the center of the driven element is at X = 0,
	Y = 0, Z = 5.  The main lobe of the beam points directly at you,
	in the +X direction.  The elements extend in the -Y direction to
	your left, and in the +Y direction to your right.  The director
	has a +X coordinate, while the reflector is in -X territory.
	Positive azimuth angle is to your left (clockwise as viewed from
	above), and positive elevation angle is above your head.


	---- WIRES, SEGMENTS, AND PULSES -------------------------------

	        A wire is always straight in AO.  You model a bent wire
	by connecting two or more straight wires.  Wires are considered
	to be connected when ends have identical coordinates.  For
	example, each loop of a cubical quad antenna is described by
	four individual wires whose endpoints lie at four points.  Thus
	you model a 2-element quad with eight wires, even though a real

				       3

	antenna has just two continuous wires strung on the spreaders.
	You can model a Yagi element using tapered, telescoping tubing
	with several connected wires of different diameter.

	        Connections are allowed only at wire ends.  Wires that
	cross or terminate at midpoints of other wires are not
	considered to be connected.  To model such geometries, subdivide
	the wires into shorter wires with coincident endpoints.  When
	modeling antennas over ground, wire ends with Z = 0 are always
	considered to be grounded.  You can define wires in any order in
	the antenna file.

	        Wires are divided into segments for analysis.  You can
	let AO segment wires automatically or you can specify the number
	of segments for one or more wires yourself.  Increasing the
	number of segments generally improves accuracy, but analysis
	then takes longer.  Using between 4 and 40 segments per halfwave
	of wire provides useful results.

	        AO uses the wire segmentation to model conductor current
	in sections called pulses.  AO models current as uniform within
	each pulse.  Pulses constitute the entire representation of wire
	current for the antenna model.  They're distributed as follows:
	Pulses are centered at segment junctions and have the same
	length as segments.  A pulse is placed at each wire junction and
	overlaps onto two wires; a junction of n wires has n-1 pulses.
	A half-length pulse is placed at each grounded wire end.  No
	pulse overlaps the last half-segment of an unconnected wire end.

	        The maximum number of pulses is limited only by
	available memory.  See the Memory section for details.  The
	number of wires is limited only by the number of pulses.

	        To check that you're using enough pulses, increase the
	number of wire segments and reanalyze to see whether results
	change significantly.  Until you develop a feeling for the
	segmentation necessary under various conditions, this
	convergence check is very important.

	        Whenever you change the analysis frequency with the
	Frequency Menu, the number of segments is not altered.  If you
	increase frequency by a large amount, the segmentation density
	may no longer be adequate.

	        For valid results, segments should not be too short.
	The length of each segment should be greater than .0001
	wavelength and greater than the wire diameter.  AO displays a
	warning message when either of these limits is exceeded.

	        As AO builds an internal representation of the antenna,
	it may reorder, subdivide, or ignore wires due to special
	processing described in subsequent sections.  The Wires and
	Pulses section of the RUN file reflects this internal
	organization.  There may not be a one-for-one correspondence
	between the wires listed and those in the antenna file.

				       4

	---- AUTOMATIC WIRE SEGMENTATION -------------------------------

	        You can control wire segmentation from the Options Menu
	by specifying the number of segments allocated per half-
	wavelength of wire.

	        AO allocates 10 segments per halfwave by default.  This
	gives good results in most cases.  For quicker results, use
	sparse segmentation, for example, 5 segments per halfwave.  For
	increased accuracy, use dense segmentation, like 20 segments per
	halfwave.

	        Until you're familiar with the segmentation necessary
	for various types of models, it's important to test convergence
	by increasing segmentation density until results no longer
	change significantly.  10 segments per halfwave is adequate for
	most models, but not for all.

	        You can disable automatic segmentation by setting the
	segmentation density to 0 in the Options Menu.  AO then
	allocates the number of segments specified for each wire in the
	antenna file.

	        When automatic segmentation is used, the number of
	segments specified on wire lines still affects segmentation in
	two ways.  When the number is positive, automatic segmentation
	allocates for the wire at least the number of segments
	specified.  This feature can be used to ensure adequate
	segmentation for special cases like embedded transmission lines.
	When the number of segments is preceded by a minus sign, AO
	disables automatic segmentation for the wire and allocates the
	number of segments specified.  This feature can be used to
	allocate fewer segments than would be allocated automatically.
	With guy wires, for example, sparse segmentation of the guys and
	automatic segmentation of the antenna can be used to reduce
	model size without loss of accuracy.

	        Except for special cases, it's best to leave automatic
	segmentation enabled and specify one segment for each wire in
	the antenna file.  This provides the widest possible range for
	automatic segmentation.  Whenever you encounter a file with a
	number other than 1 specified, you'll know that special
	precautions are being taken.


	---- AUTOMATIC SEGMENT-LENGTH TAPERING -------------------------

	        Whenever the lengths of two connected segments differ by
	more than a factor of about 2, you may get inaccurate results.
	All segments on a wire have the same length.  Problem junctions
	typically occur when a short wire with one short segment is
	connected to a much longer wire with long segments.  Allocating
	more segments to the longer wire will fix the problem, but this
	can waste a lot of segments.

				       5

	        AO can automatically subdivide a long wire into several
	single-segment shorter wires with geometrically increasing
	(tapered) segment-lengths.  The shortest wire is placed at the
	problem junction.  The length of succeeding wires is gradually
	increased until it is about the same as the original segment
	length.  A single wire with multiple segments is then used for
	the remainder of the span.

	        When automatic segment-length tapering is enabled, AO
	analyzes the entire antenna for problem junctions.  All
	junctions are corrected.  After the process completes, no
	connected segments will differ in length by more than a certain
	ratio.  The ratio is 2 by default.  See the SET Commands section
	to change this value.

	        Automatic segment-length tapering normally should be
	enabled.  It can be disabled from the Options Menu for testing.
	This feature works even when automatic segmentation is disabled.


	---- PULSE IDENTIFICATION --------------------------------------

	        Sources and loads must be located at pulses (segment
	junctions).  A particular pulse can be identified by specifying
	its pulse number.  Pulse numbers are assigned by AO and are
	listed in the Wires and Pulses section of the RUN file.  To
	specify pulse number, use the following phrase following a
	number-of-sources or number-of-loads line:

	                               Pulse 1

	        Pulse number 1 is used here for illustration.  The
	keyword "pulse" can use upper or lower case (in many contexts,
	this keyword can be omitted entirely).

	        Sources and loads usually are located at the center or
	end of a wire.  When this is the case, there is an easier way to
	identify a pulse.  In place of the pulse-number phrase, use one
	of the following:

	                            Wire 1, Center
	                            Wire 1, End1
	                            Wire 1, End2

	        Wire number 1 (the first wire specified) is used here
	for illustration.  End1 refers to the first end specified on the
	wire line, and End2 to the second.  You can use upper or lower
	case.  The comma is optional.

	        When a wire center is specified but no pulse is located
	there, AO increments the number of segments for the wire by one
	to provide a center pulse.  This trick almost always works, but
	it may fail if the wire undergoes automatic segment-length
	tapering at just one end.  If AO cannot fabricate a center
	pulse, it selects the pulse nearest the center and displays a
	warning message.  Usually a change in segmentation density will
				       6

	provide a center pulse.  For a permanent fix, divide the wire
	into two equal-length wires and specify a wire end.

	        If a pulse is specified at an unconnected wire end, AO
	displays an error message.  No pulse exists at a free wire end
	and there is no current-return path for a source or load.  You
	can specify a grounded wire end since ground provides the return
	path.

	        In addition to being highlighted in the geometry
	display, source and load pulses are identified in the RUN file
	in the Wires and Pulses section and in the Currents section.


	---- SOURCES ---------------------------------------------------

	        You can apply a source of power at any pulse.  Think of
	the wire as cut in two at the source location and a feedline or
	RF generator inserted in the gap.  At a multiwire junction,
	think of the specified wire as fed against the junction.
	Antenna excitation in AO uses an infinitely small feedpoint gap.
	If you try to imagine feedpoint gaps with finite separation,
	you'll become hopelessly confused.

	        You must take into account the feedpoint location when
	manually segmenting driven wires because an antenna can be
	excited only where a pulse is located.  In the Yagi example
	given earlier, the driven element uses an even number of
	segments.  This results in a pulse at the exact wire center at
	the junction of the two middle segments.  Specifying an odd
	number of segments for the driven element would permit only an
	off-center feedpoint.  Automatic segmentation normally takes
	care of this problem automatically.

	        You can specify either voltage or current sources.
	Voltage sources have zero source impedance, while that of
	current sources is infinite.  To specify current sources, put
	"current" anywhere on the number-of-sources line.  Current is
	specified in amperes and degrees.  Current feed is very handy
	when designing phased arrays.  Many phased-array designs specify
	element-current ratios rather than the complex phasing systems
	required to achieve those ratios.  Unless you're interested in
	absolute near- or far-field values, it makes no difference
	whether you use a voltage source or a current source for single-
	source models.

	        When you omit source magnitude and phase, 1 volt (or 1
	amp) and 0 degrees are used.  This default is convenient for
	single-source models.

	        You can change source characteristics either by editing
	the antenna file or by using the Sources command.  The latter
	does not make any changes to the antenna file and does not cause
	recalculation of the mutual-impedance matrix.  This command is
	handy when experimenting with phased-array systems.  The first
	ten sources can be modified from the Sources Menu.
				       7

	        When modeling designs with multiple sources, you may get
	negative values for input resistance.  This simply means that
	power is flowing into a source rather than away from it.  The
	power originates at other sources and is coupled to the
	negative-resistance source by the near field.  This situation
	often occurs for multiple-source models and is normal.  However,
	negative input resistance for a single-source model is a sure
	sign of improper segmentation.

	        MININEC normally does not allow a source (or load) to be
	defined on the first wire of a junction of three or more wires.
	However, AO automatically reorders wires to accomodate this
	condition.

	        The number of sources is limited only by the number of
	pulses.


	---- LOADS -----------------------------------------------------

	        AO can model antennas with lumped loads.  Loaded
	antennas include multiband Yagis, dipoles, and verticals with LC
	traps, shortened dipoles and whips with loading coils, and
	traveling-wave antennas with resistors (rhombics, V-beams, and
	Beverages).  Loads must be located at pulses.  You can subdivide
	a wire into two wires to accomodate a load where no pulse exists
	since a pulse is placed at every wire junction.  You can
	identify loads by pulse number or by specifying a wire center or
	end.

	        AO provides several ways to specify loads.  It's easiest
	to define simple loads as RLC loads.  Here's an example of three
	RLC loads:

	3 loads
	Wire 1, Center   27 pF  4.7 uH  3 ohms
	Wire 2, End2     4.7E-6 H  Q=300
	Wire 7, Center   600 ohms

	        This load syntax specifies RLC component values for
	parallel traps.  R is in series with L, and C is in parallel
	with RL.  For simpler loads, use fewer components.  For example,
	the second load defines a lossy inductor, while the third
	defines a resistor.  Component values not specified are set to
	0.  (When just C is specified, R and L are removed from the
	circuit to avoid shorting C.)  Use Q to automatically determine
	the series R for a lossy inductor.  When L is not specified, Q
	determines the parallel R for a lossy capacitor.  (Once R is
	determined from Q, it does not vary with frequency.  Q is
	defined at the file frequency.)

	        The RLC syntax can model a lossy trap, a lossless trap,
	an inductor, a lossy inductor, a capacitor, a lossy capacitor,
	or a resistor.  AO recognizes the following units:  pF, F, nH,
	uH, H, and ohms.  You can use upper or lower case.  Units must
	be separated from values by at least one space.  You can specify
				       8

	RLC components in any order on a line.  You can separate entries
	by spaces, commas, or tabs.

	        Internally, AO contains two types of load models,
	impedance loads and Laplace Transform loads.  AO translates RLC
	loads into Laplace Transform loads.  The RLC syntax is adequate
	for all common loads.  For special cases, you can specify
	impedance loads or Laplace Transform loads directly.

	        An impedance load is a resistance in series with a
	reactance.  Unlike RLC loads, the reactance does not vary with
	frequency.  Here is a syntax example for impedance loads:

	2 loads
	Impedance Loads
	Wire 2, End2   740  122
	Wire 4, Center  44

	        "Impedance Loads" is not a keyword.  AO recognizes an
	impedance load whenever this line begins with anything other
	than "Laplace", "Wire", or "Pulse".  This lets you use a
	descriptive phrase (for example, "Capacitive Reactance") to
	invoke an impedance load and simultaneously annotate the file.
	The two values on the wire line are resistance and reactance
	(resistance first), both in ohms, without units.  When the
	reactance value is omitted, it is assumed to be zero.

	        Laplace Transform loads permit complex networks to be
	modeled.  A Laplace Transform is a special polynomial
	representation of a lumped circuit.  Laplace Transforms are too
	complicated to explain here, but the following example
	illustrates the syntax:

	2 loads
	Laplace Transform
	Wire 1, End2                             ; First load location
	0   8.2E-6                     ; Numerator coefficents s^0 s^1
	1   0   4.92E-16        ; Denominator coefficients s^0 s^1 s^2
	Wire 4, End1                            ; Second load location
	0   8.2E-6         ; The second load is identical to the first
	1   0   4.92E-16

	        The maximum transform order is 6.  Enter coefficients in
	order, starting with s^0.  Magnitudes smaller than 5.8775E-39
	cannot be used.  Higher-order zero coefficients need not be
	entered, as illustrated.  When calculating coefficients, use
	ohms, Henries, and Farads.  Consult a textbook on circuit
	analysis for information on Laplace Transforms.

	        A load coincident with a source appears in series with
	the source.  The number of loads is limited only by the number
	of pulses.




				       9

	---- FREE-SPACE SYMMETRIC ANTENNAS -----------------------------

	        AO provides a special technique for modeling symmetrical
	antennas in free space.  Twice the usual number of pulses are
	available with this technique.  In addition, when the same
	number of pulses is used, the analysis runs up to three times
	faster.  Free-space symmetric mode should be used whenever
	possible to speed computation.  (Simple, single-wire models with
	less than 90 pulses run faster using regular analysis mode.)

	        In free-space symmetric mode, AO mirrors the antenna
	horizontally about the X-Z plane.  Wires or portions of wires in
	the -Y half-space are ignored.  A mirror-image of each wire and
	wire section in the +Y half-space is substituted.  The direction
	of instantaneous current flow in image wires is set by mirror-
	imaging the flow, then reversing it.  For example, the image of
	a horizontal wire parallel to the Y axis will have in-phase
	current, while the image of a vertical wire parallel to the Z
	axis will have out-of-phase current.

	        The complete antenna often is defined in free-space
	symmetric files even though everything in the -Y half-space will
	be ignored.  This makes it simple to modify the file later for
	analysis over ground.  However, you can save some typing by
	entering just the +Y half for large, symmetric models.

	        When n pulses have been allocated for half of a free-
	space symmetric antenna, n more effectively are being utilized
	by the antenna image.  This yields a total of 2n pulses for the
	complete model (except for pulses located in the image plane;
	these pulses are never imaged).

	        To invoke free-space symmetric mode, just add the
	keyword "symmetric" after "free space" on the second line of the
	antenna file.  (You don't have to remember how many ms are in
	symmetric; AO looks only for "sym".)  If the file identifies
	sources and loads by specifying wire centers or ends, nothing
	else needs to be changed.

	        However, if sources or loads are identified by pulse
	number, the numbers must be changed.  Pulse allocation is
	different for free-space symmetric models since only half of the
	antenna is actually modeled.  Examine the RUN file to determine
	new pulse numbers.  The notation "Image" in the Wires and Pulses
	list indicates the connection of a wire to its symmetric mirror-
	image at the Y = 0 image plane, where a pulse is always placed.
	If necessary, AO increments the number of segments for a wire by
	1 to fabricate a pulse at Y = 0.

	        Sources and loads specified at -Y coordinates are
	ignored, while those with +Y coordinates are mirrored into the
	-Y half-plane.  Sources and loads located at the Y = 0 image
	plane are modeled normally.  If a wire lies in the image plane
	of a free-space symmetric model, an error results.


				       10

	---- GROUND ----------------------------------------------------

	        AO models an antenna over ground whenever the second
	line of the antenna file begins with anything other than "free
	space".  You can put a short description of ground
	characteristics on this line to cause modeling over ground and
	to annotate the file at the same time.

	        You can use simple default ground characteristics or you
	can define an elaborate ground model in the antenna file.  You
	can define up to ten concentric ground zones, each with its own
	dielectric constant, conductivity, radius, and height.  Ground
	zones can be used to model antennas on hilltops (steep drop-offs
	only), in swamps or marshes, or mounted on vessels above the
	water line.  Ground zones are centered at the origin.  The outer
	zone extends to infinity.  A perfect-conductivity groundplane
	extending to infinity can be defined as well.

	        You can model a ground screen composed of radial wires.
	When you specify a ground screen and one ground zone, soil
	characteristics are assumed to be uniform.  For two or more
	zones, zone 1 defines the soil under the screen and zone 2
	begins where the radials end.  This model is useful when the
	soil under the ground screen has been chemically treated to
	increase conductivity and exhibits different properties from the
	earth beyond.  The effective RF impedance of a ground screen is
	combined with that of the earth beneath.  This impedance is used
	to calculate the ground-reflection factor for the far-field
	tabulations and patterns.  The ground-screen model has high
	accuracy only for dense screens employing more than 100 radial
	wires.

	        The height of the first ground zone is always zero.  You
	can use the height of additional ground zones to roughly
	approximate elevation variation in local terrain.  However, this
	technique will not model sloping ground accurately, even if many
	stairstep ground zones are used.  Moreover, when the height
	drops from one zone to the next, diffraction from the cliff-edge
	formed is not modeled.  When the height increases, blockage of
	radiation from the wall formed is not modeled.

	        The antenna need not be located at the center of the
	ground zones.  The ground zones are concentric with the origin
	but you can place the antenna anywhere.  You might do this to
	model reflection from a small lake near the antenna, for
	example.

	        You should exercise caution when modeling antennas over
	ground.  AO uses ground characteristics only to determine the
	ground-reflection factor for the far-field tabulations and
	patterns.  It uses a perfect-conductivity groundplane when it
	calculates wire currents.  This implies that ground-current
	losses are not accounted for.  While these losses normally are
	negligible for horizontal antennas higher than about 0.2
	wavelength, they can be significant for low horizontals and for
	verticals fed against poor ground systems.  In these cases, the
				       11

	calculated gain will be too high and the calculated impedance
	too low.  You can add a resistive load at the feedpoint to
	simulate ground-current losses.

	        Often the most suitable reference for an antenna over
	ground is not a dipole or isotropic radiator in free space, but
	a dipole, monopole, or single array element substituted for the
	antenna under analysis.  The reference antenna should be modeled
	with the same frequency, polarization, and ground conditions as
	the antenna under study.

	        Keep in mind that dBd and dBi gain figures refer to
	free-space reference antennas, not to reference antennas in the
	same environment as the antenna under analysis.  At small
	elevation angles, expect to see negative gain values over ground
	for antennas that exhibit gain in free space.  In particular,
	vertical antennas have very little response at elevation angles
	near the horizon unless they radiate over salt water.  The same
	is true for horizontal antennas over all types of ground unless
	they are very high.  Conversely, expect up to 6 dB gain over
	free-space values at lobe maxima in the elevation plane.  Gain
	occurs where the ground-reflection wave adds constructively with
	the direct wave.

	        It takes up to three times longer to analyze an antenna
	over ground as in free space.  When optimizing a design, it's
	quicker to do preliminary runs in free space and leave final
	optimization over ground for last.

	        Here's an example of an antenna file using a simple
	ground model:

	Quarter-Wave Vertical
	Ground Mounted
	7.15 MHz
	1 wire, feet
	1   0 0 0   0 0 33   1"
	1 source
	Wire 1, End1

	        When the second line of an antenna file does not begin
	with "free space", AO models the antenna over ground.  In this
	example, no additional ground information follows the second
	line, so a single ground zone with default ground
	characteristics is used.  Default characteristics can be
	predefined using a DOS SET command as follows:

	                          SET GND=D C

	where D = dielectric constant and C = conductivity in mS/m.  AO
	uses average ground constants (D = 13 and C = 5) if GND has not
	been defined.  To define a perfect-conductivity groundplane as
	the default, use SET GND=0.

	        Here's an example of an elaborate ground model involving
	several ground zones:
				       12

	Quarter-Wave Vertical
	Ground Mounted
	3 zones, 120 radials #18  ; # of zones, # of radials, wire gauge
	13 5    0 100   ; Zone #1 diel. const., conduct., height, radius
	17 7  -50 500   ; Zone #2
	19 2 -200       ; Last zone goes to infinity so no radius needed
	7.15 MHz
	1 wire, feet  ; These units apply to the ground dimensions above
	1   0 0 0   0 0 33   1"
	1 source
	Wire 1, End1

	        You can specify up to 10 zones.  The height of the first
	zone is always 0.  In this example, 100 feet from the origin the
	ground drops 50 feet.  It then drops an additional 150 feet at a
	500-foot radius.  When you specify a single ground zone, you can
	omit height and radius.

	        The radial specification is optional.  You can omit
	everything after the word "zones" (which can be singular or
	plural) when radials are not modeled.  Radial length is equal to
	the radius of the first ground zone.  When you specify radials
	and a single ground zone, the zone radius value determines
	radial length and the zone itself extends to infinity.

	        To specify a perfect-conductivity groundplane, the zone
	line should read "0 zones" and there should be no additional
	ground lines.

	        Here's a table of U.S. ground constants:


	                 Earth                   Diel  Cond    Ground
	                Surface                 Const  mS/m   Quality
	  -------------------------------------- -----  ----  ---------
	  Salt water                               81   5000  Excellent
	  Fresh water                              80    1
	  Pastoral, low hills, rich soil           20   30    Very Good
	    (Dallas TX and Lincoln NE)
	  Pastoral, low hills, rich soil           14   10
	    (OH and IL)
	  Flat country, marshy, densely wooded     12    7.5
	    (LA near Mississippi River)
	  Pastoral, medium hills and forestation   13    6
	    (MD, PA, NY, except mtns & seacoast)
	  Pastoral, medium hills and forestation,  13    5    Average
	    heavy clay soil (central VA)
	  Rocky soil, steep hills (New England)    14    2    Poor
	  Sandy, dry, flat (coastal)               10    2
	  Urban and industrial areas (average)      5    1    Very Poor
	  Urban and industrial areas (worst)        3    0.1  Extr Poor


	        A detailed map of U.S. ground conductivities appears in
	recent editions of the ARRL Antenna Book on page 3-3.  As the
	map indicates, ground conductivity has much greater local
				       13

	variation than the simplified table above suggests.  Ground
	constants don't affect horizontally polarized antennas much
	(except for high-angle radiation and the impedance/losses of
	very low antennas), but they critically affect the performance
	of vertical arrays.


	---- WIRE-CONDUCTIVITY LOSSES ----------------------------------

	        AO models losses due to wire conductivity and skin-
	effect.  You can specify conductor material on the number-of-
	wires line like this:

	        1 copper wire, inches

	        AO recognizes the following materials:

	        Silver          1.59E-08      Pure silver
	        Copper          1.7241E-08    100% IACS
	        Aluminum        2.655E-08     Pure aluminum
	        6063-T832       3.25E-08      Aluminum alloy
	        6061-T6         4.01E-08      Aluminum alloy
	        Brass           6.4E-08       Yellow brass (35% zinc)
	        Phosphor Bronze 1.1E-07       Use two words (5% tin)
	        Steel           7.2E-07       Stainless steel type 302

	        Volume resistivity values in ohm-m are given here for
	reference.  You can specify resistivity or conductivity directly
	as follows:

	        1 resistivity 5.7E-08 wire, inches
	        1 conductivity 53 wire, inches

	        Conductivity is specified in IACS units.  IACS stands
	for International Annealed Copper Standard.  A material's IACS
	value is its conductivity expressed as a percentage of that of
	annealed copper.  Pure copper is 103.06 IACS.

	        For ferromagnetic material like iron or steel, you can
	specify permeability as follows:

	        1 resistivity 5.7E-08 permeability 1.5 wire, inches
	        1 conductivity 53 permeability 2.3 wire, inches

	        You can abbreviate resistivity and permeability to res
	and per.

	        AO models all antenna wires using the wire material
	specified on the number-of-wires line.  To specify a different
	material for a particular wire, append a wire-material phrase to
	the wire line after wire diameter.

	        When present, AO displays wire losses in dB.  Load
	losses also are displayed in dB.  When both kinds of losses are
	present, AO displays the total loss in dB and the individual
	losses in percent of input power.  AO also displays antenna
				       14

	efficiency in percent.  This figure includes the effects of both
	wire and load losses.


	---- STANDING-WAVE RATIO ---------------------------------------

	        AO calculates SWR in two different ways.  The first way
	gives SWR in a 50-ohm feedline attached to the antenna
	feedpoint.  This SWR value is useful for single-frequency
	analysis (use the Options Menu to change the 50-ohm default
	value).  The alternate SWR calculation uses the complex
	conjugate of input impedance at one frequency as the SWR
	reference impedance.  This calculation reveals SWR variation
	with frequency once the antenna has been perfectly matched at a
	particular frequency.

	        The alternate SWR value is displayed during automatic
	frequency sweep (it's also displayed during regular analysis
	after the frequency has been changed).  This SWR value results
	from a perfect match to the antenna input impedance at the
	middle of the sweep range using an idealized matching network
	(for unswept analysis, the match is done at the first frequency
	analyzed after loading the file or after frequency sweep).

	        The matching network is modeled as a fixed series
	reactance followed by a broadband transformer.  The series
	reactance cancels input reactance, while the transformer matches
	input resistance to the feedline impedance.  The matching
	network is idealized:  The series reactance and transformer
	impedance ratio do not vary with frequency.

	        The alternate SWR value gives the inherent SWR-bandwidth
	properties of the antenna independent of any particular
	narrowband matching network.

	        To keep things simple, SWR is not computed for antennas
	with multiple sources.


	---- SYMBOLIC DIMENSIONS ---------------------------------------

	        Regular antenna structures often have identical
	coordinate values for several wires.  Here's an example:

	2-Element Cubical Quad
	Free Space Symmetric
	24.94 MHz
	8 wires, inches
	1   -35.5 -64.75 -64.75   -35.5  64.75 -64.75   #12
	1   -35.5 -64.75 -64.75   -35.5 -64.75  64.75   #12
	1   -35.5 -64.75  64.75   -35.5  64.75  64.75   #12
	1   -35.5  64.75  64.75   -35.5  64.75 -64.75   #12
	1    35.5 -61.25 -61.25    35.5  61.25 -61.25   #12
	1    35.5 -61.25 -61.25    35.5 -61.25  61.25   #12
	1    35.5 -61.25  61.25    35.5  61.25  61.25   #12
	1    35.5  61.25  61.25    35.5  61.25 -61.25   #12
				       15
	1 source
	Wire 5, center

	        Unless your text editor provides a search-and-replace
	function, it can be tedious to make changes to such a file.  AO
	provides a simple solution to this problem with symbolic
	dimensions:

	2-Element Cubical Quad
	Free Space Symmetric
	24.94 MHz
	8 wires, inches
	seg = 1
	sp = 35.5
	ref = 64.75
	de = 61.25
	diam = #12
	seg   -sp -ref -ref   -sp  ref -ref    diam
	seg   -sp -ref -ref   -sp -ref  ref    diam
	seg   -sp -ref  ref   -sp  ref  ref    diam
	seg   -sp  ref  ref   -sp  ref -ref    diam
	seg    sp  -de  -de    sp   de  -de    diam
	seg    sp  -de  -de    sp  -de   de    diam
	seg    sp  -de   de    sp   de   de    diam
	seg    sp   de   de    sp   de  -de    diam
	1 source
	Wire 5, center

	        Here, the symbols seg, sp, ref, de, and diam are defined
	and then used in place of actual dimensions.  Geometry changes
	become easy.  For example, to change the reflector size, simply
	edit the single line that defines ref.

	        Symbol names cannot begin with a digit but they can be
	any length.  You can define up to 50 symbols.  Wire symbols are
	defined after the number-of-wires line and before the first
	wire.  You can use symbols for sources and loads, too.  Define
	these symbols just after the number-of-sources and number-of-
	loads lines (don't include load units when you define load
	symbols; put the units on the load line).  It's not necessary to
	use a symbol for every dimension as shown above, nor must you
	leave spaces around the =.  Any symbol can be preceded by a
	minus sign.  If you attempt to use an undefined symbol, AO
	displays an error message.

	        The optimizer requires that symbols be used.

	        NOTE:  Unlike everything else in AO, symbol names are
	case-dependent.  The symbol LENGTH is not the same as the
	symbols Length or length.







				       16

	---- SYMBOLIC EXPRESSIONS --------------------------------------

	        For certain antenna geometries, it's very handy to use
	expressions in symbol definitions.  For example, you might do
	something simple like this:

	                             w = x + 5

	        You might also do something complicated like this:

	                     w = (x + y / (z - 17')) * 23

	        Symbols like x, y, and z must have been previously
	defined.  You can use dimension overrides (like 17') with
	numbers, but not with symbols (z' is not allowed).  Unless you
	use parentheses, * and / are done before + and -.  You can
	separate expression items with zero or more spaces or tabs, but
	don't use commas within expressions.  Numbers can use scientific
	notation (12300 can be represented as 1.23E+04, 1.23e4, etc.,
	while .000123 is 1.23E-04, 1.23e-4, etc.).  Unary plus and minus
	are supported.  Items like #12 indicate wire gauge.

	        In addition to the + - * / operators, AO recognizes
	sine, cosine, and square root functions.  Sines and cosines are
	very handy for defining certain antenna geometries.  For
	example, here's the geometry definition for an 80-meter, three-
	square array:

	                x = 40'
	                a = SIN(30) * x
	                b = COS(30) * x
	                1    x  0 0    x  0 61'   1"
	                1   -a -b 0   -a -b 61'   1"
	                1   -a  b 0   -a  b 61'   1"

	        Here, AO computes the sine and cosine of 30 degrees, a
	constant.  You can also use an expression as a function
	argument, for example, SIN(x + y).  You can use upper or lower
	case for function names.  Use the SQR operator for square roots.

	        Expressions are very useful when optimizing a design.
	They can be used to maintain antenna geometry while allowing a
	single independent variable to determine structure size.

	        Expressions can't be used in place of symbols on wire
	lines; they're allowed only in symbol definitions.  However, you
	can precede any symbol with a minus sign on wire lines, as
	illustrated above.


	---- SHIFT AND ROTATE ------------------------------------------

	        You can shift and rotate an object comprised of one or
	more wires.  You can use symbols for the shift and rotate
	amounts.

				       17

	        To shift an object, put a shift command immediately
	before the wires that define the object.  A shift command looks
	like this:

	                           Shift X 30

	        X can be the letter X, Y, or Z and indicates the axis
	along which the shift occurs.  Here, the shift amount of 30 is
	in the units defined on the number-of-wires line.  You can
	append dimension overrides (' " ft in mm cm m) to use different
	units.  You cannot use an expression for the shift amount; it
	must be a constant or a symbol (however, a symbol can be
	preceded by a minus sign).  You can use upper or lower case, and
	you can use spaces, tabs, or commas to separate line items.

	        To designate the end of a shifted object, put this
	command immediately after the last wire in the object:

	                           Shift End

	        All wires between the shift and shift-end lines will be
	shifted.  Often you won't need a shift-end command.  If no wires
	follow the object to be shifted, you can omit the shift-end
	command.

	        To shift an object along more than one axis, use
	multiple shift lines or simply add more items to a single shift
	line like this:

	                    Shift X 30, Y -18, Z 55

	        X, Y, and Z can appear in any order on a shift line.
	The commas are optional; they're used here just to improve
	readability.  Shift End terminates all shifts (this means that
	shifts cannot be nested).

	        To rotate an object, do this:

	                          Rotate X 30

	        This rotates the object 30 degrees about the X axis.
	Positive rotation about the X axis is defined as +Z rotated
	toward +Y.  For Y, it's +X toward +Z, and for Z it's +Y toward
	+X.  Positive rotation is the prior axis rotated toward the next
	axis in the circular sequence XYZXYZ....  If you can't remember
	this convention, just try a positive number.  If the object
	rotates the wrong way, add a minus sign.

	        To rotate an object about more than one axis, use
	multiple rotate lines or simply add more items to a single
	rotate line.  Unlike shifts, the order of rotates affects final
	geometry.  Rotates are executed in reverse file order; the




				       18

	rotate command nearest the object wires is done first.  If you
	use multiple rotates on one line, the last one is done first.

	        To define the end of a rotated object, use this command:

	                           Rotate End

	        Rotate end terminates all rotates (this means that
	rotates cannot be nested).

	        You can mix shifts and rotates.  The commands are
	executed in reverse order of file appearance.  Order matters.
	Usually it's most convenient to define the antenna at the
	origin, rotate it there, and then shift it.

	        To rotate an object about an axis and then rotate
	another object a different amount about the same axis, you don't
	need an intermediate rotate-end command.  For example, here's
	how to define an inverted-V dipole with a 120-degree apex angle
	and an apex height of 60 feet:

	                Inverted V
	                Over Ground
	                7.15 MHz
	                2 wires, feet
	                Shift Z 60
	                Rotate X 30
	                1   0 0 0   0  33 0   #12
	                Rotate X -30
	                1   0 0 0   0 -33 0   #12
	                1 source
	                Wire 1, end1

	        The second rotate just modifies the rotation value
	already in effect; no rotate end is needed.

	        Shifts and rotates let you define antennas with oblique
	wires using actual wire lengths.  You don't need to use
	trigonometry to calculate the coordinates of wire ends.


	---- POLARIZATION COMPONENTS -----------------------------------

	        AO normally obtains results using the total
	electromagnetic field.  This is equivalent to using the
	polarization component (possibly elliptical) that maximizes
	electromagnetic-field magnitude at each azimuth and elevation
	angle.  The total field provides results that are independent of
	wave polarization.  Since polarization is variable for
	ionospherically propagated signals, the total field provides a
	realistic way to characterize HF-antenna performance.

	        However, sometimes it's important to isolate a
	particular polarization component.  You can do this with the
	Options Menu by specifying one of the following:

				       19

	        Total         Total Field
	        H             Horizontal Component
	        V             Vertical Component
	        RC            Right-Circular Component
	        LC            Left-Circular Component
	        Major         Major-Axis Component
	        Minor         Minor-Axis Component
	        Ellipticity   Polarization-Ellipse Circularity

	        At each azimuth and elevation angle, the major-axis
	component is the linear-field magnitude at the polarization tilt
	that maximizes the field.  The minor-axis component is the
	linear-field magnitude at the polarization tilt that minimizes
	the field.  The minor-axis component is always zero for linearly
	polarized antennas.

	        AO treats ellipticity as if it were a field component,
	but it's really the ratio of minor- to major-axis components
	expressed in dB.  Ellipticity is -infinity dB at all angles for
	linearly polarized antennas and 0 dB at angles where
	elliptically polarized designs have perfect circularity.

	        When you select a field component, all analysis results,
	including beamwidths and sidelobe levels, refer to that
	component.

	        AO saves all field components internally during
	analysis.  You can plot any component without recalculation of
	the mutual-impedance matrix.  To compare 2-D polarization
	patterns, use the upper-case P plot command to generate a new
	plot file after selecting a different polarization component.
	Then use the overlay or compare commands while viewing the plot.


	---- NEAR FIELD ------------------------------------------------

	        AO can calculate the electric and magnetic fields
	produced by an antenna at any point in space.  The field close
	to the antenna is called the near field.  This field is quite
	different from the far field that characterizes propagated
	electromagnetic waves.  Near-field intensities are useful for
	EMI studies, TVI minimization, and RF-hazard analysis.

	        AO computes the near field over a 3-D grid of points.
	Enter starting, ending, and step (increment) values for the
	coordinates to be varied.  Set step to 0 for coordinates that do
	not vary.  (Although the peak-field value appears on the
	Z-component line, it applies to the entire field vector.)

	        When AO calculates the near field over ground, it uses
	perfect-conductivity ground regardless of the ground
	characteristics defined in the antenna file.




				       20

	---- FAR FIELD -------------------------------------------------

	        The Far Field command tabulates absolute or relative
	far-field response in the RUN file.  For the absolute far field,
	AO tabulates the real and imaginary parts of the horizontal and
	vertical field components in volts/meter.  (At nonzero elevation
	angles, the "vertical" component is the component upward-
	perpendicular to a line to the antenna.)  You specify distance
	from the antenna and input power for the absolute far-field.
	For the relative far field, AO tabulates the magnitudes of all
	polarization components in dBi or dBd; distance and power are
	not relevant.

	        The absolute far field is useful for signal-coverage
	studies and interference-level analysis.  (NOTE:  AO calculates
	only the space wave and the ground-reflection wave; it does not
	calculate the surface wave.)  The relative far field can be used
	to obtain response at angular resolutions not available in the
	far-field plots.  It also provides a convenient way to output
	pattern data to other programs.

	        Relative far-field values less than -120 dB appear in
	the tabulation as -infinity.  This notation suppresses low-level
	computational noise and makes the listing easier to read.  See
	the SET Commands section to change the threshold value.


	---- 3-D GEOMETRY/CURRENT DISPLAY ------------------------------

	        After loading or editing an antenna file, press V to
	view the antenna geometry in 3-D to check for model errors.
	Press F1 or any nonfunctional key for a list of viewing
	commands.  You can enable or disable display of segments,
	sources, loads, wire currents, axes, and annotation.  The
	spacebar switches between two viewpoints that can have different
	scale and viewing perspective.  You can use one viewpoint for an
	overall view of the antenna and the other for close inspection.

	        Wire currents are displayed after impedance is
	calculated.  You can interrupt pattern generation to view the
	currents.  Pattern generation resumes without recalculating
	prior points.

	        AO graphs wire currents directly on the wires in the 3-D
	geometry display.  You can display currents either as phasors or
	using magnitude only.  For phasors, the distance from a wire to
	the current trace represents magnitude, while the rotation angle
	of the trace around the wire represents phase.  Wires with
	traveling-wave currents exhibit non-planar current traces that
	become spirals for pure traveling waves.  Use the W key to sight
	down a wire axis to examine the circularity of a spiral.  You
	can toggle between phasor currents and magnitude-only currents
	with the M key.  Use magnitude-only to visualize current
	distribution in complex antennas.


				       21

	        Note that current phase changes 180 degrees when you
	interchange wire ends in the antenna file.  This ambiguity
	affects both tabulated currents and current traces.  If you
	define parallel wires consistently, always heading in the same
	direction from End1 to End2, you can avoid this ambiguity.


	---- 3-D PATTERN DISPLAY ---------------------------------------

	        Press T from the Main Menu or geometry display to
	display 3-D polar radiation patterns.  Press F1 or any
	nonfunctional key for a menu.  The viewing commands are like
	those in the geometry display.

	        Press R to change angular resolution from the default of
	2 degrees.  You can select 2 to 6 degrees for any model and 1
	degree for models with X-axis pattern symmetry.  Be sure to take
	advantage of X-axis pattern symmetry whenever possible; it can
	greatly reduce pattern-calculation time.  Resolution refers to
	the elevation-angle increment.  Azimuth-angle resolution varies
	automatically with elevation angle (it's the same as elevation-
	angle resolution at 0 degrees elevation).  AO saves the current
	3-D resolution in AO.INI whenever you save colors or select Save
	Options & Settings from the Options Menu.

	        Press S to slice the pattern at a selected elevation
	angle.  This is useful for examining the inside wall of complex
	patterns.

	        Press T to thin a pattern.  This opens up the mass of
	white at the center of dense plots.  It also speeds pattern
	rotation.  This function works just like a barber's thinning
	shears.  As the plot surface contracts, lines are selectively
	deleted to reduce pixel density.  Experiment with the thinning
	threshold.  Often there's one value that's just right for best
	plot appearance.  Thinning a dense plot is particularly useful
	before printing it.

	        Press V to bring up the geometry display.  You can press
	V and T to alternate between the two displays.

	        As in the geometry display, you can use the spacebar to
	switch between two viewpoints.  The viewpoints can have
	independent perspective and scale.  Two viewpoints are very
	useful with 3-D patterns.  For example, you can zoom in with one
	viewpoint to examine close pattern detail and switch back and
	forth with a second viewpoint showing the overall pattern.

	        Use the Home and End keys to change the center of
	rotation for patterns with high F/B.  This lets you rotate large
	patterns without clipping.




				       22


	---- 2-D PATTERN GENERATION ------------------------------------

	        AO can generate 2-D azimuth and elevation radiation
	patterns.  It saves the pattern data in files with the extension
	".PLT".  AO searches the data to find beamwidth and maximum
	sidelobe level.

	        By default, AO assumes that antenna patterns are
	symmetrical about the X-axis.  This allows it to generate
	patterns from 0 to 180 degrees and then mirror-image this data
	into the remaining half-plane.  This cuts pattern-calculation
	time in half.  For antenna patterns that aren't symmetrical
	about the X-axis, use the Options Menu to disable this option to
	obtain correct patterns.  Always align antennas with symmetrical
	patterns with the X-axis to take advantage of this time-saving
	feature.

	        With the Options Menu you can specify the elevation
	angle to use during azimuth-pattern generation and the azimuth
	angle for elevation patterns.  For free-space models, the
	elevation angle for azimuth patterns defaults to 0 degrees to
	obtain on-axis response.

	        You can generate 2-D patterns with 1-, 2-, or 4-degree
	resolution.  4-degree resolution can speed pattern-generation
	for large models.  1-degree resolution is useful for complex
	patterns.  Use the Options Menu to change 2-D pattern
	resolution.

	        When an azimuth or elevation search finds maximum
	response at 0 degrees, AO computes the 3-dB pattern beamwidth.
	AO uses interpolation to obtain accurate beamwidth values even
	for 4-degree patterns.

	        When AO finds a sidelobe with amplitude larger than that
	of the rear lobe, it displays its level and angle.  This is
	useful when manually optimizing an antenna pattern.  It's easy
	to get carried away and lose sight of the overall pattern when
	optimizing F/B, a measurement that involves just a single point
	to the rear of the antenna.  You might achieve 45 dB F/B but
	have a sidelobe just 15 dB down at 135 degrees.  The sidelobe
	display will alert you to this problem.  (The automatic
	optimizer uses a more generalized definition of F/B to avoid
	this difficulty.)

	        When the main lobe occurs at an angle other than 0
	degrees, AO does not display beamwidth or sidelobe information.
	Instead, it displays the angle of the lobe and its gain.


	---- 2-D PATTERN PLOTS -----------------------------------------

	        You can view 2-D radiation patterns of the current
	antenna or you can review plots made previously.  To review a
	plot, start AO by typing AO PLOT.  AO will list plot files in
	the current directory and you can select one.  Alternatively,

				       23

	you can specify a plot file on the command line, for example, AO
	PLOT DIPOLE.  This displays the plot immediately.

	        AO normally begins by drawing a polar plot of the
	azimuth pattern using the log-dB scale (the linear-dB scale is
	used when set with the POLAR environment variable).  You can
	start differently by entering up to three option letters after
	the filename as follows:

	                E       Elevation pattern
	                R       Rectangular plot
	                L       Linear-dB scale

	        For example, AO PLOT DIPOLE ERL will cause AO to draw
	the elevation pattern in rectangular coordinates.  If you later
	select a polar plot, it will use the linear-dB scale.  You can
	use upper or lower case for option letters.

	        While viewing a plot, press F1 or any nonfunctional key
	for help.  The help panel lists only currently active keys.  For
	example, the X and Y keys are listed only in rectangular mode.

	        The A key turns off overlay and compare modes.  This
	lets you examine the patterns of a single antenna more rapidly.
	The O and C keys can be used repeatedly to compare multiple
	antennas, two at a time.


	Polar Plots

	        AO displays azimuth polar plots with the +X direction to
	the top of the screen and +Y to the left.  Elevation plots have
	+Z to the top and +X to the right.

	        AO provides two radial scales for polar plots.  The ARRL
	log-dB scale causes lower-level sidelobes to be compressed
	toward the center of the pattern.  This emphasizes major-lobe
	shape.  The ARRL log-dB scale is widely used in amateur
	publications.  It provides a convenient scale to compare the
	patterns of antennas you develop with those of existing designs.
	It also yields patterns with familiar shapes.  The center of a
	log-db plot is -infinity dB, but there isn't much area below -40
	dB.

	        The other polar scale uses linear dB.  This scale cuts
	off at -50 dB at the center of the plot.  It provides much more
	area between -20 and -50 dB than the log-dB scale.  It's useful
	for examining low-level sidelobes that may be hard to see in a
	log-dB plot.

	        The dots forming the radial lines (the ones that are not
	multiples of 30 degrees) are spaced every 2 dB.  The dots
	forming the circles are spaced at multiples of 1 degree.  These
	calibrations allow directivity values to be read with good
	accuracy.

				       24

	        AO draws a 360-degree elevation polar plot for free-
	space patterns.  For patterns over ground, just the upper 180-
	degree hemisphere is shown.

	        The polar plots are perfectly circular on monitors with
	standard 4:3 aspect ratio.  If the plots appear elliptical,
	adjust your monitor's vertical height control.


	Rectangular Plots

	        Use the R key to select a rectangular plot.  This
	coordinate system can resolve small sidelobe detail even better
	than a linear-dB polar plot, but the overall pattern shape is
	not quite as apparent.  The X axis is azimuth or elevation
	angle, and the Y axis is antenna response in dB on a linear
	scale.  Use the Y key to change the Y-axis cutoff (lower limit
	in dB).  This parameter is always a negative number, but you can
	enter it without a minus sign for convenience.  The Y-axis
	cutoff is constrained between -1 and -100 dB.  The Y axis is
	easiest to interpret when the cutoff value is a multiple of 10
	degrees, but AO draws the scale and grid correctly for any
	value.  Use the X key to change the X-axis cutoff.  AO rounds
	the value entered to the nearest multiple of 10 degrees.


	Overlaying Plots

	        You can overlay a second pattern with the O key.  You
	can distinguish the traces and annotations of the two patterns
	by their individual colors.  AO substitutes filenames for file
	titles when overlaying plots.


	Comparing Plots

	        The C key selects a second plot file for pattern
	comparison.  The plot is drawn into a separate screen buffer
	(unless only one VGA display page is used).  The screens can be
	switched instantly with a single keystroke.  This permits a
	precise instantaneous pattern comparison, and can reveal
	differences that may be lost in screen clutter in overlay mode.


	Generating Plot Data

	        The p or P commands cause AO to generate and display a
	plot file. AO saves the patterns in a file so that you can
	review and compare plots later.  A plot file uses the antenna
	filename with the extension .PLT.  If you make changes to a
	design and press lower-case p, the plot file is overwritten with
	new patterns and displayed.  If you press upper-case P, AO
	generates a new plot file.  Each new plot file has an
	incrementing digit appended to its name.  Use of new filenames
	prevents previous plot files from being overwritten.  This
	preserves plot files for later comparison.
				       25

	        Use the Options Menu to select 1-, 2-, or 4-degree
	pattern resolution.

	        If you just want to view the azimuth response, you need
	not wait for elevation data to be generated.  Press Esc while
	generating elevation data to draw the azimuth plot.  The plot
	file will contain only azimuth data.  Later, if you press p and
	wait for data generation to complete, AO adds the elevation data
	to the plot file and displays the plot.

	        AO annotates the electromagnetic-field component on
	plots when it is different from the usual total field.  AO
	generates plot data with a dynamic range of 100 dB and with a
	resolution of .01 dB.  AO annotates the elevation angle on
	azimuth plots only when nonzero, and the same holds for the
	azimuth angle on elevation plots.


	---- SCREEN COLORS AND DEFAULT MENU SETTINGS -------------------

	        You can customize the screen colors used by AO.  Use the
	C key to change colors in the main screen and the optimizing
	screen.  Press F5 to change colors in other screens.  It's easy
	to experiment with various color schemes.  To make changes
	permanent, use the S command from one of the color menus or
	select Save Options & Settings in the Options Menu.  These
	functions write codes for the current color sets into the
	initialization file AO.INI.  Whenever AO finds AO.INI in the
	current directory on start-up, it sets all colors accordingly.

	        AO.INI also contains codes for most menu settings.  If
	AO.INI is present when you start AO, all menus are initialized
	to the saved AO.INI settings.  To customize AO to your way of
	working, change the menus to your preferred settings and then
	save them with Save Options & Settings.

	        To reset colors and menu settings to the defaults,
	simply delete AO.INI.


	---- PRINTING THE SCREEN ---------------------------------------

	        AO prints text and graphics screens to HP LaserJet/
	DeskJet printers and to Epson-compatible, dot-matrix printers.
	Press PrtSc to print any screen.  See the SET Commands section
	for details on configuring AO for your particular printer.

	        Unless you're in LaserJet landscape mode, AO does not
	eject the page after printing a screen.  This lets you print two
	screens on one page (24-pin prints are too big to allow this).
	You might print two plots for comparison or print a graphics
	screen accompanied by a listing of the antenna file.  To list
	the antenna file, press PrtSc while viewing its image in the RUN
	file.  Alternatively, you can save the current design to PRN or
	LPT with the Save Menu in the optimizer.

				       26

	        To correctly print extended-ASCII characters like the
	degree symbol, set your printer for the PC-8 symbol set.


	---- GRAPHICS IMAGE FILES --------------------------------------

	        AO can save the image of any graphics screen in the .PCX
	file format.  This feature lets you add AO graphics to desktop-
	publishing and word-processing documents.  If your computer has
	a fax/modem, you can fax AO images.

	        Press F9 to generate a .PCX file.  The output filename
	is the antenna filename with the extension .PCX.  Whenever you
	press F9 again, an incrementing digit is appended to the
	filename and another file is created.  You can control image
	size and centering with a DOS SET command.  See the SET Commands
	section for details.


	---- ABORTING CALCULATIONS -------------------------------------

	        You can abort calculation by pressing the Esc key.
	Since the most fundamental results are calculated first, you can
	use the abort feature to terminate unwanted calculations as well
	as to escape from command mistakes.  For example, if you are
	interested only in obtaining input impedance, SWR, gain, and
	F/B, you can press Esc after these results are displayed to
	terminate pattern generation.  If you later press G, AO
	redisplays results already calculated and resumes pattern
	generation where it left off.

	        If it takes more than one minute to fill and factor the
	mutual-impedance matrix, AO beeps when it's done to alert you
	that results are ready.


	---- AUTOMATIC FREQUENCY SWEEP ---------------------------------

	        AO can automatically analyze a design at discrete
	frequency steps across a frequency range.  It can generate a
	plot file at each frequency.  To set this up, use the Frequency
	Menu to enter start, stop, and step values for the sweep range.
	Then press G for analysis only or P for analysis and plot-file
	generation.  When you press P, AO displays the last-frequency
	plot after generating all data.  When you press G, AO does not
	compute analysis parameters that require pattern generation,
	like beamwidth.  This allows very rapid frequency sweep.  AO
	records swept-frequency results in the RUN file.

	        To return to single-frequency mode, set the stop or step
	frequency to zero.




				       27


	---- BATCH MODE ------------------------------------------------

	        Sometimes it's very convenient to run AO unattended from
	a DOS batch file.  For example, you might do lengthy analysis of
	several complex antenna models overnight.

	        In batch mode AO runs and terminates automatically
	without keyboard intervention.  You can do several AO runs from
	one batch file by saving results in separate RUN and plot files.

	        Here's an example of batch mode:  AO BATCH=G DIPOLE.
	Here, AO loads DIPOLE.ANT, runs as if you had pressed G but
	aborted the pattern search, and then terminates.  BATCH=G yields
	the same results as pressing G during automatic frequency sweep:
	AO computes impedance, SWR, losses, efficiency, gain and F/B,
	but not beamwidth or maximum-sidelobe level.  Results are saved
	in the RUN file.  AO BATCH=P DIPOLE runs as if you had pressed
	P:  AO calculates analysis parameters, generates a 2-D plot file
	(but does not display it), and then terminates.  AO BATCH DIPOLE
	is the same as AO BATCH=P DIPOLE; it's the form you'll probably
	use most often.

	        You can execute more than one command per run, for
	example:  AO BATCH=GDN DIPOLE.  Here, AO calculates gain, etc.,
	tabulates the far field and near field, and then terminates.
	Only the G, P, D, and N commands can be used with BATCH=.

	        In batch mode the D and N commands use default menu
	settings for angle limits, field locations, etc.  To change the
	default values, modify the D and N menus before entering batch
	mode.  Then use Save Options & Settings to save these values in
	AO.INI, perhaps renaming it.  You can recopy saved settings to
	AO.INI during a batch run like this:

	        AO BATCH DIPOLE          Use existing settings
	        COPY RUN DIPOLE.RUN      Save dipole results
	        COPY QUAD.INI AO.INI     New settings for quad run
	        AO BATCH=D QUAD          Tabulate far field for quad
	        COPY RUN QUAD.RUN        Save quad results
	        COPY YAGI.INI AO.INI     Settings for Yagi run
	        AO BATCH=ND YAGI         Compute near and far fields

	        To perform automatic frequency sweep with BATCH=G or
	BATCH=P, specify start, stop, and step frequencies on the
	frequency line in the antenna file.  Frequency sweep does not
	work for BATCH=D or BATCH=N.

	        If AO detects an error during batch mode, it records it
	in the RUN file and then terminates so that processing can
	continue.  It also sets ERRORLEVEL to 1.  You can use the IF
	ERRORLEVEL statement in a batch file to take alternate action
	whenever an error occurs.  You can abort the current analysis
	during batch mode by pressing Esc.

	        Due to its complex setup, the optimizer cannot be run
	from batch mode.

				       28

	---- ALGORITHM LIMITATIONS AND CORRECTIONS ---------------------

	        The MININEC antenna-analysis algorithm is reasonably
	accurate when modeling thin, straight-wire antennas at HF.
	However, the algorithm does have inherent biases and limitations
	that can affect accuracy in many circumstances.  AO contains
	special compensation for several sources of error that are
	particularly important.


	Frequency Offset

	        MININEC exhibits an inherent frequency offset for all
	models.  This error is small for very thin wires but increases
	rapidly with wire diameter.  This effect can bias results at HF,
	can cause serious error at VHF, and renders MININEC useless for
	practical work at UHF and above.  Gain, pattern, and impedance
	characteristics of real antennas occur lower in frequency than
	calculated.  Frequency offsets ranging from 0.25% to 2% have
	been observed for Yagi models with element diameters from .0001
	to .01 wavelength.

	        AO includes special compensation for this frequency-
	offset error.  The correction frequency-calibrates AO to NEC,
	the accurate but complex Numerical Electromagnetics Code that
	inspired the simpler MININEC algorithm.  Frequency-offset
	compensation normally should be enabled.  It can be disabled
	from the Options Menu for testing.

	        With the frequency-offset correction enabled, AO gives
	accurate, low-offset results for most Yagis when just 10
	segments per halfwave are used.


	Bent Wires

	        Wires joined at an angle cause an internal geometric
	distortion in MININEC.  This distortion can result in large
	frequency-offset errors unrelated to the frequency-offset effect
	described above.  Bent-wire junction effects can cause serious
	inaccuracy when modeling loops, cubical quads, rhombics,
	groundplanes, etc.  AO contains a special correction for this
	problem that works without altering segmentation or increasing
	computation time.  The correction can be disabled from the
	Options Menu for testing.

	        With the frequency-offset and bent-wire junction
	corrections both enabled, groundplanes and cubical quads with 16
	segments/halfwave and equilateral-triangle delta loops with 22
	segments/halfwave show little or no frequency offset from NEC
	results.  (However, you may need to increase segmentation
	density when a source is located at a bent-wire junction.  It's
	important to test convergence for unfamiliar geometries.)



				       29

	Dissimilar Segment-Lengths at Wire Junctions

	        As described earlier, AO can exhibit gross errors when a
	short segment is connected to a much longer one.  Connected-
	segment lengths normally should differ by no more than a factor
	of about 2.  To overcome this limitation, enable automatic
	segment-length tapering from the Options Menu.


	Ground Losses

	        AO calculates wire currents assuming perfect-
	conductivity earth for antennas modeled over ground.  Horizontal
	antennas located less than about 0.2 wavelength above real earth
	incur ground losses not calculated by AO.  Because of these
	losses, the actual impedance will be higher than calculated and
	the realizable gain lower.  A similar situation occurs for
	ground-mounted vertical antennas with poor ground systems.


	Element-Mounting Effects

	        The method used to mount elements to a boom can affect
	antenna performance, particularly for parasitic arrays at VHF/
	UHF.  Conductive mounting plates or through-the-boom mounting
	increase element effective diameter and raise resonant
	frequency.

	        To model conductive mounting plates, use the YO Yagi
	Optimizer program or refer to "Physical Design of Yagi Antennas"
	by Dr. David B. Leeson, W6QHS, published by the ARRL.

	        To account for through-the-boom mounting of Yagi
	elements, use the formula below to shorten measured element
	lengths before modeling.  Conversely, when constructing a Yagi,
	use the formula to increase calculated element lengths.

	        For noninsulated through-the-boom mounting, measurements
	by Guenter Hoch, DL6WU, have been curve-fitted by Ian White,
	G3SEK, to yield the following boom-correction formula for Yagis:

	                      C = 25.195B - 229B^2

	        C is the correction factor expressed as a fraction of
	boom diameter B in wavelengths.  B^2 means B-squared.  For
	example, a .01-wavelength diameter boom requires an element-
	length correction of 23% of the boom diameter.  The experimental
	data underlying this formula were derived from booms with
	diameters less than .055 wavelength.  G3SEK indicates that
	correction factors for insulated, through-the-boom mounting are
	close to 50% of C.





				       30

	Closely Spaced Wires

	        With some care, you can model antennas with closely
	spaced wires like embedded transmission lines.  Higher-than-
	normal segmentation density may be required for these models.

	        Experiments with a terminated, 600-ohm transmission-line
	model and with a gamma-match model yielded accurate, converged
	results only when segment lengths of the closely spaced wires
	were no more than twice the wire spacing.  This segmentation was
	required even though automatic segment-length tapering was used
	to smoothly transition the segmentation to the rest of the
	structure.

	        On the other hand, Zepp and Lazy-H models with embedded
	transmission lines give accurate results even with segments
	considerably longer than twice the line spacing.  LPDA models
	give reasonable results under the same conditions but converge
	better with more segments.

	        Be skeptical of results for any model with closely
	spaced wires until you've carefully checked both algorithm
	convergence and wire currents.  Check convergence by increasing
	the segmentation density and noting changes in the results.
	Look for erratic current magnitude along closely spaced wires.
	This is a sure sign that they have too few segments.  You can
	set a minimum number of segments for closely spaced wires by
	specifying a positive value on the wire lines.  Be sure to
	enable automatic segment-length tapering.


	---- MEMORY ----------------------------------------------------

	        The maximum number of pulses for an antenna model is
	limited by available memory.  The following terms are used in
	this section:

	                Conventional memory     0 to 640KB
	                Upper memory            640KB to 1MB
	                Extended memory         1MB up
	                High Memory Area        1MB to 1MB+64KB

	AO-Amateur

	        The amateur version of AO uses conventional memory.  A
	maximum of 225 pulses is available (450 for free-space symmetric
	models).  The actual number available on your particular system
	depends on the amount of free conventional memory.  To make more
	memory available, you can remove TSRs, device drivers, DOS
	shells, or other items loaded by CONFIG.SYS or AUTOEXEC.BAT.
	You also can free-up conventional memory by using a memory
	manager to load items into upper memory.  AO-Amateur runs at
	full speed when used with a memory manager.



				       31

	AO-Professional

	        The professional version of AO can use either
	conventional memory or extended memory.  To access extended
	memory, an XMS driver must be installed.  The HIMEM.SYS program
	that comes with DOS is such a driver.  XMS allows multiple
	programs to share extended memory without conflict.  For
	example, a disk cache, a RAM disk, and AO-Pro can all use
	extended memory under XMS.  If you've installed HIMEM.SYS to
	load DOS high (in the High Memory Area), it's all set for use
	with AO-Pro.  See your DOS manual for information on installing
	HIMEM.SYS.

	        When XMS is not available, AO-Pro uses conventional
	memory.  AO-Pro also uses conventional memory whenever more of
	it is available than extended memory.  This might occur, for
	example, when a disk cache has been allocated all of extended
	memory.  All caches and RAM disks provide some means to control
	the amount of memory they use.  Often you can dynamically reduce
	their memory usage from the DOS command line without rebooting.
	When running AO-Pro with large models, you may wish to minimize
	other uses of extended memory to maximize model capacity.

	        AO-Pro is not compatible with memory managers that put
	the CPU into virtual-8086 mode.  V86 mode prevents AO-Pro from
	using protected mode to access extended memory.  Examples of
	incompatible memory managers are EMM386, QEMM-386, and 386MAX.
	Windows and OS/2 also run DOS programs in V86 mode.  If AO-Pro
	detects V86 mode on startup, it displays "Incompatible operating
	system or memory manager" and terminates.  You must either
	disable your memory manager or install one that does not use V86
	mode.

	        The Last Byte Memory Manager from Key Software Products
	can load TSRs, device drivers, and many other items into upper
	memory without forcing the CPU into V86 mode.  TLBMM can make
	upper memory available for many (but not all) motherboard chip
	sets.  TLBMM is totally compatible with AO and imposes no speed
	penalty whatsoever.  TLBMM is shareware.  You can verify
	compatibility with your computer before licensing the program.
	You can download TLBMM from the KSP BBS at (415) 364-9847.
	Alternatively, you can obtain the software by sending a blank,
	formatted, high-density disk and a self-addressed, stamped, disk
	mailer to Key Software Products, 440 Ninth Ave., Menlo Park, CA
	94025.  After verifying that it works on your computer, you can
	license the software for $29.95.

	        The following table gives the maximum number of pulses
	for AO-Pro as a function of total memory.  These numbers assume
	that AO-Pro has access to everything above the High Memory Area.
	Double the number of pulses for free-space symmetric models.





				       32

	                  Memory         Pulses

	                   64 MB           2872
	                   32              2013
	                   16              1399
	                    8               953
	                    4               620
	                    2               350


	---- SET COMMANDS ----------------------------------------------

	        DOS provides a convenient way to specify configuration
	information to AO.  The DOS SET command places information you
	supply into the DOS environment in memory where it can be
	retrieved later by a program.  It's convenient to put SET
	commands in your AUTOEXEC.BAT file so they'll be executed
	automatically every time the computer boots.  You can enter SET
	commands in upper or lower case.  It's best not to leave spaces
	around the = in SET commands.

	        You can see what's in the DOS environment by typing the
	following:

	                              SET

	        You can eliminate a SET parameter P by typing this:

	                             SET P=

	        If you use lots of SET commands, it's possible to run
	out of environment space.  To enlarge it to 1024 bytes, put a
	line similar to this one in your CONFIG.SYS file:

	                 SHELL=\COMMAND.COM /E:1024 /P


	1.  Text Editor

	        AO is set up to automatically call the ED program
	whenever the E command is given to edit an antenna file.  This
	simple, easy-to-use text editor works with small files like
	antenna files.  However, you may prefer to use your own editor
	or wordprocessor.  To identify a different editor to AO, do the
	following:

	                     SET EDITOR=Editorname

	        The EXE or COM extension is not needed.  You can specify
	a drive and path.  When the E command is given, AO appends the
	name of the current antenna file to Editorname and calls the
	program.  After the file has been edited, AO rereads it.  All
	antenna parameters are reset to those in the edited file.  The
	RUN file is overwritten.


				       33

	        Many editors permit a line number to be specified along
	with the filename.  The editor will initialize its cursor at the
	specified line.  When AO detects an antenna-file error, it can
	pass the offending line number to your editor.  This is very
	handy for quick fixes.  AO automatically passes a line number to
	ED.  To set this feature up for a different editor, specify the
	required string in Editorname.  Put # where your editor expects
	the line number.  AO will substitute the actual line number for
	# when it calls the editor.  For example, to invoke the Norton
	Editor this way, use:

	                        SET EDITOR=ne +#

	        For BRIEF use:

	                  SET EDITOR=b -m"goto_line #"

	        If you use a wordprocessor for editing, be sure to
	generate a standard DOS text file instead of a file in
	wordprocessing format.  Files in wordprocessing format contain
	embedded control codes that will confuse AO.


	2. Subdirectories

	        Once you accumulate many antenna and plot files, it's
	nice to organize them into subdirectories.  You might use the
	current directory for antenna experiments, saving optimized
	antenna files and plots elsewhere.  You can specify
	subdirectories to AO with SET commands.  The subdirectories are
	used only for reading files; AO always writes the RUN file,
	OUT.ANT (from the optimizer), and plot files to the current
	directory.

	        Define subdirectories like this:

	                   SET ANT=BEAMS LOOPS VHF .
	                   SET PLT=. HFPLOTS VHFPLOTS BEAMS

	        The dot represents the current directory.  It can appear
	anywhere (or nowhere) in the list.  You can define as many
	subdirectories as you like.  Often it's convenient to use the
	same set of subdirectories for both antenna and plot files.  You
	can organize subdirectories by antenna type, frequency range,
	designer, etc.

	        Without a SET command, each subdirectory list defaults
	to just the current directory.

	        When you start AO, it begins by listing files in the
	first subdirectory on the list.  The remaining subdirectory
	names are listed along with the files.  You can select one to
	list its files.



				       34

	        You can specify a file from one of the subdirectories on
	the AO command line without typing the path.  AO searches the
	subdirectories in the order of the subdirectory list.

	        You can specify any subdirectory (not necessarily from
	the list) on the command line and AO will list its antenna
	files.  Use AO . to force AO to list files in the current
	directory when it's not first on the subdirectory list.


	3. Reference dB

	        To display gain figures in dBd rather than dBi, do the
	following:

	                           SET DB=dBd

	        To use dBd for free-space models and dBi over ground, do
	this:

	                          SET DB=dBdi

	        Gain in dBd is referenced to the peak gain of a halfwave
	dipole in free space.  Gain in dBi is referenced to an isotropic
	radiator in free space.  An isotropic antenna radiates equally
	in all directions.  A dipole has 2.15 dB gain over an isotropic
	antenna, so AO converts from one gain reference to the other by
	adding or subtracting 2.15 dB.


	4.  Printer Type

	        AO defaults printed output for HP LaserJet or DeskJet
	printers in portrait mode.  To try a landscape plot (not
	available on all printers), use:

	                        SET PINS=HPLJ L

	        For a bigger landscape plot, use:

	                        SET PINS=HPLJ LB

	        AO normally draws LaserJet/DeskJet plots with a border.
	To eliminate the border, add an X to other options, for example:

	                       SET PINS=HPLJ LBX

	        AO can transfer compressed data to PCL 5 laser printers
	for faster output.  To try this, add a C to other options, for
	example:

	                       SET PINS=HPLJ LBXC

	        For a 9-pin dot-matrix printer, do the following:

	                           SET PINS=9
				       35

	        Some 9-pin printers don't recognize the special line
	spacing command AO uses to print an exact screen image.  If your
	printer won't print AO screens, try the following:

	                         SET PINS=9ALT

	        Even with 9ALT, AO may be unable to print to certain old
	9-pin printers.

	        For a 24-pin dot-matrix printer, use:

	                          SET PINS=24


	5.  Printer Port

	        Graphics screens print on LPT1 by default.  To print
	graphics on another port, use one of the following:

	                          SET LPT=LPT2
	                          SET LPT=LPT3


	6.  .PCX Output

	        By default, AO generates .PCX screen images with 640 by
	480 pixels.  These images can be directly incorporated in
	documents.  However, the images won't be centered on fax pages
	because they have no margins.  You can add margins with the
	following command:

	                    SET PCX=left top bottom

	        The three numbers specify margin size in inches.  The
	bottom margin is optional.  You can use it to pad the image to
	form a complete page if your fax software doesn't do this
	automatically.  Try left = .8 and top = 2.35 for 8.5" x 11"
	pages.

	        You can create a double-size .PCX image by adding the
	keyword BIG to the line (upper or lower case).  Use BIG for
	high-resolution fax mode.


	7.  Automatic Segment-Length Tapering Ratio

	        To change the length ratio for automatic segment-length
	tapering from the default value of 2, specify a decimal number
	as follows:

	                        SET RATIO=Ratio


	8.  Critical-Command Confirmation

	        Normally AO asks for confirmation to load another
				       36

	antenna file, to edit a file, or to quit only after analysis has
	taken longer than 5 minutes to complete.  This yields quick
	command response for small models while providing protection
	against accidental destruction of time-consuming results.  You
	can change the 5-minute default by specifying a decimal number
	of minutes as follows:

	                        SET CONFIRM=Time


	9.  Linear-dB Polar Plots

	        AO normally displays polar patterns using a log-dB
	scale.  If you prefer the linear-dB scale, you can make it the
	default with:

	                        SET POLAR=Linear


	10.  Display Typeface

	        You can use the standard serif typeface built into your
	VGA card with the following:

	                       SET TYPEFACE=Serif


	11.  VGA Compatibility

	        AO can use a dual-paging scheme to provide smoother
	graphics animation by fully buffering all screen changes.  This
	scheme requires a compatible VGA card with 512K video memory.
	If your card is incompatible or has less memory, dual paging
	will cause extraneous graphics images.  Dual paging eliminates
	flicker during geometry-screen rotation and the blink when 2-D
	plot screens change.  To try dual paging, do the following:

	                          SET PAGES=2

	        AO programs the VGA overscan register to extend the
	background color into the overscan region surrounding the active
	screen.  This improves the appearance of text at screen edges.
	However, some VGA cards generate the wrong overscan color,
	yielding a distinct, off-color border.  To fix this problem, try
	the following:

	                        SET OVERSCAN=FIX


	12.  Screen Bounce

	        When switched from graphics to text mode, the screens of
	many monitors bounce or break up during resynchronization.  AO
	can blank the screen for a short period while the monitor


				       37

	settles.  To try this, specify a short delay in seconds (like
	0.1) as follows:

	                        SET BOUNCE=Delay


	13.  Default Ground Constants

	        To change the default single-zone ground constants, do
	the following:

	                          SET GND=D C

	        D is dielectric constant and C is conductivity in mS/m.
	D = 13 and C = 5 when GND is not set.  For a perfect-
	conductivity default, use:

	                           SET GND=0


	14.  Far-Field, Low-Level Threshold

	        To change the level below which relative far-field
	values are tabulated as -infinity dB, enter a negative dB value
	as follows:

	                        SET THOLD=Level


	15.  Typematic Control

	        When you press and hold a key, the key repeats after a
	short delay.  Default PC values are 500 ms delay and a repeat
	rate of 10 characters per second (10.9 cps for PS/2).  When
	scrolling the RUN file or rotating 3-D graphics, many users
	prefer a shorter delay and a faster repeat rate.  AO sets the
	delay to 250 ms and the repeat rate to 24 cps.  You can change
	these values as follows:

	          SET TYPEMATIC=delay rate exitdelay exitrate

	        Delay codes are 0-3 as follows:

	                0    250 ms     2   750 ms
	                1    500        3  1000

	        Rate codes are 0-13 as follows:

	                0   30.0 cps    7   16.0 cps
	                1   26.7        8   15.0
	                2   24.0        9   13.3
	                3   21.8       10   12.0
	                4   20.0       11   10.9
	                5   18.5       12   10.0
	                6   17.1       13    9.2

				       38

	        The exitdelay and exitrate parameters are optional.
	When present, AO sets these typematic values on exit to DOS.
	When not present, the values on exit are those in effect while
	running AO.

	        To keep AO from modifying your system's typematic
	values, do this:

	                        SET TYPEMATIC=0


	---- ADDITIONAL INFORMATION ------------------------------------

	        NOSC Technical Document 938 contains a detailed
	description of the original MININEC program, including
	mathematical formulation of the analysis algorithm, model
	validation against empirical data, error analysis, and BASIC
	program listing.  This highly technical document is intended for
	readers with a background in advanced mathematics and
	theoretical electromagnetics.  The 100+ page manual is available
	from the U.S. Dept. of Commerce for $22.95.  It can be ordered
	by mail with a check or by phone with a credit card.  Request
	NTIS document number ADA181682 from:

	                U.S. Dept. of Commerce
	                National Technical Information Service
	                5285 Port Royal Rd.
	                Springfield, VA 22161
	                (703) 487-4650



























				       39

	                             INDEX


	#  3
	+ - * / operators  17
	-120 dB  21
	.PCX file  27
	.PLT  23
	2-D Pattern Generation  23
	2-D Pattern Plots  23
	225 pulses  31
	3-D Geometry/Current Display  21
	3-D Pattern Display  22
	386MAX  32
	6061-T6  14
	6063-T832  14

	A command  2
	A key  24
	Abbreviations  3
	Aborting Calculations  27
	Absolute far field  21
	Accuracy  4, 5, 29
	Additional Information  39
	Algorithm Limitations  29
	Align antennas  23
	Aluminum  14
	Amateur version  31
	Ambiguity  22
	American Wire Gauge  3
	Analysis over ground  10
	Annealed bare copper wire  3
	Annealed copper  14
	Annotate  9, 11, 26
	ANT  2
	Antenna file  1
	Antenna Files  2
	Antenna image  10
	AO BATCH  28
	AO PLOT  1, 23
	AO.INI  26
	Array element  12
	ARRL  30
	ARRL log-dB scale  24
	Arrow keys  1
	Aspect ratio  25
	Assembly language  1
	Assumed to be zero  9
	AUTOEXEC.BAT  31, 33
	Automatic frequency sweep  15, 27, 28
	Automatic Segment-Length Tapering  5, 31
	Automatic Segment-Length Tapering Ratio  36
	Automatic segmentation  5
	Automatic Wire Segmentation  5
	Azimuth-angle convention  3

				       40

	Barber's thinning shears  22
	Batch file  28
	Batch Mode  28
	Beamwidth  23
	Beep  27
	Bent wire  3, 29
	Bias  29
	Bias results  29
	BIG  36
	Blank lines  3
	Blink  37
	Blockage of radiation  11
	Boom  30
	Boom-correction formula  30
	Brass  14
	BRIEF  34
	British spellings  3
	Bypass the Main Menu  2

	C  8
	C = 25.195B - 229B^2  30
	C key  25
	Calculated gain  11
	Calculated impedance  11
	Calibrations  24
	Cartesian coordinate system  3
	Caution  11
	Center  6
	Center pulse  6
	Chemically treated  11
	Circular  25
	Circularity of a spiral  21
	Closely Spaced Wires  31
	Clutter  25
	Code optimization  1
	Coefficients  9
	Coincident endpoints  4
	Colors  26
	COM  33
	Comma  3, 6
	Command key  1
	Command line  1, 23, 32, 35
	COMMAND.COM  33
	Comment lines  3
	Compare command  20
	Compare mode  24
	Comparing Plots  25
	Compass bearings  3
	Complete record of the analysis session  2
	Complex conjugate  15
	Computational noise  21
	Conductive mounting plates  30
	Conductivity  11, 12, 14, 38
	Conductor material  14
	CONFIG.SYS  31, 33
	Connected segments  5
				       41

	Connected-segment lengths  30
	Connections  4
	Conventional memory  31
	Convergence  4, 5, 29, 31
	Coordinate System  3
	Copper  14
	Correct patterns  23
	Correction factor  30
	COS  17
	Cosine  17
	Critical-Command Confirmation  36
	Current  4, 10, 11, 31
	Current directory  1, 23, 26, 34
	Current distribution  21
	Current phase  22
	Current sources  7
	Current trace  21
	Current-return path  7
	Currents section  7

	D and N commands  2
	DBd  12, 35
	DBdi  35
	DBi  12, 35
	Default  6, 7, 23, 36
	Default ground characteristics  11, 12
	Default Ground Constants  38
	Default Menu Settings  26, 28
	Delta loops  29
	Dense segmentation  5
	Descriptive phrase  9
	DeskJet  26
	Device drivers  31, 32
	Dielectric constant  11, 12, 38
	Dielectrics  1
	Diffraction  11
	Dimension overrides  3, 17, 18
	Dimensions  3, 16
	Dipole  12, 35
	Directory  1
	Disk cache  32
	Display Typeface  37
	Dissimilar Segment-Lengths  30
	Distance from the antenna  21
	DL6WU  30
	DOS  32
	DOS environment  33
	DOS shells  31
	Dot  24, 34
	Dot-matrix printers  26
	Dual-paging scheme  37
	Dynamic range  26

	E command  2, 33
	Earth  11, 30
	ED program  33
				       42

	Edit  2, 37
	Editor  33
	Effective diameter  30
	Efficiency  14
	Eject  26
	Electric and magnetic fields  20
	Electromagnetic field  19
	Electromagnetic-field component  26
	Element-length correction  30
	Element-Mounting Effects  30
	Elevation angle  23
	Elevation variation  11
	Elliptical  25
	Ellipticity  20
	Embedded transmission lines  31
	EMI  20
	EMM386  32
	End  1
	End1  6, 22
	End2  6, 22
	Endpoints  3
	Enter  1
	Environment  12
	Environment space  33
	Epson-compatible  26
	Erratic current  31
	Error  7, 10, 16, 21, 28, 29, 34, 39
	ERRORLEVEL  28
	Esc  1, 26, 27
	EXE  33
	Experiment  26
	Expressions  17
	Extended-ASCII characters  27
	Extraneous graphics images  37

	F  8
	F5  26
	F9  27
	Far Field  21
	Farads  9
	Fax  27
	Feedline  7, 15
	Feedpoint  7
	Ferrites  1
	Ferromagnetic material  14
	Field components  20
	Filename  1
	First wire of a junction  8
	Flicker  37
	Formula  30
	Free conventional memory  31
	Free space  10
	Free wire end  7
	Free-form  3
	Free-space symmetric  31
	Free-Space Symmetric Antennas  10
				       43

	Frequency  4, 8, 9, 15, 28, 30
	Frequency Offset  29
	Frequency steps  27
	Frequency sweep  28

	G3SEK  30
	Gain figures  12
	Gamma-match model  31
	Generating Plot Data  25
	Geometric distortion  29
	Graphics animation  37
	Graphics Image Files  27
	Graphics screens  36
	Ground  4, 11
	Ground Losses  30
	Ground screen  11
	Ground zone  11, 12
	Ground-current losses  11
	Ground-reflection factor  11
	Ground-reflection wave  12, 21
	Grounded wire end  4, 7
	Groundplane  11, 13, 29

	H  8, 20
	Half-length pulse  4
	Half-segment  4
	Half-wavelength of wire  5
	Halfwave dipole  35
	Halfwave of wire  4
	Height  3, 11, 13
	Help  24
	Henries  9
	HF  29
	HF-antenna performance  19
	Hide the Main Menu  2
	Highlighted file  1
	Hilltops  11
	HIMEM.SYS  32
	Hoch  30
	Home  1
	Horizon  3, 12
	Horizontal antennas  11, 30
	Horizontal Component  20
	Horizontal plane  3
	HP LaserJet  26

	IACS  14
	Ignore wires  4
	Image wires  10
	Impedance  11, 30
	Impedance loads  9
	In-phase current  10
	Inaccurate results  5
	Incompatible operating system or memory manager  32
	Infinity  11, 13, 21, 24, 38
	Input impedance  15
				       44

	Input power  21
	Input resistance  8
	Interference-level analysis  21
	International Annealed Copper Standard  14
	Interpolation  23
	Isotropic  12, 35

	Junction  4, 6, 7, 29

	Key repeats  38
	Key Software Products  32
	Keyword  3, 6, 9

	L  8
	Laplace Transform loads  9
	Lazy-H  31
	LC  20
	Leeson  30
	Left-Circular Component  20
	Length of each segment  4
	Lengthy results  2
	Level  38
	Lightbar  1
	Limit  4, 8, 9, 31
	Limitations  29
	Line number  34
	Linear-dB scale  24, 37
	List plot files  23
	Load another antenna file  36
	Load coincident with a source  9
	Load losses  14
	Load syntax  8
	Loads  6, 8, 10
	Lobe maxima  12
	Log-dB scale  24, 37
	Logan  1
	Loops  29
	Losses  11, 14
	Lossy capacitor  8
	Lossy inductor  8
	Lossy trap  8
	LPDA  31
	LPT  36
	Lumped loads  8

	M key  21
	Magnitude  21
	Magnitude-only currents  21
	Major  20
	Major-Axis Component  20
	Manually segmenting driven wires  7
	Mathematical formulation  39
	Maximum number of pulses  4, 31, 32
	Maximum sidelobe level  23
	Memory  4, 31
	Memory manager  31
				       45

	Menu settings  26
	Midpoints  4
	MININEC 3.13  1
	Minor  20
	Minor-Axis Component  20
	Mirror-image  10, 23
	Mirrors the antenna  10
	Model validation  39
	Motherboard chip sets  32
	Mounting plates  30
	MS/m  12
	Multiple sources  8
	Multiwire junction  7
	Mutual-impedance matrix  7, 20

	National Technical Information Service  39
	Near Field  20
	Near field over ground  20
	NEC  29
	Negative gain values  12
	Negative input resistance  8
	Networks  9
	NH  8
	Norton Editor  34
	NOSC Technical Document  39
	Number of sources  8
	Number of wires  4
	Number-of-loads line  6
	Number-of-sources line  7
	Number-of-wires line  14
	Numerical Electromagnetics Code  29

	O key  25
	Oblique wires  19
	Off-center feedpoint  7
	Ohm-m  14
	Ohms  8, 9
	On-axis response  23
	Optimizing a design  12
	Optimizing an antenna pattern  23
	Optimizing capabilities  1
	Option  23, 24
	Options Menu  5, 6, 15, 23
	Order  4, 8
	Origin  11
	OS/2  32
	Other  1
	Out-of-phase current  10
	Overlap  4
	Overlay  20, 24, 25
	Overwritten  2

	P command  25
	Path  33
	Pattern generation  21
	Patterns over ground  25
				       46

	PCX file  27
	PCX Output  36
	Perfect-conductivity ground  12, 20
	Permeability  14
	PF  8
	PgDn  1
	PgUp  1
	Phased arrays  7
	Phasors  21
	Phosphor Bronze  14
	Physical Design of Yagi Antennas  30
	Plot file  1, 25
	PLT  25
	Plural  3
	Points in space  3
	POLAR environment variable  24
	Polar plot  24, 37
	Polar Plots  24
	Polarization Component  19
	Polarization tilt  20
	Polarization-Ellipse Circularity  20
	Polynomial  9
	Poor ground systems  11, 30
	Preferred settings  26
	Preserves plot files for later comparison  25
	Print  26
	Print graphics  36
	Printer Port  36
	Printer Type  35
	Printing The Screen  26
	Problem junctions  5
	Professional version  32
	Protected mode  32
	PrtSc  2, 26
	Pulse  4, 10
	Pulse allocation  10
	Pulse Identification  6
	Pulse number  6, 10
	Pulses  10

	Q  8
	QEMM-386  32
	Quads  29
	Quicker results  5
	Quit  37

	R  8
	R command  2
	R key  25
	Radial lines  24
	Radial scales  24
	Radial wires  11
	Radius  11, 13
	RAM disk  32
	Ratio  36
	RC  20
				       47

	Reactance  9
	Rear lobe  23
	Recalculation  7
	Rectangular Plots  25
	Reference antenna  12
	Reference dB  35
	Relative far field  21
	Rename  2
	Reorder  4
	Reorders wires  8
	Resistance  9
	Resistive load  11
	Resistivity  14
	Resistor  8
	Resolution  22, 23, 26
	Review plots  23
	RF generator  7
	RF-hazard analysis  20
	Rhombics  29
	Right-Circular Component  20
	RLC loads  8
	RLC syntax  8
	Rockway  1
	Rotate an object  17
	Rotation angle  21
	Run AO unattended  28
	RUN file  2, 7, 10, 33, 34
	Running AO  1

	S command  26
	Salt water  12
	Save Options & Settings  26, 28
	Scientific notation  17
	Screen Bounce  37
	Screen Colors  26
	Search-and-replace  16
	Segment junctions  4, 6
	Segment lengths  31
	Segmentation  5, 8
	Segmentation density  31
	Segments  4, 29
	Segments per halfwave  5
	Serif typeface  37
	SET ANT  34
	SET BOUNCE  38
	SET CONFIRM  37
	SET DB  35
	SET EDITOR  33
	SET GND  12, 38
	SET LPT  36
	SET OVERSCAN  37
	SET PAGES  37
	SET PLT  34
	SET POLAR  37
	SET RATIO  36
	SET THOLD  38
				       48

	SET TYPEFACE  37
	SET TYPEMATIC  38
	Settings  26
	SHELL  33
	Shift an object  18
	Shift and Rotate  17
	Shift-end command  18
	Short wire  5
	Sidelobe  23, 24
	Signal-coverage studies  21
	Silver  14
	Simulate ground-current losses  12
	SIN  17
	Sine  17
	Single-frequency mode  27
	Single-source models  7
	Single-wire models  10
	Singular  3
	Singular or plural  13
	Slice the pattern  22
	Sloping ground  11
	Soil  11, 13
	Source magnitude and phase  7
	Sources  6, 7, 10, 15
	Sources command  7
	Space wave  21
	Spaces  3
	Spaces, commas, or tabs  9
	Sparse segmentation  5
	Special cases  5
	Speed  32
	Speed computation  10
	Speed increase  1
	Speed pattern-generation  23
	SQR  17
	Square root  17
	Standing-Wave Ratio  15
	Steel  14
	Step  20
	Subdirectories  34
	Subdivide  4, 6
	Submenu  1
	Summary printout  2
	Surface wave  21
	Surfaces  1
	Sweep range  27
	SWR  15
	SWR reference impedance  15
	Symbol definitions  17
	Symbol set  27
	Symbolic Dimensions  15
	Symbolic Expressions  17
	Symbols  16
	Symmetrical about the X-axis  23
	Symmetrical antennas  10
	Syntax  9
				       49

	Tabs  3
	Telescoping tubing  4
	Terminate data entry  1
	Terrain  11
	Text Editor  33
	The Last Byte Memory Manager  32
	Thin a pattern  22
	Thin wires  29
	Threshold  21, 38
	Through-the-boom mounting  30
	Tilt  20
	Time-saving feature  23
	TLBMM  32
	Too short  4
	Total  20
	Total Field  20, 26
	Trace  21
	Transform order  9
	Transformer  15
	Transmission lines  5, 31
	Traps  8
	Traveling-wave antennas  8
	Traveling-wave currents  21
	Trick  6
	TSRs  31, 32
	TVI  20
	Typematic Control  38

	U.S. Dept. of Commerce  39
	U.S. Naval Ocean Systems Center  1
	UH  8
	UHF  29
	Unconnected wire end  7
	Unfamiliar geometries  29
	Unidirectional antennas  3
	Unit of length  3
	Units  8, 9
	Upper case  1
	Upper memory  31, 32
	Upper or lower case  3, 6, 8, 33
	Upper-case P  20, 25

	V  20
	V86 mode  32
	Valid results  4
	Vertical antennas  12, 30
	Vertical Component  20
	VGA Compatibility  37
	VHF  29
	VHF/UHF  30
	Video memory  37
	View the whole screen  2
	Viewing commands  21
	Viewpoints  22
	Voltage  7

				       50

	Volts/meter  21
	Volume resistivity  14

	W key  21
	W6QHS  30
	Warning message  4, 6
	Wave polarization  19
	White  30
	Windows  32
	Wire axis  21
	Wire currents  21
	Wire diameter  3, 29
	Wire ends  4, 22
	Wire gauge  3
	Wire is always straight  3
	Wire junction  4, 8
	Wire line  6
	Wire list  3
	Wire losses  14
	Wire segmentation  5
	Wire spacing  31
	Wire-Conductivity Losses  14
	Wire-material phrase  14
	Wires and Pulses  10
	Wires and Pulses section  4, 6, 7
	Wires that cross  4
	Wires, Segments, and Pulses  3
	Wordprocessor  33

	X  3
	X key  25
	X-axis  23
	X-axis cutoff  25
	X-axis pattern symmetry  22
	XMS driver  32

	Y  3
	Y key  25
	Y-axis cutoff  25
	YO Yagi Optimizer  30

	Z  3
	Zepp  31
	Zone  11, 13, 38












				       51

