United States
Environmental Protection
Agency
Office of Radiation
P.O. Box 18416
Las Vegas NV 89114-5027
EPA 520/1-84-024
October 1984
Radiation
AEPA
An Automated
TEM Cell
Calibration
System
-------
AN AUTOMATED TEM CELL CALIBRATION SYSTEM
Edwin D. Mantiply
October 1984
U.S. Environmental Protection Agency
Office of Radiation Programs
Nonionizing Radiation Branch
P.O. Box 18416
Las Vegas, Nevada 89114
-------
DISCLAIMER
This report has been reviewed by the Office of Radiation Programs,
U.S. Environmental Protection Agency, and approved for publication. Mention
of trade names or commerc i a 1 products does not consti tute endorsement or
recommendation for use.
-------
FOREWORD
The Environmental Protection Agency, Nonionizing Radiation Branch
develops Federal Guidance on acceptable environmental exposure levels of
electromagnetic radiofrequency radiation and conducts an environmental
assessment program. These functions require, among other things,
determi ni ng the avai 1 abi 1 i ty of accurate i nstrumentati on to measure
electromagnetic field strengths and thus, the performance evaluation of
commerci a land experimenta 1 hazard instruments. Thi s report documents an
automated transverse el ectromagneti c (TEM) cell system which is used to
evaluate broadband radiofrequency survey meters.
Office of Radiation Programs
i i i
-------
AS STRACT
A design and evaluation project was initiated with the overall objective
of creating an automated transverse electromagnetic (TEM) cell
electromagnetic field calibration system. This system has been constructed
at the Environmental Protection Agency to evaluate the performance of
commerical and experimental probes which are used to determine environmental
electric and magnetic field strengths. The TEM cell calibration system is
composed of computer controlled signal generators, TEM cells, high power
attenuators, computer read power meters, and probe rotation and data
acqui siti on hardware. It is capabl e of produci ng hi ghly accurate
radiofrequency (RF) fields in the frequency range of 10 kHz to 220 MHz with
root-mean-square electric and magnetic field strengths of up to 2000 volts
per meter and 5.3 amperes per meter respectively.
Expressions for the electric and magnetic fields at the test location in
the cell s were developed whi ch account for the effect of an imperfectly
matched load and the resultant reflected wave, assuming the TEM cells are
uniform transmission lines. These expressions are then generalized
throughout a vertical cross section of the cell for the TEM mode of
operation. Measurements were made of hi gher order TE and TM mode resonant
frequencies, field strengths, and return losses for the two cells included
in the system. Measured parameters were compared wi th theoreti ca 1
calculations to predict resonant frequencies.
An error analysi s of the various measured cal ibration parameters is
presented. These errors are propagated through the equation for field
strength to give the probable error in establishing field strength.
Genera tor noi se and harmonic di stort i on are al so treated in the system
eval uation.
System configuration, software control, and operation are documented to
show how frequency, field strength, and probe rotation are controlled.
Sample representative applications of the system and sample results obtained
from tests of RF measurement probes are given.
iv
-------
CONTENTS
Foreword. . .
. . . .
. . . .
. . . . .
............
. . . .
Abstract. . . . . .
. . . . . .
. . . . . . .
. . . . . .
. . . . . . .
Conten ts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
F i gu re s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ac know 1 ed gmen t. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 . I n t ro d u c t ion. . . . . . . . . . . . . . . . . . . . . . . . .
2. Conclusions and Recommendations. . . . . . . . . . . . . . . .
3.
System Description. . . . . . . . . . . . . . . . . . . . . .
Configuration. . . . . . . . . . . . . . . . . . . . . . . .
Software development and operation. . . . . . . . . . . . .
4.
Basic Field Strength Expressions and Modal Characteristics
ofT EM cell s . . . . . . . . . . . . . . . . . . . . . . . .
TEM fields at the test point in a TEM cell with reflections.
TEM fields throughout the TEr~ cell cross section. . . . . . .
Modal characteristics of TEM cells. ............
5.
Calibration Parameter Measurements and Error Propagation. . .
Cell characteristics. . . . . . . . . . . . . . . . . . . .
Load characteristics. . . . . . . . . . . . . . . . . . . .
Po we r mete r s .. . . . . . . . . . . . . . . . . . . . . . .
Si gna 1 pu ri ty . . . . . . . . . . ". . . . . . . . . . . . . .
Error propagation. . . . . . . . . . . . . . . . . . . . . .
6.
Testing and Sample Applications. . . . . . . . . . . . . . .
Comparison of probe response with and without impedance
cor rec t ion. . . . . . . . . . . . . . . . . . . . . . . .
Comparison of probe response using two different TEM cells
Probe orientation changes. . . . . . . . . . . . . . . . . .
Probe potential sensitivity. . . . . . . . . . . . . . . . .
Modal effects in large cell .................
Re fere nce s
. . . . . .
. ... ........... .........
Appendices
A. Load impedance program. . . . . . . . . . . . . . . . . . . .
B. System operating program. . . . . . . . . . . . . . . . . . .
C. Detailed system diagrams. . . . . . . . . . . . . . . . . . .
D. Probe calibration results compared with and without load
impedance correction. . . . . . . . . . . . . . . . . . . .
v
i i i
iv
v
vi
viii
ix
1
2
4
4
5
8
8
19
23
44
44
49
53
54
57
60
60
60
64
65
67
68
70
77
95
99
-------
Fi gures
Number
1 TEt4 cell cal ibration system block di agram . . . . . . . . . . . . .
2 System computer control structure. . . . . . . . . . . . . . . . .
3 Configuration of a circular air coax. . . . . . . . . . . . . . . .
4 Cross-sectional geometry of a TEM cell. . . . . . . . . . . . . . .
5 Standing-wave error as a function of reflection parameters. . . . .
6 Squares of electric and magnetic field for a shorted CC-105 cell. .
7 Squares of electric and magnetic fiel d for a shorted CC-101.5 cell .
8 Squares of electric and magnetic field for an open CC-105 cell. . .
9 Squares of electric and magnetic field for an open CC-101.5 cell. .
10 Cross section through the center of a TEM cell with coordinate
sy s te m show n . . . . . . . . . . . . . . . . . . . . . . . . . .
11 Field magnitude extremes for the CC-105 cell. . . . . . . . . . .
12 Fiel d magnitude extremes for the CC-101.5 cell. . . . . . . . . .
13 Nonnal i zed fiel d strength IFI CC-105 TEM cell. . . . . . . . . .
14 Normal ized fiel d strength IFI CC-101.5 TEM cell. . . . . . . . .
15 Effective lengths defined. . . . . . . . . . . . . . . . . . . .
16 CC-105 cell TE cutoff frequencies, range of resonant frequencies,
and observed resonant frequencies. . . . . . . . . . . . . . .
17 CC-10l.5 cell TE cutoff frequencies, range of resonant frequencies,
and observed resonant frequencies. . . . . . . . . . . . . . .
18 CC-105 cell TM cutoff frequencies, range of resonant frequencies,
and observed resonant frequencies. . . . . . . . . . . . . . .
19 CC-10l.5 cell TM cutoff frequencies, range of resonant frequencies,
and observed resonant frequencies. . . . . . . . . . . . . .. 29
20 ~1oda 1 measurement system block di agram . . . . . . . . . . . . .. 31
21 CC-101.5 cell TE measurements. . . . . . . . . . . . . . . . . . 35-36
22 CC-105 cell TE measurements. . . . . . . . . . . . . . . . . . . 38-39
23 CC-101.5 cell TM measurements. . . . . . . . . . . . . . . . .. 41
24 CC-105 cell TM measurements. . . . . . . . . . . . . . . . . .. 41
25 Isotropic probe measurement of TEOll resonance. . . . . . . . .. 43
26 Cross sectional dimensions of the CC-101.5 cell. . . . . . . .. 45
vi
Page
4
6
9
11
15
18
18
18
18
20
20
20
22
22
27
28
28
29
-------
27 Cross sectional dimensions of the CC-105 cell. . . . . . . . . .
28 CC-101.5 cell dimensions. . . . . . . . . . . . . . . . . . . . .
29 CC -105 cell 1 e ng th s .......................
30 CC-101.5 time domain ref1ectometry measurement. . . . . . . . . .
31 CC-105 time domain ref1ectometry measurement. . . . . . . . . . .
32 Attenuation characteristics of high power load system. . . . . .
33 Load impedance. . . . . . . . . . . . . . . . . . . . . . . . . .
34 Time domain noise at 100 watt output. . . . . . . . . . . . . . .
35 Frequency domain noise at 100 watt output. . . . . . . . . . . .
36 Time domain noise at 1 kW output. . . . . . . . . . . . . . . . .
37 Frequency domain noise at 1 kWoutput . . . . . . . . . . . . . .
38 Field strength probe response without impedance correction. . . .
39 Fie1 d "S trength probe response wi th impedance correction. . . . .
40 Field strength probe response in large cell. . . . . . . . . . .
41 Fie1 d strength probe response in small cell. . . . . . . . . . .
42 Field strength probe potential sensitivity. . . . . . . . . . . .
43 Modal effects in CC-101.5 cell. . . . . . . . . . . . . . . . . .
vii
45
46
46
48
48
51
53
55
56
56
57
61
61
62
63
66
67
-------
Number
1
?
,-
3
4
5
6
7
8
9
T ABL ES
Lower order TE or TMm modes. . . . . . . . . . . . . . . .
mn n
Predicted TE cutoff frequencies. . . . . . . . . . . . . . .
Predicted TM cutoff frequencies. . . . . . . . . . . . . . .
Magnetic field probe calibration. . . . . . . . . . . . . . .
Electric field probe calibration. . . . . . . . . . . . . . .
CC-101.5 Cell TE measurements. . . . . . . . . . . . . . . .
CC-105 Cell TE measurements. . . . . . . . . . . . . . . . .
Characteristic impedance. . . . . . . . . . . . . . . . . . .
Probe orientation changes. . . . . . . . . . . . . . . . . .
vi i i
Page
24
25
26
32
33
37
40
49
65
-------
ACKNOWLEDGMENTS
I would like to thank M. R. Molony for his programming assistance and
program documentation. Graciela Martucci was of great 11elp in editing this
report. M. L. Crawford, C. M. Weil, D. A. Hill, S. J. Allen, P. C. Gailey,
and R. A. Tell provided many useful review comments on an initial draft of
thi s document.
ix
-------
SECTION 1
INTRODUCT ION
Several different broadband electromagnetic (EM) radiation hazard survey
meters have been cOl1ID1ercially produced or buil t as prototypes by government
laboratories. Field experience with some of these instruments has revealed
the need for determining their absolute calibration and evaluating their
general operation in radiofrequency (RF) fields.
Interaction with various government and industrial workers has led to
EPAls use of three approaches to RF calibration: transverse electromagnetic
(rEM) cell s, waveguides, and standard gain horn anechoic ranges. All three
of these approaches have been previously used by others with varying degrees
of uncertainty in the calibration results. Apparently the greatest
uncertainty results from the use of waveguide systems. Depending on the
size of TEM cell, waveguide, or horn antennas used, the systems can overlap
in frequency with physical size inversely proportional to frequency.
Thi s report di scusses an automated TEM cell RF cal ibration system. It
documents the field strength expressions applied, theoretical and
experimental determination of modal characteristics of the two cell s used,
measurements of calibration parameters of system components, a system
description, system testing, and sample applications of the automated TEM
cell calibration system.
1
-------
SECT ION 2
CONCLUSIONS AND RECOMMENDATIONS
An automated TEM cell electromagnetic field generation system was
constructed from commercially available components. This system is
particularly useful and flexible for performing isotropic field strength probe
eval uations. The system can compensate for si gnificant standi ng wave errors
which would be impractical to perform manually. Fitted curves and
interpolation for load attenuation and input impedance allow the choice of any
frequency in the operati ng range of 10kHz to 22011-lz so that narrow-band
characteristics or resonances of fiel d strength probes can be easily
examined. The wide dynamic range makes possible the evaluation of probes at
any field level of interest up to 2000 VIm electric field or 5.3 AIm
magnetic fiel d. The system can be setup by technici ans to run unattended
through any valid set of frequencies at a given field strength. Upon
completion, the system generates and stores uniform reports of calibration.
The system configuration shoul d continue to evol ve. The upper frequency
limit may be increased by implementing a smaller cell and higher frequency
loads and amplifiers. Polar plotting routines have been implemented for probe
rotati on data. Because of the temperature coeffi ci ent of the attenuati on
system, a high power directional coupler as opposed to an attenuator to
measure power should be considered. Finally, eliminating the cable between
the load and cell woul d make the load impedance more stabl e~
Expressions for the field strength in the cell including the effect of
reflections were 'derived. A comparison of probe calibration data obtained
using these expressions and using the simpl er expressions which assume no
reflection from the load resulted in the conclusion that a useful increase in
accuracy is acheived by using the more complex formulas. Even if the load
were exactly 50 ohms resistive there would be standing wave error if the cell
characteristic impedance were not also 50 ohms.
TEM cell s can be used in an open or short circuit configuration with a
directional coupler to generate known high or low wave impedance fields by
applying formulas derived in this report.
2
-------
The high frequency resonant characteristics of the two TEM cells used were
measured and modeled. The lowest frequency resonance for the Instruments for
Industry (IF!) model CC-105 cell was observed at 320 MHz and at 120 MHz for
the CC-101.5 cell. High power low pass filters are used to reduce harmonics
above these frequencies. Future work toward a more complete theory may
involve numerical solutions to the three-dimensional theoretical problem of
the cell viewed as a resonant cavity using moment methods; see for example
[1,2J.
The absol ute
including effects
(TEM) operation.
by theory, with a
probable error in the field strength at the test point
of reflections was determined to be ~0.79 dB for single mode
Variations in the field away from the test point are given
practical upper limit of ~0.4 dB.
A field strength probe was tested in both cells which resulted in readings
typically 0.2 dB higher in the smaller cell. The increase was probably due to
the loading effect of the probe.
BelO\'I approximately 10 MHz a problem was observed in which the entire
fi el d strength measuri ng instrument appears to act as an antenna or
alternatively the instrument senses the potential difference between the meter
and probe. This problem was termed "potential sensitivity" in reference to
potential probes commonly used near 60 Hz power transmission lines. This
problem makes some meters practically useless near standard AM broadcast
transmitters. A solution in this situation is to use a meter having its
antenna and readout in a single chassis.
r~ore extensive
probe measurement
include field work
testing of available meters will result in needed data on
accuracy and operati onal probl ems. Thi s testi ng must
to document operational problems unique to the real world
environment.
3
-------
SECTION 3
SYSTEM DESCRIPTION
The automated TEM cell system built by EPA is a convenient, accurate
system for establishing CW electromagnetic fields in the frequency range 10
kHz to 220 MHz. The frequency limitation is that of the linear amplifier
used; the TEM cell s can be set up manually above and below thi s frequency
range.
CONFIGURATION
A block diagram showing the basic system configuration is given in
Figure 1. Two different synthesizers are used and are selected under
program control; the synthesizer output power and frequency are al so under
program control. The linear amplifier has two outputs which are manually
selected, the low-level (100 watt) output is used up to about 30 watts and
the final section is used from about 30 watts to its maximum output of 1 to
2 kW depending on frequency. Two commercially available TEM cells are used,
a larger one, an Instruments for Industry (IF!) model CC-101.5 with a
measured first resonance at 120 MHz and a small er cell (IFI model CC-105)
having less uniform fields over a survey probe's volume but an upper
frequency 1 imit of 320 MHz where the first resonance occurs. The pl ate
separation, characteristic impedance, and electrical length are the
DEVICE UNDER TEST
SYNTHESIZER
LINEAR
AMPLIFIER
10KHz 22QMHI
HIGH POWER
A TTENUA TOR
POWER
METER
ISELECT SYNTHESIZER I
IAUTOMATIC!
(SELECT OUTPUT)
{MANUALI
(SELECT CELl)
[MANUAl)
CALIBRATION ATTENUATION
COMPLEX IMPEOENCE
(SELECT POWER METER I
[AUTOMATICI
CALIBRATION
ELECTRICAL LENGTH
CHARACTERISTIC IMPEDENCE
PLATE SEPARATION
Fi gure 1.
TEM cell calibration system block diagram.
4
-------
important cal ibration parameters of the TEM cell s and are stored in the
operating program. A 2 kW attenuator of about 43 dB is constructed of a
high power coaxial cable, a 2 kW 30 dB attenuator, 20 feet of RG-214/U
cable, and a 13 dB low-power attenuator. The calibration parameters of the
cable-attenuator (load) system are the total attenuation A and complex input
impedance Zl which are storedi n the program as a function of frequency.
Finally three different power meters are selected by the computer, depending
on the specified power level and frequency.
SOFTWARE DEVELOPMENT AND OPERATION
Fi gure 2 gives an over vi ew of the control structure through whi ch the
computer can control and/or read the synthesizers, power meters, probe
rotator, and hazard meter recorder output. The synthesi zer, power meters,
and printer are interfaced directly through the IEEE-488 interface bus. An
analog to digital converter (AOC) on the bus is used to read the hazard
meter output and determine the angular position of the probe rotator. A bus
relay actuator is used to control the rotator motor and to operate a
microwave switch to select power meters. A VHF switch on the bus selects
the synthesizer. Detailed system diagrams are contained in Appendix C.
Appendix B is a listing of the system operating program called IICELLGOII
written in BASIC. The program is divided into five basic components, each
of which may be somewhat independent of the other. Comments at the
beginning of the program give the bus addresses of the various instruments
used. A select code of 7 is always used, for example,
OUTPUT 722, IIAlIl
is a command to addres s 22, the VHF switch.
The program first initializes and defines program variables, such as
plate separation, characteristic impedance, and electrical length for each
of the two cells. Program defaults and ADC calibration settings are
displayed on the screen at the end of initialization.
5
-------
0'\
Figure
B
Rom's: 1/0
Advanced Programming
Graphics
<-- HP 98034A IEEE-488 Interface
HP 98035A Real Time Clock
2.
System computer control
structure.
-------
The program then will check the status of certain peripherials to
guarantee that they are operating and are on-line. It is at this point the
operator is allowed to interact with the computer in order to define the
measurement. As can be seen from the program listing in Appendix B, the
operator must first identify himself, and then identify the equipment
involved in the measurement. The field characteristics may then be
described by specifying 'electric' or 'magnetic' as valid selections, and
the units used to enter the desired field.
The power meter and synthesizer are selected by the computer by means of
the bus relay actuator and bus VHF switch, depending on the specified power
level and frequency. After calculating the desired setting the computer
reads the power meter and can adjust the generator power output unti 1 the
power meter reads within 30.1 dB of to the calculated setting corresponding
to the desired field strength in the cell. Once the field is established
the hazard meter probe is automatically rotated and the hazard meter
recorder output is read by an ADC and compared to the known fiel d strength
to present a plot of dB error versus rotation angle at each frequency. A
manual check of the meter display is made to insure that the recorder output
is operati ng properly.
After the 1 ast frequency is campl eted a summary tabl e and graph of
rotati on stati sti cs is presented whi ch i ncl udes frequency, average readi ng,
average error, high and low error, and error range. These plots show the
hazard meter errors as if the field absolute uncertainty is OdB. Typical
plots are given in a later section on testing and sample applications, and
in Appendix D.
7
-------
SECTION 4
BASIC FIELD STRENGTH EXPRESSIONS AND MODAL CHARACTERISTICS OF TEM CELLS
TEM FIELDS AT THE TEST POINT IN A TEM CELL WITII REFLECTIONS
This section describes derivations of expressions for the electric and
magnetic fields at the test point within a TEM cell when the load on the
cell output is not perfectly matched to the cell's characteristic
impedance. The "test poi nt" is defi ned throughout thi s report as the poi nt
halfway between the cell sidewalls, input and output connectors, and halfway
between either the upper or lower cell wall and the septum. High power
loads or attenuators generally have fairly large variations in impedance as
a functi on of frequency (typi cally resu1 ti ng in vol tage standi ng-wave rati 0
(VSWR1s) as high as 1.1) and can cause standing wave errors (typically up to
:1:0.4 dB) if the following expression [3J for the electric field with a
perfectly matched load is used.
..1""j)"R
E = V r n~o
(V 1m) where
(1)
E is the electric field at the test point in volts per meter (Vim). Pn is
the net forward or transmitted power through the cell in watts, R is the
o
real part of the cell characteristic impedance in ohms, and b is the distance
in meters between the septum and the upper or lower cell wall.
The approach described in this report for detennining the fields assumes
the TEM cell is: (a) driven by a RF power source, (b) is being operated in the
TEM mode only, (c) is used for the calibration of essentially non-perturbing
field probes, and (d) is terminated in an impedance matched 50 ohm (0)
absorptive load. Under these conditions the net forward power may be obtained
either by (a) a measurement of the forward and ref1 ected power sensed at the
cell input or (b) a measurement of the power delivered to a power sensor
through an attenuator connected to the output of the TEM cell. In practice we
use the latter technique because of its simplicity. The shorted or open cell
8
-------
is addressed in a later section. An alternative method of determining the
field eXisting within the TEM cell involves a measurement of the RF voltage at
the input terminal of the cell. Although this meti10d is not discussed in this
report, it is also subject to many of the same considerations of cell mismatch
for assigning uncertainty to the resulting electric field within the cell.
Because of reflection, the TEM condition is not met for the standing wave
at the test point, and the electric and magnetic fields no longer maintain the
free space magnitude ratio of 3770. [4]. If the magnetic field (H) can be
expressed as a function of current and the electric field (E) as a function of
voltage at the test point, all that is required is solving the transmission
line problem for the current (I) and the voltage (V) at the test point. This
is illustrated in the following example for an air filled circular coaxial
transmission 1 ine.
Example Problem for an Air Filled Circular Coaxial Transmission Line
This example shows
impedence can be used to
defi ni ti on for the well
definition is
how an alternative definition
obtain the same impedence formula
known case of air circular coax.
of characteristic
as the conventional
The conventi ona 1
Zo ={f
for a lossless line where Land C are the distributed inductance and
capacitance per unit length, respectively [5].
Figure 3.
Configuration of a circular air coax.
9
-------
For a 10ssless air coax, shown in Figure 3, the field magnitudes are given
by magnetostatics and electrostatics [6J to be
I V
H = 21Tr and E = r 1 n(b/a)
(2 )
at any point along the transmission line. The quantities a, b, and rare
defined in Figure 3. Both fields are inversely proportional to the radius (r).
If Zo is defined as the ratio of VII, such that E/H = 377 r.! = Z , where
wfs
Z is the impedance of a plane wave in free space. (Z is pure real, since
wfs wfs
the electric and magnetic fields are in phase for such a wave) then for an
air filled circular coaxial line,
E V 21T
H = T 1n(b/a)
V 377
T = -r; 1n(b/a) = Zo
= 377 r.!, so
(3 )
(4 )
which is the standard equation for air coax characteristic impedance. This
result can also be obtained from the conventional definition. Z is real
o
and a function of geometry only. This alternative definition is used in the
next secti on.
E as a function of V, and H as a function of I for TEM cells
For a TEM cell, consider the following equations for H as a function of I
and E as a function of V at the test point. The cross sectional geoliletry of
the rectangular section of the cell is shown in Figure 4, which defines a, b,
and g.
H " [4b ~
V
E = b
I
~ 1 n(sinh *~
(5 )
(6 )
10
-------
I"
2a
-,
b
~
2b
-....
9
I
I
I
~
I
I
I
Fi gure 4.
Cross-sectional geometry of a TEM cell.
To determi ne the characteri sti c impedance of the cell, set E/H = 377 Q when
VII = Z :
o
~:: t. 4b [t - ~ In(sinh *~ = 377Q,
I
(7)
and
Z V - 377Q
o = T = 4[t - ~ 1 n(sinh * ~ (8)
This is the approximate expression given by Crawford withou1; a typically
negligible correction for the inter-edge effect which is only significant
for a narrow center conductor [7, 8J. Eg. (5) was actually derived from Eg.
(8), however, this order of presentation seems more logical since Eq. (5)
could have been derived from magnetostatics. A convenient expression
obtained from Egs. (5) and (8) is:
IZ
H :: 37~b (Aim),
(9 )
11
-------
where
cell.
of th e
[9] .
Z may be determined from Eq. (8) by geometrical measurements of the
o
Exact theoreti ca 1 determi nati on of Zo for the particul ar geometry
cells used in this analysis yields insignificantly different results
Since E = V/b and H = IZ /377b, the problem is now to find the voltage
o
(V) and current (I) at the cell center from the power flow.
Voltage and Current from the Power Flow
The voltage (V) and current (1) can be determined from the net forward
or transmitted power and the 1 ine impedance at the center of the cell. The
complex line impedance (Zi) can be found from the complex load impedance
(Zl)' the cell characteristic impedance (Zo)' half the cell electrical
length (1 /2), and the frequency (t). This analysis assumes that the
e
transmission line impedance of the cell (Zo) is constant along the length
of the cell, which is not strictly true; however, the technique does yield
an approximate solution. The variation in Z is treated in a later
o
section on error analysis.
Zl must be measured as a
1e/2, and f are known, Zi can
transmission line formula [10].
function of frequency.
be determined by the
Once
Zl ' Zo'
lossless
corrrnon
Zl
Z. = Z
1 0
1 1
cos(s-f) + jZo sin(s-f )
(10 )
1 1
Zo cos(s 2 e) + jZ1 si n(s ~ )
where the phase constant S = ~ = 21ff radians.
;\. c
12
-------
Since the net radiofrequency power flowing through the midpoint of the
cell is the same as the net power measured at the output, assumi ng no
i nterna 1 1 osse s,
P = V I Ri
n -
I Zil
(11)
where I Zi I = ~R~ + x~
A1 so
dR.
an 1
jZij
V =
is interpreted as the power factor [llJ.
I Zi I I,
where V and I are root-mean-square (rms)
used.
(12)
val ues and only the magni tudes are
2
P = I R., from which I =
n 1
~
and
v = IZ; ~
1
(13)
Electric and Magnetic Fields from the Power Flow
Combining equation 6, 9, and 13 yields:
E = t = I z;/ ~
1
b
IZ - [p; Z
and H :: 37~b = ,,~ 37~b
(14 )
The electric field in Eq. 14 may also be written as:
~P / G i
E = b
since
R.
1
G. =? 2
1 R.-+X.
1 1
where G. is the rea 1 part of the 1 ine admittance.
1
The inverse expressions are used to determine the value of power in watts
needed to obtain a given field strength; they are
P = ~R.b2
n 1
IZi 12
and
P = H2R. 3772b2
n 1
Z 2
o
13
(15 )
-------
Note tha t when there is a perfect match Zl = Z , and Z is
- 0 0
(Z = R). Eq. (14) then reduces to Eq. (1) for the electric
o 0
magnetic field can be found using the free space ratio:
pure rea 1
field and
~-
P Z
E-~
- b
E
an d H = IDa:
(16 )
Also, P = P . A, neglecting mismatch loss «0.01 dB), where Pm is the
n m '
power meter reading in watts and A is the attenuation factor, which must be
measured as a function of frequency. If the attenuation is 43 dB, A :: 20000.
Note that the complex wave impedance Z is given by
w
Zw = Zi (3~:~) where Zo is the real characteristic impedance.
(17)
It is useful to see the magnitude of the maximum standing wave error as
a function of vari ou s ref1 ecti on parameters for re1 ative1y well matched
loads, (see Figure 5). Even though reflected power may be small the
standing wave error can be significant. Thus the error (mismatch loss) due
to not compensating for reflected power can be negligible while the standing
wave error is large. For example with a VSWR of 1.1, the mismatch loss is
only 0.01dB. This might lead one to believe that the standing wave error is
small, however, it is *0.4dB.
14
-------
Magnitude of
J voltage
0,2 reflectIOn
'.. P coeffiCient
1'45 1'5. VSWR~[~'~I
. . . Return loss= (~20 log"p)
15dB ~~dB (difference
......... between forward
....... and reflected
........ 50% power In dB)
.........
...... -.II'"
....... ."",. 40%
....... ."",,.
...... .""
.... """
.....- ."",
....e«l.. ."",,,
... "",
...... ."",
..:;"""
,
11
.
1 15
.
1 2
.
1 25
.
1 3
.
1 35
.
1 4
o 0 01 0025
005
0075
01
0125
015
0175
'1 ,
1 0 1 02 1 05
cO 40d B
30dB
.
25dB
.
22dB
, . ,
20dB 19dB 1 BdB
.
17dB
,
16dB
., 5dB
+ dB error
10 log" (1 +p)2
30%
+% efror In
equivalent power
density or
field squared
[(1 +p)' ~ 1] 100%
~IIIIIII~
-, OdS
...............
.-05dB
20%
10%
-05dB
iI:-:............
........ ""'................ .'0%
..... ~...........
.... ~............~
...... ................................~
... ~...........~
..... ~............................-:
.... ~.............4':;
...
....
-...
.......
...
........
.........
-.......
-20%
-dB error
10 log" (l-p)'
-30%
-% error In
equIValent power
density or
field squared
[(1~p)'-1] 100%
............... .' OdS
-40%
'IIIIIII~
-50%
.' 5dB
o '001 dB
o '01 dB 0 '02dB
O'05dB
O"OdB
O"5dB
......: %
.
% reflected power
(p' 100%)
mismatch 1055
[~1010g,o (1 p') ]
(error due to
not measuring
reflected power)
o
o 05% 0 1 %
025%
05%
1%
2%
3%
Fi g u re 5.
Standing wave error as a function of reflection parameters.
The Shorted or Open Cell
In the case where a cell is terminated with a short or open circuit the
previous analysis cannot be applied, since the forward and reflected power
are equal and the net power P n is zero. Al so, at certain frequencies, the
line impedance is infinite. Thus, another objective of this analysis is to
detennine the absol ute el ectri c and magneti c fi el d strengths at the test
point when the load end of the cell is either open or short circuited.
In this case, the forward power (Pf> is equal to the reflected power
(P > for a lossless cell and either one may be measured with a directional
r
coupl er at the cell input. The coupl er shoul d have the same impedance as
the cell characteristic impedance.
15
-------
For total reflection, the maximums in the voltage and current for the
standing-waves which repeat every half-wavelength, are:
v = 2 Vf = 2 ~rp-:JfZ
max ~. f-o
(8)
and
Imax = 2 If = 2 ~ fPf
"z:-
that is, twi ce the voltage or current for the forward or ref1 ected wave.
Z is the cell characteristic impedance.
o
Backing away from the short:
V = V Is i n e I
max
(19)
I = I Ico s e I
max
Or from the open:
V = V Icos e I
max
(20 )
I = I Is i n e I
max
where e is the electrical angle back to the cell center; for a cell with
electrical length le'
e = 1ff1 e
c
(21 )
where we have moved back toward the generator 1 /2 to the center of the cell
e
and f is the frequency (MHz) and c the speed of light is 300 MHz.m
(equivalent to 3 x 108 m/sec) [12].
Combining equations 18 through 21 with
E = V/b and H = IZo
377 b
(22 )
16
-------
where b is the
plate separation, gives:
E = 2~ls;n (~)I
b
H = 2{Pj; Icos(~)
377 b
Short
(23)
E = 2~lcos (~)
b
H=2~lsin(~)
377 b
Open
(24 )
The wave impedance, which is always pure imaginary such that E and H are 90~
out of time phase, is given by:
(1ffl )
lw = j377tan ~
Short
(25)
j377
tan (1f;le )
Both of these expressions approach infinity at nulls in H.
l =
w
Open
(26)
From Eqs. 23 and 24 the squares of the electric and magnetic fields are given
as:
(1ffl )
E2 = 4 P Z s i n2 ~
f 0 c
b2
and
2 2 (1ffle)
H = 4P flocos ----c
3772b2
2 2 (1ffl )
H = 4Pflo sin \ c e
3772b2
Short
(27)
E2 = 4P flocos 2
b2
(1f;le)
Open
(28)
Eqs. 27 and 28 are plotted in Figures 6,7,8, and 9 as a function of
frequency, with the power normalized to Pf = 1 watt for two specific TEM
cell s used for instrument calibrati-ons in the EPA laboratory. These cells
.
are the Instruments for Industry (IFI) models CC-105 and CC-101.5.
17
-------
Cell Ch.,'act.riatlC'ii.
Zo c 52 Ohl"lli
18'" 1 1-4 '"
b ,. .153 1"1
PO"'.r 1 Wan
E"'2 curvl!
H"'2 curve
9000
8000
7000
~ 6000
N
(
E
" 5000
N
<
:: ~000
N
<
w 3000
2000
1000
Figure 6.
C.ll Charact.risticfi'
Zo" 52 Ohl"l5
II!''' 1 1-41"1
b ,. 153,..
Pow.r 1 Wan'
l."'~ curve
H"'2 curve
9000
8000
7000
;;; 6000
<
E
" 5000
N
<
:: ~000
N
< 3000
w
I
2000
1000 I
;
0
0 ~0
\.
IFI 105 (Shorted)
SHORTED IFI 105 CELL
I
,/
/
\
/
80
120
200
160
Frequency (MHz)
Squares of electric and
magnetic field for a
shorted CC-105 cell
IF! 105 (Open)
OPEN IFI 105 CELL
Fi gure 8.
;
.I
\
\
\
80
200
2~0
280
F"requency (MHz)
Squares of electric and
magnetic field for an
open CC-105 cell
0.065
0.060
\
\.
0.055
0.050
0.045
N
0.040 (
E
0.035 :::.
<
0.030 a:
0.025 N
(
0.020 I
01,015
0.010
0.005
0.000
320
.0.065
0.060
0.055
0.050
0.045
0.040 'i'
E
0.035 :::.
<
0.030 !:
.I
0.025 N
(
0.020 I
0.015
0.010
0.005
0.000
320
Cell Char.lct.rlst1c'ii
Zo = 51 3 Ohl"tli
1 eo c 2 ,78 '"
b :z: "01,..
IFI 1015 (Shorted)
Po"'er 1 Watt
E~2 curve'
H~2 curve
1~00
1200
1000
;;;
(
E 800
"
N
<
:: 600
N
(
w
400
200
o
o
SHORTED IFI 101.5 CELL
0.010
"--"". ..]
0.008
0.009
0.007
C\J
0. 006 ~
"
'"
0.005 <
a:
0.004 C\J
(
I
0.003
0.002
0.001
,
40 50 60 70 80
Frequency (MHz)
Fi gure 7.
Squares of electric and
magnetic field for a
shorted CC-101.5 cell
C.11 Charact.ri.Uc..
20 - 51.J Oh",.
1. - 2 79 '"
b - .401,..
lfl 101 II) (Open)
Po",.r 1 Watt
E"'2 curve
H"'2 curve
1400
OPEN IFI 101.5 CELL
0.010
0.009
0.008
0.0137
;;;
13.13136 (
E
"
N
13.13135 <
!:
0.13134 N
<
I
., 13.1303
" '
, 0.002
1200
13.001
-'-""
,.
1000
;;;
(
E 800
"
N
<
:: 600
N
<
w
400 '
.'
21313 ,.
,
,
\,
o
o
'..' 0.000
90 100 110 120
10 213 30 ~0 50 60
Frequency (MHz)
Figure 9.
Squares of electric and
magnetic field for an
open CC-101.5 cell
18
-------
This type of field set-up with TEM cell s is useful when evaluating, for
example, the electric field response of magnetic field sensors or the
magnetic field response of electric field sensors.
TEM FIELDS THROUGHOUT THE TEM CELL CROSS SECTION
The two-dimensional electrostatic problem for the electric field
distribution of the TEM mode has been solved by Tippet [13J. The present
analysis will illustrate these field variations for the particular geometry
of the two cell s used in the EPA 1 aboratory and extend the analysi s to
provide explicit expressions for the magnetic field as a function of current
at the test point.
Electric Field Distribution of the TEM Mode
The electric field distribution was solved by Tippet [13J using the
Schwarz-Christoffel transformation resulting in elliptic integrals and
Jacobian elliptic functions in the final result. A program similar to
Tippees was written in BASIC for a desktop computer which determines
E (x,y) and E (x,y) normalized to V/b, which are the components of the
x y
transverse E field. The coordinate system is shown in Figure 10. The
z-axi s (not shown) woul d come out of the paper wi th the z-coordinate equal
to zero at the cell IS longitudinal center. The test point coordinates are x
= 0, y = *b/2, and z = O. The fields are symmetric in the four quadrants so
calculations are made only in the first quadrant with x and y positive.
The field strength variation over the measurement probe1svolume, with
the probe center at the test point is relevant to the calibration of survey
probes. Calculations of the field magnitude extremes along circles centered
on the test point were made and are displayed in Figures 11 and 12 for the
two IFI cells. The field strength magnitude is normalized to V/b and given
in dB. For example, a spherical field probe with a radius of 5 cm in the
smaller IFI 105 cell would have a field variation of '*'0.4 dB over its
volume. This analysis neglects small field variations in the z-direction
due to standi ng wave s.
19
-------
0.7
.D 0.6
'-
>
E 0.5
.g 0.4
c
o
';
.:;;
Q)
o
.r: -
... !D
g> ~ 0.0
~
en
1:)
Q;
u::
E -0.4
::J
E
.x -0.5
'"
~ -0.6
y
I
2b
o
o
2w
~g-
x
2a
Fi gure 10.
Cross section through the center of a TEM cell with coordinate
system shown.
a = 22.4 em
IFI CC-105 Cell
b = 15.0 em
w = 16.8 em
0.3
XI
if
w'
u' W
W rB-"'
Le--' ~
i.AY
Do- "¥
~- '-B..
--e..., P"-R
\'-6- hoc
~ ~
~
"'B-, '--e...
"'., r'",
~
0.2
0.1
-0.1
-0.2
-0.3
-0.7
o
1 2 456
Radius of Circle About Test Point (em)
Fi gure 11.
Field magnitude extremes
for the CC-105 cell.
0.7
.D 0.6
'-
> 0.5
E
.g 0.4
c 0.3
o
'; 0.2
.:;;
Q)
0 0.1
.r: 00
C, 0.0
c ~
~ -0.1
en
:Q -0.2
Q)
u:: -0.3
E -0.4
::J
E -0.5
.x
'"
~ -0.8
Figure 12.
20
a = 60.1 em
IFI CC-1 01 5 Cell
b = 39 7 em
w = 45.8 em
.&
.&~
~~
"" ~ ~
a:k ~ ~.".
-'"" ~ ~"""
~ ~ ~
~ e..
~ ~ ~
"'" 9:10 ~
B.,
""" I'Jjj ~
fA:
~~
~ ~
'<.
-0.7
o
18
20
2 4 6 8 10 12 14 16
Radius of Circle About Test Point (em)
Field magnitude extremes
for the CC-101.5 cell.
-------
Perspective views of the normalized magnitude of the field strength
within the t\'IO TEM cell s are given in Figures 13 and 14. Results for only
the first quadrant are shown. The field is slightly higher below the test
point at the septum (x = 0, y = 0) and is lower above the test point at the
shield (x = 0, y = b). The field goes to zero at the corner (x = a, y = b)
and goes to infinity at the septum edge (x = w, y = 0) because of the zero
thickness septum model used.
Magnetic Field Distribution of the TEM Mode
For either the for\yard or reflected wave the magnetic field is given by
the TEM requirement from the electri c fiel d. If the normal i zed el ectri c
~ .... V
field is E(x,y), the actual electric field is E(x,y) . b' and the magnetic
field without reflections is
~ ~~ V
H(x,y) = *Y'E(x,y)' b
(29)
The expression with reflections is
Z I
~ ~....i 0
H(x,y) = *Y'E(x,y)' b
(30 )
&.J.
where Y is the dyadi c wave admittance [14 ]
'" Ii
Y Y (f...f... f...f...) d Yo
= J' - 1 J an = -
110
1
= 377rJ..
The above expression illustrates that without reflections the magnetic field
is transverse to the electri c fiel d and has the °magnitude given by the
electric field magnitude divided by the impedance of free space. The
voltage V and current I are the same as in the previous section on fields at
the test point.
21
-------
a = 22.4 em
b = 15 em
w = 16.8 em
;~'~0=0Y
,~T
~
'/.'/. '/.0 \~ \'0 \10.
\'/. \0
~
'0
10.
'/.
o
X (em)
Figure 13.
Nonnalized field strength IFI CC-105 TEM cell.
a = 60. 1 em
b - 39.7 em
w = 45.8 em
'.>
.IJf:$~~
",<0 ",0 .,<0 .,0
Y (em)
X (em)
Figure 14.
Normalized field strength
3.0
2.0 .Q
'-
>
'-
w
1.0
0.0
3.0
2. B .Q
'-
>
'-
w
I.B
B.B
IFI CC-lOl.5 TEM cell.
22
-------
MODAL CHARACTERISTICS OF TEM CELLS
There is some controversy about the modal and resonant characteri sti cs
of TEM cell sin genera 1 and how these characteri sti cs affect the useful
upper frequency 1 imi t of the cell s [15]. There are two general types of
boundary value problems which are relevant: (type 1) the two-dimensional
uniform waveguide problem in which the waveguide has an indefinite length
and (type 2) the three-dimensional resonant cavity problem in which definite
resonant frequencies are determined. Tippet and others [16, 17. 18] have
solved the shielded rectangular coaxial stripline problem. These analyses
are of type 1 since they consider only the cross sectional dimensions of the
cell. The type 2 problem applied to the full three-dimensional tapered
shield and septum of a TEM cell does not appear to have been solved and is
beyond the scope of this report. The uniform waveguide solutions (type 1)
i mpl y by thei r resul tant cutoff frequenc i es that, beyond some frequency.
higher modes propagate and the cell cannot be used simply as a TEM
transmission line. The resonant cavity view (type 2) suggests that between
resonant frequencies the cell will operate normally and can be used to much
higher selected frequencies until the resonances begin to overlap.
tt1easurements made of the longitudinal electric and magnetic fields (E and
z
H ) suggest that resonant frequencies predominate starting at 120 MHz in
z
the CC-101.5 cell and 320 MHz in the CC-105 cell. Isolated wide-band fields
characteristic of propagating modes are also seen (Figure 22c). Since the
cavity problem is unsolved, the best that can be done when comparing
theoretical results to experimental return loss and field measurement data,
is to use the common expressi on for the resonant frequency of rectangul ar
waveguides, with flat conductive ends, as a function of cutoff frequency and
length. Three lengths between the beginnings, middles, and ends of the
tapers were tried as possible effective lengths. The data suggest different
effective lengths for different modes.
Hill [19] was able to more positively identify the TE resonances by
measuring the y component of the electric field at several points in the
cell. He also found different effective lengths for different modes. Here,
only the z component of the magnetic field was measured to establish the
23
-------
exi stence of a resonance conditi on.
was present was usually ambiguous.
Cutoff Frequencies
Identifyi ng whi ch resonance condi ti on
Thi s secti on addresses the uniform wavegui de cutoff frequency sol utions
for the geometries of the two cell s used. The relevant parameters are the
cross section dimensions a, b, and w; i. e., the half-width, half-height,
and half-septum width for the CC-101.5 larger cell and the CC-105 smaller
cell. The TE and TM modes fall into two classes: those perturbed by the
existence of the center conductor or septum (n odd), and those unperturbed
(n even) which have only vertical or normal electric fields and horizontal
or tangential magnetic fields at the septum [20]. Table 1 lists some of the
lower order TE and TM modes.
TABLE 1. LOWER ORDER TE or TM MODES.
mn mn
Perturbed Modes circled.
TMmn modes are not allowed in the first
row or column with m or n = O.
n
o
1
2
3
o
1
2
3
4
10
20
30
40
m
02
12
22
32
42
The two lowest perturbed TE modes are
perturbed TM modes are TMU and TM21'
Appendix L of Tippet's dissertation [21]
24
@)
@
13
33
43
TE01 and TEll' the two lowest
A program similar to that in
wa s used to determi ne the cutoff
-------
frequencies of these perturbed modes.
for hollow rectangular waveguide with
are determi ned in MHz from
300 ~4m2b2 + 4n2a2
f -
c(mn) - 8a b
The unperturbed modes are the same as
dimensions 2a x 2b. Their frequencies
(MHz)
(29)
where 300 is the speed of light in vacuum in MHz.m (m x 106/sec) units
derived from [22J. For the CC-101.5 cell, a = 0.60lm, b = 0.397m, w =
0.458m; for the CC-l05 cell, a = 0.224m, b = 0.15Om, w = 0.168m. Table 2
gives the predicted TE cutoff frequencies and Table 3 gives the predicted TM
cutoff frequencies.
T ABL E 2.
PREDICTED TE CUTOFF FREQUENCIES
CC-l05 cell, Unperturbed modes:
TEIO
TE20
TE02
Perturbe d mode s:
TEOl
TEll
CC-lOl.5 cell, Unperturbed modes: TEIO
TE20
TE02
Pe rturbe d mode s :
TEOI
TEll
fc(lO) = 335 MHz
fc(20) = 670 MHz
fc(02) = 1000 MHz
f c( 01) = 281 r~Hz
fc(ll) = 528 MHz
fc(10) =
fc(20) =
fe(02) =
125 MHz
2 50 r~Hz
378 MHz
f c( 01) = 105 MHz
fc(ll) = 197 MHz
25
-------
T ABL E 3.
PREDICTED TM CUTOFF FREQUEt£IES
CC-105 cell, Unperturbed modes:
Perturbed modes:
CC-101.5 cell, Unperturbed modes:
Perturbed modes:
TM12
TMz2
TM32
TM11
Tfv2 1
TM12
T~2
TM32
Tt411
~1
fc(12)
f c(2 2 )
f c(32 )
= 1055 MHz
= 1204 MHz
= 1417 14Hz
f c( 11) = 1052 t,1Hz
f c(21) = 1194 MHz
fc(12) =
f c(2 2) =
f c(32) =
398 MHz
453 MHz
532 r~Hz
fC(ll) = 397 MHz
f c(21) = 450 MHz
Only those perturbed modes for which Tippet's program was developed have
been calculated; hi gher order p'erturbed modes may have cutoff frequencies
less than the higher cutoff frequencies for unperturbed modes given. The
accuracy of these cutoff frequencies is detenni ned by the accuracy of the
cell demensions or ~ 1%.
Resonant Frequencies
The equation for the resonant frequencies
wavegui de, turned into a resonant cavity by putti ng
end [23J, is:
f
res
= ~f/
2
+££.
2d
(MHz)
of hollow rectangul ar
flat conductors on each
(30 )
givi ng a series of resonances for each cutoff frequency. c = 300 MHz.m,
d = resonant length (m), and p = 1, 2,3... for TE modes or p = 0, 1,2 ...
26
-------
for n~ modes. Si nce d is not we11-defi ned for the cell wi th tapers, three
values of d are used: d1' d2' and d3 as defined in Figure 15.
d3
d2
I"" d, -,
Figure 15.
Effective lengths defined.
The percentage of taper used in the effective lengths d1' ~, and
d3 are 0%, 50%, and 100% respectively. The cutoff frequencies, range of
resonant frequencies using 0% to 100% of the tapers, and observed resonant
frequencies found in the next section are shown in Figures 14, 15, 16, and
17. The resonant frequency ranges were calculated for p = 1, 2, and 3 for
TE modes, and p = 0, 1, 2, 3 for TM modes. The horizontal lIerror barsll
represent the range of resonant frequencies with the left end using 100%,
the circle 50%, and the right end using 0% of the tapers. The short
vertical bars give the cutoff frequencies and the vertical lines extending
from the bottom to top show the resonant frequencies whi ch were observed.
If an observed resonance frequency crosses more than one range of resonance
frequencies the identification is ambiguous. Hill [19J found for a cell
havi ng 2b = 0.6m that -50% of the tapers, for TE10 and TEll resonances,
and -81% of the tapers, for TE01 resonances, shou1 d be i nc1 uded in the
effective length.
27
-------
Figure 16.
Fi gure 17.
~1 ~ -< TE"11
TE"12
TE013
~l H - f--< TE,.,
TEUt2
TEU!3
~1 -< TE",
TEI12
TEI13
~f -t TE20'
TE282
TE203
~~ TE.2,
t 8 r TE022
w
'" ~ f- , . I TES23
~
~ ~ ~ H ~E
.. ~ ~ ~ ~ M M
~c
]~ .. M ...
~ i
. . ' .
200
400
600
800
1000
1200
1400
1600
Frequency (MHz)
CC-105 cell TE cutoff frequencies, range of resonant frequencies,
and observed resonant frequencies.
I f-+-- ---i TE011
TEel2
TEet3
JI >-< -----i TE,.,
TEUR
TEI03
JI ~ --i TEll,
TE112
TEII3
~I ~ TE.:!01
I . I TE202
I , I TE203
NI~TE""
~ I e I TE022
f- " I TE823
~ ~ ~ ~ ~ ~
~ ~
W
f-
100
600
200
300
400
500
Frequency (MHz)
CC-10l.5 cell TE cutoff frequencies, range of resonant
frequencies, and observed resonant frequencies.
28
-------
Fi gure 18.
Figure 19.
. TMZ11i!1
tl f+----< TM2" TM212
'I ~ TM221 ' 1----< TM",
; 9 . TM222
.- ~ ITM223
I : TM",
~I t-9----i1 TM~2J
~ I
~" ~
~; ~
~ ~ ~
~ - -
1000
1100
1200
TM] 13
TMl23
I TM3i!2
,
I TMn]
1300
1400
1500
1600
1700
1800
Frequency (MHz)
CC-105 cell TM cutoff frequencies, range of resonant frequencies,
and observed resonant frequencies.
TMII)
TMI2]
. THZIS
il~]:"I,~' ITM212 TM",
. TMZ20
ru II-&--i TH221
N TM222
~ TM22]
~
~
350
~~
450
. THa2e
lie-< TM",
~ I-----&-~TMJ22
;: I 8
I TMn)
~
~
550
650
750
Frequency (MHz)
CC-10l.5 cell TM cutoff frequencies, range of resonant
frequencies, and observed resonant frequencies
29
-------
The modes and associated resonances which have field distributions
similar to the TEM field will be strongly excited by driving the cell with
coax; for example, the TEll (perturbed) mode which has vertical electric
fields pointed up above the septum and down below it at some instant in
time. The TElO (unperturbed) mode is poorly excited because the electric
fiel d is in the same directi on above and below the septum [24, 25J. The
possible existence of a mode or resonance does not mean that it is
necessarily excited by a given driving field. Here the septum is driven to
create a TEM driving field; if a particular mode is anti-symmetric with
respect to the TEM field some nonsymmetry may have to be introduced for the
mode to be exci ted. If a transmitti ng antenna were pl aced in the cell wi th
the septum passive, a mode might or might not be excited depending on
location and ori entati on of the antenna. Thu s, if the cell were used for
emission testing (i.e., detecting the emitted fields caused by a device
inserted in the cell), one woul d expect different hi gher modes to be
strongly excited since the coax TEM fields are not the fields causing
excitati on. Emi ssi on testi ng or "radi ated emanati ons measurements" are
discussed by Crawford [26J.
Modal Measurements
Us i ng a measurement system that s imul taneously di sp 1 ays return loss and
the magnetic or electric field detected by a single polarization field
detection probe, a number of measurements were made of the longitudinal
fields. Measurements were made of H for TE modes and E for TM modes,
z z
principally at the test point with the cell s terminated in 50 r2. (When
swept insertion Joss measurements were compared to swept return loss
measurements, the return loss method appeared to detect modal
characteristics different from T~1 operation with greater sensitivity.)
Measurement System
A block diagram of the modal measurement system is shown in Figure 20.
The spectrum analyzer, tracking generator, and directional coupler form a
swept return loss measurement and displ ay system. On the spectrum analyzer
OdB return loss is at the top reference line; 20 dB return loss, for
example, would be two divisions below the reference line. Return 10s5 is
30
-------
RF OUT
H P 8444A TRACKING GENERATOR ATTENUA-
TOR
AMPLIFIER
0 FILTER
(changes de-
pending on
frequency
and power)
HP 8554L RF IN
B SPECTRUM
ANALYZER NARDA
RF SECTION 3020A
BI-DIRECTIONAL TEM CELL
HP 8552B 0 COAXIAL COUPLER
IF SECTION PENLIFT t
HP 141T 0
DISPLAY SECTION
W SCAN OUT
--'
TEKTRONIX
7623A
OSCILLO-
SCOPE
Fi gure 20.
Z BLANKING
15K Q
REFLECTED SIGNAL
~
HORIZONTAL
7A22
7A18
DIFFERENTIAL
INPUT
o
o
VERTICAL
ITHACO 4251
ELECTRONIC
FILTER (Low Pass)
TEKTRONIX
AM 502
DIFFER-
ENTIAL
AMPLIFIER
Modal measurement system block diagram.
D
HP 8481A SENSOR
-------
defined as -20 10910P, where P is the magnitude of the voltage reflection
coeffi ci ent. Di fferent ampl ifi ers and fi lters, dependi ng on frequency, are
used between the tracking generator and directional coupler to increase
power and decrease harmonic distortion. The field measurement system used a
single axis of a modified Holaday production model three axis E or H probe
(Holaday model HI-3001 used without associated read-out meter), a
differential amplifier, a low pass filter, and an oscilloscope. The
oscilloscope horizontal axis is driven by the spectrum analyzer scan out and
pen 1 ift si gnal s, so there is correspondence between frequency and the
horizontal axis of both the spectrum analyzer and oscilloscope. CRT
photographs were taken of the spec trum analyzer and osci 11 oscope. Scan
rates were slow (typically 10 see) with low pass filter values of 10 Hz to
increase signal to noise ratio. The cell was terminated in 50~; alterna-
tivelya power sensor was used to determine absolute power out of the cell.
Return 10ss measurements are not corrected for variations in coupling of
the directi onal coupl er withi n each measurement, so that wi deband
measurements at lower frequenci es are 1 ess accurate than the narrow band
measurements.
Probe Calibration
The probe connected to the differential ampl ifier, filter, and
oscilloscope was calibrated at 50 MHz only, in the CC-101.5 cell operated in
the standard, automated confi gurati on to establ ish known fi el d strengths.
The electric field probe axis is lined up with the TEM E field and the
magnetic probe axis with the Hx field. The probe dCYoutput voltage
corrected for the differential ampl ifier gain is measured on the
oscilloscope. The results are given in Table 4 and 5.
TABL E 4.
MAGNETIC FIELD PROBE CALIBRATION
Applied Field
Probe Vo 1 tag e
(A2/m2)
0.001
0.01
0.1
1
(mV)
0.2
2.0
14
62
32
-------
T ABL E 5.
ELECTRIC FIELD PROBE CALIBRATION
Applied Field
Probe Voltage
(V2/m2 )
0.1
0.5
1.0
10
100
500
1000
10,000
(mV)
0.02
0.1
0.2
1.9
15.2
52
82
340
Up to 0.01 A2/m2 and 10 V2/m2 the probes, when operating into the
1 M ~ input impedance of the differential ampl ifier, have a square law
response. The same calibration is used at all frequencies. This calibration
combi ne d wi th the power measurement allows an ap proximate compari so n between
the expected TEM fiel ds (H and E ) and the measured resonant fiel ds (H
x Y z
and E ).
z
Resul ts
To detect the resonances associated with the TE modes, the Hz component
of the fiel d was measured. Figure 21 shows the results for the H
z
measurements in the large CC-101.5 TEM cell. The location in the cell in
(x,y,z) coordinates is given. The vertical scale is given in ~V/div at the
input to the differential amplifier. The horizontal axis is frequency, either
start and stop frequency or center frequency and span per division are given
for both return loss and field plots. For up to 2000~V the conversion factor
for magnetic field is 5~A2/m2/~V see Table 4. From the measured power the
expected TEM fiel d strengths are calculated from the matched cell expressions
at the test point:
2 PZo
E --
y - b2
E 2
H 2- -L.
x - 3772
(31 )
33
-------
Using Z = 52.0n and b = 0.153 m for the CC-105 and Z = 51.3n and b =
o 0
0.401 m for the CC-101.5 results in the following expressions in convenient
units.
CC - 105
CC - 101.5
E/ (~) = 2.22 .PCmW)
E/ (~) = O.319.PCmW)
( 2)
2 ~A
Hx m2 = 2.24.P(mW)
(32 )
( 2)
2 ~A
Hx m2 = 15.6.P(mW)
Large changes in H were seen as a function of position in the cell,
z
thus some measurements were made away from the test point. The (+w,+b/2,0)
measurements were made at the point halfway between the septum and cell side
wall, and directly above the septum edge. The CC-101.5 Hz measurements
were made between 100 and 300 MHz. Figures 21 a and b give the fields and
return loss from 100 to 200 MHz at the septum edge position. The second
field peak at 153 MHz was not observed at the test point. Only amplifier
noise was seen between the two peaks, implying that the cell is usable
between resonances. The return loss val ues are not corrected for coupl er
directivity, which was specified to be better than 35 dB. In all cases some
change in return loss is seen for each field strength peak. Figures 21 c
and d are narrow-band measurements around the peak at 120.27 MHz; the probe
was at the test point with the peak Hz field strength 14 dB above the
expected TEM Hx value, even though the return loss was 30 dB. Figure 21
e,f,g show two measurement locations from 200 to 300 MHz. Strong return
loss peaks are observed with relatively weak fields at the test point,
however above the septum edge the field is very strong at 213 MHz. Figure
21 h-m are narrow band examinations of the peaks at 213, 224.7, and 264.5
MHz. Table 6 summarizes the results.
The Hz peaks must be TE resonances. On ly the predi cted
f re s (011) = 122.55 MHz is low enough infrequency to be the measured peak
at 120.27 MHz. The peak at 213 MHz is probably the fres (111) resonance
since it is expected to be strongly excited and has been predicted and
measured by Tippet [27] at the National Bureau of Standards. Note that in
Figure 21 m the return loss reference at 0 dB is one division lower than
normal.
34
-------
w
(:)
<>::
t-;>
-' .-
oT.)
»
w::!.
ro'"
o
cr:
CL
ill
(.')
<>
':;13
0--
»
ill::!.
ro~
o
cr:
CL
ill
(:)
«
':;:::
o:?
»
ill::!.
ro'"
o
a:
CL
a.
OV
c.
6.
120.27 MHz
50 kHzldiv
OV
200 MHz
f.
300 MHz
ill
(:)
<>::
t->
-' .-
O:?
»
w::!.
roo
0'"
cr:
CL
OV
200 MHz
300 MHz
b.
OdS
(j)
(j)
0>
....Jij
Z --.
a:'{5
:Ja
i--
ill
a:
100 MHz.
200 MHz
OdS
(j)
(j)
0>
-1:.0
Z --.
cr:'{5
:JO
t--
ill
cr:
120.27 MHz
50 kHz/div
g.
OdS
(f)
UJ
0>
-1:.0
z'-
cr:'{5
:JO
t;j-
cr:
200 MHz
300 MHz
Figure 21(a-g).
CC-10l.5 cell TE measurements.
35
-------
ill
--3.-.
~~
W=:\.
CON
o
g: OV
I
213 Mhz
0.5 MHz/div
ill
-' .-
O:?
»
w:::t
m'"
o
cr:
a.
OV
224.7 MHz
200 kHz/div
UJ
-' .-
~~
W=:\.
CDL()
o
cr:
a.
OV
I
264.5 MHz
05 MHz/d,,/
Fi gure 21 (h-m) .
OdS
(f)
(f)
0>
-''0
z~.
cr:cg
:JO
I-~
UJ
cr:
213 MHz
0.5 MHz/div
k
OdS
(f)
(f)
0>
-l:C
Z "
a::m
:Jg
I-~
UJ
a::
224.7 MHz
200 kHz/div
m
OdS
(f)
(f)
0>
-''0
z "
a::m
:Jg
I-~
w
a::
,
264.5 MHz
0.5 MHz/div
CC-10l.5 cell TE measurements.
36
-------
TABLE 6. CC - 101.5 CELL TE MEASUREMENTS
See H2 H2 Return
z x
Fi g. 21 Location Power Frequency Peak Field T EM Fie 1 d Loss
(1 etter) (x,y , z) (dBm) (MHz) (~A2 /m2) (~A2/m2) (dB)
c,d (0,b/2,0)* 10 120.27 525 22 30
a,d ( w, b/2 , 0) 7 120 88 11 30
a,b,d (w,b/2,0) 7 153 37 11 35
e,g ( 0, b/2 ,0) 6 213 15 9 65
e,g (0,b/2,0) 6 224.7 25 9 33
e,g ( 0 , b/2 ,0) 6 264.5 25 9 0
e,g (0,b/2,0) 6 276 13 9 25
f,g ( w, b/2 ,0) 6 213 1250 9 5
h, i (0,b/2,0) 8 213 46 14 8
j,k (0, b/2, 0) 8 224.7 45 14 37
1, m (0,b/2,d/4) 8 264.5 145 14 0
* te s t poi n t = (0, b/2 , 0 )
HZ measurements in the smaller CC-105 cell were all made at the test
point between 300 and 914 MHz. No resonances were seen below 300 1'~Hz. Both
cells were penetrated with the probe through an oblong hole (b/2,b,b/2) with
dimensions approximately 3 cm x 5 cm, such that the probe could be rotated
at its "analyti C" angl e of 35.26~ between the probe handl e and the upper
cell wall [28J. Normally the cell doors were shut, however for Fi gures 22 a
and b, one door remained open. If the doors are both closed the 320 MHz
small dip in return loss and peak in H are not seen. It is sufficient to
z
cover the 1/3 of the door closest to the septum with conductive material, in
order to suppress the peale The 120.27 MHz Hz peak is seen in the large
cell regardl ess of whether the doors are open or shut. Presumably, the
TE01 mode requires some non-symmetry to be excited and the 1 arger cell has
more deformations than the more rigid small cell. The effective length for
the TE011 resonance is approximately d3; i.e., to the ti ps of the
tapers. Figures 22 c and d show an interesting gradual rise in the field
strength starting at 420 MHz and leveling off at 480 MHz with Hz =
300 ~i /m2 compared to an expected Hx TEM fiel d val ue of 175
~A2/i. This rise implies a modal propagation with a lower cutoff
frequency rather than a resonant character. In Figures 22 e and f the strong
peak at 565 IvtHz is probably the f res (111) frequency for the TEll mode.
37
-------
w
c:J
«>
1-'6
....I '-
0>
>:::t.
WO
Q)g
0'-
cr
a..
a.
w
c:J
«
~.~
o~
»
w:::t.
IDN
o
cr
a..
OV
300 MHz
c.
w
c:J
«>
~.-
O~
»
W:::t.
ID;:'
o
cr
a..
ov
400 MHz
500 MHz
e.
ov
440 MHz
640 MHz
Figure 22 (a-f).
(f)
(f)
0>
....1'6
z~
crlD
::>g
1-'-
w
a:
(f)
(f)
0>
-' .-
'0
Z '-
a:f/5
::>0
I-~
W
cr
b.
OdB
(f)
(f)
0>
....1'-
'0
Z -,
crf/5
::>0
1-'-
UJ
a:
300 MHz
400 MHz
d.
OdB
400 MHz
500 MHz
f
OdB
440 MHz
640 MHz
CC-105 cell TE measurements.
38
-------
g.
UJ
<:J
«>
I- .-
...J"O
0"""
>~
UJO
COO
o'
-J'S
z"
a:fg
::>0
I-~
UJ
a:
700 MHz
750 MHz
i.
UJ
<:J
«>
I- .-
-J""Q
0>
>:::t
lJO
.no
ON
a:
CL
j.
OdS
U)
U)
0>
-J'6
z"""
a:fg
::>0
I-~
UJ
a:
OV
814 MHz
914MHz
814 MHz
914 MHz
Fi g u re 22 (g- j) .
CC-105 cell TE measurements.
Again, in Figures 22 g and h there is a gradual rise and leveling off of
field beginning at 715' r'4Hz and ending at the 733 MHz resonance. Between 640
and 700 MHz, and 750 to 814 MHz no marked changes in return loss and no
measured Hz occurred. Table 7 summarizes the results for peaks seen in
Fi gure 22.
39
-------
T ABL E 7. CC - 105 CELL TE r.£ASUREMENTS
(all measurements at the test point)
See Return
Fi g. 22 Power* Frequency H2 Peak Fi el d H2 T EM Fie 1 d Loss
z x
(letter) (dBm) U4Hz) (IlA2 /m2) ( IlA2 /rrf ) (dB)
a,b 5.5 320 38 55 35
c,d 10.5 408 10 175 23
c,d 10.5 446 40 175 26
e,f 10.0 565 18,000 156 9
e,f 10.0 569.6 7,500 156 22
e,f 10.0 595 2,000 156 27
g,h 10.0 703 720 156 6
g,h 10.0 733 7,600 156 36
i,j 9.5 864 5,000 139 0
i,j 9.5 885 5,600 139 17
i,j 9.5 893 2.200 139 18
* :r(). 5 dB
TM fields were measured using a single axis of a Holaday electric field
probe (Holaday model HI-3001 used without associated readout meter) pointed
in the z direction to measure E. The probe calibration is given in Table
z
5. The first TM resonance is at the TM cutoff frequency and is independent
of the resonant length and is thus well-defined. The similarity between
results in Figure 23 for the larger CC-101.5 cell and in Figure 24 for the
CC-105 cell is encouraging. In both figures part (a) is the field
measurement at the test poi nt with the three pri nci pa 1 peaks correspondi ng
well to: fres (110) and fres (120) for the first peak. fres (210) and
fres (220) for the second peak which is split. and fres (320) for the
third peak. Part (b) of both figures is at the same frequency range except
the probe is moved to (0,b/2,-d/4) longitudinally down the cell. At this
location two new peaks appear with the previous peaks still visible. The
frequencies for the peaks seen in Figure 23 bare 403. 427, 433, 472, 477.
and 529 MHz. The frequencies for peaks in Figure 24 bare 1061, 1136, 1153.
1255, 1268. and 1409 MHz. A search was made for E peaks below these
z
frequencies and none were found.
Automated runs were made between 100 and 150 ~1Hz with an isotropic
magnetic probe at the test point (see Section 6). the peak at 120 MHz in the
larger cell was wider and less intense than in Figure 21 c. The same
40
-------
a
Figure 23. CC-10l.5 cell TM
measurements.
LU
"
~
...J
0
>
LU.:!:
ID~ C.
0> _.
a:E OdS '''~
Q..q-
ov
(/)
350 MHz (/)
550 MHz 0>
...Jij
b. z-'"
a:1D
:J;g
I-~
LU
a:
LU
"
~
...J
0 350 MHz 550 MHz
>
LU.~
ID'O
0>
a:E
Q.(\I
OV
350 MHz 550 MHz
a.
LU Figure 24. CC-105 cell TM
"
~ measurements.
I-
...J
0
>
LU.~ C.
ID'O
0> OdS -
a:E
Q.~
(/)
OV (/)
0 >
1000 MHz 1500 MHz ...Jij
z"
b. a:1D
:J;g
I-~
LU
a:
LU
"
~ 1000 MHz 1500 MHz
o
>
LU.~
ID~
0>
a:E
Q....,.
OV
1000 MHz 1500 MHz
41
-------
runs made with an isotropic electric probe show no peak or dip at the test
point, which is reasonable since in a resonant cavity the magnetic field is
strong where the electric field is weak, and vice versa [29]. If the
electric field is zero due to the resonance, the TEM electric field is not
affected at that point and no dip is seen.
Measurements were made of the electric field uniformity at several
frequencies around and at 120 MHz in the CC-101.5 cell. An isotropic E
field probe was moved in the x direction in the cell at y = b/2 and z = O.
The percent deviation in the field from its value at the cell center is
plotted in Figure 25.
42
-------
+60%
+50%
+40%
+30%
::!:.
0
0
0
...
0 +20%
LLJ
,
)( 0
LLJ
LLJ
+10%
!:
o
',j:
co
'50
(I)
Q
'0
Q)
u::
LLJ
....
!:
(I)
(.)
...
(I)
0..
-10%
-20%
-30%
-40%
0%
120 MHz
,.,,~
"
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
"" 90MHz
. .................
..... ~: .".'..- r II~.....
-,-.-.---..,.
130 ~Hz
110MHz
........
90MHz
110M Hz
120MHz
130M Hz
.............................
"
\ '
\ "
\ "
~ ,
~ ,
~ ,
~ ,
~ ,
~ ,
~...._,'
_'_I.
----
Q)
-50% U
...
(I)
....
!:
(I)
U
-60%
-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30
x (em)
Fig u re 2 5 .
Isotropic probe measurement of TEOll resonance.
43
-------
SECT ION 5
CALIBRATION PARAMETER MEASUREMENT AND ERROR PROPAGATION
The expressions for the electric and magnetic fields at the test point
of an empty unperturbed cell, operating in the TEr1 mode only, including
ref1 ecti ons, require vari ous parameters whi ch are measured characteri sti cs
of the TEM cell and attenuator-1oad system. This section documents how
these parameters were measured and how their errors propagate through the
equations to give a probable error in the field strength. Absolute power
measurement accuracy and signal purity at the amplifier output are also
addressed.
CELL CHARACTERISTICS
Relevant characteri stics of the TEM cell are: dimensional measurements
particularly the plate separation (b), characteristic impedance (Zo)' and
electrical length (1). The characteristic impedance can be obtained by
e
measurement or theoretically from the cell dimensions.
Dimensional Measurements
The measured cross section dimensions of the cells near the longitudinal
center of each cell are given in Figures 26 and 27. Accuracy is probably
better than 1%. Figures 28 and 29 give the values of length used in the
resonance calculations, as well as the taper angles for the larger cell. Up
to 5mm vari ati ons in b are seen in the 1 arge cell over the regi on between
the shiel d and septum, due to the warped surface of the shield.
Characteristic Impedance (Z )
o
The cell
Theoretically
ref1 ectometry
characteristic impedance ~"as determined by three methods:
from the cross sectional geometry, from frequency domain
(FOR), and from time domain reflectometry (TOR).
Geometry
Using the approximate Eq. (8) and the values of a, b,
characteristic impedance of 51.H2 for the CC-101.5 cell
and g results in a
and 52.4 S1 for the
44
-------
--
2w=
91.6em
I
I
9 = --.J
14.3em I
I
I
.1
~
I.
2b=79.4em
--
14
2a=120.2em
.1
Figure 26.
Cross sectional dimensions of the CC-101.5 cell.
--
2w=
33.6em
I
I
g= I
---..I
5.6em I
I
I
-I
..-
I.
2b=30.0em
--
14
2a=44.8em
-I
Figure 27.
Cross sectional dimensions of the CC-105 cell.
45
-------
Fi gure 28.
Figure 29.
1450
Sideview
~ ::::.:::: ~
d3=2.349m
CC-101.5 cell dimensions.
l1=d1=o.450m::Y1
L-= d2=O.670 m=J
d3=O.890m
CC-105 cell lengths.
46
Top view
of
Septum
1340
Top view
of
Cell
-------
CC-105 cell. Results obtained using the exact elliptic functions
significantly different for these geometries; i.e., the difference
than 0.05 ohm.
are not
is 1 ess
Frequency Oomai n Reflectometry
A \'iell known technique called frequency domain reflectometry (FOR) is
often used where the transmission line equation simplifies at 1/4 wavelength
back from the load, and Eq. (10) becomes Z = .rz:ll. Z. is measured
oV L.iL.l . 1
using a vector voltmeter and a directional coupler and Zl is the known
termi nati on impedance. The resul ts of usi ng the techni que for detenni ni ng
the characteri stic impedance were 50.6 Qfor the CC-101.5 cell and 51.0 Qfor
the CC-105 cell. These may be compared with the manufacturer's specified
characteristic impedance of 50.0Q.
Time Oomai n Refl ectometry
A fast rise time domain reflectometry (TOR) system «35 psec) was used
wi th a 50 ohm standard air 1 i ne. The measurements setup consi sted of a
Hewlett-Packard TOR unit, cable, General Radio 50 ohm air line, cell, and a
50 ohm termination in series. Figures 30 and 31 illustrate the results of
TOR tests applied to the two TEM cells studied. A visual average line was
drawn through the data between the tapers to resul tin 51.5 Q for both
cells. Since the reflections due to nonunifonnity at the connector end go
both positive and negative these should approximately cancel at normal
operating frequencies. The large peaks seen at the far end of the plots are
probably due to the fact that the fast rise time pulse includes frequencies
above the cutoff frequencies for higher order modes. The same result was
obtained if the cell was turned end for end and the TOR test repeated.
The resul ts of characteri sti c impedance of the cell s are summari zed in
Table 8. The uncertainty in Z (0.9Qfor the CC-105 cell) includes some
o
effect of nonunifonnity of the cell because the FOR method uses the entire
.
cell while TOR and theoretical methods use only the central section of the
cell between the tapers.
47
-------
+020
+015
+010
Q
g +005
,g
Q;
o
()
c
"
t
Begin Cell
2 6 nsec
u
~
a: --005
-010
-015
-020
Call this cell P = +0015 - 51 50 = Zo
t
End Cell
20 7 nsec
10
12
14
16
18
20
Round Tnp Time (nsec)
Fi gure 30.
CC-10l.5 time domain reflectometry measurement.
+015
+010
+005
Q
c
~
u
'"
Q;
o
()
c
~ -005
~
a:
-010
-015
-020
o
.
AIr line
500
Begin Cell
2.86 nsec
Round Tnp TIme (nsec)
Fi gure 31.
CC-105 time domain reflectometry measurement.
48
500 Termmatlon
22
500 Termmatlon
10
11
12
-------
TABLE 8.
CHARACTERISTIC IMPEDANCE
Theoreti ca 1
FOR
TOR
CC-105
(ohm s )
52.4
51. 0
51.5
51.6 :I: 0.9
CC-101.5
(ohms)
51.1
50.6
51.5
51.0 :I: 0.5
Mean
Electrical Length (1 )
e
Using a vector voltmeter and a dual directional coupler and shorting the
output port of the cell, the lowest frequency where the forward and
reflected voltage waves are in phase at the input port was found. When this
condition exists, there is 1/4 wavelength from end to end of the cell. One-
fourth the free space wavelength associated with that frequency is the
measured electrical length of the cell. Measured results were: 1e =
1.138m for the CC-105 cell, and 1 = 2.78m for the CC-101.5 cell. The TOR
e
plots give approximate confirmation with 1.lOm for the CC-105 cell and 2.72m
for the CC -101. 5 cell. The FOR results are used because of thei r greater
intrinsic accuracy. The FOR method is more accurate because it is a nulling
techni que usi ng a known frequency source. The TOR method depends on the
time domain accuracy of an oscilloscope.
LOAD CHARACTERISTICS
The present high power load or attenuator system used is built from two
(2) attenuators and two (2) cables (see Appendix C for details). These
components fOYTtl a two-port device having a nominal attenuation of 43 dB,
maximum continuous power handling capability of 2 kW, and a maximum measured
input VSWR of 1.10, output VSWR maximum is 1.06 below 220 MHz. A
directional coupler is not used to measure the power through the cell. The
attenuator output is connected to three (3) possible power sensors throughea
pair of microwave switches. Insertion loss through the switches is
negligible (< O.02dB) below 220 MHz. The absolute power meter reading with
49
-------
the known attenuation is used to detennine the power through the cell. The
measured load impedance (Zl) is used to correct the fi e1 ds in the cell for
standi ng waves.
At tenuati on
The most critical measurement is the attenuation or insertion loss of
the attenuator system. The power ratio method was used. Mismatch error and
non-1 ineari ty of the power meter used are the prilnary sources of error in
such a measurement. Variation in source power during the measurements was
<0.02 dB. Us ing techni ques descri bed in a Hew1 ett-Packard App1 i cati on Note
[30J the maximum mismatch error for the ca1 ibration and insertion measure-
ment can be determined from the maximum source, detector, input, and output
VSWR's. Above 10 MHz, to reduce error due to power meter non-linearity, the
system was considered as two parts: (1) the high-pm.,rer input cable into the
30dB, 2kW attenuator, and (2) the output cable into a low power 13dB
attenuator. Thus, two attenuation measurements are made and added together.
The source VSWR was reduced to d.04 by usi ng ampl ification and a 20 dB
pad. The detectors were selected to give a maximum detector VSWR of 1.08 at
10 MHz. For the high power section, input VSWR was <1.10 and output VSWR
was <1.12 resulting in a maximum mismatch error of *0.03 dB. The low power
section input VSWR was d.06 and output VSWR was <1.02, givi ng a maximum
mismatch error of *0.02 dB. Adding these together gives a maximum total
mi smatch error of zO. 05 dB. Typi ca 1 val ues are much 1 es s. Measurements
made with a Fluke model 5101A calibrator indicated a maximum power meter
non-linearity of zO.05 dB for each of these measurements.
The total attenuation measured is shown in Figure 32 as a function of
frequency. Significant mismatch errors would be seen as oscillations in the
measured value and are not observed indicating small mismatch errors. Below
10 MHz a terminated RF mi11ivo1tmeter was used and accuracy was greater. A
least squares fitted attenuation curve, the dashed line in Fig. 32, was used
in the system computer control program. Confidence in the attenuation
obtained by this approach is *0.2 dB. The attenuator system was sent to the
National Bureau of Standards (NBS) for measurement of attenuation and
complex input impedance. NBS attenuation measurements between 500 kHz and
220 MHz matched EPA measurements to within zO.15dB worst case.
50
-------
All of these attenuation measurements were made at low power. Measure-
ments by the manufacturer and confirmed in this study, showed, however, a
decrease in attenuation of up to 0.28 dB at high power levels (up to 2
kilowatts) and high frequency. High power levels result in significantly
el evated attenuator temperatures and a changed attenuati on. An addi ti ona 1
error of :1:0.28 dB is included in the attenuation to give :1:0.48 dB total
attenuation error at high power. Such changes of attenuation with
temperature suggest the desirability of changing to a directional coupler in
the next generation system to measure power for reducing calibration
uncertainties.
45.4
44.4
y;
Iff
pJ
,j
,,;/
-,
I
I
.r
l'
I
I
/
A, 11\ II
f'\ ~~ i--P
-~. -- '.# ~
VIi ~-
''I
45.2
45
""
~ 44.8
c:
o
+-'
'"
::>
c:
~ 44.6
+-'
CI:
44.2
44
10 kHz
100 kHz
1 MHz
10 MHz
(30 MHz/DIV)
220 MHz
FREQUENCY
---------- Fit
Data
(yellow system)
Fi gure 32.
Attenuation characteristics of high power load system.
51
-------
Complex Impedance
A vector voltmeter was semi-automated with an analog to digital
converter interfaced through the IEEE-488 bus to a desktop computer. A
program listing is given in Appendix A. The measurement technique was as
described in a Hewlett-Packard Application Note [31J for 100 I~Hz to 1000
MHz. This technique involves two calibration measurements and one load
measurement of complex reflection coefficient. A synthesizer under
computer control drives an amplifier and attenuator to give a nominal signal
power of 100 mW and low source VSWR. This source drives a dual directional
coupl er, either a 0: Narda model 3020A or a 180: Anzac model CD-920-4. The
first calibration measurement using a coaxial short establishes the coupling
difference between the two sidearm ports, and a 180: phase reference. The
termination (second calibration) measurement combines with the known maximum
VSWR of the termination (d.0027)* to determine the directivity complex
reflection coefficient (fd). Finally, the load measurement at the high
power attenuator cable input was made to give fl. The true reflection
coefficient ( ft) can be determined by ft = fl - fd. The reflection
coefficient of the calibration termination is called 0, its near zero value
results in a final error of Zl of %0.14rl for reasonably small values of
ft. The Smith chart equations [32J are used to determine Zl from ft.
The Anzac coupl er was used from 1 lv1Hz to 60 MHz and the Narda coupl er from
50 MHz to 220 MHz; overlapping values of Zl at 50, 55, and 60 MHz were not
significantly «0.2 ohm) different. A plot of the real and imaginary
components of Zl = Rl + jXl is given in Figure 33, numbers beside the
points indicate frequency in MHz. The ratio accuracy and phase error of the
vector voltmeter are two other significant errors. The manufacturer1s
specified phase error is typically %1", while ratio accuracy is %0.3 dB.
For typical values of Zl' the ratio error results in an error of :r0.25rl
in I Zll resulting in probable total error of %0.4rl. Comparison with NBS
measurements on the same load resul ted in a maximum di fference in the
angular component of the reflection coefficient e of l7~ and a maximum
difference in the magnitude of the reflection coefficient p of 0.004. The
worst case difference in Zl had a magnitude of 0.85rl. This value of
%0.85 rl is used as a conservative estimate of the error in Zl. Temperature
stability for the load input impedance has not been determined.
* All references to maximum VSWR apply to frequencies ~ 220 MHz.
52
-------
4.0 130
180 185
3.0
2.0
1.0 10
140
en
-E 170 30 195
X..c: 0.0
0 35
-1.0
145
-2.0
-3.0
-4.0
155
-5.0
47 48 49 50 51 52 53 54 55 56
Ohms
RI
Figure 33. Load impedance. Number s beside the points indicate frequency
in MHz.
POWER METERS
For absolute power (Pm) measurement at the attenuator output port
above 10 MHz a Hewl ett-Packard model 436A power meter with a model 8481A
power sensor is used between -20 and + 20 dBm. The 8484A power sensor is
used below - 20 dBm. Thi s error ana lysi s foll ows closely Hewl ett-Packard
"Power Meters and Sensors" technical data 8/1/81. Maximum mismatch error is
:1:0.6%. Cali brati on factor vari ati ons below 220 MHz are 1 ess than :1:1%.
Power reference uncertainty is ~O.]"Io. A calibration pad used only with the
8484A sensor adds ~1.1%. Instrumentation uncertainty and sensor po~er
53
-------
linearity adds *1.7%. Zero set, noise, and power reference mismatch
uncertai nty are 0.2% each. Probab1 e total error (root-sum-square) is gi ven
by
~0.62 + 12 + 0.72 + 1.72 + (0.2)2,3
= *2.46% or *0.11 dB
A Boonton 9200 RF millivoltmeter is used with a 50r2 feed-through
termination from 10 kHz to 10 MHz. This meter was tested using a Fluke
model 5101A voltage calibrator with a wide band option which indicated a
maximum error of *0.10 dB.
Thus all
traceab 1 e to
converters and
power measurements are probably within ~0.11 dB (~2.5%)
NBS through Hewlett-Packard or Fluke. Thermal voltage
thermi stor mounts wi 11 be used in the future for di rect
traceability to NBS.
SI GNAL PU RITY
Implicit in much of the analysis of this report is the requirement that
the signal from the generator (linear amplifier) is a pure sine ~"ave at a
single fixed frequency. However, harmonic distortion at high-power levels
and noise at low-power level s limit the dynamic range over which a sine wave
is well approximated.
Harmonic Distortion
A high power 110 MHz low pass filter is always used at the input of the
CC-101.5 cell to keep harmonic distortion small at mu1timode resonant
frequencies. The limited bandwidth of the amplifier is adequate to keep
such harmonics small when using the CC-105 cell. Low pass filters at 30 MHz
and 60 MHz are also used if the operating frequencies are low enough.
Second harmonics within the TE~~ bandwidth of the cells may be as large as
17 dB below the fundamental leading to a :1:2% or :1:0.08 dB power error worst
case.
5[t
-------
Noise
Harmoni cs of the 60 Hz power 1 i ne frequency are more important than
thermal noise in limiting dynamic range. This power supply noise is
documented in Figures 34 through 37. The linear amplifier input was
terminated in 50 nand gain was set to maximum. Figures 34 and 35 show the
noi se at the 100 watt output; Fi gures 36 through 37 show the 1 kW output
noi se. At each ampl ifi er output a 20 dB 50n attenuator was used and both
the oscilloscope and spectrum analyzer have 50n inputs. To obtain the
actual output voltage, multiply the scope voltage by 10. To obtain output
power, add 20 dB to the values indicated in dBm. Figure 34 a and b show the
time domain noise, into a 50 MHz bandwidth oscilloscope, at the 100 watt
output. Peak vol tage is about 260 mV. Spectrum analysi s at the 100 watt
output is given in Figure 35 a, b, c, and d; band c are samples over a
3 kHz span; d shows the thermal noi se over the full bandwi dth of the
amplifier. Figures 36 and 37 give similar information at the 1 kWoutput.
A practical check is to operate the amplifier with the input terminated and
see if the survey instrument being calibrated in the TEM cell responds.
a
b
Fi gure 34.
Time domain noise at 100 watt output.
55
-------
a
05/lt'l/82. 1: 1212 PM
REF -1121. adEm ATTEN l~ dB
Ie dB/
STffiT 0 Hz
R~S BhI 1 kHz
STOP S~0 kHz
SWP 1.513 see
VBI-! 3 kHz
c
05/1~V82. 1:121121 PM
REF -Ie 13 dEm ATTEN 1121 dB
Ie dB/
STFIH 1121121.12113 kHz
RES BH Ie Hz
vaw 30 Hz
STOP 103.00 kHz
SHP 98.13 ..0
Figure 35.
b
1215/10/81. 12:43 PM
REF -113.13 dEm ATTEN U3 dB
Ie dB/
STffiT 0 Hz
RES B~ I e HZ'
STOP 3.121121 kHz
SWP 913.0 see
V8W 38 H;z:
d
05/1121/81. 12:36 PM
REF -1"121.121 dEm ATTEN 1121 dB
10 d.
STffiT 0 Hz
RES BH 1 Hl-lz
vaw 3 Hl-lz
Frequency domain noise at 100 watt output.
a
Fi gure 36.
Time domain noise at 1
b
kW output.
56
-------
a
b
L\, 10 ~:1. ]2:32 P~1
hEF 1 (I (1 dB~-, ATTEt~ 28 o:'IB
135/10/92. 1 :06 PM
REF 10 0 dBm ATTEN 20 dE
--- ---- ---
10 dB
113 dBI'
~. --- -..
- -- --
-~
-:TAF-T (I Hz
~'E'" E'll ! ~ Hz
YEW ~ hH7
ST!)P 500 I.Hz
SWF' 1. SG 1.....:
STffiT 0 Hz
R~S girl Ie Hz
YEW 313 Hz
STOP 3.013 kHz
SHP 98. e s.c
c
d
0':.1082. 1:0:3 Pl1
REF! 0 0 dBro-, ATTEN 20 dB
0':, I C1 :~ I. 1'::,;:-1 F'11
REF !O (1 dB,,, ATTEI! 21] dB
I (1 dB
I (I dE.
STFF'T 10£1.130 ~Hz
t=;'ES Sirl 113Hz
VBW 313 Hz
STOP 1133.00 hHz
SHP 90. B $.C
'':::.TART i3 Hz
RES B~I ] NHz:
YEH 3 MHz
STOP 3013 NHz
SHP 213.13 ",$.:
Figure 37.
Frequency domain noise at 1 kWoutput.
ERROR PROPAGATION
From equation 14 and Pn = Pm' A, the formula for the electric field.
strength squared (proportional to power) is
E2 = I Zi 12 . Pm . A
2
b . R.
1
at the test point,
(33)
where each parameter has some percentage error. When the factors are
.
mul ti pl ied or divi ded the error is propagated usi ng the root-sum-square
(RSS) of the individual percentage errors. The error is multiplied by the
power when powers or roots are taken [33]. Thi s error propagati on step
57
-------
assumes all errors are independent whi ch is not strictly true because Z.
1
and Ri are related, however, most of the terms going into it are "worst
case". This method should result in a realistic estimate of error.
Previous sections give the needed individual errors. Adding, for worst case
error in I Zi I or Ri' the :1:0.90 error in Zo and the :1:0.850 error in
Zl ' for values of Zi near 500, this gives a 4% or :1:20 error in IZil
or Ri. A maximum error in b is :1:1%. Absolute power measurement
uncertainty is :1:2.5%. The ability to measure power (Pm) also includes a
z2% error due to harmonic distortion and a maximum of z2.3% (zO.l dB) error
allowed in computer setting the power meter to read what its calculated
reading should be. This gives a P uncertainty of z6.8% worst case.
m
Uncertai nty in the attenuati on A is zO .48 dB or :1:12%. If!! represents
x
the percent error in x the error equation corresponding to Eq. (33) is:
!! 2 =
E
? 2 2 2
(2'!!lzil( + !!Pm +!!A + (2'!!b)
2
+ !!
R.
1
(34)
=~2.4)2 + (6.8)2 + (12)2 + (2.1}2 + (4)2
= 16.56% or zO.79 dB
for the electric field squared.
Similarly for the magnetic f'ie1 d squared:
2 ~ . Pm' A
H -
- b2. 3772. R.
1
(35)
so 'fi! = ~Z .'Zo)Z + '~m +,~ + (Z .'bIZ + '~i
=~2.2)2 + (4.5/ + (12)2 + (2.1/+ (4)2
= 15.04% or :1:0.71 dB
58
-------
So, the probable error in the fields is .T0.79 dB at the test point of
either empty unperturbed cell operating in the single TEM mode only. The
section on TEM fields throughout the cell cross section (Figures 11 and 12)
gives information on the field variation in a circular zone occupied by the
probe due to mi spl acement and pI\Ysi ca 1 si ze of the probe. If the radiu s of
this circle were 5 cm this error would be :1:0.4 dB for the small cell and
.T0.15 dB for the large cell. This error is in addition to the .T0.79 dB
error at the test point and should be evaluated for the particular probe,
position accuracy, and cell used. The fiel d increases on approachi ng the
septum and decreases when approaching the shield vertically from the test
point, so the effect of the field gradient is probably smaller than the
above would imply.
There are practical problems due to the non-ideal nature of survey
probes whi ch affect the abil ity to use them as transfer standards between
different calibration systems. For example, depending on the angle of
insertion into a cell, an electric field probe will usually read higher as
the portion of the handle near the probe end is aligned with the electric
field. Also a given probe normally reads higher in a smaller cell as if a
conductive slab were between the pl ates. Both probl ems are caused by the
probes perturbation of the empty cell field and its tendency to detect that
peturbed field. These are not problems in the calibration systems but
characteristics of the device under test. These effects are illustrated in
the final section on testing and sample applications.
59
-------
SECT ION 6
TESTING AND SAMPLE APPLICAT IONS
This section includes a few important test runs on the system operation
and some sample outputs which illustrate the system1s applicability to
measurement probe evaluation. It is not an attempt to give extensive test
data on all available meters.
COMPARISON OF PROBE RESPONSE WITH AND WITHOUT IMPEDANCE CORRECTION
A first approximation for the line impedance (Zi) in the cell is to
set the cell characteri sti c impedance (Zo) and the load impedance (Zl)
equal to 500 pure real. This considerably simplifies the expressions for
fiel d strength and e 1 imi nates the need for impedance measurements. Fi gures
38 and 39 show the resul t of not usi ng and usi ng the standi ng wave
corrections respectively. For this particular system, deviations up to
o .45dB can be seen and the probe response woul d be incorrectly concl uded to
oscillate with frequency. The complete outputs are given in Appendix D.
COMPARISON OF PROBE RESPONSE USING TWO DIFFERENT TEM CELLS
Since the field is less uniform in the smaller cell different results
mi ght be expected over a frequency range in whi ch both cell s are operati ng
properly. Fi gures 40 and 41 show results for the 1 arge and small cell s
respectively. The meter response is typically 0.2dB higher in the smaller
cell. This response may be explained by the loading effect of the probe on
the cell. This effect is equivalent to reducing the septum to shield
separation distance thus increasing the field strength.
A simple model useful as a first attempt at quantifying this effect is
an infinite conductive slab of thickness (t) between infinite parallel
plates of separation (b) and applied voltage (V). The probe or meter is
modeled as a slab having some constant equivalent thickness (t). The
e 1 ec tri c fi e 1 d (E) i s give n by:
V
E = b-t
60
-------
NBS EFM-5 SIN 1 Z i =Zo=50+0j
0
"J
.:J ...
I '"
- cci
LJ"] ..
00 .... I I
... - ~
III m If) If) "" III I
Ii1 I[) si I ~ I
I
If) IS! ~ 1Il I CSI In I
(E) I I I ""
-------
HOLADAY 26026
Friday
NoveMber 20, 1981
6.28 PM
Operator. MANTIPLY, E. D
TEM Cell. CC-101.5 (Large)
Driver' AMpllfier Output
Rotation Statistlcs
Freq
(MHz)
Avg. Reading
.(V)/M»
Avg Error
( dB)
Low Error
(dB)
High Error
(dB)
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
100.00
5005.630
5077.690
5070.340
4938.130
5018.620
5093.860
5284.110
5193.410
5256.690
5437 . 190
0.00
0.07
0.06
-0.05
0.02
o 08
0.24
0.16
0.22
0.36
-0 41
"0.31
-0.21
-0.34
-0 21
-0.18
0.03
-0.03
0.03
0.12
0.29
0.34
0.37
0.17
0.34
0.37
0.50
0.34
0.37
o 64
Applied Field
5000 V)/M)
HOLADAY 25026
eI.64
!!I. 50
0.34 eI.37 0.34 eI.37 I Ii!. 34 eI. 37
0.~9 1 I
0.17
'- 1-0." -0"
0 0.12
'-
'- 0 121 1213 121.1213
W -0.03
In
1) -".21
-0.31 -0.34
-0.41
-1
10
20
30
,
40 50 60 70
Fr-equency (MHz)
130
90
100
Fi gure 40.
Fiel d strength probe response in 1 arge cell.
62
-------
HOLADAY 26046
Fr'lday
NoveMber 27) 1981
4.04 PM
Operator, MANTIPLY, E
D
TEM Cell, CC-105 (SMall)
Dr i ver AMp 11 f 1 er 0 tJ l ~J U l'
RotatIon Statistics
Freq. Avg. Reading Avg Error Low Error High Error
(MHz) (V'2/M'2) (dB) (dB) (dB)
10.00 5315 100 0.27 -0.06 0 50
20.00 5408 500 0.34 0.09 0 58
~O.OO 5523.120 0,43 0 20 0.69
40.00 5515 950 0.43 0.15 0 71
50.00 5499.390 0.41 0.17 0 69
60,00 5464.940 0.39 0.12 0 58
70.00 5510.470 0 42 0 . 12 0.66
80.00 5487.390 0.40 0.12 0.61
"'0.00 5510,710 0.42 0 17 0.61
100.00 5364.870 0.31 0.09 0.56
Applied Field ~ 5000 V"2/M"2
HOLRDRY 2b04G
'-
o
'-
'-
W
III
'U
0.58 T .." 10." '" '-', T O.S<
'.5. ! '.", IS 0" f I 8." I
121.1119 . 0.12 0.12 0.12 111.1119
10
-111.00
-I
10
210
310
410 50 813 70
Fr-equency (MHz)
810
90
1010
Fi gure 41.
Fiel d strength probe response in small cell.
63
-------
If bl and b2 and V1 and V2 are the plate separation and voltages
for two systems and t is assumed to be zero and the fields are set equal
then:
V 1 b1
'12=1>2
Now, if t is not zero the ratio of the field El in system 1 to E2 in
system 2 is given by:
E1 bl (b2 - t)
---
E2 b2 (b 1 - t)
solving for t gives:
b1 b2 (E1 / E2 - 1)
t =
b2 (E1/E2) - b1
Modeling the two cells as parallel plates with system 1 the smaller cell
and system 2 the 1 a rger cell and usi ng 0.2 dB for the rati 0 of E1/E2 =
1.(£3 and b1 = 15 cm, b2 = 40 cm then the equivalent thickness of the
probe is t = 0.53 cm. If this model is valid the calibration error in the
large cell versus free fields would be [b2/(b2-t)-l] 100 or 1.3%.
PROBE ORIENTATION CHANGES
Usually the probes are inserted through the side door of either cell
with the shaft parallel to the septum. A more standard procedure is to
insert at the "analytic" angle such that when the shaft is rotated each
probe element is at some point parallel to the field. A Holaday Industries
model HI-3001 electric field probe with 5,000 V2/m2 applied at 100 14Hz
was rotated in the smaller CC-105 cell at a parallel and an analytic angle.
Four repeats were made at each orientation and are summarized in Table 9.
64
-------
TABLE 9. PROBE ORIENTATION CHANGES
Para 11 e 1 Rotati on Ana lyti c Rotati 0 n
Average Error Low Error Hi gh Error Average Error Lo\'/ Error Hi gh Error
(dB) (dB) (dB) (dB) (dB) (dB)
0.30 0.09 0.56 0.53 0.34 0.79
0.28 0.09 0.50 0.52 0.32 0.79
0.27 0.09 0.50 0.53 0.32 0.79
0.30 0.09 0.56 0.55 0.34 0.79
The meter read higher on the average with the analytic rotation but the
isotropicity was not significantly different. This difference between
parallel and analytic position is probably due to perturbation of the E
field by the probe handle, or pickup on the leads parallel with the E field
at the analytic angle insertion.
PROBE POTENTIAL SENSITIVITY
It has been observed that diode based meters having a separate probe and
metering instrument may operate poorly toward the lower end of their
specified frequency range. The entire instrument appears to act as an
antenna or al ternative1y the instrument senses the potential difference in
the field between the meter and the probe. The problem appears to be
negligible above 10 MHz in frequency. This may be caused by the radiation
source impedance of the probe element becoming comparable in magnitude to
the sou rce impedance of the hi gh resi stance 1 eads at low frequency [34].
Using high resistance lead as a replacement for the wire element in an
act i ve monopole antenn a seems to conf i rm thi s concept. At low frequency
(100 kHz) the antenna performs as if the wire element is in place, at VHF
(100 MHz) it performs as if no element is in place.
In order to measure this potential sensitivity a modified version of the
system operati ng program was created. In thi s version the probe was not
rotated but was manually moved vertically in the CC-101.5 cell such that. it
was at the approximate center or test point (C), touching the shield or
ground (G), at the center again (C), touching the septum (S), returned to
65
-------
center and repeated at
fie1 d is approximately
the meter variation
frequencies.
each frequency. The actual vertical variation in the
O.8dB (see Figure 14). As can be seen in Figure 42,
approaches the actual field variation at higher
NBS
1'-10DEL En'1-S
SiN 1
Rppl,ed F,eld: 30 dBV/m
500 kHz
Meter Full Scale: 40 dBV/m
J MHz
4
3
e
'2
~ 1
~
~ e
~=: !
-; \j
'-
'"
L
-~
C G C S C G C S C
-3
C G C S C
G C S C
3 MHz
J0 MHz
'"
111
~
a
L
~-l
'"
"
L
D
'-
~l
I:tI
"0
-2
C G C
S C G C S C
-2
C G C
S C G
c s C
30 MHz
J 00 11Hz
\3
o
L
a
~
'-
'"
L
W-l
!XI
"'IJ
~-1
IQ
15
-~
-~
'-
C G C S C
G C S C
C G C S C
G C So C
Figure 42.
Field strength probe potential sensitivity-
66
-------
MODAL EFFECTS IN LA~E CELL
Using the nonna1 operating program, the effect of the TE01 resonance at
120 MHz in the large cell can be seen with an isotropic magnetic field probe
(see Figure 43). This peak is not seen with the same probe in the small
cell. Also the 120 MHz peak is not seen with an electric field probe at the
large cell test point. The peak is about 2.5dB above the surrounding data at
other frequencies. Thi s resu1 t emphasizes the need to determine the modal
characteristic of cells used for probe calibrations.
~,JRF-' DR
3
r"10 DEL BE 15./ 8t, 3 l
fu
;.::
'-
0
'-
W
!.II
v f:I
w
(0 (')
~ m ~ - m ~ - ru ~ ~
~ ~ n w ~ ~ w ~ m
D ~ ~ ~ ~ D Q ~ I
I I I 1 I I I I
:Iifi!EIJ::I:!I:
~ m ~ ru m Q ~ 00
~ ~ ~ ~ ~ ~ ~ ~ ~
t;JQ6~m~,:;;' I
I I I I I I I 9
-[
-2 J
,
100
,
110
Fi gure 43.
Modal effect in CC-I01.5 cell.
v
d
ID
r'-
~ is'
III
ISI
I '~ m
.::;1 ~ U)
LD IS! ~
~ .
6II?~ro
I Q ~ ('":, V
Li3, I 1 ISI, r-. r-. 1('1 ,:v)
. '.-o;,wfWOJUl
IS' GI '. - en CD ']1 ~
I I I i!I'!J."';: .(TJ,~
~~ "if I 1r:;JL:.J,:sI~'
9; u-.1IIII-~~I' 7"
I cO:: ~ I~J ~ f]) r'- '.1. f'- J I I
m . ~ ~ ~ ~
IISI "~'~~""~--N
117 - - ~ 7 - ~ N
I I I I"': 7 "
, !
140 150
I
120
Fr-equenc)'
,
130
(MHz)
67
-------
REFERENCES
1. Stutzman, W. L., and Thi e1 e, G. A.
Wiley and Sons, Inc., New York, 1981.
2. Numerical Electromagnetic Code (NEC) - Method of Moments. Naval Ocean
Systems Center, San Diego, CA 92152, latest revision January 1981 NOSE
Technical Document 116, Vol. 2.
Antenna Theory and De si gn.
John
3. Crawford, M. L., and J. L. Workman. Using a TEM cell for EMC
Measurements of Electronic Equipment. NBS Technical t~ote 1013, U.S.
Department of COlTlllerce, Boulder, Colorado, 1979. p. 8, Eq. (6).
4. Collin, R. E. Field Theory of Guided Waves.
York, 1960. P p. 119- 123.
McGraw-Hi 11 Book Co., New
5. White, J. F. The Smith Chart:
p. 51, Eq. (21), November 1979.
An Endangered Species.
Mi crowave J.,
6. Lorrain, Paul and D. R. Corson. Electromagnetic Fields
W. H. Freeman and Co., San Francisco 1970. pp. 566-568.
7. Crawford, p. 2, Eq.(1).
and Waves.
8. Weil. C. M. liThe Characteri stic Impedance of Rectangular Transmission
Lines with Thin Center Conductor and Air Dielectric.'. IEEE Trans.
Microwave Theory and Techniques, Vol. MTT-26(4), pp. 238-242 (April 1978).
9. Tippet, J. C. Modal Characteristics of Rectangular Coaxial Transmission
Line. University Microfilms International, Ann Arbor, Michigan, 1981.
p. 36.
10. Hewlett Packard. Using the Vector Impedance Meters.
86, 1968, p. 29, Eq. (1).
App1 ication Note
11. Tang, K. Y. A1 ternati ng-Current Ci rcuits.
Scranton, Pennsylvania, 1961, p. 92.
Internati ona1 Textbook Co.,
12. White, November 1979.
13. Tippet, pp. 47-52.
14. Collin, p. 120.
15. Kuci a-Korni ewi cz, H. Conunent on IIFrequency Range of Large Sca1 e TEM Mode
Rectangular Strip Linesll by Weil, Joines, and Kinn; Microwave J.,
November 1981, pp. 93-100. Microwave J., p. 113, April 1982.
16. Tippet, Chapter VI.
17. Gruner, L. Higher Order Modes in Rectangular Coaxial Waveguides.
Trans. Microwave Theory and Techniques.
IEEE
68
-------
18.
20.
21.
22.
23.
Gruner, L. Estimati ng Rectangul ar Coax Cutoff.
pp. 88-92, April 1979.
Microwave Journal, 22,
19.
Hill, D. A. Bandwidth Limitations of TEM Cells due to Resonances.
Microwave Power, 18(2). 1983.
J. of
Tippet, p. 123.
Ti ppet, pp. 219-223
Co 11 in, p. 189, E q . ( 55) .
Reference Data for Radio Engineers, 5th Ed., Howard W. Sams and Co. Inc.,
1968, Chapter 23, p. 19.
24.
Brackelmann, W. Wellentypen auf der Streifenleitung mit rechteckigem
Schinn. Archive Fur Electronik Und Ubertragungs-Tecknik. Vol. 21,
December 1967, p. 647.
Saad, T. S. (Editor), R. C. Hansen and G. J. Wheeler (Co-editors),
Microwave Engineers Handbook. Artech House Inc., Dedham, Mass., 1971,
VoL 1, p p . 1 45- 14 7 .
25.
26.
27.
Crawford, Section 5.
Ti ppet, p. 95.
28.
Bowman, R. R., Some Recent Developments in the Characterization and
Measurement of Hazardous Electromagnetic Field in Biological Effects and
Health Hazards of Microwave Radiation, Polish Medical Publishers, Warsaw
1974, pp. 221-222.
29.
Ishii, T. K., Microwave Engineering.
1966.
The Ronal d Press Co., New York,
30.
Hewl ett Packard. Hi gh Frequency Swept Measurements.
183, December 1978, Apendix C.
Ap p1 i cti ons Note
31.
Complex
1-1000
MHz.
Hewlett Packard. Measurement of
Application Note 77-3. April 1,1967.
White, November 1979.
Impedance
32.
33.
Bevington, P. R. Data Reduction and Error Analysis for the Physical
Sciences. McGraw-Hill Book Co., 1969. pp. 61-62.
34.
Schelkunoff, S.A., and H.T. Friis. Antennas Theory
John Wiley and Sons, Inc., New York, 1952. Chapter 10.
and
Practice.
69
-------
APPENDIX A
LOAD IMPEDANCE PROGRAM
! "LOADZV, F"
!
Storet$=" ,F8,0"
Qu 0 te$=CHR$ (34)
NsaMp=10
OUTPUT 9j"Recall tiMe"
ENTER 9jA$
Date$=" "&A$[1j2J&"/"&A$[4j2J&"/82,"
MASS STORAGE IS Storet$
DEG
PRINTER IS 16
EXIT GRAPHICS
PRINT PAGE,"PROGRAM, LOADZV",Date$jLIN(1)
PRINT" Mass Storage defaults to, "jQuote$&Storet$&Quote$
PRINT" SaMples per Meter reading' "jNSaMpjLIN(1)
SHORT Ckyl(8),Rho(500),U(500),V(500),A1(500),A(500),Thetac(500),Rl(500),Rl
SHORT Theta,Phi,Freqx(500),Start,Stop,Inc,Freq,Rho_t,Theta_t,Swr,Xl(500),X
REV 08/04/82
1u
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
1
180 SHORT Size
190 DIM Meas$(1,3)[55J,A$[65J,Z$[40J,Title$[80J
200 INTEGER Ch,Ck
210 MAT READ Ckyl
220 DATA ,1,,3,1,3,10,30,100,300,1000
230 N$=Nu 11$
240 Dg$=CHR$(179)
250 Clrend$=CHR$(27)&"K"
260 Ch$(O)="A,"
270 Ch$(1)="B,"
280 Meas$(1)="
290 Heas$(2)="T
300 Heas$(3)="
310 C(2)=1,2
320 C(1)=1
330 S=1
340 PRINT "A/D Calibration Settings"
350 PRINT" Channel t1'"jC(1)j" V"jTAB(25)j"(PHASE)"
360 PRINT" Channel t2'"jC(2)j" V"jTAB(25)j"(AMPLITUDE)"
370 PRINT" Channel .4, HI-HI"jTAB(25)j"(PHASE UNLOCKED Light)"jLIN(1)
380 INPUT "Enter the default Phase Range to use for this MeasureMent",Phs_def
390 IF (Phs def=6) OR (Phs def=18) OR (Phs def=60) OR (Phs def=180) THEN 440
400 BEEP - - - -
410 DISP "Error in Default Phase Range" ,Re-enter"
420 WAIT 1 50 0
430 GOTO 380
440 Enter_freq, Tf=500
450 Nf=O
"60 Z$="R"
470 INPUT "Do you want to enter seperate frequencies or a range(S or R)",Z$
480 IF Z$[1j1J="R" THEN Range
490 IF Z$[1j1J()"S" THEN' 460
500 Z$=Null$
510 Freqx(Nf)=O
520 DISP "Enter Frequency (MHz) t"jNf.1,Z$j
530 INPUT Freqx(Nf)
540 IF Freqx(Nf)=O THEN Exit
550 IF (Freqx(Nf»=,01) AND (Freqx(Nf)(=2400) THEN 580
560 CALL Dcol("Specified frequency is out of range, 0(f(=2400, try again,")
570 GO TO 510
S H 0 R T MEA SUR E HEN T"
E R M I N A T ION MEA SUR E M
LOA D MEA SUR E HEN T"
!calibration for ch .1 for
!calibration for ch .2 for
E NT"
the 59313A
the 59313A
70
-------
580 Nf=Nf+1
590 Z$~"Press CONTinue to Exit"
600 PRINT" Frequency 1 ",Nfi" = "iFreqxCNf-1)
610 GOTO 510
620 Range' Inc=1
630 INPUT "Enter Start, Stop, and IncreMent Frequencies",Start,Stop,Inc
640 IF CStart<' Oil OR CStop)2400) OR CStart>2400) OR (Stop<. Oil THEN Range.
650 IF ABSCStop-Start)/Inc+1C=Tf THEN 680
660 CALL DcolC"Too Many frequencies, change diMension size.")
670 GOTO Range
680 FOR Freq=Start TO Stop STEP Inc
690 FreqxCNf)=Freq
700 Nf=Nf+1
710 NEXT Freq
720 Exit, IF Nf)O THEN 750
730 CALL Dcol( "No frequencies defined, try again.")
740 GOTO Enter_freq
750 Tf=Nf-1
760 REDIM FreqxCTf),RhoCTf),UCTf),VCTf),A1CTf),ACTf),Th~tacCTf)
770 INPUT "Enter AMplitud~",AMpl
780 IF AMpl)-10 THEN AMpl=-10
790 CALL CfreqCFreqxCO»
800 CALL SetaMpCAMpl)
810 PaMpl=AMpl
820 INPUT "Enter a new aMplitude or press CONTinue to proceed",AMpl
830 IF AMpl()PaMpl THEN 800
840 PRINT" AMplitude, "iAMpl," dBM"
8S0 LINPUT "Enter the title for the MeasureMent",Title$
860 ForMat, IMAGE 1X,4D.DD,3X,DDZ.DD,2X,MDZ.DD,2X,Z.3D,2X,SDDZ.D,3X,DZ.3D,6X,D
Z.DD,8X,DZ.4D
870 D$="N"
880 P$=Null$
890 INPUT "Do you want to keep th~se XL's and RL's? CYES or NO)",D$
900 IF D$[1,1]="N" THEN 1060
910 INPUT "Enter a unique 6 character file naMe",F$[1i6]
920 P$=Quote$~TRIM$CF$)~Quote$
930 ASSIGN Ii TO F$,New_file
940 IF NOT New_file THEN 1010
950 Size=INTC20*Tf/256)+4
960 PRINT "File, ",P$," will be CREATEd for storage of",Tf*S,"NuMbers."
970 PRINT "CoMputed size is",Sizei"Records."
980 CREATE F$,Size
990 ASSIGN Ii TO F$
1000 GOTO 10S0
101 0 D$= "N "
1020 DISP "WARNING, File ",P$i" already cataloged on disk. REPLACE? CYES or
NO) "
1030 INPUT D$
1040 IF D$[1,1]="N" THEN 870
10S0 PRINT 11,Title$
1060 PRINTER IS 0
1070 PRINT LIN(2)i"LOADZV Run on "iDate$
1080 IF P$[1,1]=Quote$ THEN PRINT "Data will be SAVEd on, ",P$
1090 PRINT LIN(2)iTABC(80-LENCTitle$»/2),Title$
1100 PRINT LIN(3)i"Frequency RI Xl Rho
oss MisMatch Lossl'
1110 PRINT" (MHZ)
(dB) "
1120 PRINTER IS 16
1130 FOR I=S TO 3
(OHMS)
(OHMS)
CDeg)
71
Theta
VSWR
Return L
(dB)
-------
1140
1150
1160
1170
1180
1190
1200
1210
1220
1230
1240
1250
1260
1270
1280
1290
1300
1310
1320
1330
1340
1350
1360
1370
1380
1390
1400
1410
1420
1430
1440
1450
1460
1470
1480
1490
1500
1510
1520
1530
1540
1550
1560
1570
1580
1590
1600
1610
1620
1630
1640
1650
1660
1670
1680
1690
1700
1710 Nn 1 ,
1720
1730
DISP "Press CONTinue to begln with"jN$j" MeasureMent."
PAUSE
N$=II the next n
PRINT PAGEjLIN(6)j" Vector
PRINT" AMplitude, CH
PRINT" Phase, Range,
CALL CenterC1,40,Meas$CI»
CALL PositC5,51,"Frequency'")
CALL PositC6,51, "Meter Val,")
CALL PositC19,1,"SysteM Status")
FOR J=O TO 2
CALL Set aMp CAMpi)
CALL CfreqCFreqxCO»
GOSUB Setprd
Cony=CCCh)/1022
Ch=2AINTCCh-1)
Instr$="H"~VAL$CCh)~"AJ"
DISP "Press STOP to terMinate the prograM."
K=O
Freq=Freq xC K)
CALL CfreqCFreq)
CALL PositC5,63,VAL$CFreq)~" MHz
WAIT 1000
R ec = 0
Ayg=O
FOR Ck=1 TO NsaMp
OUTPUT 706,"H8AJ"
Dig=SHIFTCREADBIN(706),-B)+READBINC706)
IF Dig<-1022 THEN 14BO
BEEP
CALL PositC20,5,"Phase unlocked. .reset freq range.
PAUSE
CALL PositC20,5,A$)
GOTO 13BO
OUTPUT 706 USING ".,K"jlnstr$
Dig=SHIFTCREADBIN(706),-B)+READBINC706)
Ayg=Ayg+Dig
NEXT Ck
Volts=Ayg/NsaMp*Cony
IF J=2 THEN 1590
IF CVoltsC.25) OR CVolts>1.05) THEN GOSUB Recoy
IF Rec THEN 1370
AMp=AMp_rangex*Volts
CALL PositC6,63,VAL$CPROUNDCAMp,-4»~"
GOTO 1640
IF CVolts<-.5) OR CVolts>.5) THEN GOSUB Recoy
IF Rec THEN 1370
Phi=Phs_offset+Phs_range*2*Volts
CALL CkangCPhi)
CALL PositC6,63,VAL$CPROUNDCPhi,-3»~Dg$~"
IF IC>1 THEN Nn1
IF J=O THEN A1CK)=AMP
IF J=1 THEN A1CK)=AMp/A1CK)
IF J<>2 THEN Nnx
ThetacCK)=1BO-Phi
CALL CkangCThetacCK»
GOTO Nnx
IF J=O THEN ACK)=AMP
IF J=1 THEN RhoCK)=AMp/ACK)/A1CK)
IF J<>2 THEN Nnx
VoltMeter ParaMeters"
. Range/MV, "
OHset/dg' ..
!set aMplitute AMp I dBM
" )
" )
" )
72
" )
-------
1740
17S0
1760
1770
1780
1790
1800 Nn2,
1810
1820
1830
1840
18S0
1860
1870
1880
1890
1900
1910
1920
1930
1940
atch_loss
19S0
1960
1970
1980 Nnx,
1990
2000
2010
2020
2030
2040
20S0
2060
2070
2080
2090
2100
2110
2120
2130
2140
21S0
2160
2170
2180
2190
2200
2210
2220
2230
2240
2250
2260
2270
2280
2290
2300
2310
2320
Theta=Phi+Thetac(K)
CALL Ckang(Theta)
IF 1=3 THEN Nn2
U(K)=Rho(K)*COS(Theta)
V(K)=Rho(K)*SIN(Theta)
GOTO Nnx
UM=Rho(K)*COS(Theta)
VM=Rho(K)*SIN(Theta)
Ut=UM-U(K)
Vt=VM-V(K)
DenoM=1+Ut*Ut+Vt*Vt-2*Ut
Rl=(1-Ut*Ut-Vt*Vt)/DenoM*SO 0
Xl=2*Vt/DenoM*SO.0
Rho t=SQR(Ut*Ut+Vt*Vt)
Swr~(1+Rho_t)/(1-Rho_t)
Theta t=ACS(Ut/Rho t)*SGN(Vt)
Rtn_loss=-20*LGT(Rho_t)
MisMatch loss=-10*LGT(1-Rho t*Rho t)
IF D$[1IfJ="Y. THEN PRINT IfIFreq;(K),Rl,Xl,Rho_t,Theta_t
PRINTER IS 0
PRINT USING ForMat,Freqx(K),Rl,Xl,Rho_t,Theta_t,Swr,Rtn_loss,MisM
Rl(K)=Rl
Xl
-------
2330 Setprd. IF J>1 THEN 2490
2340 A$="AMplitude MeasureM@nt on channel ".Ch$CJ)
2350 CALL PositC20,5,A$)
2360 CALL PositC8,23,Ch$CJ»
2370 Ch=2
2380 INPUT "Ent@r AMplitude range in MV?" ,AMp_range
2390 FOR Ck=O TO 8
2400 IF CkuICCk>=~Mp_range THEN 24~0
2410 NEXT Ck
2420 GOSUB Err
2~30 GOTO 2380
2440 Cf=1 . 033
2450 IF Ck/2=INTCCk/2) THEN Cf=.98
2460 CALL PositC8,37,VAL$CAMp]ange)."
2410 AMp_rang.x=AMp_range*Cf
2480 GOTO 2720
2490 Ch=1
2500 CALL 'ositC8,23," ")
2510 CALL PositC8,37," ")
2520 A$-"Phase M.a.ureMent.".Clr.nd$
2530 CALL PositC20,5,A$)
25~0 Ph._rang..Phs_d.f
2550 LINPUT "Enter Phase Meter offset",Phase$
2560 IF Phase$=Null$ THEN 2550
2570 Cpos=POSCPhase$,",")
2580 IF Cpos=O THEN 26~0
2590 Phs_rang.-VALCPha..$11,Cpos-1])
2600 Ph$_offs@t-VALCPhas.$ICpos+1])
2610 IF CPhs_range=6) OR CPhs_rang.-18) OR CPhs_rang@=60) OR CPhs_range=180) TH
EN 2650
2620 GOSUB Err
2630 GOTO 25~0
2640 Phs offset=VALCPhase$)
2650 FOR-Ck=O TO 18
2660 IF ABSCPhs offset)/10=Ck THEN 2700
2670 NEXT Ck -
2680 GOSUB Err
2690 GOTO 2550
2700 CALL PositC9,22,VAL$CPhs]ange)." ")
2710 CALL PositC9,38,VAL$CPhs_offset).Dg.."
2720 CALL PositC20,5,A$)
2730 RETURN
2740 END
27~1 !
2750 SUB CfreqCSHORT F_Mhz)
2760 Hz=F_Mhz*1E6
2770 T=10-LiNCVAL$CHz»
2780 CMd$="/".REV$CRPT$C"O",T).VAL$CHz»."C"
2790 OUTPUT 719;CMd$
2800 SUBEX IT
2810 SUBEND
2811 !
2820 SUB SetaMpCAMp_dbM)
2830 N=3
2840 AMp=AMp_dbM-13
2850 IF AMpCO THEN N=4
2860 T=N-LENCVAL$CAMp»
2870 CMd$=REV$CRPT$C"0",T).VAL$CA8SCAMp»)."C"
2880 OUTPUT 719;CMd$
2890 SUBEX IT
" )
" )
!prograM center frequency for 86608
!prograM aMplitude for 86608
74
-------
2900
2901
2910
2920
2'930
2940
2950
2960
2961
2970
2980
2990
3000
3010
3011
3020
3030
3040
3050
3060
3070
3080
3090
3100
3110
3120
3130
3140
3150
3160
3170
3180
3190
3200
3210
3220
3230
3240
3250
3251
3260
3270
3280
3290
3300
3310
3320
3330
3340
3350
3360
3370
3380
3390
3400
3410
3420
3430
3440
3441
SUBEND
!
SUB Posit(X,Y,Q$)
A=INT180 THEN Theta=Theta-360
IF Theta(-180 THEN Theta=Theta+360
SUBEXIT
SUBEND
!
SUB Lgrid(XMin,XMax,Xdiv,YMin,YMax,Ydiv)
DEG
PLOTTER IS "GRAPHICS"
LORG 2
CSIZE 3.3
LOCATE 20,120,13,93
SCALE XMin,XMax,YMin,YMax
Y corr=(YMax-YMin)*.04
G~id x=(XMax-XMin)/Xdiv
Grid=y=(YMax-YMin)/Ydiv
GRID Grid_x,Grid_y,XMin,YMin
LORG 8
LDIR 90
FOR X label=XMin TO XMax STEP Grid_x
MOVE X_label,YMin-Y_corr/2
LABEL USING ".,K"jX_label
NEXT X_label
LDIR 0
FOR Y label=YMin TO YMax+1E-6 STEP Grid_y
MOVE XMin,Y_Iabel
LABEL USING ".,DZ.D,A"jY_label," "
NEXT V_label
SUBEXIT
SUB END
!
SUB Label
DEG
SETGU
LOR G 6
MOVE 71.50,5.2
LABEL USING "',K"j"OhMs"
MOVE 71.50,2.5
LABEL USING ".,K"j"RI"
MOVE 71.50,98
CSIZE 4.55,9/15
LABEL USING "',K"j"Load
LDIR 90
CSIZE 3.3
MOVE 0,50
LABEL USING "',K"I"XI"
MOVE 3,50
LABEL USING.".,K"j"OhMs"
SUBEXIT
SUBEND
!
IMpedance"
75
-------
3450
3460
3470
3480
3490
3510
3520
3530
3540
3550
3560
3570
3580
3590
3600
3610
3611
3620
3630
3640
3650
3660
3670
3680
SUB Dcol(A$)
Size=LEN(A$)
FOR 1=0 TO Size-1
DISP TAB(Size-I);A$[1;I+1J
WAlT 75
NEXT 1
FOR 1=1 TO 3
WAlT 350
DISP
WAlT 350
DISP A$
BEEP
NEXT 1
WAIT 40*Size
SUBEXlT
SUBEND
!
SUB Center(X,Y,Text$)
A=INT(X)-1
B=INT(Y)-1
L=INT«LEN(Text$)-1)/2)
PRINT USING ".,K"jCHR$(27)&"'a"'VAL$(A)&"r"&VAL$(B-L)'"C"&Text$
SUBEX lT
SUBEND
!center Text$ on line X coluMn Y
76
-------
10
20
30
40
SO
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
250
260
270
280
290
300
310
320
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
480
490
500
510
520
530 I
540
550
560
570
580
585
COM SHORT Freq(45),Rl(45),Xl(45),Ri,Xi,A~PQain
COM SHORT Rl,Xl,Atten,Celln,Pt,Pdb~,Pd(1),El(1),Ro(1),P~val,PQ
COM INTEGER A(730),X2,Prcnt_a~
COM TitletI80J,OpertI80J,Sys'11J
SHORT Freqz(500),Scale,Power,Volts,Valx,X1,X,SaMp(900),flotdb(900)
SHORT AYQi(500),Errdbi(500),P~ini(500),P~axi(500),Xlabel(4)
SHORT Pre_err(500),Post_err(500),Rw(300),Xw(300),Status
INTEGER Np~,I,Year,B(900),Dir(900),Px1(3),Px2(3),Py1(3),Py2(3),Nsg,Paddr
INTEGER Cond,N
DIM Unitst(1,4,1)125J,ZtI50J,Cell_idtI1J,8yste~'11J
,
! 1.2
I
READ Pd(1),Pd(0),El(1),El(0),Ro(1),Ro(0)
DATA 0.401, 0.153
DATA 1 .39, 0 . 569
DATA 51.3, 52.0
READ Xlabel(*),Px1(*),Px2(*),Py1(*),Py2(*)
DATA 180,270,360,90,180,15,75,15,75,50,110,50,110,55,55,10,10,85,85,40,40
S'.CHRt(190) ! ** May be used as squared sy~bol **
S'="'2" ! ** Current definition for squared sy~bol **
Esct=CHRt(27) I ** ASCII code 27 used for escape sequences to
! 2631G printer **
I
!
!
!
!
!
!
!
! 0.0
!
! Model'
!
! 2631G
! 59313A
! 9200
! 436A
I 436A
! 59306A
! 3325A
! 86MB
! 59307A
!
I
! 1. 0
!
!
I 1. 1
!
APPENDI X B
SYSTEI~S OPERATING PROGRNvt
CCCCC EEEEEE L L GGGGG 00000
C C E L L G G 0 0
C E L L G 0 0
C EEE L L G GG 0 0
C E L L G G 0 0
C C E L L G G 0 0
CCCCC EEEEEE LLLLLL LLLLLL GGGGG 00000
HPIB DEVICE ADDRESSES
Select Code .7
01112/83
Description
Address Code
HP-IB Graphics Character Printer
Analog to Digital Converter
Boonton Power Meter
Power Meter SIN 1930A06579
Power Meter SIN 1930A06521
Relay Actuator
Synthesizer IF unction Generator
Synthesized Signal Generator
VHF Swi tch
01
06
08
13
14
16
17
19
22
COM M 0 H,
D A T A
T Y P E
D E FIN I T I 0 H S
AND
PRO G RAM
I NIT I A LIZ A T ION
S TOR AGE
ALL 0 CAT ION
ASS I N G MEN T
o F
V A R I A B L E S
!
Unitst(1,0)="Vn&St&"/~"&S'
Units'(1,1)."A"&S'&"/~"&S'
77
-------
S87
S89
S90
600
610
620
630
640
6S0
660
670
680
690
700
71'0
720
730
740
7S0
760
770
780
790
800
810
820
830
840
8S0
860
870
880
890
900
910
920
930
940
9S0
960
970
980
990
1000
1010
1020
1030
1040
10S0
1060
1070
1080
1090
1100
1110
1120
1130
1140
11S0
1160
Units.12,O)="V/M"
Units$(2}1>="A/M"
Units'13,O)="MW/cM"~S'~"
Units'13,1)="MW/cM"~S'~"
Units.14,O)."dBV/M"
Units.14,1)="dBA/M"
!
Slen=6S0
Storet$=11 :FBjOIl
Quote'=CHR'(34)
Card=7
Printer=1
Version'="01/12/83 "
!
electric"
Magnetic"
I ** Initial value for saMple length **
! ** Default Mass Storage Device **
I ** HP-IB 98034A Select Code **
! ** Address of Printer **
! 1. 3
I
SET
FOR
E X E CUT ION
DEFAULTS
MASS STORAGE IS Storet'
EXIT GRAPHICS
PR INTER I~ 16
ON ERROR GOSUB Err20
!
I 2.0
!
CALL TiMeIDate',T1Me',DaY',Month',Year,2)
PRINT PAGEi"PROGRAM, CELLGO Version, "iVersion'iLIN(1)
PRINT" Mass Storage defaults to, .iQuote'~Storet'~Quote$
PRINT" HP-IB Select Code is, "iCard
PRINT" HP-IB Printer is on Address, "iPrinteriLIN(1)
PRINT "AID Calibration Settings"
PRINT" Channel '1, 1 V"iTABI2S)i"Meter Output"
PRINT" Channel '2, 3.S V"iTABI2S)i"Meter Output"
PRINT" Channel '3, 3.S V"iTABI2Sli"Rotator Reading"ILIN(1)
PROGRAM
I N T ROD U C T ION
I
! 2.1
! .
C H E C K
S TAT U S
o F
D E V ICE S
I 2.11 2631G Line Printer
I
SET TIMEOUT Cardi1
ON INT ~Card,Printer GO TO 970
STATUS Card,PrinteriCond
IF BITICond,OI{)1 THEN 1000
DISP "NO POWER APPLIED TO 2631G PRINTER"
GOTO 960
OUTPUT Card,Printer USING "',K"iEsc$~"n"
STATUS Card,PrinterlCond
IF BITICond,6) THEN 10S0
DISP "TURN PRINTER ON-LINE"
GOTO 1010
OUTPUT 7161"A124B3S6"
!
!turn printer on-line
'hard off-line,
user Must turn on-line
!set AID switches to default state
! 2.12
!
Probe Rotator
OUTPUT 7061"H4AJ"
N=SHIFTIREADBIN(706),-8)+READBINI706)
IF ABSIN»=1 THEN 1140
DISP "NO POWER APPLIED TO ROTATOR, OR OUT OF CALIBRATION."
GOTO 1090
OFF INT .7
SET TIMEOUT 7110000
I
78
-------
1170 ! 2 2
1180 !
1190 CALL Se1aMp(-90)
1191 OUTPUT 717 USING "t,K"j"AM-30DB"
1200 !
1210 ! 2.21 Idenhfy 1he MeasureMen1
1220 !
1230 EDIT "En1er opera10r Iden1ifica110n",Oper$
1240 EDIT "En1er a 1i1le for 1his MeasureMen1",Ti1Ie$
1250 GOSUB Ge1_freq 'en1er frequencies in10 array Freq(*)
1260 !
1270 ! 2.22
1280 !
1290 Sys1eM$="Y"
1300 INPUT "En1er 1he A11enua10r Sys1eM being used (Yellow or Red)",Sys1eM$
1310 IF (Sys1eM$="R") OR (Sys1eM$="Y") THEN 1340
1320 CALL Dcol("Only Y or Rare accep1ed responses, 1ry again.")
1330 GOTO 1290
1340 IF (Sys1eM$="R") AND (Sys$C)"R") THEN LINK "RFIT",9820
1341 IF (Sys1eM$="Y") AND CSys$C>"Y") THEN LINK "YFIT",9820
1350 SYS$=Sys1eM$
1360 !
1370 ~ 2.23 Selec1 1he TEM Cell 10 use
1380 !
1390 Cell id$="S"
1400 INPuf "En1er TEM Cell (S or L)",Cell_id$[1j1J
1410 IF (Cell id$="S") OR CCel1 id$="L") THEN 1440
1420 CALL Dcof("Only L or S are-accep1ed responses, 1ry again.")
1430 GOTO i390
1440 Celln=O
1450 IF Cell_id$="L" THEN Celln=1
1460 !
i470 ! 2.24 Selec1 1he AMplifier
1480 !
1490 AMp$="D"
1500 INPUT "Are you using 1he Driver or Final AMplifier? (D or F)",AMp$[1j1J
1510 IF (AMp$="D") OR CAMp$="F") THEN 1540
1520 CALL Dcol("Only D or Fare accep1ed responses, 1ry again.")
i530 GOTO 1490
1540 AMpgain=60
1550 IF AMp$="D" THEN AMpgain=49
1560 !
1570 ! 2.3
1580 !
1590 ! 2.31 Selec1 Elec1ric or Magne1ic
1600 I
1610 A=O
1620 INPUT "Do you wan1 10 se1 ElectricCO) or Magne1ic(1) field?",A
1630 IF (A=O) OR (A=1) THEN 1690
1640 CALL Dcol("Only selec1ions 0 or 1 are allowed, 1ry again.")
1650 GOTO 1610
1660 !
1670 ! 2.32 Selec1 1he Uni1s 10 use
1680 !
1690 IF A=O
$(3,O)A,", or
1700 IF A=1
$(3,1 ),1,", or
1710 N=1
1720 INPUT N
D E FIN E
MEA SUR E MEN T
PAR A MET E R S
!initial set1ing for 8660B genera10r
Selec1 1he A11enua10r Sys1eM 10 use
D E FIN E
THE
FIE L D
T 0
EST A B LIS H
THEN DISP "En1er
(4) dBV/M";
THEN DISP "En1er
(4) dBA/M" j
uni1s,
(1) ",I,Uni1s$(1,O),I,", (2) VIM, (3)
II~Units
units:
(1) ",I,Uni1s$(1,1),I,",
(2) AIM, (3) "A,Uni1s
79
-------
1730
1740
1750
1760
1770
1780
1781
1782
1800
1810
1820
1830
1840
1850
1851
1860
1870
1880
1890
1900
1910
1930
1940
1950
1960
1970
1980
1990
2000
2010
2020
2030
y."
2040
2050
2060
2070
2080
2090
2100
2110
2120
2130
2140
2150
2160
2170
2180
2190
2200
2210
2220
2230
2240
2250
2260
2270
2280
2290
IF (N(1) OR (N)4) THEN 1690
!
! 2.33
I
SeleCT The Field and ConverT To appropriaTe UniTs
DISP "EnTer Desired Field in "'UniTs$(N,A)j
INPUT Valx
IF (N()4) AND «Valx{O) OR (Valx)1.0E8» THEN 1690
IF (N=4) AND (Valx)100) THEN 1690
Inval=Valx
IF N=2 THEN Valx=Valx*Valx
IF (N=3) AND (A=O) THEN Valx=3767*Valx
IF (N=3) AND (A=1) THEN Valx=.026b*Valx
IF N=4 THEN Valx=10A(Valx/20)A2
INPUT "EnTer MeTer Full Scale VolTage",FsvolTS
IF Fsvo1TS)5 THEN 1850
DISP "EnTer MeTer Full Scale Field in "'UniTs$(N,A)j
INPUT Fsval
IF N-4 THEN INPUT "EnTer scale range in dB",Fsrange
!
! 2.34
!
Cn=2
CnseT=3.5 !VoITS
IF ABS(FsvoITS»1 THEN 1980
Cn=1
CnseT=1 !VOITS
Con v=CnseT 111122
InsTr$="H"'VAL$(2A(Cn-1»'"AJ"
PRINT LIN(3)j"CAUTION"ILIN(1)
PRINT" 1. Level The Hazard MeTer and,"
PRINT" 2. Check The AMplifier STaTus Before CONTinuing."
DISP "ConnecT MeTer recorder OUTpUT To Cht"jCnj" Press CONTinue
COMpuTer SelecTs The proper AID Channel
!OUTpUT cOMMand To saMple above channel
PAUSE
!
I 3.0
!
! 3.1
I
! 3.11
!
PRO G RAM
E X E CUT ION
I NIT I A LIZ E
E QUI P MEN T
FOR
E X E CUT ION
PosiTion Probe
2300
Mdir=1
OUTPUT 70b."H4AJ"
Dig=SHIFT(READBIN(70b),-8)+READBIN(70b)
IF Dig{15 THEN 2210
DISP "PosiTioning Probe."
GO TO 2130
!
! 3.12 SeTup CELLGO Screen
I
STarT, DISP
Gp=O
GOSUB SeTup_screen
FOR 1=0 TO Nf-i
Cf=Freq z (1)
CALL FiT(SysTeM$,Freqz(I»
CALL COMplx(A,Freqz(I),Valx)
F=FNC(Cf*1E6,F$,2)
Rw(I)=377*Ri/Ro(Celln)
Xw(I)=377*Xi/Ro(Celln)
80
when read
-------
2310
2320
2330
2340
2350
2360
2370
2380
2390
2400
2410
CALL
CALL
CALL
CALL
CALL
CALL
PositI7,16,VAL$IF)&" "&F$&" ")
PositI8,55,VAL$IPROUNDIPt,-3»&" Watt ")
Positl10,25,VAL$IPROUNDIPMval,-2»&" dBM
Poslt(11,32," ")
Posit(12,33,11 II)
Positl14,1,RPT$I" ",48»
")
! 3 13
I
COMputer Selects Power Meter and Frequency Synthesizer
CALL SelectpMINpM,Nsg,Paddr,PMval,FreqzII»
I
2420
2430
2440
2450
2460
2470
2480
2490
2500
2510
2520
2530
2550
2560
2570
2571
2580
2590
2600
2601
2610
2620
2621
2630
2640
2645
26S0
2660
2670
2671
2680
2690
2700
2705
2710
2720
2721
2730
2740
2750
2760
2770
2780
2790
2800
2810
2820
2830
2840
I 3.14
I
PrograM Frequency and AMplltude on Selected Synthesizer
IF Nsg THEN 2550
OUTPUT 717j"FR"&VAL$IFreqzII»&"MH"
OUTPUT 717,"AM"&VAL$IPg)&"DB"
CALL ReadpMINpM,Paddr,Power,Status)
I ** CF on 3325A **
I ** Set AMplitude to calculated
I setting froM SUB Fit **
Pg=Pg+IPMval-Power)
IF ABSIPower-PMval»
OUTPUT 717; "FR"
GOTO 2770
!
CALL CfreqIFreqzII»
High=Low=O
.02 THEN 2460
! ** CF on 8660B **
! ** Set AMplitude to calculated
I setting froM SUB Fit **
CALL SetaMIO)
CALL SetaMpIINTIPg»
CALL ReadpMINpM,Paddr,Power,Status)
DISP n Setting Power Meter: IljPowerj"
IF High AND Low THEN Flne
IF Power)=PMval THEN 2640
Pg=Pg+1
Hlgh=1
GOTO 2580
IF High THEN Fine
Pg=Pg-1
Low=1
GOTO 2580
Fine. High=99
Low=O
X2=IHigh+Low)/2
CALL SetaMIX2)
CALL ReadpMINpM,Paddr,Power,Status)
DISP II Setting Power Meter' "jPowerJ'1 (FIne)
IF IABSIPMval-Power)( .03) OR IABSIHigh-Low)(2) THEN 2770
IF Power)PMval THEN Low=X2
IF Power(PMual THEN High=X2
GOTO 2680
!
! 3.2 ROT ATE
!
Pre errII)=PROUNDIPower-PMval,-2)
CALL Po sit I 11 , 33, VAL $ I Pre _er r (1 ) ) &" dB")
CALL PositC14,1,"Field Establlshed in Cell' "&VAL$IInval)&" "&UnitsIIN,A»
DISP " SaMpling data froM rotation"
IF Mdlr)O THEN OUTPUT 716;"B4A6"
IF Mdir(O THEN OUTPUT 716,"B6A4"
REDIM Dlr(Slen),B(Slen),Plotdb(Slen)
MAT Dlr=ZER
(Coarse) 11
PRO B E
AND
SAM P L E
RESPONSE
'probe clockwise
Iprobe counter-clockwise
!rediMenslon arrays
81
-------
2850
2860
2870
2880
2890
2900
2910
2920
2930
2940
2950
2960
2970
2980
2990
3000
3010
3020
3030
3040
3050
3060
3070
3080
3090
3100
3110
3120
3130
3140
3150
3160
3170
3180
3190
3200
3210
3220
3230
3240
3250
3260
3270
3280
3290
3300
3310
3320
3330
3340
3350
3360
3370
3380
3390
3400
3410
3420
3430
3440
MAT B=ZER
!
! 3.21
!
SaMple Da1a froM Hazard Probe and R01a10r Box
FOR J=O TO Slen
OUTPUT 706jInS1rS
BIJ)=SHIFTIREADBINI706),-8)+READBIN(706)
OUTPUT 706j"H4AJ" !A/D
DirIJ)=SHIFTIREADBINI706),-8)+READBINI706)
NEXT J
!
! 3.22 Check for Power Me1er Drif1
!
CALL ReadpMINpM,Paddr,Power,S1a1us)
PoS1 errII)=PROUNDIPower-PMual,-2)
CALL -Pai 1 I 12,34 ,VALSIPos1_err (1) ).\." dB
Hdir=-Mdir
IF ABSIPos1_errII»>.1 THEN 2250
Power=O
Se1db=10*LGTIValx)
!
! 3.23
!
!A/D Channel for Probe
Channel for R01a10r Posi1ion
" )
De1erMine Array Leng1h and Check for Errors
FOR J=Slen-1 TO 500 STEP -1
IF ABSIDirIJ)-DirIJ+1»>5 THEN 3150
IF BIJ>I=O THEN Err21
NEXT J
Err19, DISP "Sys1eM Malfunc1ion Check r01a10r connec1ions."
PAUSE
GOTO 2560
IF JISlen-2 THEN 3190
E~r18, DISP "PrograM Malfunc1ion SaMple size larger 1han"jSlen
PAUSE
GOTO 2560
SaMp_si ze=J
Slen=SaMp_size+40
!
! 3.24
!
REDIM DirISaMp_size>,B(SaMp_size),Pl01dbISaMp_size)
DISP " Conuer1ing ualues for pl011ing. ISaMpie ="jSaMp_size,")"
IF N<>4 THEN 3330
FOR J=O TO SaMp_size
TMp=Fsual-Fsrange+B(J)*Conu/FsuoI1s*Fsrange
Pl01dbIJ)=TMp-Se1db
Power=Power+10'ITMp/20)'2
NEXT J
GOTO 3450
Chu=1
IF IN=3) AND IA=O) THEN Chu=3767
IF IN=3~ AND IA=1) THEN Chu= 0266
FOR J=O TO SaMp_size
TMp=BIJ)*Conu/Fsuolts*Fsual*Chu
IF (N=3) OR (N=1> THEN TMp=5QRITMp)
Pl01dbIJ)=20*LGTITMp>-Se1db
!
Conuer1 Values 10 Pl011ing Uni1s
! 3.25
!
SUM up Powers and COMpu1e Auerage
Power=Power+TMp*TMp
NEXT J
82
-------
3450
3460
3470
3480
3490
3500
3510
3520
3530
3540
3550
3560
3570
3580
3590
3600
3610
3620
3630
3640
3650
3660
3670
3680
3690
3700
3710
3720
3730
3740
3750
3760
3770
3780
3790
3800
3810
3820
3830
3840
3850
3860
3870
3880
3890
3900
3910
3920
3930
3940
3950
3960
3970
3980
3990
4000
4010
4020
4030
4040
Avg=Power/Sa~p_size
Errdb=10*LGT(Avg)-Setdb
IF (N=2) OR (N=4) THEN Avg=SQR(Avg)
IF N=3 THEN Avg=Avg/Chv
IF N=4 THEN Avg=20*LGT(Avg)
,
, 4.0
!
G RAP H 1 C S
AND
TAB L E S
MAT SEARCH Plotdb(*),MIN,PMin
MAT SEARCH Plotdb(*),MAX,PMax
Avg i( 1) =Avg
Errdbl(I)=Errdb
P~inl(I)=P~in
PMaxi( l)=PMax
DISP
GRAPHICS
DEG
LDIR 0
IF Gp)O THEN 3880
PLOTTER IS "GRAPHICS"
!
! 4.1
!
! 4.11 Plotting Titles
I
LORG 6
CSlZE 4.55,7/15
MOVE 60,100
LABEL USING "t,K",Titl~$
CSlZE 3.1,8/15
MOVE 29,94
LABEL USING "t,K","Applied Field, "~VAL$(Inval)~" "~Units$(N,A)
MOVE 93,94
LABEL USING "t,K","Meter Full Scale, "~VAL$(Fsval)~" "~Units$(N,A)
IF 1)0 THEN 3880
!
! 4.12
,
I NIT 1 A LIZ E
GRAPHICS
PrograM Printer for Perforation Skip, and DUMp Title
OUTPUT 701,CHR$(12)~CHR$(10)~Esc$~"~166p56f1L"
DUMP GRAPHICS t7,1,90,100
!
I 4.13
I
Setup Plotting ParaMeters, and Scales
LOCATE Px1(Gp),Px2(Gp),Py1(Gp),Py2(Gp)
SMin=PROUND(PMin-.5,O)
S~ax=PROUND(PMax+.5,O)
SCALE O,Sa~p_size,SMin,SMax
FRAME
LORG 4
CSIZE 2.25
Y corr=(SMax-SMin)*.11
X=corr=SaMp_size*.13
Grid_x=SaMp_size/4+1E-4
Grid_y=SMax-S~in-1
FOR J=O TO 4
MOVE Grid_x*J,SMin-Y_corr
LABEL USING "t,K",VAL$(Xlabel(J»~CHR$(179)
NEXT J
AXES Grid_x,Grid_y,O,S~in
Mid=ABS(PMin+PMax)/2
! ** Y correction factor **
! ** X correction factor **
! ** X Grid spacing **
! ** Y Grid spacing **
! ** Draw X labels **
83
-------
4050 LORG 8
4060 FOR L=S~in TO S~ax
4070 MOVE O,L
4080 LABEL USING ".,MDZ,X"iL
4090 NEXT L
4100 MOVE O,Errdb
4110 DRAW Sa~p_size,Errdb
4120 MOVE O,S~in
~130 IF Mdir=1 THEN 4210
4140 !
4150 ! 4.15 Plot conuerted data
4160 !
4170 FOR J=O TO Sa~p_size
4180 PLOT Sa~p_5ize-J,Plotdb(J)
4190 NEXT J
4200 GOTO 4240
4210 FOR J=O TO SaMp_size
4220 PLOT J,Plotdb(J)
4230 NEXT J
4240 LORG 6
4250 MOVE SaMp_size/2,S~ax+Y_corr
4260 LABEL USING ".,K"iVAL.(F)'" "'F$
4270 MOVE -X_corr,(SMin+SMax)/2
4280 LDIR 90
4290 LORG 5
4300 LABEL USING ".,K","dB Error"
4310 Gp=Gp+1
4320 SETGU
4330 IF Gp(4 THEN 4370
4340 DUMP GRAPHICS .7,1,2 25,90
4350 IF 1=7 THEN OUTPUT 701 USING ".,K",Esc$'"'166p49f1L"
4360 Gp=O
4370 EXIT GRAPHICS
4380 NEXT I
4390 GRAPHICS
4400 DUMP GRAPHICS .7,1,2.25,90
4410 EXIT GRAPHICS
4420 !
4430 I 4.2
4440 I
4450 OUTPUT 701 USING ".,K",Esc$'"'166F"
4460 PRINTER IS 7,1,WIDTH(132)
4470 PRINT LIN(4),TAB«81-LEN(Title$»/2) 'Title$ LIN(1)
4480 PRINT Day"" ",Month$," ",Date'[5,~J,",",tear,TAB(65),Ti~e$,LIN(1)
4490 PRINT "Operator' ",Oper$,TAB(48) i
4500 IF Celln THEN PRINT "TEM Cell, CC-101.5 (Large)"
4510 IF NOT Celln THEN PRINT "TEM Cell, CC-105 (S~all)"
4520 IF AMp$="D" THEN PRINT TAB(48),"Driuer A~plifier Output"
4530 IF AMp$="F" THEN PRINT TAB(48)i"Final A~plifier Output"
4540 IF Syste~$="Y" THEN PRINT TAB(48),"Running Yellow SysteM"
4550 IF Syste~$="R" THEN PRINT TAB(48),"Running Red Syste~"
4560 I
4570 ! 4.21 Rotation Statistics
4580 !
4590 PRINT LIN(3),TAB(35),"Rotation Statistics"iLIN(2)
4600 PRINT" Freq. Aug. Reading Aug. Error Low Error
igh-Low Error"
4610 PRINT" (MHz) (",Units$(N,A)[1,7J,")",TAB(31),"(dB)
(dB) (dB)" ,LIN( 1)
4620 FOR 1=0 TO Nf-1
! ** Draw Y labels **
! ** Label the graph **
PRINT
R E P 0 R T
SUMMAR¥
!return printer to on-line state
High Error
H
(dB)
84
-------
4630 PRINT USING 4660iFreqz(I),PROUND(Avgi(I),-3);Errdbi(I);P~ini(I);P~axi(I)jP
~axi(I)-P~ini(I)
4640 NEXT I
4650 PRINT LIN(1)i"Applied Field = -;PROUND(Inval,-3),Units$(N,A)
4660 IHAGE DDD DDD,4X,5DZ.DDD,5X,HDZ.DD,8X,HDZ.DD,7X,HDZ.DD,10X,HDZ.DD
4670 !
4680 ! 4.3
4690 !
4700 Nf=Nf-1
4710 GRAPHICS
4720 PLOTTER IS "GRAPHICS"
4730 !
4740 I 4.31
4750 !
4760 LOCATE 20,120,10,93
4770 HAT SEARCH Freqz(*),HIN;F~in
4780 HAT SEARCH Freqz(*),MAX,F~ax
4790 HAT SEARCH P~ini(*),MINjS~in
4800 MAT SEARCH P~axi(*),MAX;S~ax
4810 S~in=PROUND(S~in-.5,0)
4820 S~ax=PROUND(S~ax+.5,0)
4830 IF Nf)O THEN Fc=(F~ax-F~in)/Nf
4840 IF Nf=a THEN Fc=F~ax/2
4850 F~in=F~in-Fc
4860 F~ax=F~ax+Fc
4870 IF F~in=F~ax THEN 5400
4880 SCALE F~in,F~ax,S~in,S~ax
4890 FRAHE
4900 CSIZE 2.55
4910 Y_corr=(S~ax-S~in)*.02
4920 X corr=(F~ax-F~in)*.02
4930 CLIP F~in-X_corr,F~in,S~in,S~ax
4940 AXES Fc,1,F~in,S~in
4950 UNCLIP
4960 LORG 8
4970 FOR I=S~in TO S~ax
4980 MOVE F~in-X_corr*2,I
4990 LABEL USING "t,MDZ"iI
5000 NEXT I
5010 IF Nf)15 THEN X_corr=O
5020 FOR 1=0 TO Nf
5030 LORG 5
5040 LDIR 0
5050 HOVE Freqz(I),S~ln-Y_corr
5060 DRAW Freqz(I),S~in
5070 HOVE Freqz(I),S~in-Y_corr*2
5080 IF Nf)20 THEN CSIZE 2.20
5090 LABEL USING "t,K"jFreqz(I)
5100 LINE TYPE 10
5110 IF Nf)20 THEN CSIZE 2.55
5120 HOVE Freqz(I),P~lni(I)
5130 DRAW Freqz(I),P~axi(I)
5140 DRAW Freqz(I),P~ini(I)
5150 LINE TYPE 1
5160 I
5170 I 4.32
5180 !
5190 HOVE Freqz(I),Errdbi(I)
5200 LABEL USING "t,K"; .0"
5210 IF Nf<16 THEN 5270
SUM MAR' Y
G RAP H
Setup Scaling, and Plotting Li~its
! ** Label the Y axis **
! ** Draw a Box Plot of error ll~its **
Mark the Average Error with a Clrcle
85
-------
5220
5230
5240
5250
5260
5270
5280
5290
5300
5310
5320
5330
5340
5350
5360
5370
5380
5390
5..00
5410
5420
XWIl
5430 PR INT "
ohM")" iLIN( 1)
5440 IMAGE 6X,3D.5D,6X,MDZ.DD,12X,MDZ.DD,10X,3D.3D,4X,2DZ.3D
5450 FOR 1=0 TO Nf
5460 PRINT USING 5440iFreqz(I),Pre_err(I),Pos1_err(I),Rw(I),Xw(I)
5..70 NEXT I
S..80 COSU» Ge1_freq
S..90 COTO S11r1
5S00 END
5510 !
5520 ! 5. 0
5S30 ,
5540 Ge1_freq, ! En1er Frequencies froM 1he Keyboard
5550 !
5S60 EXIT GRAPHICS
5570 Nf=SOO 'MaxiMuM NUMber of Frequencies
SS80 REDIM Freqz(Nf),Avqi(Nf),Errdbi(Nf),PMini(Nf),PMaxi(Nf),Pre-err(Nf)
5590 REDIM Pos1_err(Nf)
5600 CALL En1erf(Nf,Freqz(*»
5610 Tf=Nf-1
5620 REDIM Freqz(Tf)tAvqi(Tf),Errdbi(Tf),PMini(Tf),PMaXi(Tf),pre_err(Tf)
5630 REDIM Pos1_err( f)
5640 RETURN
5650 !
5660 Se1up_screen' ! Se1up 1he CELLGO Screen
5670 ,
5680 PRINTER IS 16
5690 EXIT'GRAPHICS
5700 PRINT PAGE
5710 Sub_se1up' CALL Cen1er(1,40,"TEM Cell Au10Ma1ic Con1rol Sys1eM")
5720 PRINT LIN(3)iDay$i" "iMon1h'i" "iDa1e$[5i2Ji","iYeariTAB(65)iTiMe$iLIN(1
)
5730
57..0
5750
5760
5770
5780
!
! 4.33
!
Label 1he Pl01 and DUMp i1 10 1he Graphics Prin1er
LDIR 90
LORG 2
MOVE Freqz(I)-X_corr,PMaxl(I)+Y_corr
LABEL USING ".,MDZ.DD"iPMaxi(I)
MOVE Freqz(I)-X_corr,PMini(I)-Y_corr
IF Nf)15 THEN LORG B
LABEL USING ".,MDZ.DD"iPMini(I)
NEXT I
CALL Label("Frequency (MHz)","dB Error",Ti11e$)
PRINT PAGE
DUMP GRAPHICS .7,1
I
! 4.34
!
Se11ing Error and Wave IMpedanc@ SUMMary
PRINT P.AGEiLIN(1)iTAB«B1-LEN(Ti11e$»/2)iTi11e'
~RINT LIN(2)iTAB(20)i"Power M@1er Se11ing Error and Wave
PRINT" Pre-R01a1ion Pos1-R01a1ion
PRINT" Fr@quency S@11ing Error Se11ing Error
IMpedance"iLIN(3)
Rw
(MHz)
(OhMS)
(dB)
(dB)
I N
L I N E
SUB R 0 UTI N E S
CALL Posi1(6,1,"Sys1eM ParaMe1ers")
IF Celln THEN CALL ~osi1(6,4B,"TEH Cell, CC-t01.5 (Large)")
IF NOT Celln THEN CALL Posi1(6,4B,"TEM Cell, CC-105 (SMall)
CALL Posi1(7,5,"Frequency''')
CALL Posi1(B,4B,"Power,")
CALL Posi1(',1,"Power Me1er ParaMe1ers")
")
86
-------
5790
5800
5810
5820
5830
5840
5850
5860
5870
5880
5890
5900
5910
5920
5930
5940
5950
5960
5970
5980
5990
6000
6010
6020
6030
6040
6050
6060
6070
6080
6090
6100
6110
6120
6130
6140
6150
6160
6170
6180
6190
6200
6210
6220
6230
6240
6250
6260
6270
6280
6290
6300
6310
6320
6330
6340
6350
6360
6370
6380
CALL Poslt(10,5,"Calculated Setting ")
IF SysteM$="Y" THEN CALL PO.ltl10,48,"Runnlng Yellow SysteM")
IF SysteM$="R" THEN CALL Positl10,48,"Running Red SysteM")
CALL Posit(11,S)"Pre-Rotatl0n Setting Error.")
CALL Posit(12,S,"Post-Rotatlon Setting Error: ")
CALL Posit(20,l,"SystsM Status")
IF AMp$="D" THEN CALL Positl9,48,"Driuer AMplIfier Output")
IF AMp$="F" THEN CALL Positl9,48,"Final AMplifier Output")
RETURN
!
! Miscellanous PrograM Errors
!
Err21:
PAUSE
RETURN
Err20' DISP .PrograM Malfunction PrograM Error, "IERRM$
GOTO 5940
, END
I
SUB COMplxlQ,SHORT
!
1
!
,
!
!
!
!
!
!
!
!
DISP "SysteM Malfunction Check Meter connections."
Freq,Valx)'
SUB COMplxlQ,SHORT Freq,Valx)
Q (1)Magnetic/10)Electric Indicator
Freq Operating Frequency
Valx Desired Field in A~2/MA2 for Magnetic,
for Electric
ilnd V"'2/MA2
Returns in
Pt
PMval
Pg
COMMON
Power IWatts)
Desired Power Meter Reading
ApproxiMate Generator Power
! Description
! Uses RI, Xl, and Ro to find cell center iMpedance Ri and Xi
! Also uses fIeld equations to deterMine operating power IPt) froM
I desired field (Valx) and uses attenuation (Atten) to deterMine
! power Meter settIng IPMval)
!
COM SHORT Freq(45),RI145),XI145),Ri,Xi,AMpgain
COM SHORT RI,XI,Atten,Celln,Pt,PdbM,Pd(1),EII1),Rol1),PMval,Pg
RAD
Xo=O
Ro=Ro ICeUn)
Leng=ElICelln)
D=Pd I CeUn)
X=2*PI*Leng
Beta_Iength=X*Freq/300
A=COSIBeta_length)
B=SINIBeta_Iength)
DenoM=IRo*A-XI*B)A2+IXo*A+RI*B)A2
F=IIA*RI-Xo*B)*IRo*A-XI*B)+IXI*A+Ro*B)*IXo*A+RI*B»/DenOM
G=IIXI*A+Ro*B)*IRo*A-XI*B)-IA*RI-Xo*B)*IXo*A+RI*B»/DenOM
Ri=Ro*F-Xo*G
Xi=Xo*F+Ro*G
TMp=Valx*Ri*D*D
IF Q=1 THEN Pt=142129*TMP/IRo*Ro)! ** Magnetic **
IF 1;1=0 THEN Pt=TMp/IRi*Ri+Xi*XI) ! ** Electnc **
PdbM=10*LGTIPt*1000)
PMval=PdbM-Atten
Pg=PdbM-AMpgain
***CoMplx
!Cell Modeled as Lossless TransMission Line
87
-------
6390
6400
6410
6420
6430
6440
6450
6460
64'70
6480
6490
6500
6510
6520
6530
6540
6550
6560
65'70
6580
6590
6600
6610
6620
6630
6640
6650
6660
6670
6680
6690
6700
6730
6740
6750
6760
6770
6780
6790
6800
batG
6820
MJ30
6B40
"'850
~.870
&\100
6910
t.no
109,0
69,,0
6950
6SOl_D
6970
6geu
6990
7000
7010
702C
7030
SUBEXIT
SUBEND
!
SUB Dcol (A$)!
!
I SUB Dc 0 l( A$ )
I A$ String to Display
!
Size=lEN(A$)
FOR 1=0 TO Size-1
DISP TAB(Size-I),A$[1,I+1J
WAIT 75
NEXT I
FOR 1=1 TO 3
WAIT 350
D1SP
WAF 350
D 1SP A$
BEEP
NEXT I
WAIT 40*Si ze
SUBEXIT
SUBEND
!
SUB Cfreq(SHORT F_Mhz)!
!
! SUB Cfreq(SHORT F_Mhz)
! F_Mhz Frequency in MHz
!
! Descrlp'tion
! Set the Center Frequency on the 8660B
I
CMd$=VAl$(lE-l0*INT(F Mhz*1(6»
OUTPUT 719; "II ".!.REV$(CMd$£2J ).!."("
SliBEXIT
SUBEND
!
SUB SetaMp(AMp_dbM)!
!
! SUB SetaMp(AMp_dbM)
! AMp_dbM AMplitude in dBM
!
! Description
I Set .he AMplitude on the 8660B to AMp_dbM
AMp$=VAl..$(ABS(1E-3*(AMp_dbM-13»)1."000.
CMd$="I"I.REV8(AMp$£2,4J)I."C"
OUTPUT 7t9,C"dil
SUItE>:: IT
SUBEND
I
TUB Posi.,X.Y,~~) I
,
I
!
!
A=1N'f(X)-1
B=INT(Y,-l
PRINT U,.HIG
SUB Posit(X,Y,Q$!
X,Y Line and ColuMn on 'the 9845 Screen
Q$ String to Display at that Position
'1.)Kt1jCHRG(~7)&'I&~"&VAL$(A)&"~I'&VA~~(B'~~C'\&Q~
f)()
( ~CI
***Dcol
**J!i
-------
7040
70'00
7060
7070
7080
7090
7100
7110
7120
7130
7140
7150
7160
7170
7180
7190
7200
7210
7220
7230
7240
7250
7260
7270
7280
7290
7300
7310
7320
7330
7340
7350
7360
7370
7380
7390
7400
7410
7420
7430
7440
7450
7460
7470
7480
7490
7500
7510
7520
7530
7540
7550
7560
7570
7580
7590
7600
7610
7620
7630
SUBEXIT
SUBEND
,
SUB Center(X,Y,Text$)'
,
I SUB
!
!
!
Center(X,Y,TEXT$)
X/Y Line and ColuMn PosItion to center Text$
Text$ StrIng to Display
A;INT(X)-l
B;INT(y)-1
L;INT«LEN(Text$)-1)/2)
PRINT USING "t,K",CHR$(27)'"'a"'VAL$(A)'"r"'VAL$IB-L)'"C"'Text$
SUBEXIT
SUBEND
!
SelectPM'
SUB SelectpM(INTEGER
!
I
I
,
,
,
I
,
!
!
!
NpM,Nsg,Paddr,SHORT PMval,Freq)'
SUB SelectpMIINTEGER NpM,Nsg,Paddr,SHORT PMval,Freq)
NpM NUMber of the Power Meter
Nsg NUMber of the Slgnal Generator
Paddr HP-IB Adress of the Po~er Meter
PMval Power Meter Value froM COMplx
Freq Operating Frequency
***Center
***SelectpM
Description
Based upon the Frequency (Freq) and the Po~er Meter Value (PMval)
eIther the Boanton or HP Power Meters are used to Measure the
and deterMine the establlsh fleld The Signal Generator 15 also
selected to provide for the proper frequency source.
NPM;O
Paddr;713
IF Freq(10 THEN NpM;1 !select
IF (NpM;O) AND CPMval);-20) THEN OUTPUT 716J"B1A2"
IF (NpM;O) AND CPMval(-20) THEN OUTPUT 716j"A1B2"
IF (NpM;O) AND (PMval(-20) THEN Paddr;714
IF NpM THEN OUTPUT 716,"A12"
IF NpM THEN OUTPUT 708,"D-70L25H"
Nsg;1
IF Freq(;20 THEN Nsg;O
IF Nsg THEN OUTPUT 722,"A2" ! ** 8660B **
IF NOT Nsg THEN OUTPUT 722,"A1" '** 3325A **
SUBEXIT
SUBEND
!
SUB SetaM(INTEGER Prcnt)'
!
! SUB SetaM(INTEGER Prcnt)
Prcnt Setting for 8660B Percent AMplitude ModulatIon
BOONTON po~er Meter
!HP power Meters
P$;VAL$(ABS(Prcnt»
IF LENCP$)(2 THEN P$;"O"'P$
CMd$;"48$"'REV$(P$)'"Z"
OUTPUT 719jCMd$
SUBEXIT
SUBEND
!
SUB ReadpMIINTEGER NpM,Paddr,SHORT Po~er,Status)'
89
***SetaM
***ReadpM
-------
7640
7650
7660
7670
7680
7690
7700
7710
7720
7730
7740
7750
7760
7770
7780
7790
7800
7810
7820
7830
7840
7850
7860
7870
7880
7890
7900
7910
7920
7930
7970
7980
7990
8000
8010
8020
8030
8040
8050
8060
8070
8080
8090
8110
8111
8112
8113
8114
8115
8120
8130
8140
8150
8160
8170
8180
8190
8200
8210
8220
I
! SUB
!
!
I
IF NpM THEN 7720
CALL Readhp(Power,Paddr)
SUBEXIT
CALL Readb(Power,Status)
SUBEXIT
SUBEND
ReadpM(INTEGER NpM,Paddr,SHORT Power,Status)
Power Value read froM the Power Meter
Status Power Meter Status
I ** NpM=O, Read HP Meter **
I ** NpM=1, Read Boonton Meter **
SUB TiMe(Datel,TiMe$,DayS,Monthl,INTEGER Year,Type)!
!
! SUB TiMe(Date$,TiMe$,Day$,Month$,INTEGER
! Date$ The current Date in the forM
I TiMe$ The current TiMe in the forM
I or "HH,MM xM" when Type=2
Day$ The day of the week
Year,Typ11 THEN "$="PM"
IF (Hour)12) OR (Hour=O) THEN Hour=ABS(Hour-12)
OUTPUT TiMe$ USING TiMe_forMat2;Hour,Minute,M$
Year_=Year-(Month(3)
Month=Month+12*(MonTh(3)
Year =INT(1.2S*Yea, )-INT(Year /100)+INT(Year /400)
Weekday=INT(Day+2.6i(Month+1)+Year_) MOD 7 -
g~~:~~~Y$(Weekday)
I
Date_forMaT,
TiMe_forMat! ,
TiMe forMaT2,
SUBEND
IMAGE i,1X,ZZ,"/",ZZ,"/",ZZ," "
IMAGE t,1X,ZZ," .. ,ZZ," .. ,ZZ," "
IMAGE ~.1X,DD," ",ZZ," ",AA
DEF FNC(V,F',Pr)!
!
!
,
!
!
!
! DescripTion
! Use inTernally
!
DEF FNC(V,F$,Pr)
V NUMeric value
F$ STring equivalent
Pr rower Rounding Factor
to convert Hz to kHz, Mhz, CH:, etc.
90
J!:**TiMe
***FNC
-------
8230
8240
8250
8260
8270
8280
8290
8300
8310
8320
8330
8340
8350
8360
8370
8380
8390
8400
8410
8420
8'130
8440
8450
8460
8470
8480
8490
8500
8510
8520
8530
8540
8550
8560
8570
8580
8590
8600
8610
3620
86.30
8640
8650
8660
8670
8680
8690
8700
8710
872D
8739
8740
8750
e7~.0
8770
8780
879C
8800
8810
DHI N..$(-1 ;j)[3]
READ N..S(*)
DATA ..HI,HI,kHI,MHI,GHI
I=INT(LGT(V)/3)
FS=N..S(I)
RETURN PROUND(V/10'(I*3),-Pr)
FNEND
I
SUB Label(XS,YS,T$)!
I
I
I
I
!
I
SUB Label(XS,YS,TS)
XS X aX1S label
YS Y axis label
TS Plot label or
SUB Readb(SHORT Power,Status)
pq~@r Power:n dBM returned froM the BODntcn
Status Boonton Power Meter Status
SETGU
!...DIR 0
CSIZE 3.5
LORG 6
MOVE 71. 5,4
LABEL USING "*,K",XS
HOVE 7L5,98
CSIlE 4.5,6/15
LABEL USING "i,K"jTS
DEG
LDIR 90
CSIZE 3.5
tlOVE 0,50
LABEL USING ".,K"jY8
SUBE)(IT
SUI/END
!
SUB Readb(SHORT
I
!
I
Title
Power ,Status) J
! Description
! Routine to take ~ reoding froM the Boonton Power Meter.
I and allow for settli~g tiMe.
DIM Err$[5iJJ
SHORT T
INTEGER I
FOR 1=1 TO 25
T=P o...."r
ENTE~ 7~8iPower,S1atus,~ange
IF 1=1 fHEN 8740
IF ABS(T-Power)C=.Oi THEN 8770
WAIT 250+300*~Power<-2D)*INT(ABS(Power/10»
~!EXT I
~lSP "Power Met~r not Settled D
PAUPE
IF St21'US~O
IF StLl'(US=]
IF Sta.. u,,=4
IF St"us=5
bEEP
vISP I(B()ont~n
8'32iJ
THEN
1 HEN
THEN
THEN
SUBE~r~
ErrS="Meter Underrange."
Err$;IIMetRr Overrarlge."
Err~="Posslble Hardware MalfuncTion. ,.
, ,[rr$
91
***Label
***Readb
-------
8830
8840
88S0
8860
8870
8880
8890
8900
8910
8920
8930
8940
89S0
8960
8970
8980
8990
9000
9010
9020
9030
9040
90S0
9060
9070
9080
9090
9100
9110
9120
9130
9140
91S0
9160
9170
9180
9190
9200
9210
9220
9230
9240
92S0
9260
9270
9280
9290
9300
.
9310
9320
9330
9340
93S0
9360
9370
9380
9390
9400
9410
PAUSE
DISP
GOTO 8680
SUBEND
!
SUB Readhp(SHORT Power,INTEGER Paddr)!
I
***Readhp
! SUB Readhp(SHORT Power,INTEGER Paddr)
Power Power read froM the HP Power Meter in dBM
Paddr Adress of the Power Meter
DIM Err$!6S1
Start. !
FOR X=1 TO 20
OUTPUT Paddr,"D-V"
WAIT (Range=73)*4000
ENTER 7,Paddr USING "B,B,X,F"IStatus,Range,Power
IF Status=80 THEN 9090
Errt="Power Meter Malfunction."
IF (Status=81) OR IStatus=83) THEN Err$="Meter Under Range."
IF St~tu$=82 THEN Err$="Meter Over Range.11
BEEP
DISP "HP436A
PAUSE
DISP
GOTO Start
IF (X(>1) AND IABSIPower-P1)(=.02) THEN 9130
P1=Power
NEXT X
DISP "POWER METER NOT SETTLED"
LOCAL 7,Paddr
SUBEXIT
SUBEND
I
SUB Enterf(Nfreq,SHORT Freqx(*»1
I
I
!
I
!
! DescrIption
SubroutIne to
".\.Err$
***Enterf
SUB Enterf(Nfreq,SHORT Freqx(*»
Nfreq NUMber of Frequencies that were entered
Freqxl*) The actual frequency values
enter frequencIes froM the keyboard intI the prograM
DIM Z$!2S1,A$!1601
PRINTER IS 16
En ter _freq' Nf=O
Z$=" R "
INPUT liDo you want 10 enter seperate frequencIes or a range? (S or R)",Z
IF Z$!111J="R" THEN Range
IF Z$!1j1J()"S" THEN 9290
ON ERROR GOTO Bad_nuMber
!
! Enter frequencies Seperatly
!
Z$=NuIU
M=NuIU
DISP "Enter
LINPUT A$
Slze=LEN(A$)
Frequency IMHz) t"INf+1IZ$,
92
-------
9420
9430
9440
9450
9460
9470
9480
9490
9500
9510
9520
9530
9540
9550
9560
9570
9580
9590
9600
9610
9620
9630
9640
9650
9660
9670
9680
9690
9700
9710
9720
9730
9740
9750
9760
9770
9780
9790
9800
9810
9820
9830
9840
9841
9842
9850
9860
9870
9880
9890
9900
9910
9920
9930
9940
9950
9951
9960
9970
9980
IF Slze=O THEN Exit
FOR 1=1 TO Size
C=POSCA$[I Size] " ")
IF C=O THEN 9490' ,
X=C+I-2
FreqxCNf)=VALCA$[I,X])
GOTO 9500
FreqxCNf)=VALCA$[I])
IF CFreqxCNf»=.01) AND CFreqxCNf)C=2400) THEN 9530
g~~~ ~~:~C"Specifled Frequency is out of Range, Try Again.
Nf=Nf+1
ZS=IIPress CONTInue to Exit"
PRINT" Frequency 1 ",Nf," - ",FreqxCNf-1)
I=X+1
IF C=O THEN 9380
NEXT I
I
! Enter frequencies as a range
I
" )
Range' Inc=1
INPUT "Enter Start, Stop, and IncreMent Frequencies (MHz)",STart,Stop,Inc
IF CStart(.01) OR CStop>2400) OR (Start>2400) OR CStopC.01) THEN Range
IF ABSCStop-Start)/Inc+1(=Nfreq THEN 9680
CALL Dcol("Too Many Frequencies, Change diMension size. II)
GOTO Range
FOR Freq=Start TO Stop STEP Inc
FreqxCNf)=Freq
Nf=Nf+1
NEXT Freq
Exit, IF Nf>O THEN 9750
CALL DcolC"No Frequencies Defined, Try Agaln .")
GOTO Enter_freq
Nfreq=Nf
SUBEXIT
Bad_n u,.,ber ,
IF ERRN(>32 THEN 10559
CALL DcolC"Illegal Nu,.,erlc Response Try Again .")
GOTO 9380
SUBEND
SUB FltCSyste,.,$,SHORT Freqx) 'yellow systeM
COM SHORT Freq(45),RIC45),XIC45),Ri,Xi,AMpgain
COM SHORT RI,XI,Atten,Celln,Pt,PdbM,Pd(1),EIC1),RoC1),PMoal,Pg
SHORT Rho, Theta
DIM IMp_title$[160]
DEF FNAtten1CSHORT X)=44.126-3.26E-2*LOGCX)
DEF FNAtten2CSHORT X)=44.2+ 1034*X
DEF FNAtten3(SHORT X)=44.28+.03*X-.00282*X*X
DEF FNAtten4CSHORT X)=44.24+.0082*X-1.4E-5*X*X
IF FreqCO)()O THEN 9990
ASSIGN Ii TO "ZY1",A
IF NOT A THEN 9950
CALL DcoIC"ZY1 is not found, please supply correct disk")
PAUSE
GOTO 9900
DISP
READ 11,IMp_title$
FOR 1=0 TO 45
READ 11,FreqCI),RICI),XICI),Rho,Theta
NEXT I
93
-------
9990
10000
10010
10020
10030
10040
10050
10 060
10070
10080
10090
10100
10110
10120
10130
10140
10150
10160
10170
10180
10190
10200
10210
Atten=-9E10
IF IFreqx)=.01) AND IFreqxC.1) THEN Atten=FNfr,en1IFreqx)
IF IFreqx)=.1) AND IFreqx(1) THEN At'en=FNAt,~~~IFreqx)
IF IFreqx)=1) AND (FreqxC10) THEN A'ten=~NAt'~n3IFreqx)
IF IFreqx)=10) AND CFreqxC=220) THEN Atten=FN~'len4(Freqx)
R1=So-
X1-'0
IF IFreq"C=O) OR IFreqx}220} THEN Err
IF FreqxC=1 THEN SUBEXIT
FOR 1=0 TD 45
IF FreqxC=FreqII) THEN 10130
NEXT I
Err' R1=Xl=-9E10
SUBEXIT
Freqd=FreqII)-FreqII-1)
Mr1=IRIII)-RlI1-1»/Freqd
Mx1=CX1II)-XIII-1»/Freqd
Br1=R1(I)-Hr1*FreqII)
Bxl=XlII)-Mx1*FreqCI)
R1=Mrl*Freqx+Brl
Xl =Mx HeFr e"l x +8.1
SUBEXIT
SUB END
94
-------
30" Ns" Hellax Cable
Amplifier
d\search
0":) (1 'wi
Model 1 GOOL
/
-----f(~,
! Linear (1QOW)
l Amplifier
:oo~~ /
;/
I,Q
01
Modl~i",i:on Ilirul
\ < (15 VUe.,
P~-l
L=J
:':'€2 D€tCiIi1
~
~
~
~
~~.B""~ffi~~~~
~I
g ~I
VI VI I
(Ch11 (Ch2J (Ch31
HP 59313A
A/D Convert'3r
:-------~----- ---I
!
H"'wle!\-':>a,(:k2.rrJ
IF'
1,-'::,\rIJrTle'lls
Indl.Js!r\
f---f=1---1
--- System ---
- - (See Det,::III) 7' :-
":;-
(6)
HP 59306A
Relay
Actuator
(C1)
16
IEEE-488 Interface bus
16 bit 1/0 bus HP 9845 B
In n - Me...
'1- ! - ,'t- HP 9&034 1E::t'i::' :~GfJ Irllerfsce
0" _.r' ,.
\ '- HP '28035';:" [-',02'<:1 Til!18 Ciock
1- HP ~Oi}32P tb 1.::>lt I/O
0)
c=) ~~:~t: 4%
CJ
,r]
-I
)::0
.......
r
M
0 ~
V> -0
-< rrl
~ Z
o
rrl ......
3: X
o n
......
):0
'"
;0
~
V>
-------
Hewlett-Packard 86608 Synthesizer
Signal Generator (Including Plug-In's)
I Ej8660B Synthesizer Signal Generator
HP Hewlett-Packard
I Ceo"'"""'""
o KYBD
o STEP
o SWP WIDTH
Line
DON
SWEEP MODE
RATE
\.0
0"1
@
Single
o
OUTPUoO - 5V)
7 8 9 GHz CF
4 5 6 MHz STEP
1 2 3 kHz STEP
o . Clear Hz SWP
KYBD Width
Manual Mode
r;::;pI86632A ModulatIOn Section r;:;p186602A RF Section
D AM-FM L:.:J 1-1300 MHz
~
@@@
Manual Sweep
o
FM
.
CF CAL
MODULATION
LEVEL
o
Sorenson
Power Supply
INPUT
OUTPUT
OUTPUT
SOURCE
VERNIER
(.2V-2V) 0
RMS
I
I
OUTPUT
@
+ 0.5 Vdc
Model QHS40-.5
-------
f,! Malt Ccn
-------
N
1.0
Male Connector
!
~- 0
AG 22S/u
Red System
C
Connector
)
6 Ft. Coaxial Cable
N
Female Connector \
30 dB, 2kW High Power Attenuatar
Electro-Impulse Model AX-2000-3
Wooden Blocks
N
Connector
RG 214/u
13dB, 5 Watt Attenuator
N
Connector
20 Ft. Coaxial Cable
-------
APPENDI X D
PROBE CAL IBRAT ION RESUL TS COMPARED WITH AND WITHOUT LOAD
IMPEDANCE CORRECTION
NBS EF~1-5 SIN 1
Applied Field: 30 dBl//ro Meter Full Scale: 40 dBl//m
L0 MHz 1S MHz
€I 0
~ '-
" '"
L L
~-1 w-l
III ~
l! "U
-i! -z
lsa' 27iQ< 3S~' ga' I SiQ' lElIa' 27a' ~SiQ> 91a' I Ella'
20 MHz 2S MHz
~I ~I
I I
L L
"
a '-
~ '-
~-l W
II:! III
"'0 ~
-2 -l
180' 27e< 3613' 80' 18e' l813' 270' 36e> 913' L8S'
30 MHz 3S r~Hz
€I 0
~ '-
" '"
L L
~-1 w-L
ItI ~
l! "U
-i! -z
lsa' 27iQ> 3SIa' ga' I SiQ' I Ella" 27a' ~ S iQ> 91a' I Ella'
99
-------
413 MHz 45 MHz
'" a
~ ..
a c
L '-
~-l ~l
II> III
" ."
-2 -2
1 8 a" 27a" 3613' 9a" 18a" l813' 2713" 36a" 913' 188'
5a MHz 55 MHz
a 13
L ...
co "
~ ..
ki-l w-L
IQ ~
1! ;::;
-2 -2
Hila" 2 7 t3~ 381a" 91a" Hljd" l81a" 271J4 :I S jd" 91a' 18B'
613 MHz 65 MHz
'" a
~ ..
a c
L '-
~-l ~l
II> III
" ."
-2 -2
18a" 27a" 380" :30" J 80" l813' 2713" 360'. 90'. 188"
100
-------
70 MHr 75 r1Hz
13 0
L ~
0 "
~ L
~-1 w-l
IQ I:Q
'U "U
-2 -2
HIli!' 27~' 3f;1iJ. 91i!' Hj~' LBIiJ. 2713' ~S~,. 91iJ. l8id.
80 MHr B5 MHz
0 eJ
L
a a
L <..
L <..
I.J W
lEI IJ:I
" ..,
-1 -L
180' 270' 360. 90' 180' l813. 2704 360' 90. l80.
90 MHz 95 MHz
13 0
L ~
D '"
~ L
~-1 L
w
III I:Q
'U "U
-.. -l
~
1IH3' 27~' 3f;1iJ. 91iJ' I B~' lBIiJ. :27kP ~s~,. !;I1iJ. LBid'.
101
-------
D
L
L
W
III
."
100 MHz
"
-1
180'
2713-
360.
80'
102
1 813'
-------
NBS
EFM-5
SIN 1
Friday
October 01, 1982
4,0:3 PM
Operator, MANTIPLY, E. D.
TEM Cell. CC-101.5 (Large)
Driuer AMplifier Output
Running Yellow SysteM
Rotation Siatistics
Freq. Aug. Reading Aug. Error Low Error High Error High-Low En'or
(MHz) (dBV/M) (dB) (dB) (dB) (dB)
10.000 29.050 -0.95 -1.16 -0.75 0.41
15.000 29.063 -0.94 -1.13 -0.75 0.38
20.000 29. 117 -0.88 -1.06 -0.68 0.38
25.000 29 . 142 -0.86 -0.99 -0.75 0.24
30.000 29.048 -0.95 -1.10 -0.82 0.27
35.000 29.033 -0.97 -1.13 -0.82 0.31
40.000 29 . 019 -0.98 -1.10 -0.86 0.24
45.000 29.072 -0.93 -1.06 -0.82 0.24
50.000 29.012 -0.99 -1.10 -0.86 0.24
55.000 29.036 -0.96 -1.06 -0.86 0.21
60.000 29.035 -0.96 -1.10 -0.82 0.27
65.000 29.080 -0.92 -1.03 -0.82 0.21
70.000 29.092 -0.91 -1.03 -0.82 0.21
75.000 29.068 -0.93 -1.06 -0.82 0.24
80.000 29.157 -0.84 -0.99 -0.75 0.24
85.000 29.192 -0.81 -0.96 -0.72 0.24
90.000 29 . 175 -0.83 -1.06 -0.62 0.45
95.000 29.218 -0.78 -0.92 -0.68 0.24
100.000 29.217 -0.78 -0.96 -0.65 11.31
Applied Field = 30 dBV/M
103
-------
L
o
L
L
W
III
\J
NBS
S/N 1
ErH-5
o
r\J
lD
. aJ
501 r:.c
I
In
(£0
ID
I
-[
00
W N
~ ~ n ~ ~
~ ~ , ~ N N N ru N ru N ~ m
~ ~ ~ I ~ ~ ~ ~ ~ ~ ~ 7 7 ~ I ~
- I ...... - I ...... I I I
I I I
IS!
1
- J
I
-2
11'] ]5 20 25 30 35 -HI 15 50 55 61!! 65 70 75 80 85 91'] 95 11']1']
Frequency (MHz)
104
-------
NBS
EFM-5
SIN 1
Power Meter Setting Error and Wave IMpedance
Pre-Rotation Post-Rotation
Frequency Setting Error Setting Error Rw Xw
(MHz) (dB) (dB) (OhMS) (OhMS)
10.00000 -0.01 0.04 377.913 7.271
15.00000 -0.01 0 .00 383.912 5.322
20.00000 0.00 0.00 386.071 '-.739
25.00000 0 . 0 ~~ 0.05 383.928 -7.100
30.00000 -0.01 -0.05 376.633 -it.100
35.00000 0 01 -0.01 368.998 -8 . 179
40.00000 -0.02 -0.03 364.250 0.238
45.00000 0 .00 0.00 368.025 10.556
50.00000 -0 .01 0.00 379.341 18.657
55.00000 -0.01 -0.02 393.368 Ii .986
60.00000 0.00 -0.02 397.626 -4.085
65.00000 -0.02 0.03 387.727 -18.432
70.00000 0.01 -0.06 370.569 -19.696
75.00000 -0.02 0.00 357.968 -8.078
80.00000 -0.01 0.01 358.079 9 481
85.00000 0.01 0 .01 372.200 21.897
90.00000 -0.02 0.10 392.178 18.920
95.00000 -,0.01 -0.03 402.998 421
100.00000 -0.01 0.02 393.599 -21.109
105
-------
NBS EFM-5 SIN 1 Zi=Zo=50+0j
Rpplied Field: 30 dBV/m Meter. Full Scale: 40 dBVln1
La MHz 15 MHz
a a
L. ..
" '"
.. L
.:J-1 L
w
III ~
~ -0
-~ -l
li11a' 27ji1+ 3100" 91a' ! 9ji1' l80" <171a' 36111<. 90" l8121"
213 MHz 25 MHz
" 111
.. L
a c
L. '-
L. '-
!.J W
III ID
... ~
-1 -l
1813' 2713+ 3613" 913' 180" L813" 2713' 3613. 913" L8e.
313 MHz 35 MHz
a a
L. ..
" '"
.. L
~ w-l
III ~
~ -0
-1 -.,
"-
Hila' 27111+ 36; 0" 91a' ! 913' l80. ~7k3G 3613<. 90. l8121'.
106
-------
413 MHz 45 MHz
a eJ
~ L
D "
L. '-
~-1 ~-l
'" CQ
" ...,
-2 -;2
1813" 2713. 360. 913" 1813" l80. 2713" 3613. 90. l88.
513 MHz 55 MHz
a a
L. ..
" "
~ L
~ L
ILl W
III ~
"jj "U
-1 -l
1B"'" 27a. 3601a. 90~ ! Ba" l81a. 27"'" 3sa. 91a. l89.
613 MHz 1>5 MHz
a eJ
~
D
L.
L.
ILl
'"
"
L
"
'-
~
CQ
...,
-1
1813"
2i' a.
360"
313"
J 813"
-l
l80"
2713'
3613.
90"
l80'
107
-------
713 MHz 75 MHz
13 13
L. ..
" "
~ L-
~ W-l
It! j:Q
11 "U
-1 -~
...
IS0' 27ki< 310.," 90' I Ski' le.," 270' 3 Ski'. 9.," lBIa"
813 MHz B5 MHz
eI eJ
~ L-
a c
L. L.
~-l L.
W
II:! III
"D ..,
-2 -l
lSeI' 2713< 360" 913' I se' 180" 270' 3613" 90" L88"
913 MHz 9S MHz
13 13
L. ..
" "
~ L-
~ L-
W w
III j:Q
1J "U
-A -l
1913' 271iJ< 310.," 90' ISki' lB.," 270' 3Ski< 9.," l81a"
108
-------
100 MHz
!!I
..
a
L
L
W
III
."
-1
180'
270-
3613-
90'
109
180'
-------
NBS
EFM-5
SIN 1
Zi=Zo=50+0j
Friday
October 01, 1982
459 PM ( I)
Operator: MANTIPLY, E. D.
TEM Cell CC-101.5 (Large)
Driuer AMplifier Output
Running Yellow SysteM
Rotation Statistics
Freq. Aug. Reading Aug. Error Low Error High Error High-Low Error
(MHz) (dBV/M) (dB) (dB) (dB) (dIn
10.000 29.142 -0.86 -1.03 -0.65 0.38
15.000 29.266 -0.73 -0.92 '-0.55 0.38
20.000 2'9.330 -0.67 -0.86 -0.48 0.38
25.000 29.313 -0.69 -0.82 -0.55 027
30.000 29.202 -0.80 -0.96 -0.65 0.31
35.000 29. 071 --0.93 -1.06 -0.79 O.2?
40.000 28.968 -1.03 -1.16 -0.89 0.27
45.000 29 068 -0.93 -1.06 -0.79 0.27
50.000 29.165 -0.84 -0.96 -0.72 0.24
55.000 29.354 -0.65 -0.75 -0.51 0.24
60.000 29 414 -0.'59 -0.72 -0.45 0.27
65.000 29.323 -0.68 -0.79 -0.55 0.24
70.000 29.163 -0.84 -0.96 -0.68 0.27
75.000 28.966 -1.03 -1.16 -0.92 0.24
80.000 29.027 -0.97 -1.13 -0.86 0.27
85.000 29.258 -0.74 -0.92 -0.65 0.27
90.000 29.504 -0.50 -0.65 -0.41 0.24
95.000 29.667 -0.33 -0 48 -0.24 0.24
100.000 29.536 -0.46 -0.65 -0.34 0.31
Applied Field 30 dBV/M
110
-------
NBS EFM-S SiN 1 Zi=Zo=50+0j
0
(\1
~ V
I M
~
LI1 111 I
111 tD In si I 111 I
lS") ~ I ~ 111
I CSI 10 1
<.I:! I I (D I I co w
,s) I ~ '" roO
I'- I ~ IIi III
I I I I ()') oJl ro I 1 .0-
1'-- I'- I
I (D CSI
1 I tD m ~ 'D I
I II) I I N I
I ~ ? I/') I/')
r;j UJ LtI
r\J
I 1 I ~1 "- IS) IS! ~
L r~ IT! I I I
r" IS! r\. I
0 w II) I IS) I
L 0) I IS! I
-[ .:\1 !!i @ I 0J
- 01 I (D w (i) (])
W c:i I IJ1 '" '" ~
J;T>
IS) I IS! W ~ t>j t>j I
P=I I ~ I I
\J "" ,"'
I ~ LD ~ LD ~
I ~ I ~
~ ~ I
I I
-2
HJ ]5 20 25 30 35 4~ 15 50 55 60 55 70 75 80 85 9~ 95 l~~
Fr-equency (MHz)
111
-------
NBS
EFM-S
SIN 1
Zi=Zo=SO+Oj
Power Meter Setting Error and Wave IMpedance
Pre-Rotation Post-Rotation
Frequency Setting Error Setting Err.or Rw Xw
(MHz) (dB) (dB) (OhMS) (OhMS)
10.00000 0.01. -0.01 377 000 0.000
15.00000 0.01 0.02 377.000 0 000
20.00000 0.01 0.01 377.000 0 . 000
2'";.00000 0.00 0.03 377.000 0.000
30.00000 0.00 0.01. 3'77 . 000 O. 000
35.00000 0.00 0.03 377.000 0.000
40.00000 -0.02 -0 . O;~ 377 000 o. 000
4~;.00000 0.00 0.02 377 000 0 000
50.00000 -0.01. -0.03 3'77.000 0 000
55.00000 0.00 0.00 377.000 0.000
60.00000 0.01. 0 . O~~ 377.000 0.000
65.00000 '-0.02 0.01 377.000 0.000
70.00000 -0.01 0.01 377.000 0 000
75.00000 -0.01 -0.02 3'77 . 000 0.000
SO.OOOOO -0 01 -0 01 377.000 0 . 000
S~;.OOOOO -0.01 0 .00 377.000 0.000
90.00000 -0.01 -0.01 377.000 0.000
95.00000 0.01 0.02 377.000 0.000
100.00000 0 00 -0.01 377.000 O. 000
112
------- |