Académique Documents
Professionnel Documents
Culture Documents
1955
Library
B(^o
"Broadband
THE
OPEN SLEEVE
BROADBAND ANTENNA
H. B. Barkley,
Jr.
THE
OPEN SLEEVE
BROADBAND ANTENNA
by
Henry Brock Barkley, Jr.
Lieutenant, United States Navy
ENGINEERING ELECTRONICS
19
5 5
MASTER OF SCIENCE
IN
ENGINEERING ELECTRONICS
from the
United States Naval Postgraduate School
PREFACE
California,
January 1955
for the open sleeve antenna and extending and increasing the experimental data for its broadband performance.
The subject was given some consideration during the term preceding
the industrial tour, but the major portion of the material was developed
and obtained while working during the winter school term at Stanford
Research Institute.
material.
J.
T.
Institute, under whose direct cognizance the project was pursued and
11
1955
A Technical Memorandum on
BROADBAND ANTENNA
by
Lt.
H:
Barkley,,
Jr.
USN
AF 19(604)-1296
Cambridge, Mas'sachusetts
Approved:
B,
Barkley, Jr.,
S.
Program.
11
Industrial Experience
ABSTRACT
tenna.
vides
2:1 VSWR on
1X1
50-ohm
CONTENTS
FOREWORD
ABSTRACT
LIST OF ILLUSTRATIONS
Chapter
Chapter
ii
iii
vi
INTROLUCTION
....
Chapter
Chapter
............
E.
EVALUATION OF
EVALUATION OF Z^ AND EFFECTIVE LOADING
F.
LIMITATIONS OF THEORY
D.
fe
.........
6
8
10
.................
...........
20
20
22
'23
iv
24
24
27
CONTENTS
G.
H.
Chapter
Chapter
............
Chapter
34
34
35
3 5
36
3 6
3 6
43
44
50
53
CONCLUSIONS
55
LIST OF SYMBOLS
REFERENCES
SELECTED BIBLIOGRAPHY
.
32
................
Chapter
28
52
52
52
57
58
59
ILLUSTRATIONS
Fig.
2
3
5
Fig.
Fig.
14
15
16
17
18
21
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
10
Fig.
11
22
26
...
28
Fig.
12
Fig.
13
29
14
Experimental Setup
34
15
16
38
17
39
18
40
19
41
42
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
20
YX
37
ILLUSTRATIONS
Fig.
Fig.
Fig.
Fig.
Fig,
Table
21
22
23
24
25
45
47
48
4'9
51
49
ix
BROADBAND ANTENNA
CHAPTER
INTRODUCTION
parasites,
1)
T.
J.
Bolljahn,*
investigation were not very extensive and no report was published. However,
these data indicated that the antenna had good impedance characteristics
over
partly experimental.
In Chap.
These parameters
maximum band-
The theory,
ducted in
and
There are
the open sleeve to allow a rather wide range of compromise between the
dipole.
in
J.
T.
L^c:?
A-II97-52
FIG.
CHAPTER
A.
with the dipole element being driven against the ground plane
Fig.
2(a),
by
generator
V.
2(b).
r
Ll
y//////////////////
y///////////////////////
FIG. 2(a)
FIG. 2(b)
With
(1)
grounded,
F,
and
^'
in opposition maintains
the
tion permits the operation of the open sleeve to be broken down into two
modes,
This procedure is
line mode is not immediately obvious and the division of the antenna
the nature of this termination and the current distribution may be ignored
and an equivalent circuit derived for the open sleeve antenna in terms
of the two fundamental modes of operation.
Zin
(2)
where / is the total current flowing into the base of the dipole. The
I^,
Then
Generator
F.
I,
+ Ij.
(3)
antenna mode to
is called Z^,
V^
where
fe
of course,
implies
Combining Eqs
(2)
Z.
nz
_
Z^
(6)
k,
and Z^.
3.
in
consideration of the
z.
^.
^r
FIG 3
B.
of the open sleeve antenna may be obtained by analogy to the sleeve an-
tenna
Taylor
on the outside of the sleeve and on the protruding part of the center di
the restriction that the entire dipole of the open sleeve can radiate.
Transmission line currents flow within the coaxial sleeve but do not correspond directly to the transmission line mode currents of an approximately
sleeve antenna,
parasites,
sinusoidal current
as would be expected.
or antenna mode,,
More important,
It is
It should be
noted however,
line modes on the open sleeve is not the same as that on the sleeve an
tenna because of radiating currents along the entire length of the dipole
of the open sleeve antenna.
The current
Again,
to
first approximation,
C.
EVALUATION OF Z
From the preceding discussion,
pedance, Z^,
it is clear
as
long as
the spacing of the parasites from the dipole is not too great and the
current flowing in the parasites is not too large, this impedance will
not differ excessively from that of a uniform dipole.
are to be kept electrically quite small,
the imped-
ance variation of Z.A will be assumed to be the same as that of the center
D.
10.
EVALUATION OF
Although the current in the parasites is small and does not have
much effect on Z., the fact that one component of it comes from a shunting
of part of the total antenna mode current must be accounted for in the
It has
However,
it turns out
point impedance upon the choice of the relative sizes of the current
and the particular choice made gives
then,
with I
I.
sin
/3h
(7)
sin
/3s
(8)
- ^^^.
and
Ap
since
and,
from Eq.
(9)
(4)
Substituting Eqs.
(7)
and (8)
^ ^^Ap
(10)
in Eq.
(10),
sin
1
jSs
(11a)
+
2| sin. i5h
(i.e.,
for
/3h
tt)
fih
tt,
mation for k may be made on the basis of King's* theoretical current distribution curves by noting that the base current on an antenna is closely
approximated by assuming
effective electrical length 0.5 rad longer than the actual electrical
This gives for
length.
sin
-i
2|
E.
with
k,
sin
/is
^
{/3h
>
jJh
tt
0.5
(lib)
)|
mode impedance are the length and termination of the transmission line.
Although this mode is excited with antisymmetr ical currents at its driving
it does not follow that the
point,
sleeve.
It will be shown,
however, that Z
may be evaluated to
good
2)
mode current at the ends of the parasites also requires the vanishing of
the transmission line mode current,
it is
apparent that
in which all three conductors do not terminate at the same point has some
radiation.
Hence,
the cur-
rent of the dominant mode does not entirely vanish at the end of the line,
but must assume such
vanishes.
The effect of the higher order modes and radiation may then be
accounted for in
required to flow at the ends to cancel that from the higher order modes.
If the amount of radiation is small, which is the case for closely spaced
parasites,
and
the deviation of the transmission line from the open circuited condition
The value of Z-
transformed along
Z-^,
The charis
given
Z,,
207
^og(-^f\
(12)
but
ical value for Z-^ for some average case and then let it vary in accordance
with theory.
3.
short antenna,
trans-,
coaxial cable is
not present,
Radiation from
radiation of
Therefore,
is inversely proportional
to the square
is neglected.
LIMITATIONS OF THEORY
Most of the cases where the application of the theory fails have been
A brief
half wave
errors in Z.,
is not applicable
reasonably
10
3.
CHAPTER
A.
With the theory for the open sleeve antenna developed in Chap,
it is
results
The
are most readily discussed in terms of the shape of the driving point
is generally applicable,
is
half wavelength.
quarter wavelength to
fied is that of the dipole and is identical to that used for Z^.
locus is the familiar one of
large,
^,
Its
The
through multipli-
impedance, causes the driving point impedance curve to break away from
Z.
and make
is
frequencies
at
variation of
magnitude of Z_
A change in the spacing of the parasites or in the diameters of the
Z^^,,
crease of
Zqj.
to be of smaller magnitude,
The de-
while
11
Since Z is capacitive
general capacitive
shift.
The length of the parasites controls the length of the transmission
line and,
hence,
although the
VSWR is quite low and has little variation over this range. With shorter
parasites, resonance is not reached until higher frequencies; the impedance
loop is much larger in size and covers
The vari-
B.
carried out for several cases for which some experimental data were available before the theory was developed.
the establishment of
of Chap.
It was
as discussed in Sec.E
2.
gated theoretically and experimentally were picked because of the simplicity of construction of models for
also covered
They
appeared promising.
The physical dimensions for the cases considered theoretically under
the theory are,
(see Fig.
Z)
1)
in.
/Z
22 cm
in.
0,
in.
22 cm
11 cm
9,
11,
and 13 cm,
and
in
4
12
chosen for Z^^, which is used as the reference for the other computed
values,
is
Zj.^
500 ohms,
for
600 Mc
in.
and
impedance loops over the widest frequency range, but the variation of
the
the
with
The variation of Z
spacing.
is plotted as
spacing.
as a function of frequency is
plotted in Fig.
in.
with
extrapolation for higher frequencies, but show somewhat more vividly the
nature of the variation of VSWR.
It is
It
sites is to cause
is
a
theo-
retical and experimental results for 11-cm parasites are plotted on the
same Smith chart,
spacing.
Fig.
for 1-in.
is
and Fig.
for 1/2-in.
ment;
spacing,
there is also
The
13
0=f.d = f'
h= 22cm, 2^=67X2
A-II97-53
FIG.
14
A- 197- 54
1
FIG.
15
40
250
300
400
350
450
500
FIG.
650
600
550
FREQUENCY
700
750
Mc
C.
theory in Chap,
2.
There are
Above 600 Mc
no longer decreases at
Z^
were used at 80
Mc
for example,
Therefore, if
16
A- 1971
FIG.
17
56
A-
FIG.
97- 57
18
In-
j^
The theoretical
values are somewhat larger than the corresponding experimental ones, allowing the capacitive shunting by the transmission line mode to be somewhat more pronounced than it would be for lower Z
values in
..
However,
the second
differences
19
CHAPTER
/4
A.
considera-
introduction of
third wire in
This is particu
(This presupposes,
of
That is,
position of the fields of two similar two-wire lines, where the wires of
like polarity coincide, with negligible distortion of the original fields.
This is equivalent to saying that the charge distribution on the two wires
is
In order
Figure
and in-
sofar as the above relationship remains valid, the two parasites of the
open sleeve may be treated independently, and in any case as two. wire lines
excited in shunt.
20
taining
In
(13)
[Z - 2a) (Z + 2a)
lines of like polarity coincident at the origin and the other two lines
to
to
FIG. 9
and how much do they differ from those of the indiviJual two-wire
cases?
21
it
turns out that the equipotential surfaces near the outer charges are
nearly circles while those around the center charge are elliptical.
The
equipotential circles from the separate two-wire lines when the line
charges do coincide.
Thus,
does not result in the three-wire case except in the limiting case of
line charges.
However,
Knowledge of the
B.
broader in-
10.
z.
>z.
'^.
FIG. 10
22
C.
This discontinuity may effectively be ata change in diameter of one of the conductors, since it was
shown that the two-wire lines equivalent to the three-wire line consist
tributed to
Alternatively,
or
constant times
D.
The reference value for effective loading on the end of each line,
Z
and Z
1 1
'
is most
since even for equal spacings the radiation and losses from lines
with one conductor of variable length are somewhat dependent upon the
variable length.
difficult.
inversely
It should be noted
also that these values cause the generalized equivalent circuit to reduce
to the previously developed basic equivalent circuit when the parasites
23
(i.e.,
Zj.^
Zj,^
2Z j)
D.
of unequal length.
of spacing upon k,
k,
and using
the general
fe
sin
i3Sy\
7]
4
I
sin
4
sin
(5s^
+ 0.5)
sin(/3/i
(r <
600 Mc)
/3^|
E.
+
I
/iSj
rn
sm
/:JSi
sin
0.
single
parasite can be discussed from the theory without actually carrying out
theoretical computations.
predictions
With the single parasite, the characteristic impedance for the twowire line is about 1.5 times that for the three wire line with the same
parasite spacing
transmission line mode impedance is thus the same, but the impedance itself is 1.5 times as large.
Hence,
is
reduced.
Actually,
The
balanced two-wire line might well be less than from the three-wire line.
24
.,
but
ftZ.
is reduced
The resulting driving point impedance should then follow more nearly
the curve of Z^.
the first
Thus the
parasites
If an attempt is made to increase the effect of the single parasite
These shifts
result from the decrease in loading on the transmission line with the deThis case is illustrated in Fig,
creased spacing.
Z)
single parasite,
in.
/I
22 cm
in.
11 cm
11,
where theoretical
with
The computations for this case do include the fringing effect of the
transmission line.
all measurements except the very first ones for equally spaced parasites
of equal length were taken on a 50-ohm slotted line.
25
at
A-II97-58
FIG.
11
26
F.
which suggest its superiority over the equal length cases. The
Jinith chart
impedance plot shows that with equal length elements a fairly severe increase in VSWR
must be accepted near the middle of the frequency band if wide bandwidth is desired. On
the other hand,
this
frequency
curves thus resembles very closely that of the impedance curves for two
tuned,
coupled,
coupled circuits.
If one of the
shorter parasites can be used to give the general folding desired of the
"^in
locus,
VSWR or perhaps
double hump.
of energy between all three elements and various modes set up.
as
the
possible excitation of
frequency where the longer line is inductive and the shorter capacitive,
Z)j
r^2
(see Fig.
12)
in.
Sj
13 cm
in,
9 cm
1
2
27
is
normal-
13.
indicate
substantial improve-
Unfortunately,
777777777/777?
FIG. 12
fS
the
In fact,
the theory
fully adequate for indicating the types of changes which can lead to
A dis-
G.
removal,
reduction,
or
effect from the frequency band of interest while leaving undisturbed the
although
improvement possible.
accomplished,
although there is
A-II97-59
FIG.
13
29
either
The decrease of
Z^j,
causes
antiresonance,
at
to imply
quarter wavelength
This condition tends
by
detri-
mental effect on the reactance term, but the resistance can predominate
;for
theoretical curve).
There is
above
the dipole,
tion.
This,
is
a
suggestion.
However,
determined experimentally.
a
30
the antenna
broadband operation could compensate for the lower efficiency would have
to be investigated. In this regard it might be pointed out that the ultimate in broadbanding would be the addition of a resistor across the generator terminals, but effective radiation would hardly result.
Another
situation to be coped with is the power rating obtainable for resistors
in various frequency ranges.
difficult to handle analytically with the present theory, since the antenna mode of operation would be disturbed.
Evaluation of k particularly
would be difficult,
significant
sponse is changed.
However,
compromise whereby
An indication of
still
practical lower
limit is the impedance locus variation with parasite length for equal
length parasites. The length of the longer parasite can not be made shorter
than that of the equal parasite case which gives the maximum tolerable
however,
resonant frequency.
and the proper choice of spacings may be obtained by considering the case
of antiresonant,
31
Zj.
-jZ^^ ctn
/^s
(15)
^OT, ctn
/3s,
ctn
-Z^
/3s^
or
Z_ Ctn
"'i
pedances.
Hence,
It
o)
is
ctn
-Z._
"'2
(16)
c
> S,
00^
for
Zq^
>
Zqj.
is
o)^
for
Z Q^
<
Z Q^
is greater than
less than
D,
should be less
even
a
and Z
provement
Thus,
should be made in the antenna setup to give the best attainable performance with unequal lengths and unequal spacings of parasites.
The actual
H.
6.
32
single parasite
\;he
The single
pair was chosen because of its simpler analysis and physical construction,
however,
antiresonant impedance exists with the two pairs as with the single pair.
it appears that
33
CHAPTER
A.
14.
The arrangement is
slotted line was three 20-in. Hewlett-Packard lines bolted end to end,
with
by Styrofoam slabs
HP AUDIO
SIG.
GEN.
205AG
/
/
/
/
/
GR UNIT
OSC.
1209 -A
250-920MC
HP LOWPASS
LUMPED
FILTER
fc = 700Mc
TUNER
HP SLOTTED LINE
(3-20in. LINES JOINED) 805 A
CONST.
(SRI)
/
/
/ LOAD
/
/
/
/
/
/
PWR
SUPPLY
GR-I205-A
UNIT
HP
SQUARE-
LAW DETECT
4I5A
A- 1197-60
FIG.
14
EXPERIMENTAL SETUP
A 1/4-in. diameter
erected over
of parasites
7,
9,
11,
and 13 cm
were
34
Four lengths
1,
Two parasites of
B.
a
a
The short
C.
ACCURACY
The accuracy of impedance values
(/?
to 10%.
is
limits of precision.
However,
there are
slightly more error is present because of some limitations of the measuring equipment.
35
CHAPTER
A.
Fig.
17,
1)
15
and 16,
data taken with the equipment setup described in this report for some of
the same conditions as those of Figs.
15-17.
of Figs.
18 and 19.
A normalizing impedance of
15-17 are slightly different from those used throughout the rest
of the report,
The agreement of the 67-ohm data with that computed from the theory
has already been shown (Figs.
curves of Fig.
17
and 8)
3.
The VSWR
ones of Fig.
6.
The
agreement between the 50-ohm data and the 67-ohm data is seen to be quite
good when allowance is made for the different value of normalizing impedance
The main points to note with this data are the size and frequency range
of the impedance loop as
spacing.
1/2-in.
parasite
The plot of VSWR shows quite clearly the possible compromise be-
B.
13.
There is
As mentioned in Chap.
4,
the discrepancy
36
The theory
FIG.
15
37
A- 197
I
FIG.
16
38
-62
does,
however,
the form of the impedance curve and hence provides a useful guide to the
designer
4.5
250
300
350
400
450
500
550
- Mc
600
700
650
750
800
FREQUENCY
FIG.
17
13
cm long.
Generally,
tliat
ob-
impedance variation for the open sleeve with parasites of unequal length
is very similar to the variation which would be
In addition,
yielding
39
It
Z =50il
A- 197- 64
1
FIG.
18
40
A- 1197-65
FIG.
19
41
A- 1197-66
FIG.
20
42
is primarily
frequency.
together in the resistive end loading, the failure of the theory can be
attributed to errors in Z
higher frequencies before, but it was chosen to vary strictly in accordance with theory.
length,
In this case,
spacing increased, closer agreement could be had between theory and experThe important point, however,
iment.
is that one
needs only keep in mind the deficiency of the end loading while employing
the theory to direct the changes necessary for improvement of antenna
performance
C.
but did not predict much about the degree of improvement offered,
The
The measurement of
VSWR by itself is much more rapid on the slotted line than when minimum
shift is also required,
changes.
Full impedance data were taken and plotted for the one arrange-
Only the
to 1.5 in.
increased the resistive end loading by radiation and reduced the maximum
value of VSWR at antiresonance from 18.0/1 to 15.0/1.
The frequency of
maximum VSWR was reduced from 580 to 560 Mc because of the effective lengthening of the transmission line by increased fringing with the greater spacing.
43
impedance.
the parasite
lengths
D^
1/2 in.
Sj
11
D^
in.
Sg
cm
cm.
This was
decided
the
resulting performance was still not so good as some of those achieved with
There was
tion considered the optimum attainable with unequal parasites and is dis-
D.
in.
Sj
11 cm
D^
1.5 in.
Sg
cm.
Impedance data were taken for these conditions and are shown in Fig.
VSV/R curves
22.
It
21;
should be noted
that the lower frequency impedance loop still looks very much like that
for the case for parasites 11 cm long with equal
1-in.
spacing,
again
verifying the statement made previously that the lower- frequency impedance
44
A-II97-67
FIG.
21
45
Furthermore,
material improvement in the higher frequency range over the use of equal
parasites.
Thus,
equal length.
It does
appear, however,
that if
greater than two to one must be used with an antenna of this type, some
transmission line of
Accordingly,
the
impedance data for the unequal parasite case just discussed is shown in
Fig.
22.
in.
and S
11
cm.
It
is
seen
that this latter curve behaves very similarly to the case of unequal par-
asites normalized to 75 ohms, except that at the high frequency end the
rise in VSWR is not quite so abrupt.
spaced
in.
line.
It
is not
23.
The resulting
higher impe-
as compared to a dipole,
however,
that no
mid frequency range of the pass band of the antenna, with characteristic
46
those with
in general
for
is
inserted between
260
300
340
380
420
460
500
540
580
620
660
FREQUENCY-Mc
FIG.
22
47
I.
Other results
A-
FIG.
23
48
1 1
97- 69
9
1
d=
D = l",s = 9cm
h=22cm
/Zo=50il
--0
x^X
a.
-o--'
^-K, \
"'
9
1
1
'9
.^--^^
"
v\
<
.^^^\\D,=D2=D=
vv^^^
^^x>>--^
1",
s=
^^
9cm
>-.l/
1^
300
350
400
450
/y
^^IN^
^v
600
FREQUENCY-Mc
500
550
650
700
750
800
24
FIG.
TABLE
^1
(IN J
Unequal
^2
(IN.)
Wz
Si
(cm)
S2
(cm)
11
Frequency
Range (Mc)
^0
(ohms)
Max.
75
2.05/1
50
2/1
2.3/1
50
2.4/1
2.2/1
VSWR
1.96/1
(with 61 ohm
matching
section
A./4
Equal
75
8/1
2.64/1
11
11
67
2.1/1
1.96/1
49
E.
Comparison of both impedance and VSWR curves for equal and unequal
parasites indicates that the case of equally long, equally spaced parasites allows as wide control over the results obtainable, and gives as
desirable performance, as any combinations of unequal lengths or spacings
at least
Further improve-
In-
to 13 cm causes
reduction in the
general vertical
shift toward the low impedance region of the Smith chart. Changing the
spacing D causes
with smaller spacings yielding larger impedance loops having marked cap-
acitive components.
From
implying that
smaller parasites might help to match the antenna directly to 50-ohm coaxial cables.
If both d and d are changed while maintaining
with Zq_ thus remaining constant, the performance of the antenna is not
materially changed.
25 where an experimental
VSWR curve is plotted for increased d and d but the same characteristic
impedance of the antenna transmission line mode as that reported throughout the paper.
cited by
being
good figure.
However,
using
5fr
h=22cm
ZQ=50ii
s=9cm
c
>
-6
300
350
400
450
ts-
500
550
600
650
700
FREOUENCY-Mc
FIG.
25
51
750
800
CHAPTER
A.
GENERAL
There are several other avenues of approach
to the
analysis of the
approaches
B.
to
the ordinary
common to both.
transmission line
degenerate form,
metrically fed antennas when its image in the ground plane is considered,
so may the open sleeve
antenna.
King
shows how
single asymmetrically
one
monopole corresponds to the upper half and one to the lower half of the
asymmetrical antenna.
halves,
By
to frequency variation
then,
that
the superposition to the sleeve or open sleeve antenna would also exhibit
broadband characteristics.
52
and hence
con-
a
a
point that the change of antenna electrical length with frequency produces compensating changes in current distribution, because of the re-
basically sinu-
Coleman
with the sleeve nothing more than an extension of the feed trans-
mission line.
antenna itself and the transformation of this impedance to the base of the
antenna through the sleeve transmission line.
Thus there are many aspects
of
computations
G.
is
sleeve antenna,
particular obstacle.
impedance between similar elements can be found for certain rather restricted
cases
Chaney
rent distributions for the calculation of mutual impedance between like elements, and also elements of different length.
the values obtained from his and the other published data would be
questionable when applied to the open sleeve antenna because of the uncertainty of the current distributions of such closely spaced elements. Also,
53
considerable frequency
self impedance,
Even
half
wavelength.
In short,
a
mation,
Computations were carried out for one particular case, where the
dipole was very nearly
(13
spacing.
Calculated
Measured
Z.
38.0
in-_ 37.0
for f
j
?'
-'
330 Mc
8,
1.2
cause
38.0 + J 7.5
The
could be covered with the available data and the question of how much
the open sleeve would depart from the behavior of the ordinary sleeve
antenna.
Nevertheless,
at
least
S4
CHAPTER
CONCLUSIONS
The
antenna and
found theoretically,
1/4-in.
diameter,
stant.
It was
For
variation in diameter of
larger
A compromise
cable;
however,
the insertion of
a
ohms,
if requirements call
with
50
50-ohm line,
results compar-
an ordinary dipole.
band antennas.
simplicity.
It is,
however,
fication that quite readily could be made to existing v-h-f and u-h-f
antennas in aircraft or ships.
Furthermore,
55
in the h-
installations,
it
Thus,
would not be surprising to find the open sleeve antenna used for
S6
SYMBOLS USED
C
speed of light
frequency
hf
high frequency
j^
Jj.
fe
TEM
transverse electromagnetic
uhf
^^,^2
vhf
VSWR
Z^
Z^
/v'-plane
In/K
fi
wavelength constant,
)y
wavelength
ii
Hallen's constant,
<^
^^ "w
57
REFERENCES
1.
W.
2.
3.
John Taylor, "The Sleeve Antenna," Tech. Report 128, Task Order 1,
Cruft Laboratory, Harvard University, Cambridge, Mass. (20 April
J.
1951).
4.
S.
Inc
6,
A.
,
(D.
A.
D. Duncan, Jr,
"Characteristic Impedance of Grounded and Ungrounded Open-Wire Transmission Lines," Vol 18, p- 10,
R.
(June 1938).
7.
E.
Line,"
9.
F.
10.
J.
58
SELECTED BIBLIOGRAPHY
1.
A. Blasi, "The Theory and Application of the Radiation MutualCoupling Factor ," Proc IRE, 42, 7, pp. 1179-1183 (July 1954).
E.
2.
3.
Federal Telephone and Radio Company, Reference Data for Radio Engineers, Third ed.
(American Book-Stratford Press, Inc., 1949).
J.
Appl.
4.
"A Note
V.
N.
Granger,
of Folded Dipoles,"
5.
Helliwell,
A.
E.
J.
"Engineering Survey of U
CONFIDENTIAL.
6.
E.
C.
Hall,
7.
R.
W.
Antennas,
New York,
8.
J.
D.
York,
9.
10.
12.
(Prentice-
Kraus, Antennas,
1950).
First ed.
Radio Research Laboratory Staff, Harvard University, Very High Frequency Techniques First ed.
Inc.
Vol I (McGraw-Hill Book Co.
New York, 1947).
,
11.
(John
59
13.
C.
T.
15.
T.
16.
F.
Co.
E.
,
Inc.
60
First ed.
(McGraw-Hill Book
IB225
Barkley
The open sleeve
broadbanc^
antenna.
^--^
B226
3334
Parkley
The open sleeve broadband
antenna.