V.
MICROWAVE
Prof. L. D. Smullin
Prof. H. A. Haus
A.
HIGH-PERVEANCE
ELECTRONICS
Prof. A. Bers
Prof. L. J. Chu
P. A. Mandics
HOLLOW ELECTRON-BEAM
R. P. Porter
H. M. Schneider
STUDY
Direct-current and radiofrequency interaction measurements have been completed
on both a cylindrical-cathode and a conical-cathode magnetron injection electron gun.
Detailed dc measurements on the cylindrical-cathode gun have been performed previ ously, and have been described by Poeltinger.1
The guns described in this report were
operated with 1- 1 sec pulses at voltages up to Va = 8 kv.
All rf interaction measure-
ments were performed at f = 1119 mc.
Figure V-1 is a schematic diagram of the beam
tester.
with
The first
cavity is
SOLENOID MAGNET
provided
two coupling
MOVABLE
FIRST CAVITY SECOND CAVITY
loops (not shown) for the
CARBON
VIEWING SCREEN
r
VACUUM
SYSTEM
measurement
ANODE
CATHODE
MOVABLE ANODE SUPPORT
Fig. V-1.
COLLECTOR
COAXIAL LINE
/CAVITY
L
TELESCOPING DRIFT TUBES
Schematic diagram of the beam tester.
of beam loading.
Figure V-2 gives the general layout of the cathodes,
cavity gaps, and mesh target for viewing the beam cross section.
1.
Cylindrical-Cathode Gun
a.
Measurement of the Beam Dimensions
The beam dimensions
were
determined by observing the heating of a car-
bonized nylon mesh screen that was placed in the path of the beam (see Fig.V-1).
Photographs of some typical beam cross sections
are pertinent
cathode
gun.
QPR No. 70
are shown in Fig. V-3.
to the previously measured dc characteristics
1
These
of a cylindrical-
1OOr
-
80
o 60
M
40
20
CYLINDRICAL
CATHODE
GAP II
GAP I
TARGET
CAVITY
IITrA
TRAVEL
CYLINDRICAL
CATHODE
GAP IT TARGET
GAP I
!
I
CONICAL
CATHODE
.CAVITY I
TRAVEL
GAPI TARGET
GAP I
II
0
2
4
I Ir
6
I
8
I
10
I
12
CAVITY Tr
TRAVEL 1
18
16
14
POSITION
0
SMAGNET
D('-IT I(
5.25CM
5.25 CM
MAGNET
POSITION
O
I 0 1
22
20
24
L (INCHES)
Fig. V-2.
U =
I =
6
Relative magnetic flux density vs length of the magnet.
U =
KV
I
5.2 A
K° = 11.2
pAV
- 3/ 2
7
Uo=
KV
I
= 6.4 A
K = 10.9
pAV
- 3/ 2
B = 1660 G
B = 1330 G
8
KV
= 8.2 A
K: = 11.5
pAV
B
= 1590 G
a
= 0.98 cm
a
= 0.98 cm
a
= 0.98 cm
b
= 0.81 cm
b
= 0.81 cm
b
= 0.84 cm
c
= 0.66 cm
c
= 0.65 cm
c
= 0.62 cm
Fig. V-3.
QPR No.
70
Hollow-beam cross sections.
- 3/ 2
MICROWAVE ELECTRONICS)
(V.
Results of beam cross section measurements for a new gun that was used to take
all the rest of the data are summarized in the top of Fig. V-4. The accuracy of the
CYLINDRICAL
Va =
4KV
I b = 2.9 AMP
CATHODE
5
5
6
6
8
3.05
4.5
4.4
6.0
6.0
9.5
9.5
1530
1260
Bo = 1080 GAUSS 1100
MAGNET POSITION 0 CM
CONICAL
CAT HODE
Va
=
4 KV
4.0 AMP
=
CATHODE
B o = 1260GAUSS
8
4
1360
1500
0
5.25
0
5
5
5
6.8
6.0
4.15
1500
1800
0
5.25
4
4
2.8
2.1
1500
1800
0
0
MAGNET POSITION 0 CM
1310
5.25
1260
Fig. V-4.
1
I 1260
6
6
8.5
7
1800
I 1500
0
0
0
0
I
I
I
5.25
Beam dimensions.
dimensions obtained in this way was approximately 20 per cent.
b.
Beam Loading
part of the electronic
The real
first
cavity as
measurements
ured bandwidth
as
for
is
transmission
diagrammed
of the
cavity
space-charge
two different
calculated
The
the
well as
a
results
QPR No. 70
Fig. V-5.
wavelength
for all
of the beam-loading
as
measured by using the
arrangement
experimental
Gel
These
and gain measurements,
shown
in
Fig. V-2.
of the measurements
measurements
for
these
was calculated from the meas-
without the beam.
with and
magnet positions
dc parameters
The
cavity.
in
Gel was
admittance
are
are
measurements,
were performed
The measured and
presented in Table V-l.
listed in
Table V-2.
---
BEAM
TRANSMISSION CAVITY
CAVITY I
Fig. V-5.
Table V-1.
Case
V
Vb
Beam-loading measurement.
D-C characteristics of the cylindrical-cathode gun.
Ib
B
Magnet
K
f
Position
(kv)
(kv)
(amp)
(gauss)
(cm)
(microperv)
1
2
1.92
0.8
1030
0
9.0
2
2
1.86
0.8
1030
5.25
9.0
3
4
3.80
2.9
1080
0
11.4
742
4
4
3.68
3.1
1100
5.25
12.0
743
5
5
4.77
4.5
1260
0
12.7
866
6
5
4.59
4.4
1260
5.25
12.5
854
7
6
5.70
6.0
1310
0
12.9
954
8
6
5.47
6.0
1360
5.25
12.9
960
QPR No. 70
(mc)
Table V-2.
Beam loading of the cylindrical-cathode gun.
Gel
Measurement
Case
Calculation
Kinematic
Theory
((ohm) -
I
1
72.9
2
71.5
X10 )
((ohm) - 1 X10 )
Kinematic Theory
Thin Beam (b c)
((ohm) -
I
X10 )
Space-Charge Theory
Thin Beam (b=c)
((ohm) - 1 X106
-
-
-
3
129
100
104
108
4
149
121
121
124
5
120
129
126
127
6
144
138
143
139
7
-
126
133
135
164
157
155
8
TO CAVITY
WAVEMETER
MOVABLE
Fig. V-6.
QPR No. 70
Space-charge wavelength, gain and noise measurement.
(V.
MICROWAVE
ELECTRONICS)
For purposes of comparison, Gel was calculated by using the approximate kinematic 2
and space-charge
3
theories based on a thin-beam assumption (b-c <c), and also by using
the exact kinematic formulation. 2 The calculated values are also tabulated in Table V-2.
c.
Space-Charge Wavelength and Gain
The
space-charge
wavelength
and
arrangement illustrated in Fig. V-6.
two-cavity
gain were
measured by using the
With a constant power input into the first cavity,
the second cavity was moved along the beam and the power output from the second cavity
was plotted against distance.
The
space-charge
The resultant curves are plotted in Fig. V-7.
wavelength Xq was
calculated
frequency-reduction factors of Branch and Mihran.
determined from the formulation given by Bers. 3
by making
use
of the plasma
The available two-cavity gain was
These theoretical values are com-
pared with the experimental results in Table V-3.
d.
Noise
The noise power output from the second cavity was plotted against distance along
the beam for four different voltage settings.
Table V-3.
Case
Space-charge wavelength and gain of the
cylindrical-cathode gun.
I Measurement
q
Calculation
Gain
q
(cm)
(cm)
(db)
1
17.1
7. 8
17.5
2
15.0
6. 2
13.8
3
18.2
9.4
I 19.7
4
11.9
2.6
I
5
1 18.9
14.8
10.0
1 19.5
6
13.3
3.9
I 15.6
7
19.2
8.5
1 20.6
8
11.8
4. 0
QPR No. 70
The circuit used for these measurements
16.4
Gain
M2
o
((ohm)-1 X 10 3)
(db)
10.5
M2y
0. 545
1. 16
0.458
0. 701
13. 1
0. 593
1.45
7.4
0. 511
0.913
13.4
0. 627
1.58
0.554
0.92
4.74
6. 1
30-
Va = 2 KV
B0 = 1030 GAUSS
Ib = 0.8 AMP
MAGNET POSITION 0 CM
28/
.'6- /
24
E
o
I
I
22
20
I
-D
VO
18 F
15
10
5
3
'12
=
(CM)
B o = 1030 GAUSS
Va = 2 KV
28 - I b
20
MAGNET POSITION
0.8 AMP
5.25 CM
262422 20
\\
18
\ I
//
ki
I
5
I
10
I
15
1
20
12 (CM)
Fig. V-7.
QPR No. 70
Second-cavity power output vs distance between cavity gaps for the
cylindrical-cathode gun. (Cavity I, Pin
in = 20 dbm.)
3
Bo = 1080 GAUSS
2 r- V = 4 KV
MAGNET POSITION 0CM
AMPS
30 - Ib = 2.9
/
/
28
"\
\
/\
26
24
E
-.
22 20
I
I
'I
I
)
5
15
10
-j12 (CM)
20
V0 = 4 KV
Ib
.05 AMPS
3=
B o = 1100 GAUSS
26
MAGNET POSITION 5.25 CM
24
2220
/
\
_/ /
/
\ //
5
5
I
10
Il2
Fig. V-7.
QPR No. 70
I
15
(CM)
(continued).
I
20
B o = 1260 GAUSS
V = 5 KV
32 -
Ib
=
0 CM
-
30
2826-
MAGNET POSITION
4.5 AMPS
/
/
24- / /
SI
'I
22
20
\i
18
I
15
1
10
I
5
1
20
j12 (CM)
30
V a =5 KV
I
=
4.4 AMPS
Bo = 1260 GAUSS
E
.0 25
MAGNET
•
POSITION
/
5.25 CM
-/ -\
0
20
15
10
5
112 (CM)
Fig. V-7.
QPR No. 70
(continued).
51
20
V = 6KV
30
b
B o = 1310 GAUSS
MAGNET POSITION
= 6 AMPS
28-
0 CM
/0
26-
//
\
/
\/
E
o
2
4-
-v
22
\
/
-//
3.0
/\//
kJ
20 -7
18 )
I
I
I
I
5
10
15
20
12 (CM)
Va = 6KV
Ik
= 6 AMPS
B o = 1360 GAUSS
26 24
22-
MAGNET POSITION 5.25 CM
p--\
/
\ \'
/
\
C
\ /
\/
//
20
20
12 (CM)
Fig. V-7.
QPR No. 70
(concluded).
(V.
B o = 1025 GAUSS
A
Va= 2KV
A
4
08 AMPS
32
o
4
32
1100
0
6
60
1315
*
6
60
1360
*
8
95
1500
X
8
95
Ib
S-539 dbm
474
-483
-435
-446
-41.0
-42 0
m
1100
1530
A PSHOT NOISE
o
MICROWAVE ELECTRONICS)
MAGNET
POSITION
0 CM
0
5 25
0
525
0
525
Figure V-8.
Second-cavity noise power output vs distance between cathode and cavity gap for
the cylindrical-cathode gun.
26
23
26
28
30
32
34
36
1C2 (CM)
was identical to the one shown in Fig. V-6, with the exception that the power input to
the first cavity was disconnected.
presented.
In Fig. V-8 the results of these measurements are
It is quite apparent from Fig. V-8 that for a nonuniform magnetic field over
END SHIELD
ANODE
CATHODE EMITTING
SURFACE
Fig. V-9.
QPR No. 70
TRANSITION
ELECTRODE
DRIFT TUBE
Conical-cathode magnetron injection gun.
(V.
MICROWAVE ELECTRONICS)
the cathode corresponding to magnet position 5. 25 cm (consult Fig. V-2) the noise output is drastically reduced.
It was also observed that the noise output increased for an
increase in the magnetic field.
the beam.
2.
For all seven curves the noise output increases along
In one case it increases by as much as 3 db/cm.
Conical-Cathode Gun
The design data and computer results for the electrode shapes of this gun have been
described in a previous report.5
After the cylindrical corrections
6
were carried out,
the electrodes had the dimensions shown in Fig. V-9.
a.
D-C Measurements
The results of the beam.dimension measurements are given at the bottom of Fig. V-4.
In general, the beam cross section appeared to be quite symmetric.
observed only for low magnetic fields and high perveance,
Ib = 10 amps, and magnetic field Bo = 925 gauss.
Beam breakup was
for example, for V = 7 kv,
Ib
a
The perveance K = -
is plotted
V 3/2
against the magnetic field in Fig. V-10.
a
For very high magnetic fields and relatively
low voltages, V a = 2, 3, and 4 kv, the perveance seemed to reach a limiting value of
K = 7.5 microperv. On account of magnetic-field limitations, it was not possible to
determine whether or not the higher voltage curves would also approach this (or some
other) limit.
15-
L.J
a_
on
Y 1
I
500
Fig. V-10.
QPR No. 70
1000
Bo (GAUSS)
1 -~-1500
2000
Beam perveance vs magnetic field for the conical-cathode gun.
CAVITY I P.in =23 dbm
Va= 2 KV
Bo = 1330 GAUSS
Ib= 0.8 AMP
MAGNET POSITION 0 CM
\
/
E
o
0
I
15
15
I
10
20
20
12 (CM)
34-
Va = 4 KV
CAVITY I Pin = 23 dbm
Bo = 1500 GAUSS
Ib= 2.4 AMPS
MAGNET POSITION
0 CM
32
3028
E
o
o
2624
n
22
207
I
/
18
10
15
15
20
20
12 (CM)
Fig. V-11.
QPR No. 70
Second-cavity power output vs distance between cavity
gaps for the conical-cathode gun.
Table V-4.
D-C characteristics, space-charge wavelength,
of the conical-cathode gun.
and gain
Measurement
Va
Vb
Ib
B
(kv)
(kv)
(amp)
4
3.79
2.4
(gauss)
Magnet
Position
(cm)
1500
0
K
Gain
(microperv)
X
q
(cm)
9.5
21.0
9.6
Calculation
x
M2
Gain
q
(cm)
(db)
21.8
13.1
M2
((ohm)
o
0.552
- 1
f
o
3
X 10 )
1.13
Va
I
b
p
(mc)
744
075 AMPS
3.0
24
A
85
60
130
50
B o = 1330 GAUSS
o
PSHOT NOISE
1330
40
S
x
A
S
-471
1500
-476
1330
-427
1800
1660
-435
-403
~
30
-541i dbm
CIhJ
20
10
0-
28
30
32
_
Fig. V-12.
QPR No. 70
C2
34
(CM)
36
38
Second-cavity noise power output vs distance between
cathode and cavity gap for the conical-cathode gun.
(db)
(V. MICROWAVE ELECTRONICS)
Note that the design parameters (V a = 10 kv, K = 10 microperv,
not achieved.
b.
Bo= 1000 gauss) were
In fact, the gun gave much higher perveances than expected.
Cavity-Interaction Measurements
Space-charge wavelength and gain measurements were made for V a = 2 and 4 kv as
shown in Fig. V-11.
For Va = 4 kv, the measured and calculated values of .q and two-
a
cavity gain are given in Table V-4.
Because of the excessive amount of noise generated
by the beam and the limited amount of power available at the input (Pin
=
20 dbm) gain
measurements could not be made for higher voltages.
The noise data presented in Fig. V-12 indicate no growth of noise power output along
In fact, for the V a = 2 and 4 kv cases, noise space-charge standing waves
were observed with wavelengths equal to the ones measured with signal input to the first
the beam.
cavity.
For both the cylindrical- and conical-cathode guns it was shown that the noise
output from the second cavity (f = 1120 mc) had no relation to the noise observed on the
collector current which was in the megacycle range.
P.
A. Mandics, A. Bers
References
1. A. Poeltinger, High-perveance hollow electron-beam study, Quarterly Progress
Report No. 66, Research Laboratory of Electronics, M.I.T., July 15, 1962, pp. 25-29.
2. A. Bers, Kinematic theory of gap interactions for relativistic electron beams,
Quarterly Progress Report No. 66, Research Laboratory of Electronics, M.I. T., July 15,
1962, pp. 29-32.
3. A. Bers, Linear space-charge theory of gap interaction between an electron
beam and electromagnetic fields, NTF 221, 53-60 (1961).
4. G. M. Branch and T. G. Mihran, Plasma frequency reduction factors in electron beams, Trans. IRE, Vol. ED-2, No. 2, pp. 3-11, April 1955.
5. A. Bers and A. Poeltinger, High-perveance electron-beam study, Quarterly
Progress Report No. 64, Research Laboratory of Electronics, M.I. T., January 15,
1962, pp. 47-48.
6. A. Bers, L. Anderson, and K. Keller, Theory and Design of a Conical Electron
Gun for Producing a Hollow Beam, Spencer Laboratory Engineering Report No. PT-277,
Raytheon Company, Burlington, Massachusetts, 1962.
QPR No. 70