Documente Academic
Documente Profesional
Documente Cultură
_ _
_ I
W
END
8 -7 9
- —--
~~~~~ ~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
S
4
~
j
“~ \; .
~~ H
>-
S r L
La_i
-~,
-.:A’
~~~~~
JUL 12 1919~~
cI
~~~~~~
‘i 9 U’j 12 014
__________________________________________________ ________________
S E C U R I TY C L A S S I F I C A T I O N OF T H I S P A ’ ~~ (lb) ,.n O.s .Eot• •d)
~
DCD(SOT I I U I• T IE••kI Q READ INSTRUCTION S
I L~~ I ~~ I UJ r’ U
~~~ E BEFORE COMPLETING FORM
~~~
1. ~~~~~~~ ~~ u u e E R A C C ES S I O N NO. 3. R E C I P I E N T s C ATAL QG MBU
~~ ~~~~
NRL Memorandum Report 3989 .
/ I -
________ —- S. C O N T R A C T OR G RA N T NUMSERI .)
- - . —. -—~~— i
J. L./Seftor *, V. L.~Granatstein , K. R. Chu , P. ‘SPran ~lef and
M. Read . . . _ _ - -._ . -. .~ - ______
DO ‘ “°‘
, ~~~ ,, ~473 EO i0N OF I N O V A S IS O B SOLET E
102- 114- 660 1
~~~~
I S E C U R IT Y C L A S S , F I C A IOH or T H I S P A G E (UR.n 00.. INIS1IRI
V
4
have yet to be measured directly but no fundamental problems w7fre observed which will prevent
I
successful achievement of the design predIctions (viz., bandwidth ,~~~ 10%, power on the order of
10~ waLLs, efficie ncy > 10%). ,
,
\
A ’
JL -
14
- - --
~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~ _ _ _ _ _
CONTENTS
INTRODUCTION
PRINCIPLES OF OPERATION 3
DISCUSSION 16
REFERENCES 18
_ ‘ _
i ~~~ L~ .
ill
/
THE ELECTRON CYCLOTRON MASER AS A HIGH-POWER TRAVELLING-
WAVE AMPLIFIER OF MILLIMETER WAVES
INTRODUCTION
[1—5 ] and experimental [6—B] began two decades ago . These studies have
power leve ls that are or ders of ma gn itu de lar ger t han prev ious ly ava il-
able at millimeter wavelengths,and with good efficiency . These oscil-
bandwidth.
Note : Manuscript submitted March 1, 1979.
~~~~~~~~~~~~~~
In 1 975, an experiment on the electron cyclotron maser instability
in an intense relativistic electron beam demonstrated wideband amplif 1—
design predicted that the type of’ high power and efficiency which char—
ac t.er izes gyr otron osc illators woul d als o be obta ina b le in th e gyrot ron
amplifier (viz., 340 kW at 51% efficiency), while at the same t ime , pre—
first Operation of the gyrotrori travelling wave amplifier which was con—
~,
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~
~‘t ~INCIPLf~S OF OPEHAT ION
will outline for the readers ’ convenience the general principles under—
ly ing the operation of elect ron cyclotron masers. For simplicity, we
will assume that the electrons are moving transverse to a steady magnetic
field , ~~~~~, and that the unquant ized electron velocity component parallel
to B is negligible.
0
electromagnet ic wave by int eract ion with free electrons gyrat ing in a
steady magnetic field comes from the work of Twiss [1] and Schneider [2].
- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~
. ~-
.
L .
- - ‘..
-
1
[19 ]:
where the relativistic energy factor y(q) = (w +mflc2) /mc2. The form of’
q
Eq. ( 14 ) makes clear that relativ ist ic eff ec ts serve to decrea se the
The class ical descr ipt ion of the electron cyclotron maser is closely
than the streaming energy , and so follow a helical path about the mag-
net ic field lines. The Larmor radius, rL, is usually much smaller than
beam diameter may be large r tha n the wave lengt h at which amplification
occurs (v i z . , A 21Tr / B
L
where is the transvers e velocity normalized
short wavelengths.
have almost a single value of transverse energy but are randomly distrib-
lose energy while others gain energy depending on the initial phase of
and there is no net wave amplification. However , the cyclotron
phase. The resulting phase bunching will favor wave damping if the wave
freque ncy is slightly smaller than the cyclotron frequenc~, or its harmon-
’
ics in the reference frame where axial electron energy vanishes (W .
~~
the wave. ~{alf a cyclotron period later , the net azimuthal notion of the
electrons has reversed , but s ince t h e ph ase of t h e wave also reverses in
I .— . ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~
— - — — -‘- —— — -—- - — - -----—- -
— -
~
the electrons and extract energy. This synchromism between the orbiting
electrons and the field implies that in the laboratory frame , the fre—
eration becomes much more difficult when frequency and photon energy is
microw ave tubes are based on beams of free elect rons, and radiate a huge
number of’ photons per particle with little change in frequency ; however ,
the wavelength of radiation is not a characteristic of the 7 articles but
is determined by the physical dimensions of some resonant structure such
6
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -- - -
The elect ron cyclotron maser is based on a beam of’ free electrons
and thus emits many quanta per particle. In addition , the frequency is
fixed by a characterist ic frequency of the part icle (i.e. , the elect ron
cyc lotron frequency) and no small— scale resonant structures are required .
Thus , the practical development of the electron cyclotron maser has made
possible a leap in power generation capability at millimeter wavelengt hs .
7
_ _ _ ___ ---— — -
- ~ - -
-— —
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~ ~
the TE01 circular waveguide mode (azimuthal electric field )and the fun-
damental cyclotron mode of an electron beam were studied . The beam is
would favor the process [20]. With the ratio of transverse to axial
momentum chosen as p /p = 1.5 , the gun design attempted to minimize
0 ~0
momentum spread. A computer analysis of electron trajectories for the
chosen gun configuration indicated that momentum spread in the beam would
~ ~~ “ r-’~ 0 = ~
be 2 p /P10 l0 . This beam is propagated down a uniform
~
metallic tube for interaction with an injected r.f. signal. Both gun and
tube are placed within a superconducting solenoid , which provides a converg-
ing field at the gun , and a uniform field over the interaction region.
backward moving wave). The interaction between the beam and the radia-
tion can occur over the downstream length of the uniform field , which is
17 cm. Beyond this , the magnetic field rapidly decreases , and the elec—
troris , which are guided by the field lines , are collected on the wall.
0~ -
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The electromagnetic wave continues to propagate down the tube , towards
mechanism, measurements were made of the power coming through the system
with the electron beam on, and the beam off. Any effect due to the unop—
timized input coupler or wall losses in the drift tube are, thereby , sub—
The optimized design values for this device as taken from reference I
[18] are listed in Table I. The aim of the designs was to produce an
output power of ~ 100 kW at 35 GHz with good efficiency , gain , and band-
width. By tuning magnetic field it is possible to trade off efficiency
and power for gain and bandwidth. Column (a) gives predicted performance
when B is chosen to optimize efficiency and column (b) gives predicted
0
performance when B is chosen to optimize gain. Both t~p /p ~~ and
0
were considered to be negligible in these calculations. The geometry and
electron beam parameters are held constant for both design (a) and design
(b), a 71 kV , 9.5A electron beam being assumed in both cases.
Because the input r.f. does not couple all its power into the grow-
ing wave , a careful analysis of the gain process , including the transient
where E is the init i.al electric field , and r is the spatial growth rate.
0
2l~
For rL > 1 , we find that the power gain in dB is given by
The first term on the right hand side of Eq . (6) represents the cou-
pling loss due to some of the input radiation exciting waves which do not
grow exponentially .
The line through each set of data points in Fig. 3 fixes a measured
value of gain from which a growth rate may be determined according to Eq.
10
_ _
(b), using 17 cm as the value for L. The values correspond to growth
rates r )
~.14 , 0.1~ , and 0.18 cm ~ for ‘
b 3.5 , 6.0, and 7.5 A
respective ly .
imental width ~B/B = 2.9%. The output power was also observed to have
0
a broad peak as V , the electron accelerating voltage , was varied; P
0
I
changed by less than 1 dB as V was swept from 67 kV to 70 kV.
11
L ____
~~~~~~~~~~~~~~~~
CO PA ~ ISO~1 OF EXPERIMENTAL RESULTS WITh ThEORY
• 3/ 3x e (v x (1)
( 3 / 3: + V —
~~~~~ )f
( 1) 1)
e ~ + v x ~ ) . (~~/ ap)f (7)
tivelv, and ~~~~ B~1~ are the wave fields of the TE waveguide node;
0~
.( 1)
_eJf~~~v 8d 3p ; (8)
+
2, : iWt )JrJ~ l)
- - = J (~~ r)dr
~~~~~~~~~~ (9)
12
L •~~~~~~:: ~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~
To solve these three equations , one must fi rst specify the form of
Using the methods of ref. [23] , the set of equations (7) through
(10) has been solved to yield the following dispersion relation for the
interaction between the TE waveguide mode and the .S—th electron cyclo-
on
tron harmonic. -
5
2
pdp ~~~dp g (p ,p )
~
- -
yr 2E (x )
(II)
2 2 ( -
- k c )p H ( r , r
S ~~ Q ~~ L~
- ~W
f
~
22
~~m c (w - k~v~
- SW )
k v
~ ~ - SWee
13
r
The amp litude growth rate per unit length (I’) has been solved for
nume rically from Eq. (11) for the mode and cyclotron harmonic in the
g(p 2 2 2 2
= C exp ~ (p -p ) /L~p~ -
1 ~~~~~~~~~~~
~~~~~
.
~p / p , as a parameter .
growth rate peak has been marked on each curve in Fig. 5. Also marked
i ts peak value ; it may be seen from Eq. (6) that a 15% drop in I’ cor re-
It is observed from Fig. 4 that for a cold beam (Tho) the peak
v~ 1ue of i’ (falling at B 1
= 13.5 kG) is 0.40 cm while ~ B/B = 6.7%.
(viz., i’= 0.14 cm 1 at B~~~ 13.0 kG and ~ B/B = 2.9%) . However , the
0
calculated values of peak F, the corresponding value of B
0
and ~ B/B Q all
of theoretical model and the fact tha t elec t ron monentuin spread was
that growth rate will depend on beam current as F .— ~~~~~~~ [23]. This
15
• -
L. • •. • • •
~~~~~~~~~~ ~~~~~~ ~~~~~ ~~~~~ ~ ~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
—~~~ -~~~~~~~~~ •~~~ —
DISCUSSION
input power was limited and the device was not driven into saturation ,
by linear theory . This agreement between experiment and theory was found
for the absolute value of gain , the scaling of gain with current and the
variation of gain with magnetic field. Thus one has some confidence that
16
— ~~~~~~~~~~~~~ ~~~~~ -
~~~ ~~
o3~ illator~3 are already being effectively applied to controlled thermo—
The authors thai~ Dr. - . Jor y , Mr. R. Lucey and Dr. L. Barnett for
:~ e p fu l ii~~uss ions and ex~ erimental assistance .
-
REFERENCES
1965 .
[6] R.H. Pantell, “Backward wave oscillations in an unloaded waveguide”,
Proc. I.R.E., vol. 47, p. 1146 , June 1959.
[7] I.B . Bott , “Tunable source of millimeter and sub—millimeter electro-
magnet ic radiat ion ”, Proc. IEEE , vol. 52, pp 330—331 , March 1964;
also, “A powerful source of millimeter wavelength electromagnetic
radiation ” , Phys . Letters, vol. 14, pp 293—294, February 1965.
[8] J.L. Hirshfield and J.M . Wachtel, “Electron cyclotron maser” , Phys.
Rev. Letters , vol. 12 , pp 533—536 , May 11 , 19614 .
[9] See the following review papers: V .A. Flyagin , A.V. Gapanov , M.I.
d Petelin and V.K. Yulpatov , “The Gyrotron ” , IEEE Trans.; M icrowave
Theory and Technique , vol. M’rT—25, pp 514—521 , June 1977; and, J.L.
Hirshfield and V.L. Granatstein, “The Electron Cyclotron Maser —
• [10] N.I. Zaystev , T.B. Pankratova, M.I. Petelin and V.A. Flyagin,
“Millimeter and sub—millimeter gyrot rons ” , Radio Engineering and
Electronic Physics, vol. 19 , no. 4, pp 95—99 , April 1974.
[11] A.A. Andronov , V.A. Flyagin, A.V. Gapanov , A.L. Gol’denberg, M.I.
Petelin , V.G. Usov and V.K. Yulpatov , “The gyrotrori; high power
source of millimetre and submillimetre waves”, Infrared Physics,
vol. 18, pp 385—394, December 1978.
18
• —• -•
—~~~~~~- • ~~~~~~~ ~~ ~~ -•
•• •
~~~~~~~~~~~~~
• ~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~ -
-
~~~~~~~~~~~~~~~~~~~~
‘-•
-- _______ - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
13] H. Jory , S. Hegji, J. Shively , R. Symon s , ‘t Gyrot ron Deve lopments ” , Micro-
[
wave Journal , vol. 21, pp . 30-3)4 , August 1978.
117] P. Spran gle and A . Drobot, “The linear and self-consistent nonlinear theory
of the electron cyclotron maser instability” , Trans. IEEE Microwave Theory
and Technique , vol . MTT-2 5, pp 528-5~4i- , June 1977; also, P. Sprarigle and
W. M. Manheimer, “Coherent Nonlinear Theory of a Cyclotron Instability” ,
Phys. Fluids, vol. 18, pp 22~4-23O , February 1975.
f2l] P. Sprangle and B. Smith , “Forward and Backward Wave Operation of the
Electron Cyclotron Maser” , to be submitted to Physics of Fluids .
[
22~ M . Friedman , D. A . Hammer, W. M. Manheimer and P. Sprangle , “Enhanced
Microwave Emission Due to Transverse Energy of a Relativistic Electron
Beam” , Phys . Rev. L e t t ., vol . 31, pp. 752-755, September 1973 .
23] K. R. Chu and A. T. Drobot, “Theory arid single wave simulation of the
(
gyrotron travelling wave amplifier operating at cyclotron harmonics” ,
Naval Research Laboratory Memorandum Report 3788, August 1978 (unpublished).
19
-
Voltage , V 71 kV
C urrent , ‘ 9.5A
b
Freq uency , w/2 w 35 GHz
Wa ll rad ius , r 0.54 cm
~
Guiding center radius, r 0.25 cm
0
Larmor radius, rL 0.061 cm
v 0/c 0.40
20
_ _
___
E
9
(b) W vE
~~~~~~~~~~~~~~~~~~ ~
w £ W ce
E3~ ~~~~~~
‘ ELECTRON
~~
2 ,
WAV E
E9
W>W
~~ ~~~ ~~~~ ~~
AMPUFICAT}ON
PHASE OF E v
-
E9
Fig. 1 — A Classical Representat ion of the Operation of the High Power
Eiectron Cyclotron Maser. Cross—sect ion of annular electron
beam is shown . Steady state magnetic field points outward from
page .
21
I. ~~~~~~~
-- -‘
~~ ~
—— ~~
-—
~~
— --—
~~~~~ ~~
- - - -
- - -
~~~~~~~~ ~~~~
O0
~
c~ ~~
—
S O
U) ~~~~ lLJ
Lu
0
r u 1
C U I—
u_ I 0 ~~~~~~~~ — 0
l ~ n~ ~
0 ~~~~
0 ~~~~~~~~~ ~~~~~ ‘— 1
~~~
~
~~~~~~~~~~~~~~~~~ ~~
L? Jj
~ I
1~~~ E~~~ ®
LL
~~~
~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
hi
1 i’
( /
/
~~
/
~~ , ~~ ~ \~~\\ ,z~ / ‘1 ~
- / - ‘~~~ // S
~~~
Q
\~\ C-,
/ ,//
/
-,
I ‘ ~~
~ ~~~~
/~~~~‘~ Z
~~~~~~ ~~~~~~~~ ~~
~~~~~~~~
in
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~
1o
~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ r
~~~~~~~
L~~~~~~~ ~~~
22
- -
- — ~ _ - —- — -.
-
-
~~ - ---a-—-
~~~~~~~~~~~
-- —
—
~~~ ~~~~~~~~~~~
.— —
~~~~~~~~~~~~~~~
-
-
\~~~~~~~ \o
\\ \\ \\LJ (fl cI) c/)
Q Q-Q
° ~~ ~ -
\ I \ \ .
\ \ o
~~~
~
‘
\ ~ ‘
S
w
(9 LC) Q L() -
~
\ S \
,
\ \~ ii II II I
L
~~
0 0 .cJ IL.
\\ c~~~
\~
\ ‘s 0 ~~~~~~~~
‘
‘V
‘
‘s.
V
V
V -
Z 0
> 3
V V V ¼J
. \~~ \\ D\ -
~~~~\ L m
\
W
-
z ‘ ‘ ‘
~~~~~~~~~~~~
_J
-
\ \ R
Z 0c
~ \ -
O ..j z \V \) .
a. 4
C)
S U
V 4 V
4
~~~~~~~~
S LL W r,. ‘~.1 ‘V ‘~~LQ
\ \
‘
(1) < — \ \o UI
‘
-b
LC) X \ ~~~
—
\
-
Z
\ \ \
_ I1!I I ! i I t i i i i i i i \i
’
I1 F—’p5.
N)
~~~ i ~i
CsJ
~~
2 2 2
(NO V JV 9 NO LL3J1 3)Sfl VM
~~ ~
23
0
c\J
n—
S_ 0
cx:J
-
0
a-
-c
-1
0 — ~
U
~~~~~~~~~~~~~~
(0 0 ,~
r~..1i O ~~~~~~
~~ ‘
— —
-H3
Li.1
-
Z~~~ir
4 ~~~
lI
o 0 ) -
rO
-
a-
4
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
0.11
-4-,
0 ’—
o
Li ~~i_ . i f i l l f I I I _Q
r~) cJ c ’ J
~
x x
c’J c’J
(SIIVM ) ~~~
‘ èflMOd ±fld±flO
24
- -~~~~~~~
r --
~~~~~~~~~~~~~~~~~~~~~~~~
-
~~~~
0.45 —
I’ = .40 cm ’ ___
= ~~~~~~~~ - =
~~~~~~~~~~~~~~~~~
// ~—~r--N\\\ ~
-
~~~ 0.2 —
/ \
0~~:-
0 ’ I , I i I i I
~~~~ I 1 I
~~~~
13.0 13.2 13.4 13.6 13.8 14.0 14.2 14.4
B0(kG)
Fig . 5 — Theoretical Calculations of Amplitude Growth Rate , r , as a
Funct ion of B (V 70 kV , I — 3.5A , w/2~ 35 GHz).
0
25