EV Batteries Thermal Management

Home Search Collections Journals About Contact us My IOPscience
A Travelling Wave Linear Accelerator with R.F. Power Feedback, and an Observation of R.F.
Absorption by Gas in Presence of a Magnetic Field
This content has been downloaded from IOPscience. Please scroll down to see the full text.
1949 Proc. Phys. Soc. B 62 270
(http://iopscience.iop.org/0370-1301/62/4/107)
View the table of contents for this issue, or go to the journal homepage for more
Download details:
IP Address: 130.237.165.40
This content was downloaded on 20/08/2015 at 01:58
Please note that terms and conditions apply.

LETTERS TO THE EDITOR
A Travelling Wave Linear Accelerator with R.F. Power
Feedback, and an Observation of R.F. Absorption by Gas
in Presence of a Magnetic Field
A travelling wave linear accelerator usually consists of a high ,frequency power source
feeding power into a waveguide acceleratmg tube which is termmated by a dummy load’
to absorb without reflection the power reaching the output end. It has been shown
(Harvie 1948) that there is an optimum length of accelerating tube, w h c h corresponds,
to an attenuation of approximately 1.25 nepers, but unfortunately there are sometimes
practical difficulties in making accelerators as long as the optimum length. This note
describes means for overcoming some of these difficulties, in which an accelerator of less
than optimum length retains its efficiency by utilizing the power which is usually wasted
in the dummy load.
T h e system is shown diagrammatically in Figure 1, in which arrows indicate the
direction of power flow. Power is fed back from the output end of the accelerator and
combined with the power from the source in a suitable bridge circuit. It is sometimes
Thermocouple Power Monitor

?l dariable Probe Ma@tron
Feedback Line
Bridge \
$Absorbing Load
. Figure 1. Diagrammatic arrangement of Figure 2. Feedback system applied t o 4 MeV

h e a r accelerator with feedback. linear accelerator.
desirable that arms (1) and (2) of the bridge should be mutually conjugate, in order to make
the mpedance presented to the source constant during a transient building-up period and
independent of the accelerator characteristics. I t can then be shown that the bridge
circuit must include a dssipative element which is shown connected to ann (3) externally;
under steady state conditions no power is dissipated in this element. Steady state power
fluxes are shown on the diagram; the power flux entering the accelerator (l+n)W,,
is always greater than the power supplied by the source and n is determined by the attenua-
tion of the accelerator. Calculations show that the ‘I efficiency ” of the accelerator,
V*/WL,increases as the length is diminished and is slightly greater than that of an optimum
length accelerator with a dummy load at the end.
Bridges of the type required have been studied previously (Tyre11 1947, Smullin and
Montgomery 1948) for the special case where n=l, and examples will be familiar undei
the names of “ magic T ” and “ Rat-race ”. There seems to be no great difficulty in
designing bridges along similar lines for other values of n.
In a pulse operated system it is necessary for waves to make several transits of the
accelerator and feedback loop, after application of the pulse, before steady state conditions
are approached. Calculations show that the build-up time is independent of the length
and is approximately the same as in a conventional accelerator of optimum length. If
the frequency of the source is varied the power fed back from the end does not return In
the correct phase and the circulating power is reduced. It can be shown that the permissible
frequency bandwidth is again approximately the same as in an ordinary accelerator of
Letters to the Editor 271
optimum length; however, by adding an adjustable phase changer i n the feedback pathA
it is possible to maintam the correct circulating power at any frequency and the bandwidth
is then only limited by the wave-to-electron phasing in the usual way.
An advantage of the system is that the accelerator can be made appreciably shorter
than the optimum length without losing efficiency, which leads to increased dimensional
tolerances on the waveguide. A shorter waveguide also has a wider frequency bandwidth’,
as far as phasing of the wave to electrons is concerned, and it may also be an advantage
in practice to allow the frequency of the source to vary as it will within this band while
the operating conditions are kept right by adjustmg the phase changer in the feedback
loop.
For the case of a bridge circuit designed for n = l , theoretical curves can be plotted
for the power in the various arms of the bridge when the phase and attenuation in the
accelerator loop are varied. Experimental confirmation has been obtained at low C.W.
power with a mock-up system on a wavelength of 3 . 2 cm., using a normal n= 1 ‘‘ Rat-race ’’
and about 30 wavelengths of rectangular waveguide to represent in some measure the
corrugated wavegbide. A method has also been worked out theoretically and confirmedl
experimentally for obtaining a suitable signal which will reset the phase shifter i n , the
feedback arm when the magnetron frequency changes.
It has also been possible to try out the n=l system on the 4 M e V . linear accelerator
(Fry et al. 1948) since the R.F. loss in the accelerator is about 2 . 5 db. and can be made u p
to 3 db. with beam loading. T h e experiment had unfortunately to be carned out in a very-
short time and the external waveguide components were not matched as well as they should!
have been; consequently the thermocouple power monitors could not be relied upon to
within 20%, although they were partly compensated for by the presence of standing waves..
It was not therefore possible to verify the power fluxes in all parts of the system. However,
independent measurements showed that the accelerator operated as if it had an input
R.F. power flux of 2.1-2.2 MW. when the magnetron power was only about 1 . 4 MW. T h e
energy spectrum was identical to that previously obtained with the same input power
but without feedback. T h e maximum power which could be obtained in the accelerator
was 2 . 4 ~ and, ~ . with a somewhat lower frequency than normal, the mean energy was-
raised to 4.5 MeV. but with more than twice the normal spectrum width. These experiments
indicated that a well matched system would work as predicted, but with one reservation-
that a somewhat better vacuum is required than for an equivalent accelerator without
feedback.
It was noticed that the loss in the corrugated waveguide soon after applymg the R.F.
power but without the gun in operation was very greatly increased when the focusing field
was switched on, and moreover the effect happened quite critically a t about one-third of
the normal focusing field and was not confined to any particular spot along the length of
the corrugated waveguide. I t has been checked that the power leaving the accelerator
falls in the same manner for the system without feedback, hut since there is no reaction on
the input power its effect on this particular accelerator IS not serious and had therefore
not been observed before. T h e effect may be due to the presence of positive ions or electrons
which spiral under the influence of the focusing field and consequently abstract very
much greater quantities of energy from the R.F. wave than if they drifted straight to the *
walls of the tube. As the vacuum cleans u p the effect entirely disappears and the operation
is as predicted.
This work was carried out as part of the programme of the Accelerator Group of the
Atomic Energy Research Establishment at the Telecommunications Research Establishment,
Malvern, and acknowledgment is made to the Director, A.E.R.E., f o r permission to.
publish the results.
Atomic Energy Research Establishment R. B. R.-SUERSBY;-~~ARVIE.
at Telecommunications Research Establishment, L.B. MULEETT..
Malvern, Worcs.
3rd February 1 9 4 9 .
FRY, D.W., et al , 1948,Naturr, Lond., 162, 8j9.
H ~ V I ER., B. R -S.-,1948,Proc. Php.SOC., 61,255
SMULLIN, L. D., and MONTGOMERY, C . G , 1948,Microwave Duplexers (Near Yorkand London.:
McGraw Hill).
TYRELL,W. A.,1-94<-Proc.Inst. Radio Engrs., 35, 1294.

EV Batteries Thermal Management

Uploaded by

Copyright:

Available Formats

EV Batteries Thermal Management

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EV Batteries Thermal Management

Uploaded by

Copyright:

Available Formats

Home Search Collections Journals About Contact us My IOPscience

Absorption by Gas in Presence of a Magnetic Field

1949 Proc. Phys. Soc. B 62 270

Please note that terms and conditions apply.

Thermocouple Power Monitor

. Figure 1. Diagrammatic arrangement of Figure 2. Feedback system applied t o 4 MeV

You might also like