J.W. Wang, J.P. Eichner, K.H. Fant, H.A. Hoag,
R.F. Koontz,
T. Lavine, G.A. Loew, R.J. Loewen,
A. Menegat, R.H. Miller,
C.D. Nantista, C. Pearson,
R.D. Ruth, S.G. Tantawi, A.E. Vlieks,
P.B. Wilson, and C. Yoneda
SLAC, Stanford University, USA
This paper presents new results of high-gradient studies performed on a 1.8 m traveling-wave accelerator section with detuned high-order deflecting modes. This structure was designed initially for studies of detuned structures and will be installed in the Next Linear Collider Test Accelerator (NLCTA). The paper describes the test set-up in the Accelerator Structure Test Area (ASTA) including electron gun, pre-buncher, pre-accelerator, spectrometer, Faraday cups, 200 MW SLED-II power compression system, Magic-T type phase shifters and attenuators. Rf processing, detailed dark current analysis, radiation problems, and beam acceleration measurements are discussed.
The Accelerator Structure Test Area (ASTA), where all tests were performed, has been described in a previous paper [1]. Figure 1 shows the layout of the setup for the high power test of the 1.8m traveling-wave section. The accelerating mode of this detuned structure has a uniform phase velocity equal to the speed of light [2]. The electrons from an 80kV thermionic gun are bunched and pre-accelerated in order to be captured in the accelerator section. Both prebuncher and pre-accelerator use single cavities with nose-cones, in order to reduce the transient beam loading effect. The cavities are made of stainless steel to lower their Q value. The rf power for the prebuncher and pre-accelerator is obtained via directional couplers from the feed waveguide for the accelerator section. Each feed includes a Magic-T type attenuator/phase shifter to adjust rf amplitude and phase. Two main arms are assigned to be input and output ports, and two side-arms are shorted with non-contacting plungers driven by stepping motors. Moving the shorts respectively toward the T-junction and away from it changes the phase at a constant amplitude, whereas moving the shorts synchronously in the same direction changes the amplitude at a constant phase.
Several scintillator detectors are installed alongside the accelerator
structure for X-ray measurements. The detector heads are made from NaI
crystals (0.38 in. in diameter 
1.0 in. in length) with lead
collimation. The photons created by the X-rays are transmitted through
a 3-5 ft. optical fiber to a Hamamatsu HC125 PMT based-detector, which
includes a divider, high voltage power supply and signal processing
circuitry. The whole system is calibrated by using a standard
radioactive source.
The 1.8m detuned structure is the first model for the NLC Test Accelerator at SLAC. It has 204 cavities plus input and output couplers. Its characteristics and main rf parameters are shown in the table.
Figure 1: Layout of Accelerator Structure Test Area.
RF processing was performed in two stages. The first stage took place
in March of 1994, over a period of 30 hours in a two-day period, to
reach an average accelerating gradient of 50 MV/m. The second stage
took place in June of 1995 after the installation of a 50 MW peak power
klystron. The improved SLED-II system was able to produce 200 MW,
150 ns rf pulses. The maximum power delivered to the accelerator
structure was 150 MW, and the maximum accelerating gradient was limited
by the klystron output power. Figure 2 shows the
straight-ahead dark current as a function of accelerating gradient.
Significantly higher power is needed to further reduce this dark
current [3]. The dark current spectrum obtained so far was
measured using the 45 
spectrometer. In order to obtain more
information on the origin of the captured dark current, two dipole
magnets were installed alongside the accelerator section at a given
z-position. They can create about 600 Gauss of transverse magnetic
field in ``x" and ``y" to prevent any dark current originating upstream from
reaching the spectrometer. Hence, each curve in
Fig. 3 is a spectrum of the dark current transmitted and
collected from that part of the accelerator section downstream of the
magnets. Figure 4 shows the total dark current collected at
the straight-ahead Faraday cup as a function of the magnets position. From
these plots we can conclude that most of the dark current is composed
of low energy electrons. The total number of electrons captured at the
end of the accelerator section no longer increases when the magnets are
located upstream of cavity No. 90. This is probably because when the
field emitted electrons are accelerated or decelerated by the rf
fields, a majority of them strike the cavity walls. Secondary or
back-scattered electrons are then created, which are also accompanied
by X-ray radiation. During rf processing, the outputs of the
scintillator-PMT system give strong bursts if there is sparking at
nearby cells. Figure 5 gives the waveforms of four X-ray
detectors at two different rf power levels. In the test, scintillators
1, 2, 3, and 4 are located respectively at the input region, one third,
two third, and the output region of the accelerator section.
Figure 6 shows the radiation dosage as a
function of the position along the accelerator section at
two different gradients.
Figure 3: Energy spectrum of dark current at average accelerating
gradient of 58 MV/m for different positions of the dipole magnets.
Figure 4: Total dark current collected at the straight-ahead Faraday cup
as a function of the position of the dipole magnets.
A beam test was performed at an average accelerating gradient of 50 MV/m. The rf pulse length was 150 ns for the SLED-II system and the filling time of the structure was 100 ns. Because of radiation safety limitations, a pulsed beam of only 40 mA, 30 nsec was used. The maximum rf power available to the pre-accelerator was lower than expected because of the rf loss in the waveguide system. The operating frequency was adjusted slightly higher to allow the captured electrons to drift in phase toward the rf peak in order to increase the net acceleration. The electron energy gain showed that for the particles on the rf peak, the average accelerating gradient for an input power of 90 MW reached 50 MV/m.
Figure 6: Radiation dosage along the accelerator section at
two different average gradients.
Work supported by the Department of Energy, contract DE-AC3-76SF00515.
