L. Palumbo Lpalumbo@lnf.infn.it
LNF, Frascait, Italy3mm
The e
e
-factory DAFNE is presently under
construction in Frascati (Italy) [1].
It is designed as a double ring system with a maximum number of
120 bunches/beam. The short term luminosity goal is
with 30 bunches while the designed luminosity
should be achieved in a period of at least 2 years of continuous
operation.
DAFNE is provided with a full energy injector composed by a
Linac, an Accumulator ring and 
m of transfer lines
[2]. Both electron and positron beams are produced and
accelerated up to the nominal energy of 510 MeV by the Linac
and then stored into the Accumulator Ring for phase space
damping before being injected into the Main Rings. The
injection in the Accumulator is performed with a repetition
rate of 50 Hz, while the extraction is operated at 1 Hz. The
electron and positron modes are non simultaneous and the
estimated time necessary to completely fill all of the 120
buckets of the Main Rings is 
min for the positrons and
min for the electrons. Although the expected beam
life time
in the Main Rings is 
hours, a ``topping-up'' scheme is
foreseen: the integrated luminosity is optimized by injecting
approximately every hour.
The injector is completely installed and under commissioning, the main rings installation is in progress and it is expected to be complete by Spring 1997.
F. Sannibale Sannibale@lnf.infn.it
LNF3mm
The DAFNE LINAC, built by the American company TITAN BETA, is
a 
long Linac designed to produce and accelerate, with
50 pps, electron and positron beams up to 800 and 550 MeV
respectively (max.). The macro-bunch length is 10 ns FWHM and
the nominal macro bunch peak current is 150 mA for electrons
and 36 mA for positrons. The positron conversion configuration
includes a metallic target, a pulsed 5 T solenoidal lens (SLAC
Flux Concentrator) and
m of 0.5 T uniform solenoidal field. The
electron beam nominal parameters at the positron
converter, 250 MeV energy, 4 A current and 1 mm (
)
spot size, should ensure a positron yield of
0.9%. The RF network
includes 4 S-band (2856 MHz) 45 MW klystrons followed by
energy compression devices (SLED) while the accelerating
scheme is composed by an harmonic prebuncher, a five cells
buncher and 15 SLAC type 3 m long accelerating sections. A
complete set of diagnostics, including beam position, profile,
current monitors and a positron separator, permits the
transport of the beam along the Linac.
A preliminary set of electron beam tests were first performed at TITAN BETA on May 1994. The Linac was assembled at factory up to the positron converter and a 4 A , 1 mm spot size, 240 MeV electron beam was accelerated to that point.
The Linac electron tests at Frascati started in
January 1996. The beam nominal values for the Accumulator
electron injection, 510 MeV energy and current
mA (500 mA), were rapidly achieved and in
June 96 about four weeks of
shifts were dedicated to the Accumulator electron
commissioning. During this period the Linac operated in a
reliable way without presenting relevant problems.
The first positron tests were done on July 96 when in about three shifts a positron beam was accelerated at the end of the Linac with 4.3 mA of current and 435 MeV of energy. The tests were performed with one of the four klystrons seriously damaged: because of arcing phenomena it delivered just 5 MW instead of 45 MW. Moreover a failure in the electronic control, limited the electron gun output current to 4.5 A instead of the nominal 8 A. With this configuration the conversion energy and current were reduced to 80 MeV and 1.5 A respectively. By scaling the achieved results with the nominal operation sets, the obtained values fairly match with the design ones.
Future plans include continuation of the Accumulator electron commissioning and completion of the Linac commissioning.
A. Ghigo Ghigo@lnf.infn.it
LNF3mm
The DAFNE Accumulator is a 32.5 m long ring with a magnetic structure which can be easily optimized for injection: a small value of the dispersion function all over the machine enlarges the energy acceptance, and the RF frequency smaller that in the main rings widens the longitudinal acceptance. Two of the four straight sections are dedicated to the injection and extraction septa while the other two to the RF cavity and four kickers. The single bunch injection and the solution of extracting positrons from the electron injection channel (and vice-versa) have been adopted.
The energy and emittance acceptance, 11.5% and 10 mm.mrad at 510 MeV respectively, permit complete capture of the LINAC beams; moreover the low value of the energy spread and emittance of the extracted beams, 10.1% and 0.25 mm.mrad at 510 MeV respectively, make the injection into the Main Rings buckets easier.
The Accumulator commissioning started in June 1996, with the
electron beam from the LINAC at 510 MeV energy, after the
transfer lines completion. In four weeks the achieved stored
current was 75 mA limited by the vacuum (no bake-out made
after installation). The measurements of the horizontal and
vertical betatron wave number were performed and the optical
functions has been set to the design ones; the measured
chromaticity values, without and with the sextupoles
on, were in
good agreement with the expected ones. The horizontal and
vertical orbit measurements have shown that the mechanical
alignment of all the magnetic elements is very good, with a
maximum displacement of the beam from the axis, without
correctors, of 7 mm. A wide radiofrequency stability range was
measured (
) at central frequency
. The control system and
all the main and
ancillary systems worked quite satisfactorily. In the next
future the injection efficiency in the Accumulator for
electrons and positrons beams, the beam emittance and pulse
lenght will be measured.
C. Biscari Biscari@lnf.infn.it
LNF3mm
F. Sgamma Sgamma@lnf.infn.it
A3mmDONE was run for the last time in Spring 1993 and the
preparation of the hall for the installation of the DAFNE collider
rings has lasted two years. The DAFNE group, together with the
KLOE an FI.NU.DA. ones, were given the beneficial occupancy of
the DAFNE hall in January 1996.
Installation of the collider rings and of the last part of the transfer lines is being performed in parallel with the injector commissioning. Optimization of the different actions according to the availability of elements and men-power is constantly updated.
After the installation of the alignment pillars, the machine supports have been placed and aligned. Installation of the hydraulic plant, air conditioning system and general electric plant is almost complete. About 50% of the magnetic elements are already placed, and the first alignment has been performed. Cabling (about 70 km) between power supplies and magnets is in progress. Rf system is ready to be installed, apart from one of the two cavities which will be available in two months. A great part of the vacuum chambers has been already delireved, vacuum tested and is being dressed with the diagnostic elements. The chamber dressings with ion pumps, Ti sublimators, photon absorbers is a delicate task to be done in a clean area, and for this purpose one of the experimental pit will be used. The beginning of this operation is foreseen in december 1996. The diagnostic cabling layout is being positioned in the hall. The cryogenic system which will used by the two larger experiments and by the four superconducting solenoids will be installed next summer; in fact it is not necessary for the machine commissioning which will be performed without any superconducting element.
The completion of the whole installation is foreseen for spring 1997.
M. Zobov Mikhail@lnf.infn.it
LNF3mm
A preliminary study of beam-beam interaction for DAFNE, including a scan of working points in the tune areas with respect to the luminosity, has already been performed and suitable working points have been found [3]. However, as far as the DAFNE damping coefficients are very small, long beam tails can be induced that cause lifetime and background problems. One of the reasons of the tail growth is the high order resonances which can manifest only in case of weak damping.
A "brute force" beam-beam simulation would take enormous CPU time to collect a satisfactory statistics at large betatron amplitudes. In order to save the CPU time (up to several orders of magnitude) we use the special dedicated tracking technique by Shatilov [4] to define the equilibrium distribution in the beam tails. In our simulations we also took into account cubic machine nonlinearities and the parasitic crossings (PC). In this study we concentrated only on good working points which were found [3]: small areas around tunes (0.09;0.07) and (0.53;0.06).
Simulation show that DAFNE can operate safely, without substantial bunch core blow up and with a reasonable lifetime, with up to 60 bunches per each beam. However, for the maximum possible number of bunches (120) some blowup and reduction in the lifetime are observed due to the smaller separation at the PCs.
Because of that we have undertaken a study to optimize the
machine performance with 120 bunches. In particular, it has
been proposed to increase the separation (in terms of 
) by
a factor of =882 without any change in the luminosity, tune
shifts and the horizontal crossing angle itself. This can be
done by decreasing the horizontal beta-function at the
interaction point by a factor of 2 and increasing the
vertical emittance by the same factor simultaneously. A
slight change in the betatron tunes from (0.09;0.07) to
(0.08;0.06) is also necessary in order to further improve
the lifetime. This shift of the working point closer to the
integer tunes helps us to avoid resonances resulting in
particle's horizontal drift.
The performed simulation has confirmed that the beam-beam
lifetime in that case is about 10 hours without any
substantial reduction in the luminosity. Preliminary
calculations have shown that the DAFNE lattice can be adjusted
in accordance with the proposed changes. These results have
been worked out with the collaboration of Dr. D. Shatilov from
BINP-Novosibirsk
(Shatilov@vxinp.inp.nsk.su) who spent several
weeks at LNF.
L. Palumbo Lpalumbo@lnf.infn.it
LNF3mm
The coupling impedance of a small hole in a beam pipe coupled with coaxial stuctures has been studied. By applying a modified Bethe theory which takes into account also the scattered fields in the external structure and therefore fufilling the energy conservation law, the real part of the impedance corresponding to the radiated fields can be derived. Two basic structures have been considered: infinite coaxial beam pipe, and a coaxial resonator coupled to the beam pipe, for which simple expressions of the low frequency longitudinal and transverse impedances have been derived [5, 6]. Numerical calculations performed with MAFIA code are in good agreement with the analytical results.
The problem of coupling through large apertures is now being studied. The Bethe and the field matching methods are applied to long narrow slots in order to assess the reliability of the Bethe perturbation theory, and get a deeper understanding of polarizability coeffients for such apertures Measurements on a coaxial cavity prototype are planned for the near future.
M. Migliorati Migliorati@lnl.infn.it
LNF3mm
The single bunch behavior in the DAFNE accumulator machine has
been investigated using both a time domain tracking code and
an analytical study based on the Vlasov equation and mode
coupling theory [7]. The numerical simulations show that
the bunch is expected to be in the turbulent regime at the
nominal current 
. The microwave instability
threshold has been found to be at
. The bunch
lengthens from the natural value of 
cm to
cm
at the nominal current, and the relative energy spread widens
from 
to 
. Analysis of
turn-by-turn beam sizes
at the nominal current shows that the change in the bunch
length does not exceed 4%, i.e., not dangerous for the machine
operation.
Analytical study has been undertaken, and different models
have been used to predict the microwave threshold value and to
compare it with the numerical results. Application of Boussard
criterion gives for the instability threshold a factor 3 lower
than that of the simulations. The perturbation modal analysis
of Vlasov equation gives a closer result. In particular the
bunch shape can be approximated by a gaussian distribution
function, and the azimuthal mode coupling theory can be
applied. The corresponding strong microwave instability
threshold is found to be 1.5 times the one obtained with the
simulations. The double water-bag model allows the
radial mode coupling to be treatedanalytically. The threshold
of a weak
instability due to a coupling of radial modes with different
azimuthal numbers has been found at
,
coinciding with the value found by numerical simulations.
The planned bunch-lengthening measurements will allow us to check
the analytical predictions in the turbulent regime.