Trevor Paul R Linnecar

The High Frequency Longitudinal and Transverse Pick-Ups in the CERN SPS Accelerator

The different applications and consequent specifications for longitudinal and transverse high frequency pick-ups in the CERN SPS have led to the use of three separate types of pick-up; they are the electrostatic, the wall current and the directional coupler pick-up. The high frequency longitudinal pick-ups are protected from propagating waveguide modes by traps. Certain points in the design are described and a brief summary of their performance and use is given.

Non integer harmonic number acceleration of lead ions in the CERN SPS

The project to accelerate lead ions in the CERN complex has been successfully completed and physics has begun. In the SPS, the final machine in the chain, the ions are accelerated from an energy of 5.1 GeV/nucleon to 160 GeV/nucleon using the existing 200 MHz travelling-wave cavities. The change in revolution frequency during acceleration is much larger than can be accepted by the untuned cavities when operated at constant harmonic number. A technique has been developed to overcome this limitation which takes advantage of the filling time of this type of cavity which is shorter than one turn. Fast amplitude and frequency modulation of the RF waveform allows the cavities to operate at a constant, optimum frequency during the passage of a batch of particles in the structure. This frequency is not a multiple of the revolution frequency and therefore during the gaps between batches the phase of the composite RF waveform is changed to maintain synchronism from turn to turn as the beam accelerates. The technique and hardware are described in detail together with the first operational experience.

Acceleration and Storage of a Dense Single Bunch in the CERN SPS

First tests of the lifetime of a normal SPS beam stored for several hours at 200 and 270 GeV were encouraging. The natural logarithmic decay time is in excess of 24 hours. However, in the proton-antiproton scheme, 200 MHz bunches containing fifty times the normal design population of particles are to be injected into the SPS above transition at 26 GeV, accelerated and stored. Lacking the hardware to inject at so high an energy, we first injected bunches of 1011 protons at 10 GeV accelerating them through transition but found it difficult to pag transitionwith more than 40% of this design population . Nevertheless we report some interesting observations on head-tail and negative-mass effects which limited intensity during these tests.

Beam Transfer Functions and Beam Stabilisation in a Double RF System

The high intensity proton beam for LHC accelerated in the CERN SPS is stabilised against coupled-bunch instabilities by a 4th harmonic RF system in bunch-shortening mode. Bunch-lengthening mode, which could also be useful to reduce peak line density and alleviate problems from e-cloud and kicker heating, does not give desirable results for beam stability. In this paper an analysis of the limitations of these two different modes of operation is presented together with measurements of the Beam Transfer Function for the double RF system. As predicted by theory, for sufficiently long bunches with the same noise excitation, the measured amplitude of the beam response in bunch-lengthening mode is an order of magnitude higher than that for bunch-shortening mode or for a single RF system.

A Transverse Schottky Noise Detector for Bunched Proton Beams

Transverse Schottky noise analysis is a powerful standard technique for continuous proton beam diagnosis. The extension to bunched proton beams of very low intensity, as will be used in pp¿ colliders, requires new technical features to enhance the detector sensitivity and to reduce the coherent signal of the parasitic common mode. During SPS pp¿ collider experiments, we met this requirement by using a resonating transverse pick-up three meters long, the two plates being movable to match the electrode gap to the beam width at high energy. The parasitic sum signal was drastically reduced by centering the plates around the beam and by filtering the output signal. Clean betatron and longitudinal Schottky bands have been observed for a 0.3 mA continuous beam and for a 0.01 mA single bunch beam. Betatron satellites have also been detected for a 6 mA multi-bunched beam. As a future development the coil inductance in parallel to the electrode plates will be varied by a movable ferrite core, to allow continuous modification of the pick-up transverse acceptance through a change of the gap distance without changing the resonant frequency, the band-width or the filtering device.

Longitudinal Phenomena in the CERN SPS

After almost one year of operation two kinds of problems dominate the longitudinal behaviour of the SPS beam. At high energy longitudinal coupled bunch instabilities occur. They are driven by the RF cavity impedance both on its fundamental and high-order passbands. Present cures include damping the high-order modes of the cavities, Landau damping techniques and feedback systems. At the injection energy a debunching-rebunching procedure is performed in order to change from the 9.5 MHz RF frequency of the injector to the 200 MHz of the SPS. Debunched beam instabilities driven by the cavity and vacuum chamber impedances (up to the GHz region) at present limit the RF capture efficiency.

Continuous Tune Measurements Using the Schottky Detector

A continuous measurement of the betatron tune has been implemented in the SPS pp/sup -/ collider with hardware normally used for the detection of Schottky noise. Suitable values for the tunes and chromaticities are thus more easily achieved in a machine where the strong space charge effects of the two counterrotating beams severely reduce the available space in the resonance diagram. The basic idea is to monitor the frequency of one of the betatron lines together with the revolution frequency and to deduce the tune according to a formula provided in this paper. The betatron lines which naturally appear as noise signals adding incoherently are strongly enhanced by exciting the beam at the required frequency with a kicker. The whole system then forms a closed loop in which an oscillator excites the beam producing a coherent signal at the Schottky detector, this signal providing the reference to lock the oscillator frequency. The excitation, 50 W at 10.7 MHz causes particle loss (about 3%/cycle) and emittance blow-up (about 10%/cycle) restricting the use of the apparatus to setting-up periods.

Tune Measurement and Control at the CERN-SPS

Two complementary techniques have been developed which allow the betatron tunes to be controlled during energy ramping of the SPS to a precision of better than ¿0.002. By the use of these techniques the setting-up of the SPS as a collider and a fixed target machine is substantially simplified and better physics understanding has been gained. The first technique uses an electrostatic beam deflection every 60 ms. A Fast-Fourier Transformation of the beam response in a dedicated computer yields the betatron tunes with a precision better than 0.01. In the second technique the beam is continuously excited. The frequency of the excitation is measured and fed back to the beam by a Phase-Locked-Loop. This measurement is accurate to better than 0.001 but requires a reasonably well tuned machine. Tune deviations are automatically compensated by acting on the main quadrupoles through a software loop.

Acceleration in the CERN SPS Present Status and Future Developments

The hardware status of the 200 MHz accelerating system is described together with changes that are foreseen in the near future. The parameters for an 800 MHz Landau damping cavity at present under construction are given. Recent studies on the machine have been in the following areas: beam loading effects in the cavities, coupled bunch instabilities, and acceleration of intense single bunches. The results of these studies are presented.

Nominal longitudinal parameters for the LHC beam in the CERN SPS

A proton beam with the basic structure defined by the LHC requirements was first available for injection into the SPS in 1998. At the end of 2002, following a significant beam-studies and RF hardware upgrade programme, a beam having both the nominal LHC intensity and the correct longitudinal parameters was obtained at top energy for the first time. This beam, characterized by high local density, must satisfy strict requirements on bunch length, longitudinal emittance and bunch to bunch phase modulation for extraction to the LHC, where only very limited particle losses are acceptable. The problems to be solved came mainly from the high beam loading and microwave and coupled bunch instabilities which led both to beam losses and to unacceptably large longitudinal emittance on the flat top. In this paper the steps taken to arrive at these nominal beam parameters are presented.