 |
Several significant improvements to SPEAR were completed in 1997, reflecting marked progress in the development of accelerator hardware and software. Improvements to the orbit feedback system largely concentrated on the Beam Position Monitor (BPM) processor, multiplexor switch equipment, and interface to the digital control hardware. Extensive hardware development was carried out to complete the installation of wide-bandwidth, low-loss cables to all BPMs in SPEAR. Four remote multiplexor stations (MUXs) were installed around the storage ring tunnel, complete with sophisticated low loss, high isolation RF circuits and shielded boxes. A new rack was installed in the SPEAR control room to house the signal processor (Harris chip) and MUX circuits that access the four remote MUX stations in the tunnel.
To enhance the data transfer protocol, the signal processor (producing quadrature components for the individual BPM button signals) was interfaced with the DSP chip (signal processor and control algorithm) via a CPLD (Programmable Logic) which implements a communication port compatible with the one on the DSP board. The CPLD accepts commands, an initialization sequence, MUX settings, and attenuator settings from the DSP, and sends the button data from the Harris chip to the DSP. The CPLD software has not yet been fully implemented, but the DSP data acquisition and button averaging process has been tested by emulating the CPLD/Harris chip on the DSP board. The emulation test included DMA data transfer through the communication port to memory on the DSP board. Simulated turn-by-turn BPM button data were processed on the DSP retrieved by the VME crate controller (MicroVAX) and placed into the SPEAR database where the data can be accessed for machine operations. We also carried out a successful demonstration of the complete data transfer sequence from BPM buttons through the MUX circuitry to the processor. Improvements to the software are now focusing on the final implementation of the global and local (beamline) feedback components.
A new method to power the linear accelerator was put into place. For the last 4 years the SSRL injector linac system has been powered by one SLAC 5045 klystron and one XK-5 klystron. After the current upgrade, the system will be powered by one 5045 klystron at an increased RF power output. The two major changes are: 1) reconfiguration of waveguide systems for even RF power distribution to three linac sections, and 2) modification of the modulator for improved stability and reliability. This change was motivated by problems with the continued use of the old XK-5; there are no resources available to manufacture or refurbish this aging klystron. The performance of the tube has been deteriorating over time; some of the spare XK-5 klystrons are not even operational. On the other hand, the SLAC 5045 klystron is in production on site. Spare tubes are readily available, and its reliability has been continuously improved.
A number of improvements to the modulator are also being carried out. These include upgrading the high voltage power supply regulation, improving the reproducibility of the thyratron performance, and increasing system protection through an interlock revision. A new LCW control panel was also installed to allow automatic monitoring and control of the status of the system.
The present thrust of the accelerator modeling studies is to calibrate a linear model of SPEAR using a response matrix measurement. Information contained in the corrector-Beam Position Monitor (BPM) response matrix is used to fit the parameters of the model.
In the case of SPEAR, the matrix contains 3,596 measured data points. These data points are compared to a calculated model response matrix. The singular value decomposition (SVD) method is used to analyze the difference between the model and the measurements; and calibration factors can be calculated for the various elements of the accelerator.
The modeling program has the following objectives:
Thus far our analysis has produced accurate models of the linear optics of SPEAR. The accuracy of the model is determined by measuring the rms difference between the measured and model response matrices. These differences have been below 20 micrometers. Another measure of the success of the model calibrations is to assess how well the machine functions of the model match the measurements. The matching is good, and the model has been used to calculate new magnet settings that reduce the beta function beating due to individual field errors within magnet families. The modeling has also been useful as a diagnostic tool for identifying faulty magnets and position monitors. At present, measurements made during the last run are being analyzed to calculate the best configuration for the coming run.
A study of the parasitic resonances driving longitudinal instabilities has promoted interesting insights into the frequency and strengths of the modes in the accelerating cavities. With these findings it was possible to optimize the position of the cavity tuners and to suppress the longitudinal coupled bunch oscillations, thus delivering a more stable beam to the users.
A phase space monitor system for studying the non-linear dynamics in SPEAR has been operational on SPEAR since 1996. The system consists of a pair of fast kickers that excite the bunch oscillations in the transverse plane and the turn-by-turn beam position monitors that allow recording of the beam motion following the kick. The data can then be transformed to visualize the dynamics in phase and frequency space. The system has been used to analyze the time variation of the tunes caused by ground motion, power supply ripples and other effects.
A study of the electron bunch characteristics following the launching of an oscillation in SPEAR revealed that the oscillation is not damped by the tune spread (Landau damping) when the chromaticity is set at some negative values. Under these conditions filamentation does not play a role in the turn-by-turn mapping of the accelerator and the beam acts as a fine probe of phase space.
All in all, the improvements made to SPEAR and to the accelerator physics program in 1997 have increase stability and reliability, raised the level of our confidence in our programs, and enhanced our ability to implement new technologies and ideas.
|