Impedance Study with Single Bunch Beam at Taiwan Photon Source

The impedance at Taiwan Photon Source was investigated. The effects of bunch current such as a tune change, a synchronous phase shift and a bunch lengthening under operation conditions at various stages were measured; the machine impedances were deduced. This report presents the results with insertion devices in various configurations. INTRODUCTION The Phase-I beam commissioning of Taiwan Photon Source (TPS) was achieved with a stored beam up to 100 mA (limited by two normal-conducting cavities) in 2015 March. The maximum single-bunch current attained up to 12 mA. The bunch length and tune shift were measured as functions of the bunch current also in 2015 March. A preliminary analysis of the ring impedance with dummy chambers of full gap 20 mm vertically in the straight sections for insertion devices (ID) has been reported [1]. Several months were required to install ten Phase-I IDs and to replace the RF systems with superconducting modules in mid 2015. The Phase-II beam commissioning of TPS began in September [2, 3]. Abnormal vacuum events in the chamber of Cell-2 prevented an increased beam accumulation (< 300 mA.) In November, a dipole chamber of Cell-2 was removed and replaced with a new one; a burned screw and a melted plastic cap were found on the bottom of this chamber. After the replacement of the chamber, we attained 520 mA in December, above the design value 500 mA. User trial runs began in 2016 March. Of ten ID, three are elliptically polarized undulators (EPU) and seven are in-vacuum undulators (IU). These three EPU have chambers of racetrack shape; the inner vertical dimension is full gap 8 mm. The seven IU can be closed to full gap 7 mm. ID chambers have special tapers to keep the transitions as smooth as practicable. We conducted measurements of bunch length, synchronous phase, beam size (energy spread) and tune shift vs bunch current up to 10 mA (single bunch) with varied configurations of the ID gaps. The computer simulations of the TPS broad-band impedance were reported [4, 5]. The measurement results of the real machine are discussed in this article. LONGITUDINAL IMPEDANCE The longitudinal broad-band coupling impedance of the ring elements are divisible into real and imaginary components [6]. The real part is the sum of the resistive components and is characterized as the loss factor that represents the synchronous phase shift to compensate for energy loss and the skewness of the longitudinal bunch profile due to wake fields; the imaginary part perturbs the particle motion inside the bunch and results in bunch lengthening or shortening, a synchrotron tune shift and an energy spread. Synchronous Phase Shift To extract the resistive part of the impedance, a lock-in amplifier (Zurich Instruments, UHFLI) was employed in the synchronous phase-detection system, shown in Fig. 1. Figure 1: Synchronous phase detection of TPS. The loss factor of the coupling impedance is expressible as ) / ( ) cos( 0 || b s s RF I f V k, in which RF V denotes the RF voltage, s the synchronous phase, 0 f the revolution frequency and b I the bunch current. For a Gaussian bunch, ) 2 /( || a z R k , in which R is the resistive component in a R and L (inductive as most light sources) series-circuit model, z is the bunch length and a is a machine-dependent scaling parameter. Figure 2 depicts the measured synchronous phase shift vs bunch current at varied RF voltage after the IDs were installed. The measured loss factor as a function of bunch current is shown in Fig. 3; scaling factor a ranges from 1.17 to 1.21. The resistive component R is shown in Table 1. The loss factor is larger for a gap open as shown in Table 1 and Fig. 2. Measured temperature rise in the taper area with 40 mm gap was larger than 7 mm. There might be some trapped modes. Figure 2: Measured synchronous phase shift of TPS with varied RF voltage and in-vacuum undulator gap. MOPOR016 Proceedings of IPAC2016, Busan, Korea ISBN 978-3-95450-147-2 630 C op yr ig ht © 20 16 C C -B Y3. 0 an d by th e re sp ec tiv e au th or s 05 Beam Dynamics and Electromagnetic Fields D04 Beam Coupling Impedance Theory, Simulations, Measurements, Code Developments Figure 3: Measured longitudinal loss factor || k vs bunch length z of TPS; a z k ~ || , a =1.17~1.21. Bunch Lengthening In a diagnostic beam line from a bending port, there exist an X-ray pinhole camera and an interferometer to measure the beam size, and a streak camera branch in the visible light regime to detect the bunch length [7]. A dual-sweep streak camera (C10910 Hamamatsu Photonics) is equipped to measure the longitudinal motion of the beam and the bunch length with picosecond resolution. The impedance causes a distortion of the potential well (PWD); the equilibrium distribution generated by the wake fields of the bunch charges are described approximately with the Haissinski equation [8],