The Calibration of BEPC Beam Position Monitors

The basic requirement for the BEPC beam position monitor is the measurement of the beam orbit with 0.1 mm precision near the collision point. To improve the measurement accuracy, the response of the beam position monitor pickups was mapped in the laboratory before they were installed in the BEPC ring. The microcomputer-controlled test set consists of high frequency coaxial switches to select each pickup electrode, a movable antenna to simulate the beam, a signal source, a spectrum analyzer to measure the pickup signals, and analysis software. The signal source operates well below the -3 dB cutoff frequency of the pickups (buttons). We believe that the low-frequency measurement yields the same information as the real beam. The button signals were clear. This calibration technique is satisfactory for BEPC operation. INTRODUCTION Four-button type beam position monitors (BPM) are used in the BEPC 2.2 GeV storage ring. The BPM assembly is an electrostatic type with four disk electrodes and BNC vacuum feedthrough connectors. The beam pipe is a cylinder. The buttons are rotated 45 degrees off vertical and horizontal axes to avoid the fan of synchrotron radiation. A tooling ball located on the top of the vacuum chamber is used as the fiducial mark for survey and alignment of the BPM assembly, relative to an adjacent quadrupole magnet. Electrical differences in the buttons and the mechanical installation tolerances cause the BPM to report beam offsets that are not real. We measured these offsets in our test set. The button disk is made of stainless steel. The disks are welded to the center conductor of the BNC feedthroughs. The feedthroughs themselves are welded into the vacuum pipe. The disks are flush with the vacuum pipe wall. Button signals can reach a few GHz. A spectrum analyzer was used to measure button response. The BPM quality was measured with three different tests. First, we measured the button response to a fast pulse on an antenna in the center of the chamber. This measurement gave the button sensitivity to beam current. In the second test, we excited the antenna with a low-frequency field with the antenna centered. This gave us the BPM offset due to electrical and mechanical errors. If the buttons had equal capacitance and were perfectly installed relative to the center of the beam pipe, there would be no offset error and all buttons would produce identical signals. The third test we performed was an evaluation of the BPM sensitivity to antenna position. Figure 1 shows button signals from a typical antenna scan. FIGURE 1. Button outputs over 10 X 10 mm scan. The measuring setup is shown schematically in Figure 2. The antenna may be moved transversely inside the monitor along the Xand Y-axes. A SP6T rf switch is used to select signals from each button. The insertion loss of the switch was tested and corrections were made to the data. The attenuation of all button cables were measured. The mechanical error of the stepping motor was considered while the antenna moved. A personal computer (PC) controls the equipment via a PC I/O board, the RS232 port, and a GPIB board. The measurement is completely automatic. wGenerator BPM Coaxial E-104 Switch AmpPfk FIGURE 2. BPM calibration method schematic. The button electrode faces a beam, and the button senses an image current lb: * 2!c2&% b dt b dt (1) where p is the linear charge density, a is the radius of the button, and b is the radius of the duct (beam pipe). The self-capacitance of the button to the wall of the beam duct is Cb. A load resistor R, in shunt with C,, will produce a frequency-dependent coupling impedance to the beam that acts like a high-pass filter: The button output signal is as follows: v=Ibz=I R b 1 + iwRCb R l+iwRCb )I@) =$ ; )Rl(w) (2)