Study on beam position measurement based on diode detection in HLS-II

In order to improve the sensitivity and long-term stability of beam position measurement for Hefei Light Source – II (HLS-II), it is necessary to improve the measurement method. The beam position monitor (BPM) electronics is used to measure the beam position and is an important part of the beam position measurement system. In this paper, a beam position measurement system based on the compensated diode detection (CDD) technology for electron storage ring was proposed. Since HALF under construction, the parameters of HLS-II were used to design the system and simulate the system circuits to verify its feasibility.


Introduction
The beam position measurement system is one of the most important beam diagnostic subsystems in the synchrotron radiation storage ring, providing the most basic and quite important transverse position measurements.The Hefei Light Source -II (HLS-II) is a second-generation synchrotron radiation source with a circumference of 66.13 m, a revolution frequency of 4.534 MHz, and an electron beam energy of 0.8 GeV.The main parameters of the HLS-II storage ring are shown in Table 1.Improving the accuracy of the HLS-II beam position measurement requires that the rms in SA (10Hz) mode is about 700nm in x-direction and 400nm in y-direction.The system is based on the compensated diode detector technique, first proposed by M. Gasior et al. at CERN.It has been successfully applied to beam position measurements in LHC to a submicron resolution of the orbit measurements during the years 2011 -2013 [1][2].The measurement accuracy and long-term stability of this method are also applicable to the Hefei Advanced Light Facility (HALF).The HALF is a fourth-generation synchrotron radiation storage ring light source based on diffraction-limited storage rings (DLSR) conducted by the National Synchrotron Radiation Laboratory [3].The main parameters of the HALF storage ring are shown in table 1, the design goal of the HALF is to become a vacuum ultraviolet (VUV) and soft X-ray DLSR light source providing high quality synchrotron radiation.To achieve ultralow emittances, the beam orbit should be stable, with the rms fluctuations of not more than 500 nm horizontally and 200 nm vertically in SA (10Hz) mode [4].
In this paper, a beam position measurement approach based on the compensated diode detector technique for electron storage ring was proposed.The simulation of a single bunch filling was used for the prefeasibility studies.The advantages of the approach are simple circuits, low cost, slowly varying signals at the output, easy for signal processing.Then, the parameters of the HLS-II storage ring were used for preliminary system design and simulation validation, and plan to carry out online experiments.Horizontal beam size (μm) >400 >5

Measuring principle
The diode detector can be designed to process beam position monitor (BPM) signals and provide the output signals related to beam position [5].Using a simple diode detector for beam position measurements will lead to a systematic error that depends on the forward voltage of the diode Vd.This voltage drop can be compensated by a circuit similar to the diode detector.The circuit is shown in Figure 1, and it consists of two RF diode detectors connected to the same input signal Vi.The output voltage of the diode detector with a single diode is V1, and the output voltage of the detector with two diodes in series is V2.The diode voltage drop Vd is assumed to be a constant, the detector voltages V1 and V2 can be described as The output signals of the two detectors are processed by two operational amplifiers OA1 and OA2.As the voltage between the inverting and noninverting inputs of the operational amplifier is practically zero, and the current of the inputs also is practically zero.The difference of the output voltages of the two detectors can be converted into current in the resistance Roa2.The output voltage of the circuit Vo can be described as Relationship between the circuit output voltage Vo and input voltage Vi can be found by inserting the detectors output voltage equations (1) into equation (2).The derived equivalence can be then expressed as: From Equation (3), it can be seen that the voltage drop is compensated when the feedback resistance is equal, i.e.Roa1= Roa2.It can make the final output voltage equal to the input voltage, and the equation (3) can be simplified as ) But the main difficulty in measuring the beam position using the circuit is the diode forward voltage Vd.The diode forward voltage Vd is not constant but varies with the detector input signal amplitude, conversion rate, and temperature, thus making the beam position error a complex function.It is worth noting that in general the current of two detectors are not practically equal, so the forward voltage drop of the diode Vd in two detectors is different.

System design
The block diagram of the beam position measurement system based on the compensation diode detector technique is shown in Figure 2. The system mainly includes three parts: BPM, analog frontend (AFE) and analog-to-digital conversion (ADC).

Simulation Source
For simplicity, the beam signal was simulated as a voltage source with very short pulse.The stripline BPM is used for position measurement in HLS-II.The simulation source is generated according to the characteristics of the stripline BPM in HLS-II.The simulation BPM signal is shown in Figure 3.

Time Constant
In order to select the appropriate time constant τ of the detection circuit, the CDD part is separately simulated.This is mainly adjusted by the parameters of detection R and C. The simulation results with different time constants are shown in Figure 4.It can be seen that the amplitude of output signal increases with the increase of time constant or capacitance value.According to the previous work of HLS-II based on diode detection tune measurement system, the time constant should be less than 100T (T: revolution period) to avoid drag effect [6].Therefore, the time constant of CDD is set to 100T.

Low-pass Filter
RF-LPF uses the the XLF-151+ from Mini-circuits, while LF-LPF will use operational amplifiers to build 2nd order Sallen-key topology low-pass filters.The cut-off frequency of the filter was chosen to be about 100 Hz removing most of the noise above this frequency.The simulation results of its frequency characteristics are shown in Figure 5.

All Circuits
The simulations of all front-end circuits were performed in OrCAD Pspice.For simplicity, RF-LPF uses ideal low-pass filter IC, isolation transformer, calibration switch and PGA are ignored, and FGA uses ideal operational amplifier to build an amplifier with sufficient gain to meet the input voltage to CDD with optimal range.Simulations were performed using HLS-II parameters with a single bunch mode at 4.534 MHz.The cut-off frequency of RF-LFP is 80 MHz, the gain of FGA is about + 200 (≈ +46 dB), and the cut-off frequency of LF-LPF is 100 Hz.The time constant of the compensated diode detector (CDD) is 100T.The simulation results of each node in the circuit are shown in Figure 6.
As can be seen in Figures 6 (a) and (b), the RF-LPF reduces the BPM signal and extends the signal pulse width to a large extent, which reduces the performance requirements of the devices used in the subsequent multistage signal processing.For convenience, the gain of the FGA is set so that it can be amplified to the input signal amplitude, and the simulation result is shown in Figure 6 (c).
Figure 6 (d) shows the result of the signal after the CDD.Although the CDD did not strictly compensate the signal to the input amplitude (about 8 V) and there are some errors.However, this does not affect the feasibility of the system and can be compensated by other means.
Figure 6 (e) shows the output result of LF-LPF (that is, the processing result of the AFE), which can intuitively show the DC voltage related to the beam position.It can reduce the sampling rate requirements of ADC for signal acquisition, and has the advantages of simple circuit and low cost.

Conclusion
In this paper, the design of a beam position measurement system based on the compensate diode detection for electron storage ring was proposed.The AFE of the system mainly includes filtering, amplification, and compensation diode detection for processing BPM signals.While ensuring signals related to beam position, other signals are filtered and signal processing requirements are reduced.The circuit simulation of the system is carried out by using the parameters of the HLS-II storage ring.The simulation results show that the system is feasible on the electron storage ring.In the future, the influence of the non-ideal electronics and the noise contribution are considered in the design, the design of the system is optimized and carry out physical manufacturing and online experiment of the system.

Figure 2 .
Figure 2. Block diagram of beam position measurement system based on compensation diode detector.The BPM is used to obtain the beam signal.Each AFE channel can be used to process the beam signal from a pair of opposite electrodes in BPM, that is, the x direction or the y direction.The AFE adapts the signal from BPM detectors to ADC range and filters out the useless frequency content of the beam signals.The ADC digitizes the processed beam signal, and sends it to FPGA or other equipment for digital processing, and finally calculates the beam position.The AFE circuit includes:  Non-reflection RF low-pass filter (RF-LPF): reduces the BPM input signal amplitude and spreads the signal pulse width for subsequent amplification processing. Isolation RF transformer: isolate signals and transmit clean BPM signals. Calibration GaAs Switch: periodically exchanges BPM electrode signals to eliminate systematic errors introduced by the gain and offset of the cables and electronics. Programmable amplifier (PGA): The gain is automatically set by the system to maintain the optimal signal on the diode detector. Fixed gain amplifier (FGA): Complementary to the programmable amplifier, the signal is amplified at a fixed gain. Compensating diode detector: converts the rapidly changing beam pulse signal into a slowly changing envelope signal, and restore it to the input voltage amplitude. Low frequency low-pass filter (LF-LPF): Filter out signals above than 100Hz.

Figure 3 .
Figure 3. Waveform of simulation signal source, the amplitude of pulse is 12 V.(a) Single-period waveform.(b) Partial zoom of pulse signal.

Figure 4 .
Figure 4. Simulation results of CDD output voltage with different time constants or capacitance values.

5 Figure 5 .
Figure 5. Simulation results of frequency characteristics of 2 nd order LF-LPF with Sallen-key topology.

Figure 6 .
Figure 6.Simulation results of the all circuits.(a) The BPM output signal.(b) The RF-LPF output signal.(c) The FGA output signal.(d) The CDD output signal.(e) The LF-LPF output signal, which is also the final AFE output signal.

Table 1 .
Parameters of the Storage Ring of the HALF and HLS-II.