Beam Test Results of High Q CBPM prototype for SXFEL *

Transcription

1 Beam Test Results of High Q CBPM prototype for SXFEL * Jian Chen ( 陈健 ),;) Yong-bin Leng ( 冷用斌 ) ;) Lu-yang Yu ( 俞路阳 ) Long-wei Lai ( 赖龙伟 ) Ren-xian Yuan ( 袁任贤 ) Shanghai Institute of Applied Physics, Chinese Academy of Science, Shanghai 4, China University of the Chinese Academy of Science, Beijing 49, China Abstract: Aiming at high precision beam position measurement of micron or sub-micron for Shanghai Soft X-ray free electron laser (SXFEL) facility which is being built in site of the Shanghai Synchrotron Radiation Facility (SSRF), Shanghai Institute of Applied Physics has developed a high Q cavity beam position monitor (CBPM) that the resonant frequency is 4.7 GHz and relevant BPM electronics include dedicated RF front-end and home-made digital BPM (DBPM) also has been done. The cavity design, cold test, system architecture and the beam test with three adjacent pickups has been performed in Shanghai Deep ultraviolet free electron laser [](SDUV-FEL) facility are included. The beam experiment results show that the physical design of our CBPM is consistent with the expectations basically and the beam position resolution can fulfill the resolution requirements for the SXFEL project if we optimize the beam conditions. Key words: High Q CBPM, SXFEL, resolution, RF front-end, DBPM PACS: 9.7.Eg, 9.7.Fh Introduction The Shanghai soft X-ray free electron laser is a test facility was proposed in 6 to verify key FEL schemes and technologies cover two-stage cascaded High Gain Harmonic Generation (HGHG) and cascade of HGHG and Echo-enabled Harmonic Generation (EEHG) scheme, which is expected to generate output wavelength of 9 nm, fs pulse duration, Hz repetition rate and MW peak power by adopting FEL frequency doubling of ultraviolet band seeded laser of 65 nm to achieve []. The SXFEL is a fully coherent light source and currently under construction in SSRF campus, the overall length is about 3 meters and the nominal electron beam energy of the linac is.84 GeV, the linac commission will be carried out at the end of this year and first lasing in 7. The FEL facility which is based on the HGHG or EEHG scheme has a strict beam position resolution requirement of micron even sub-micron level to ensure the electron beam could overlap the generated photo beam stringently and pass through the entire undulator section together. For SXFEL project, the resolution requirement of beam position is um@.5nc. Among various types of BPMs, traditional BPM like button BPM and stripe BPM can t achieve this goal, but cavity BPM, adopt resonant cavity structure using asymmetric characteristics mode coupled from cavity to measure beam position, can reach submicron or even nanometer level position resolution requirement has been becoming a critical beam instrumentation component for FEL. For the CBPM, the quality factor (Q) is one of the most important parameters which is directly proportional to the decay time and inversely proportional to the intensity of the output signal. A cavity with a low Q factor means that the coupling structure has a very high efficiency, so the output signal will have a larger amplitude, better SNR, broad bandwidth and a short duration time. Conversely, high Q cavity output signal will have a lower amplitude, a little bad SNR, but have a narrow bandwidth and a long duration time. Our group designed a low Q cavity and tested in the last year [3,4], although it has a lot of strengths we mentioned but the performance was limited by the electronics due to the short duration time for the low Q cavity. Considering the matching with the electronics and single-bunch working mechanism of SXFEL, we choose high Q CBPM as beam position measurement scheme for SXFEL eventually. The high Q cavity BPM prototype designed for SXFEL has been fabricated and relevant electronics also have been developed. So as to conduct preliminary performance test, an array of three adjacent high Q cavity BPMs have been *Supported by National Natural Science Foundation of China (No.5758 No.3553) ) chenjian@sinap.ac.cn ) lengyongbin@sinap.ac.cn

2 installed at the SDUV-FEL facility. Table. Pre-designed parameters of the high Q CBPM. Measurement principle of CBPM For a cylindrical pill-box cavity, when beam source is along the z-axis, the bunch will not loss any energy in the transverse electric field of the TE mode, but due to the existence of the longitudinal electric field of the TM mode, bunch will loss energy in longitudinal electric field excited by itself and induce the mode be excited effectively. Therefore, only TM modes will be excited and the amplitude is determined by the bunch lost energy. Considering the asymmetric characteristics of TM dipole mode which has a strong linear dependencies to the beam offset and the beam charge, whereas TM monopole mode unaffected by the beam offset but it is proportional to the bunch charge and bunch length. So TM resonant mode usually employed in the position cavity to measure the beam offset and use TM resonant mode to eliminate the variation effect of bunch charge and normalize the amplitude from position cavity simultaneously. When a bunch of particle beam transit the cavity close to the z-axis, the TM mode will get the most lost energy so the amplitude is larger than TM mode much and both of them will have a certain amount of crossover in bandwidth. So it is important to design an appropriate coupling structure which can export TM mode well and damp most of the power from the TM mode in the meantime. It is also become a main influence for the measurement of position resolution [5]. Due to the limited quality factor of the cavity, the energy of all electromagnetic field modes excited by the cavity will have three tracks. Some will be stored inside of the cavity, small portion will lost on the metal wall and another will be coupled out by appropriate coupling structure so it can be detected by the following electronics. 3 Cavity pickup Position cavity Reference cavity Resonant frequency 4.7 GHz 4.7 GHz Loaded Q factory 8 8 Number of ports 4(X:,Y:) The three-dimension structure of the cavity which consists of seven parts is shown in Fig.. The total length of the probe is fixed in mm, in order to reduce the possibility of signal coupling between position cavity and reference cavity so the distance of the two cavities increased from 35 mm to 45 mm. The ports number of the reference cavity was raised so as to convenient for cold test and coupling structure also changes from magnetic coupling to weaker electric coupling. Fig.. Three-dimension structure of the high Q CBPM. 4 Cold test Considering the complexity of setup a platform to test the cavity, we develop a convenient and effective method to test the resonant frequency and the Q factor of the cavity by network analyzer in laboratory, in which we measured S parameters and regarded the frequency of peak as cavity working frequency, the Q factor was got by measuring the 3 db bandwidth. The results of the cold test can be seen in Fig.. Base on the defects of the low Q cavity by beam test, our group redesigned the high Q cavity BPM intentionally and strategically. The pre-designed parameters of the high Q CBPM are summarized in Table. The working frequency was chosen at 4.7 GHz due to consider the vacuum pipe radius and to avoid the dark current from the accelerating system. Cavity material also changes from stainless steel of 34 to copper so as to raise the Q value, the quality factor of the high Q is 8 in theory. Fig.. Cold test results of the cavity. From the figure mentioned above, the cold test working frequency of the cavity is GHz and GHz for

3 position cavity, GHz for reference cavity. Which front end we developed is shown in Fig.5. accord with the designed value of 4.7 GHz well. But correspondent quality factor has greater difference compared with the design. We are searching the cause for this difference and will give an ideal solution in the later stage. 5 System setup Although the results of the cold test coincide with the theoretical design value well, but it may be have some different in the case of the particle beam. On the other hand, we also want to test the whole BPM system we developed with beam. Therefore, we conduct the beam test on the platform of SDUV. The BPM system comprise of cavity BPM, a dedicated RF front end and a home-made digital BPM processor. 5. Arrangement of the CBPMs Fig.3 show an array of three BPM pickups installed at the SDUV test injector facility, the middle of the cavities was mounted on a movable stage to imitate the beam offset. The arrangement of the three CBPMs also illustrated in Fig.4. Fig.5. Schematic of the RF front end. An input band pass filter (BPF) selects the cavity signal components around 4.7 GHz to suppress other harmonics coupling from the cavity or spatial disturbance into the followed low noise amplifier (LNA). The LNA with 45 db gain is used to adjust the signal intensity from BPF and acquire a broader dynamic range. The down conversion operates with a LO convert the RF signal to IF which can range from DC to hundreds of megahertz. Generally, we convert the RF signal to IF about 5 MHz in project which depend on the performance of the DBPM. An adjustable IF amplifier accomplishes the last gain adjusting to fulfill the input requirement of ADCs and a BPF with center frequency of 5 MHz, bandwidth of MHz is used as an anti-aliasing filter to suppress other amplified harmonics. Both of them are integrated in the DBPM. Fig.3. BPM test array installed at the SDUV test injector facility. Fig.4. Arrangement of the three CBPMs. 5. RF front end Owing to the limitation of ultra-high speed ADC resolution, RF sampling methods are difficult to achieve precise measurement of sub-micrometer level. It is absolutely essential to adopt RF receiver architecture which can convert the RF signal to IF and digitized by ADC with high resolution. The simplified block diagram of the RF 5.3 Home-made DBPM The commercial BPM signal processor has a great many of advantages such as mature technology, high reliability, large users and easy to communicate. But the price is too expensive to use in smaller accelerators which limited by funds, and there also has some problem in maintenance. Our group start research on the new DBPM with independent intellectual property in 7, the performance is closed to the similar products of foreign. The DBPM based on software radio architecture and adopt band-pass sampling technology as the quantization scheme which consists of RF pre-processing module, analog to digital conversion module and a digital board module. Fig.6 is a photograph of the DBPM and Table show the basic parameters of the DBPM.

4 Figure.8. CBPMs spectrum of the reference cavity. Fig.6. Photograph of the DBPM. Sample rate Table. Basic parameters of the DBPM. Centre Frequency Band- width ENOB Dynamic gain-range 7 MHz 5 MHz MHz 6 bit 6 db More details about the DBPM can be seen in [6]. 6 Beam test The beam test has been done based on the platform of SDUV. We use the broadband oscilloscope to evaluate the performance of the cavity BPM directly, measured the calibration factor and the position resolution by whole BPM system. The detailed results were described in the following subsections. 6. Cavity evaluation For getting the RF signal of the CBPM prototype in beam conditions to evaluate the physical design accord with the expected goal or not, we use the broadband oscilloscope to process which can evaluate it directly and quickly. Also, with broadband oscilloscope we got the signal data and processed in MATLAB. The CBPM output signal waveform was shown in the Fig.7. Fig.8 and 9 illustrate the CBPMs spectrum of reference cavity and position cavity. Figure.9. CBPMs spectrum of the position cavity. From the figures mentioned above, the RF signal waveform in the time domain consistent with the theoretical expectations. Signal spectrum of reference and position cavity in accordance with the cold test very well. We also can find that the resonant frequency of three CBPMs position cavities has a good consistency but a great difference in reference cavity which may be caused by the solder melts into the reference cavity when welding. Meanwhile, the Q factor and the decay time also can be evaluated. As shown in Fig., the Q value of reference cavity and position cavity is 34 and 46 respectively which accord with the cold test. Fig.7. RF signals waveform of CBPM. Fig.. Q values of the reference cavity (a) and position cavity (b). The decay time can be calculated in frequency domain

5 also can be fit in time domain. In the frequency domain, the decay time can be calculated using the Eq. (): Q. () *pi* f L Using the Q value and resonant frequency we evaluated, we calculated the decay time of the reference cavity and position cavity is 73 ns and 45 ns respectively. In the time domain, the data-fitting method can help us get decay time directly but we need to do some pretreatment for the original data like digital filter to remove the influence of harmonics. Fig. show the result of 77 ns for reference cavity (44 ns for position cavity) processed in time domain which is close to the result processed in frequency domain. Fig.. IF waveforms of reference cavity (a) and position cavity (b) sampled by broadband oscilloscope. With the RF front end and home-made DBPM, IF signals about 5 MHz were sampled. Due to the Sub-Nyquist sampling method we used, the frequency of IF signals were transferred to the first Nyquist zone. The waveforms and spectrums are shown in Fig.3 and 4. Fig.. Decay time of reference cavity processed in time domain. From the discussion we mentioned, the results of cavity evaluation generally agree with the design except the quality factor but it also can meet the requirement for project base on the BPM system. Nonetheless, we will do our best to find out the cause and optimize it in the later. 6. IF pulse shape The RF signals output from the cavity is directed down-converted to a low intermediate frequency by this RF front end then sampled by signal processor, two different LO are used to get the IF signals about 3 MHz and 5 MHz respectively to verify the correctness of the RF front end. Considering the bad performance for DBPM about 3 MHz, so a broadband oscilloscope is used to evaluate the RF front end also. Fig. show the IF waveforms sampled by broadband oscilloscope with LO of 4665 MHz, a LPF of DC to 3 MHz also be used to suppress the higher-order harmonics generated by mixer. The waveform and spectrum are in accordance with the theory. Fig.3. IF waveforms of reference cavity (a) and position cavity (b) sampled by DBPM.

6 amplitude / a.u reference cavity vertical vertical vertical frequency / MHz Fig.4. Spectrums of the IF signals sampled by DBPM. From the figures above, the IF signal waveforms and correspond spectrums are in line with expectations that proof the RF front end can be applied in the high Q CBPM system. 6.3 Calibration The calibration factor is required to convert the position of beam offset. A two-dimensional motion platform was installed under the one of the cavity which can imitate the beam offset from - mm to mm with a step of um both on the horizontal and vertical directions. The diagram of the calibration system is detailed in Fig.5. With the motion platform, we collected the data at every step by DBPM. Then, the signal data was processed in the frequency domain with MATLAB and the calibration factor of the CBPM was shown in Fig.6. show in Eq. (), the position reading of CBPM (U ) can be estimated by the position reading of CBPM and CBPM3 (U and U3), we also get a position reading of CBPM by itself (U) at the same time. To calculate the difference between U and U ( U), and assume that all BPMs have equal position jitter then the position resolution can be get by Eq. (3). Where the GF is a geometrical factor related to the location of three cavities that can be calculated by Eq. (4), d is the position difference converted by the calibration factor and U. U ' ( D* U3 D3* U) / D3. () GF. (3) CBPM * std d GF. (4) D3 D ( ) ( ) D3 D3 UY Fig.5. Diagram of the calibration system. experiment linear fit Y =.8587X cavity position / mm Figure.6. Calibration factor of the CBPM. 6.4 Position resolution The measurements of position resolution were performed based on correlating the readings of three cavities. Schematic of position resolution measurement base on three cavities is shown in Fig.7. Using geometric relationships as Fig.7. Schematic of position resolution measurement. In the experiment, we sample one thousand sets of data by DBPM and processed offline. The bunch charge during this measurement was pc and the gain setting of RF front end was chosen to provide a linear measurement range of - mm to mm. Using signals of reference cavity to normalize the signals of position cavity and the position of beam offset was converted by calibration factor, we got the relationship of the position measured and expected values of CBPM is shown in Fig.8. position measured / mm position expected / mm.4.6 Figure.8. Relationship of the position measured and expected.

7 counts total sample = 65 STD = 8 m Y meas - Y expe / mm Fig.9. Histogram of the d with 65 samples. Then calculating the distribution of the d results in the histogram seen in Fig.9, under this condition, the position resolution of the CBPM system is 3um. Owing to the SDUV facility is a test bed only, most of time it work in the state of low charge about ~3 pc and the beam jitter also has a bit serious so we set the linear measurement range of +/- mm. But for SXFEL facility, the work of charge will be set at.5 nc and the beam jitter will be controlled within +/-5 um. In theory, if we raise the bunch charge from pc to.5 nc we will get a gain of 5 times. Meanwhile, control the measurement range within +/-5 um we also can get the electronics gain of 4 times. Consider the most ideal conditions and the noise of the whole system has not changed a lot, we can get the position resolution of.3 um. Related experiments will be conducted in subsequent. 7 Conclusion We accomplished the whole CBPM system and conducted a preliminary test in the SDUV facility. The aim is to evaluate the performance of the new designed high Q CBPM and the performance of the whole CBPM system to prepare for the construction of the SXFEL. The results show cavity performance meets the needs of the project although the quality factor less than the design value. We got the position resolution of 3 um when the bunch charge was pc and the linear measurement range of +/- mm. By theoretical calculations, we can easily fulfill the resolution requirement of um@.5nc for the SXFEL if we raise the bunch charge and control the measurement range. 8 References Z.T. Zhao, Z.M. Dai, X.F. Zhao et al, Nuclear Instruments and Methods in Physics Research A, 58(-): (4) B. Liu, J.H. Chen, J.H. Chen et al. Status of the SXFEL and DCLS. Talk of the 37th International Free Electron Laser Conference, (Korea: JAcoW, 5) 3 Yuan Ren-Xian, Zhou Wei-Ming, Chen Zhi-Chu et al, Design and teat of SX-FEL cavity BPM. Chin. Phys. C, 37(): 8(3) 4 J. Chen,Y.B. Leng, L.Y. Yu et al, Beam experiment of low Q CBPM prototype for SXFEL, in proceedings of the 7th International Particle Accelerator Conference, edited by Stefano Deiuri (Korea: JAcoW, 6), p. 5 CHU J H. Development of C band cavity beam position monitor and RF front end signal processing system, Ph.D. Thesis (Shanghai: Shanghai Institute of Applied Physics, CAS, 8) (in Chinese) 6 Yi Xing. Research of signal conditioning and high speed data acquisition techniques for particle acceleration beam diagnostics, Ph.D. Thesis (Shanghai: Shanghai Institute of Applied Physics, CAS, ) (in Chinese)