Akihiko Miura

Beam Monitors for the Commissioning of Energy Upgraded Linac

In 2009, Japan Proton Accelerator Research Complex (J-PARC) began a project to upgrade the J-PARC Linac using the annular-ring coupled structure linac cavities. The aim was to achieve a beam power of 1 MW at the exit of the downstream rapid cycling synchrotron. For the upgraded beam line, beam monitors were designed, fabricated, and laid out considering the beam commissioning strategy. This study introduces the beam monitor layout in the new beam line and the results of commissioning, which confirm appropriate functioning of the beam monitors.

Beam-Loss Monitoring Signals of Interlocked Events at the J-PARC Linac

It is important to understand why the beam loss occurs during user operation. It is understandable that the beam loss results from RF cavities failure. However, it would be still useful to study the beam loss detailed mechanism and to know which beam loss monitor (BLM) experiences the highest loss or is most sensitive. This may lead a reduction in the number of interlocked events and a more stable accelerator operation. The J-PARC Linac BLM has a simple data recorder that comprises multiple oscilloscopes. Although its functionality is limited, it can record events when an interlock is triggered. Of particular interest here are the events associated with only the BLM Machine Protection System (MPS). These may reveal hidden problems with the accelerator. INTRODUCTION The Japan Proton Accelerator Research Complex (J-PARC) is a high-intensity proton accelerator facility with three experimental hall, Materials and Life Science Experimental Facility (MLF), Hadron Experimental Facility (HD) and Neutrino Experimental Facility (NU). The accelerator parts are a 400-MeV linac, a 3-GeV Rapid-Cycling Synchrotron (RCS) and the Main Ring (MR), which is operated with 30 GeV. The designed beam power and intensity of the RCS at repetition rate of 25 Hz are 1 MW and 8.3 × 1013 protons per pulse (ppp), respectively. On a one-shot basis, this goal was achieved in early 2015 [1]. That same year, there were two MLF target failures at 500 kW. Since then, the nominal operational beam power and intensity of the RCS have been limited to 200 kW and 1.8 × 1013 ppp, respectively, for the MLF. The MR is operated with cycles of 2.48 and 5.52 s for the NU and HD, respectively. While most of the RCS beam is supplied to the MLF, four consecutive batches of two bunches each are injected from the RCS to the MR within either of these cycles. Operational MR beam powers of over 425 kW and 42 kW are achieved for the NU and HD. The designed linac beam current and macro-pulse length are 50 mA and 500 �s, respectively. However, the peak current of the linac has been kept to 40 mA so far in 2016. The linac bunch structure has also been changed at the request of users. The typical bunch structure for the MLF is 300 �s macro-pulses in the linac and one bunch in the RCS. For the HD, the macro-pulse length is the same as that for the MLF, but the intensity is typically 1.2 × 1013 ppp. For the NU, the macro-pulse is the designed 500 �s length and a typical RCS intensity is 5 × 1013 ppp. ∗ naoki.hayashi@j-parc.jp It is important to understand the over-all accelerator behavior, performance and characteristics, particularly in relation to the beam loss. The Machine Protection System (MPS) is usually triggered when a machine or instrument mal-functions or a beam loss monitor (BLM) hits its predeined threshold. The consequence in either case is that the beam is automatically stopped by the MPS. It is certainly the case that failure of a RF cavity can cause a beam loss. Hence, it is useful to study the detailed correlation between RF cavity failures and the beam-loss pattern. This requires event data from many recorders with time identiication. Sometime, a BLM will trigger the MPS without any sign of machine failure. This could be because of beam instability, accidental beam loss, or some other sources. Understanding beam losses and the entire machine characteristics further would help to reduce the number of MPS events and improve accelerator operation. LINAC AND BEAM MONITORS The linac comprises various sub-systems. Its front end is an RF-driven H− ion source [2] and a 3-MeV RFQ [3]. Three drift-tube-linac (DTL) and 16 separated drift-tubelinac (SDTL) cavities then follow, and the H− beam reaches 190 MeV at this point. After that, 21 annular-ring coupled structure (ACS) cavities that were added in 2013 accelerate the beam up to 400 MeV [4]. The linac-to-3 GeV RCS beam transport line (L3BT) has a length of 190.5 m1 and includes a 90 degree arc section in between two straight sections. The inal ACS cavity, is showing in Fig. 1, along with debunchers 1 and 2, and 0-degree and 30-degree beam dumps. There are two more beam dumps (100-degree and 90-degree) downstream of the second straight section. These four beam dumps are used during beam tuning. The arc section contains six bending magnets from the marked BM01 to BM06 in Fig. 1. A proportional chamber type BLM (BLMP) is adapted as the main BLM [5]. Its pre-ampliier is placed either in the sub-tunnel (B1F) or in the machine tunnel (B2F). The signal unit is in the klystron gallery (1F). Its high voltage (HV) is set to 2 kV. The maximum raw output is < 5 V. There are many BLMs distributed all over the linac. In particularly, after 7�h SDTL, each SDTL and ACS cavity has its own BLMP. In total, 79 BLMPs are connected to the MPS. The number of BLMP is 31 and 5 in the L3BT and in the beam dump area, respectively. ����14, ����18, and ����21 are located between the debuncher cavity 1 1 It comprises four subsections. Straight section before arc is 33.0 m, Arc section is 44.9 m, Straight section after arc is 59.1 m, and Injection section (to the RCS) is 53.5 m. Figure 1: Downstream section of the linac after the inal RF cavity ACS21. Locations of SCT and BLMP are indicated. and the beginning of the arc-section. ���12 (Slow Current Transformer, monitors the beam current) is also located at the right after the debuncher 1. ����33 is in front of BM04 and ����39 is in front of BM05. ���45 and ����55� are in the second straight section. In contrast to the RCS or MR BLMP, the linac BLMP MPS is triggered by the raw waveform and not by a signal integral. Although the integrated value might be more stable, the response time would be longer. The MPS for the linac is designed to stop the beam within 5 �s. The MPS thresholds can be changed using EPICS, most are set to 1.3 or 1.6 V. Inside the MPS unit, a comparator and two PLCs judge whether the raw BLMP signal is too wide. Presently, the threshold width is set to 340 ns. Description about the MPS and a MPS unit can be found in references [6, 7]. WAVEFORM ARCHIVING SYSTEM The raw BLM waveform archiving system comprises multiple oscilloscopes2. There are more than 50 oscilloscopes for the entire linac. At present, 12 of these actively archive the data when the MPS is triggered. The sampling rate is 100 Msample/s (10 ns/step), the record length is 100 ksample, and the sampling time is 1 ms. The scope parameters are monitored and can be modiied through EPICS. During a communication between the EPICS IOC and the oscilloscopes, the system is locked, no trigger is accepted and the data are not archived for about a second. This interrupt occurs every several seconds and this dead time is a problem of this system. The MPS stops the beam within several �s. However, the associated beam trigger from the timing system has an inherent delay. Several triggers are ired even after the MPS event, usually leading to some empty BLM data being recorded. That is why the archive system records 20 con2 Yokogawa, DL1640. secutive waveforms. The archive system records not only the BLM signals but also some SCT and fast current transformer (FCT, monitors the beam phase) waveforms.

Present Status of the Laser Charge Exchange Test Using the 3-MeV Linac in J-PARC

The Accelerator-driven System (ADS) is one of the candidates for transmuting long-lived nuclides, such as minor actinide (MA), produced by nuclear reactors. For efficient transmutation of the MA, a precise prediction of neutronics of ADS is required. In order to obtain the neutronics data for the ADS, the Japan Proton Accelerator Research Complex (J-PARC) has a plan to build the Transmutation Physics Experimental Facility (TEF-P), in which a 400-MeV negative proton (H) beam will be delivered from the J-PARC linac. Since the TEF-P requires a stable proton beam with a power of less than 10 W, a stable and meticulous beam extraction method is required to extract a small amount of the proton beam from the high power beam using 250 kW. To fulfil this requirement, the Laser Charge Exchange (LCE) method has been developed. The LCE strips the electron of the H beam and neutral protons will separate at the bending magnet in the proton beam transport. To demonstrate the charge exchange of the H, a preliminary LCE experiment was conducted using a linac with energy of 3 MeV in JPARC. As a result of the experiment, a charge-exchanged H beam with a power of about 5 W equivalent was obtained under the J-PARC linac beam condition, and this value almost satisfied the power requirement of the proton beam for the TEF-P.

Upgrade and Operation of J-PARC Linac

Kazuo HASEGAWA*, Hidetomo OGURI, Takashi ITO, Etsuji CHISHIRO, Koichiro HIRANO, Takatoshi MORISHITA, Shinichi SHINOZAKI, Hiroyuki AO, Kiyonori OHKOSHI, Yasuhiro KONDO, Jun TAMURA, Saishun YAMAZAKI, Toshihiko HORI, Fumiaki SATO, Yasuo NEMOTO, Isao KOIZUMI, Nobuo OUCHI, Nobuhiro KIKUZAWA, Akira UENO, Akihiko MIURA, Shinpei FUKUTA, Akinobu YOSHII, Koichi SATO, Akira OZONE, Yuki SAWABE, Yusuke KAWANE, Hiroshi IKEDA, Yuichi ITO, Yuko KATO, Kazuo KIKUCHI, Fumio HIROKI, Toshio TAKAYASU, Tsutomu USAMI, Munetoshi YANAI, Kazuhiko TADOKORO, Kenji OHSAWA, Fujio NAITO, Yong LIU, Zhigao FANG, Takashi SUGIMURA, Kenta FUTATSUKAWA, Kiyoshi IKEGAMI, Masato KAWAMURA, Kesao NANMO, Yuji FUKUI, Tomoaki MIYAO, Tomofumi MARUTA and Akira TAKAGI

Bunch Shape Measurement of 181 MeV Beam in J-PARC Linac

In the Japan Proton Accelerator Research Complex linac, an energy upgrade project was started in 2009 using annular-ring coupled structure (ACS) linac cavities. We decided to use bunch shape monitors (BSM) for monitoring longitudinal beam width measurement to achieve longitudinal matching using two bunchers located upstream of the ACS cavities, where the radio frequency jumps from 324 to 972 MHz. Three BSMs were fabricated and installed in the original beam line. The BSMs were commissioned with the beam and their operability was demonstrated. In this study, we introduce the mechanism of the BSMs, its operability, measurement results with the 181 MeV beam, and consistency check with the respective cavity amplitude. Furthermore, we describe the operational vacuum conditions and outline the improvements to the BSMs vacuum system.