Proceedings of IBIC2014, Monterey, CA, USA WECZB3 MEASUREMENT OF BEAM LOSSES AT THE AUSTRALIAN SYNCHROTRON E. Nebot del Busto, M. Kastriotou, CERN, Geneva, Switzerland; University of Liverpool, Liverpool, UK, M. J. Boland, ASCo, Clayton; The University of Melbourne, P. D. Jackson, University of Adelaide, Adelaide, R. P. Rasool, The University of Melbourne, Australia E. B. Holzer. CERN, Geneva, Switzerland, J. Schmidt, Albert-Ludwig Universitaet Freiburg, Freiburg, Germany C. P. Welsch Cockcroft Institute, Warrington, Cheshire; University of Liverpool, Liverpool, UK Abstract contribution coming from the particle sources, particularly the positron source, and the emittance growth budget up to The unprecedented requirements that new machines are the interaction point. The Damping Rings (DR) designed setting on their diagnostic systems is leading to the devel- to cool down the CLIC beams [2] comprise a 412 m ring opment of new generation of devices with large dynamic built of FODO cells in the long straight section and The- range, sensitivity and time resolution. Beam loss detec- oretical Minimum Emittance (TME) in the arcs with the tion is particularly challenging due to the large extension use of two half cells for dispersion suppression. Super- of new facilities that need to be covered with localized conducting wigglers aim to decrease the damping times to detector. Candidates to mitigate this problem consist of the 2 ms level, values well below the 20 ms imposed by the systems in which the sensitive part of the radiation detec- 50 Hz CLIC repetition rate. This constrains the require- tors can be extended over long distance of beam lines. In ments on time response of beam instrumentation. More- this document we study the feasibility of a BLM system over, the low foreseen longitudinal emittances triggers a set based on optical f ber as an active detector for an electron of single bunch collective effects typically not observable storage ring. The Australian Synchrotron (AS) comprises in electron machines. Hence, the measurement of beam a 216 m ring that stores electrons up to 3 GeV. The Ac- parameters on a bunch by bunch basis imposes a tighter celerator has recently claimed the world record ultra low requirement on the time resolution of the instrumentation transverse emittance (below pm rad) and its surroundings systems. are rich in synchrotron radiation. Therefore, the AS pro- vides beam conditions very similar to those expected in The Australian Synchrotron (AS) [3] is a third genera- the CLIC/ILC damping rings. A qualitative benchmark tion light source consisting of a 100 MeV linac, a 100 MeV of beam losses in a damping ring-like environment is pre- to 3 GeV Booster and a 3 GeV Storage Ring (SR). The SR sented here. A wide range of beam loss rates can be comprises 14 Double Bend Achromat (DBA) cells where achieved by modifying three beam parameters strongly cor- a 200 mA beam is circulated in pulses of approximately related to the beam lifetime: bunch charge (with a variation 600 ns when f lling 300 out of the 360 buckets. As it is range between 1 uA and 10 mA), horizontal/vertical cou- shown in table 1, there are many similarities between the pling and of dynamic aperture. The controlled beam losses parameters of the AS and the CLIC damping rings. In par- are observed by means of the Cherenkov light produced in ticular, the AS has recently measured vertical normalized a 365 µ m core Silica f ber. The output light is coupled to emittances comparable to those expected in CLIC. More- different type of photo sensors namely: Metal Semiconduc- over, the synchrotron provides good f exibility to modify tor Metal (MSM), Multi Pixel Photon Counters (MPPCs), some of its nominal parameters to approach those of the standard PhotoMulTiplier (PMT) tubes, Avalanche Photo- DR. For instance it is feasible to reduce the pulse length to Diodes (APD) and PIN diodes. A detailed comparison of CLIC like values by keeping nominal beam charge. Hence, the sensitivities and time resolution obtained with the dif- this facility provides a great opportunity to test and develop ferent read-outs are discussed in this contribution. instrumentation targeting requirements for the DR of the future collider. THE AUSTRALIAN SYNCHROTRON AS A DAMPING RING TEST FACILITY OPTICAL FIBER BEAM LOSS MONITORS: THE EXPERIMENTAL The compact linear collider CLIC [1] foresees two 20 km SETUP main linacs accelerating electrons and positrons up to 3 TeV. In order to provide an instantaneous luminosity on The use of optical f bers, once restricted to waveg- − − the order of 5 · 10+34 cm 2 s 1, the beam spot at the in- uides for the transport of information, is increasingly being teraction point shall reach unprecedentedly low nanome- adopted in the beam instrumentation f eld. The light gener- 2014 CC-BY-3.0 and by the respective authors ter sizes. This is only achievable with ultra low emittance ated inside an optical f ber due to the crossing of ionizing © beams at the entrance of the linac that account for the large radiation may be used as a tool for Beam Loss Monitoring ISBN 978-3-95450-141-0 Beam Loss Detection 515 Copyright WECZB3 Proceedings of IBIC2014, Monterey, CA, USA (a) Storage Ring scrapers (far left), Beam Loss Monitors be- (b) Optical f ber installed in the outer side of the bending mag- hind dipole (far right). net, beam from left to right (top view). Figure 1: Beam loss monitor set-up in the AS. Two 7 m long optical f bers with 365 µm diameter SiO2 Table 1: Comparison Between Some of the Parameters of core were located on the horizontal plane of the outer side the AS and CLIC DR. of the beam. A third f ber with reduced diameter core Parameter AS CLIC DR (200 µm) was also available and provided the possibility Energy (GeV) 3.0 2.86 to test three photo sensor within the same experiment. All Intensity (elec) 9.0 · 10+11 1.28 · 10+12 three f bers were enclosed into a tube to shield from ambi- Number of bunches 300 312 ent light and were situated near the f rst bending magnet in Pulse length (ns) 600 156 sector 11. This location was selected as the easiest position Circ. length (m) 216 427.5 to produced controlled beam losses since the beam scrap- ers are situated approximately 1.5 m upstream of the front frev (MHz) 1.38 0.73 Bunch spacing (ns) 2 0.5 face of this bending magnet, as shown in f gure 1(a). The f ber enclosing plastic tube run parallel for the length of the γǫx (nm rad) 58708 472 magnet, as seen in Figure 1(b) and it immediately deviated γǫy (nm rad) < 5 4.8 upwards from the horizontal plane toward the roof of the accelerator tunnel where it was extracted to the upper level. 2 (BLM) [4, 5]. Two main advantages raise from the use of A 14400 pixels 3x3 mm active area Multi Pixel Photon Optical f ber based BLM (OBLM) systems: Counter (MPPC) and a Photo Multiplier Tube (PMT) were coupled to the end of the two 365 µm f bers. The light • The active ionizing radiation detector can be dis- output of the third 200 µm f ber was directed via focusing tributed over large sections of beam lines. This pre- lens to the 25 µm active area of an Avalanche Photo Diode vents missing the observation of beam losses at other- (APD). The readout signals were observed via a 10 bit Ana- wise uncovered locations and it minimizes the number log to Digital Converter (ADC) or via fan in/fan out pulse of required sensors (and hence acquisition systems) discriminator and scaler as pulse counting mode electron- and therefore the cost of the system. ics. • OBLM systems can provide position reconstruction of SINGLE SHOT BLM CALIBRATION the original location of the beam loss with resolutions down to a few tens of centimeters [6]. The prototype OBLM system was calibrated by direct- ing a single bunch beam onto a fully closed scraper in the The light generation mechanism inside the optical f ber SR. The f rst challenge to overcome was the determination can be of various origins, namely: scintillation, f uores- of the number of charges hitting the intercepting device. cence, thermoluminescence, radioluminescence, etc. In The measurements of the DC current transformer were not this contribution, we concentrate on Cherenkov light gen- available as the beam did not even complete a single turn erated when charged particles cross the f ber with speed from the injection point. Intensity measurements from the higher than the phase velocity of light in the core medium. Booster ring and transfer lines did not provide enough ac- As a reference number, the Cherenkov light generating en- curacy due to the diff culty of a precise determination of the ergy thresholds for electrons and positrons (which account injection eff ciencies. Moreover, the lower current beams 2014 CC-BY-3.0 and by the respective authors © for the largest fraction of the charge component of the injected into the machine reached values either near or sig- showers) crossing a quartz f ber is 186 keV [7]. nif cantly below the noise level of the current measuring in- ISBN 978-3-95450-141-0 Copyright 516 Beam Loss Detection Proceedings of IBIC2014, Monterey, CA, USA WECZB3 ×106 1 VGUN = 80 V 1000 VGUN = 110 V 0.8 800 Number of Electrons Relative electron loss 6 ×10 0.6 600 18 16 14 0.4 400 12 10 8 6 200 4 0.2 2 172 174 176 178 180 182 184 0 0 20 40 60 80 100 120 140 160 180 11 11.5 12 12.5 13 13.5 VGUN (V) Slit position (mm) (a) Calibration voltage scan (b) Calibration aperture scan p0 3.067e-11 ± 1.549e-11 p0 1.365e-11 ± 2.572e-12 -9 p1 1.652e-19 ± 6.018e-20 p1 ± -12 -12 ×10 7.109e-20 9.992e-21 ×10 ×10 0.25 p2 1.334e-29 ± 4.958e-29 p2 -2.785e-29 ± 8.23e-30 0.25 3.4 PMT oBLM 3.2 0.2 0.2 Charge (C) Charge (C) 3 MPPC oBLM 2.8 0.15 0.15 MPPC integral (C) 2.6 0.1 2.4 0.1 2.2 0.05 2 0.05 1.8 0 6 1.6 3 0 ×10 ×10 200 400 600 800 1000 50 100 150 200 250 -0.05 Number of electrons Number of Electrons (c) Voltage scan.