Personal summary of work carried out at Daresbury Laboratory: ILC/CLIC, NLS, ALICE and ITER Luis Fernandez-Hernando STFC/ASTeC Daresbury Lab

ILC and CLIC collimation: Wakefields test beam Thermo-mechanical studies ILC positron target Radiation/dose studies Thermo-mechanical studies New Light Source beam dump ALICE RF instability problem BLM ITER Central Interlock System Summary

ILC and CLIC collimation design ILC 0.5 TeV – 30 km ILC 1 TeV – 50 km CLIC 3 TeV: 48 km CLIC 0.5 TeV: 13 km

J.L. Fernandez-Hernando The collimator mission is to clean the beam halo from e- or e+ off orbit which could damage the equipment and mainly to stop the photons generated during the bending of the beam towards the Interaction Point. These photons, if not removed, would generate a noise background that would not allow the detectors to work properly. The spoiler serves as protection for the main collimator body as it will disperse the beam, reducing the beam energy density by multiple Coulomb scattering, in case of a direct bunch hit avoiding severe radiation damage. CLICILC Energy1500 GeV250/500 GeV Bunches it has to resist 3122/1 Particles per bunch 3.72E92E10 σ x in the spoiler position 796 µm111 µm σ y in the spoiler position 21.9 µm9 µm ILC and CLIC spoilers

J.L. Fernandez-Hernando Long, shallow tapers to reduce short range transverse wakes High conductivity surface coatings Robust material for actual beam spoiling Long path length for errant beams striking spoilers ( Large   materials (beryllium…, graphite,...) Consider range of constructions, study relative resiliance to damage (melting, fracture, stress) Starting point ILC and CLIC spoilers

J.L. Fernandez-Hernando Beam Parameters at SLAC ESA and ILC ParameterSLAC ESAILC-500 Repetition Rate10 Hz5 Hz Energy28.5 GeV250 GeV Bunch Charge2.0 x Bunch Length 300  m Energy Spread0.2%0.1% Bunches per train1 (2*)2820 Microbunch spacing- (20-400ns*)337 ns *possible, using undamped beam ILC and CLIC spoilers – Wakefields test beam

J.L. Fernandez-Hernando T480 “wakefield box” ESA beamline ILC and CLIC spoilers – Wakefields test beam

J.L. Fernandez-Hernando BPM A run with the beam going through the middle of the collimator (or without the collimator) is used as reference for the next run where the collimator will be moved vertically. This run also serves to calculate the resolution of each BPM. BPM The analysis will do a linear fit to the upstream and downstream BPM data separately, per each pulse (bunch). For this fit the data is weighted using the resolution measured for each BPM. The slopes of each linear fit are subtracted obtaining a deflection angle. This angle is transformed into V/pC units using the charge reading and the energy of the beam. All the reconstructed kicks are averaged per each of the different collimator positions and a cubic, or linear fit of the form: y’ = A 3 ·y 3 + A 1 ·y + A 0 or y’ = A 1 ·y + A 0 (only to collimator positions from -0.6 mm to 0.6 mm) is done to the result. The error in the kick reconstruction at each collimator position weights the different points for the fit. The kick factor is defined as the linear term of the fit (A 1 ). ILC and CLIC spoilers – Wakefields test beam

J.L. Fernandez-Hernando

ILC and CLIC spoilers – Wakefields test beam

J.L. Fernandez-Hernando 1 ± 0.1 V/pC/mm 1 ± 0.2 V/pC/mm 1.4 ± 0.3 V/pC/mm 1.7 ± 0.1 V/pC/mm1.9 ± 0.2 V/pC/mm 1.7 ± 0.1 V/pC/mm2.6 ± 0.1 V/pC/mm ILC and CLIC spoilers – Wakefields test beam

Flexural Section (wakefield taper) Peripheral cooling sufficient? Angle varies from 0 at max aperture opening to 90mrad ~5 o (full included angle (or ±20mrad about axis) Precision encoded actuators with bi directional repeatability to <10  m (<5  m possible?). Note with 10  m over 300mm span, 0.03mrad angle control is possible on pitch of collimator surfaces Vented Side Grill for Wakefield continuity and pumping Vessel (wire seal UHV compatible) EntranceTransition Flare From 20mm diameter to 30(h)x40(w)mm rectangular section. Spoiler Block 21mm width Ti Inclined Wakefield Collimator Block Bulk Material – Be, semi-transparent to 500GeV electrons. Converging in 2 steps of opening angle 65mrad (3.7 o ) & 40mrad (2.3 o ) nearer the spoiler block (note: opening angle = ± 32.5mrad & ±20mrad about central axis respectively) then diverges at same angular rate downstream of the spoiler block EPAC08, WEPP168 ILC and CLIC spoilers

Energy spoilers Energy1500 GeV Bunches it has to resist 312 Particles per bunch 3.72E9 σ x in the spoiler position µm σ y in the spoiler position 21.9 µm Material length needed to spoil beam 0.05 Xo J.L. Fernandez-Hernando

Resulting stress (using FLUKA and ANSYS) right after a CLIC bunch train has hit the spoiler at 0.2 mm from its bottom (or 4.29 mm from its top). Being the normal orbit of the beam at 8 mm from the bottom of the spoiler (3.51 from the top) that represents a deviation from normal orbit of 10σ x. 950MPa, and tensile, which is way above tensile strength limit. J.L. Fernandez-Hernando

The top value of stress is ~340MPa and compressive. Meaning that there will not be fracture but there will be a permanent deformation, and in this case it is a vertical deformation of 5 µm, which represents a 0.1% of the half gap. Can we live with that? But… is a deviation of 10 sigmas even possible? I have also calculated the stresses when the bunch train hits 0.2mm from the top instead of 4.29 mm (or 4.29 mm from the bottom instead of 0.2). Which means a deviation of “just” 4.75σ x. ILC and CLIC spoilers

Silicon carbide (SiC) foam MaterialRadiation length Xo [cm] Copper1.44 Ti alloy3.56 Beryllium35.3 SiC (solid)8.1 SiC (foam 8%)337 SiC is a material with good thermomechanical properties. Used for LHC collimation phase 2, in F1 brakes, and aerospace applications. It can be used as core material for CLIC spoilers, coated with metal (Be, Cu...) Very long radiation length of the foam at 8% of nominal density allows for low energy deposition of the particle beam. ILC and CLIC spoilers J.L. Fernandez-Hernando

Pros and cons for using SiC foam as core material in CLIC energy spoilers covered by 0.05Xo (in the z direction) of beryllium: Pros: It will not matter the depth the beam hits as it will always see 0.05Xo of beryllium (the contribution of the SiC foam can be negligible). Save some beryllium. Cons: The junction of two different materials is a complicate thing, mechanically speaking. The different thermal properties can lead to dislocation or fracture of the junction when the bunch train hits. A single material spoiler is more “whole” in that aspect. ILC and CLIC spoilers J.L. Fernandez-Hernando

ILC positron target studies J.L. Fernandez-Hernando

Right after 8 hours irradiation (no cooling time) Dose rate [pSv/s] 1 hour cooling time Simulated a photon irradiation of 8 hours, bunches of 2E10 photons separated 300 ns. (6.67E16 photons/second) This edge is due to the steel vacuum vessel the target is in ILC positron target – Radiation and thermal effects J.L. Fernandez-Hernando

NLS beam dump 2.25 GeV e- 2E9 e- per bunch 1 MHz frequency meaning 450 kW in total A high repetition rate coherent FEL, a new class of machine which produces fully controlled X-ray pulses. J.L. Fernandez-Hernando

Energy density deposition given by FLUKA for a 2 mm radial beam size NLS beam into 200 cm of graphite and 10 cm of copper. Results are normalised per primary particle. GraphiteCopper Density [g/cm 3 ] Critical energy E C [MeV] Radiation Length X 0 [cm] Molière radius R M [cm]71.6 Melting Temp T melt [  C] Operating Temp T op [  C] <200 Static stress limit [MPa] at compression >40 at tension σ 0.2 ≈ (plasticity limit) Cyclic stress limit [MPa]60 at compression 30 at tension NLS beam dump design J.L. Fernandez-Hernando

Stress [Pa] Time [seconds] Equivalent stress (σ eq ) in the initial sweep spot position in the graphite core and a quadratic fit done to its peaks of stress showing how the stress stabilizes after a period of ~13 seconds at a value nearing 3 MPa Temperatures achieved in the section of the dump where the electromagnetic shower generated by the beam is at maximum intensity NLS stress studies – Steady and transient states Steady state calculation of the temperatures, in degrees C, achieved in the impacting surface of the 200 cm graphite/Cu beam dump using a beam sweep radius of 9 cm with 12 spots J.L. Fernandez-Hernando

NLS beam lines, experimental areas and beam dump J.L. Fernandez-Hernando

ALICE Nominal bunch charge on ALICE is 80pC. The bunches are produced in trains lasting from ~10ns to 100ms and the train repetition frequency can vary from 1 to 20Hz. Within the train, the bunches are separated by 12.3ns that corresponds to the laser pulse repetition frequency of 81.25MHz.

Booster Compressor IR-FEL Photoinjector Laser High brightness electron source Linac Arc FEL Acceleration LINAC Deceleration ALICE : Design, commissioning and running ALICE – Accelerators and Lasers In Combined Experiments 350 keV 8.35 MeV 35 MeV

EMMA – The first non scaling FFAG A FFAG (Fixed Field Alternating Gradient) is a type of accelerator in which the magnetic field in the bending magnets is constant during acceleration. This means the particle beam will move radially outwards as its momentum increases. A linear non-scaling FFAG is one in which a quantity known as the betatron tune is allowed to vary unchecked. In a conventional synchrotron such a variation would result in loss of the beam. However, in EMMA the beam will cross these resonances so rapidly that their effect should not be seen. EMMA (Electron Machine for Many Applications) accelerates electrons from 10 to 20 MeV. J.L. Fernandez-Hernando

ALICE commissioning and Beam Loss Monitoring system My tasks: Characterise the BLM system. Design and carry out experiments. Analyse the signal of the ionisation chambers and of the BPMs to extract a charge measurement that can be correlated with the monitor readings. Investigate the possible sources of RF phase instability. Operate the machine both for experiments I am responsible and other experiments related to ALICE.

My tasks: Characterise the BLM system. Design and carry out experiments. Analyse the signal of the ionisation chambers and of the BPMs to extract a charge measurement that can be correlated with the monitor readings. Investigate the possible sources of RF phase instability. Operate the machine both for experiments I am responsible and other experiments related to ALICE. ALICE commissioning and Beam Loss Monitoring system

My tasks: Characterise the BLM system. Design and carry out experiments. Analyse the signal of the ionisation chambers and of the BPMs to extract a charge measurement that can be correlated with the monitor readings. Investigate the possible sources of RF phase instability. Operate the machine both for experiments I am responsible and other experiments related to ALICE.

Example of the signals on a BLM sensor for different dipole currents. The base plateau moves away from zero and needs to be subtracted from the top signal plateau in order to get the real reading value. To test and characterise the BLM sensors in ALICE a test was done using BLM4 (in ST1), BLM5 (in ARC1) and BLM6 in ST2). Varying the currents on DIP-03 of ST1 and DIP-01, DIP-02 and DIP-03 of ARC1. The beam loss induced radiation is detected by a series of long ionization chambers (LIC) distributed around the machine. These chambers consist of an air-filled coaxial cable (Andrew HJ4-50, 50 Ω) with a 1 kV potential to attract the ionised gas particles forming a current flow. This current, although very small, can be measured to give an indication of beam loss.

Linearity check on the BLM response Several measurements were taken for BLM4 whilst maintaining the current setting for ST1 DIP-03 at 20 A. The train length was then incremented from 1 µs to 10 µs to vary the charge the BLM sensor would see. Example of a test on the ARC-01 BLM sensor. Analysis of the signal for different DIP-01 currents When we vary the current in DIP-01 we have a high increase of signal which decreases as we move away from nominal. This is due to the beam hitting less beam pipe and less accelerator components such as quadrupoles, sextupoles, screen vacuum vessels, etc. In the lower end of dipole current settings we start to see an increment of BLM signal, which is probably due to backscattering off the lead shielding surrounding the external face of this bending section. J.L. Fernandez-Hernando

ITER Central Interlock System J.L. Fernandez-Hernando

Follow up and manage ITER contracts with different companies and institutes (CERN, Create, Procon) for the creation of a Central Interlock System and Quench Loop prototypes. Design, prepare and carry out tests of the different prototype systems. Investigate the functions to be implemented into the Interlock system to ensure machine safety. Redundant PLC system s are used for the Central Interlock System (CIS) and the Plant Interlock System (PIS) For the prototype two Plant Interlock Systems will be used (Vacuum and Cryo). Each of them communicates with a single CIS. ITER Central Interlock System – Machine protection J.L. Fernandez-Hernando

ET200M SM 331 8AI 2xSM DI 3TK S- RELAY T T T PROFIBUS DP CPU414-2DP PROFIBUS DP network for redundant link (via CP module one at each CPU) IM xSM DO relay ITER – Quench Loop prototype The Quench Loop main target is to protect the ITER superconducting (sc) magnet system composed of a total number of 48 magnets: 18 Toroidal Field magnets, 6 Poloidal Field magnets, 6 Central Solenoid magnets and 18 Correction Coil magnets. In order to assure magnet protection throughout the different operational phases, several actors are required: the quench detection system which mission consists of detecting any imminent quench in one of the sc coils, the fast discharge units (FDUs) to extract the energy stored in the magnets, power converters that feed the magnets with current, and finally the central interlock system for magnet protection based on PLC (Programmable Logic Controller) technology. The quench detectors, the FDUs and power converters can open the Quench Loop if they detect a fault or see the status of the Quench Loop and act accordingly. ITER Central Interlock System – Quench Loop Prototype

Thank you for your attention