Fermilab is responsible for the design of the 3.9 GHz cryomodule for the LCLS-II that will operate in continuous wave (CW) mode [1]. The bandwidth of the SRF cavities will be in the range of 180 Hz. In the tuner design, the slow tuner-mechanism slim blade tuner was adopted, which was originated by INFN for the European XFEL 3.9 GHz [2]. The bandwidth of the SRF cavities for LCLS II will be in the range of 180 Hz and fine/fast tuning of the cavity frequency required. A fast/fine tuner made with 2 encapsulated piezos was also added to the design. The first prototype tuner has been built and went through testing at warm conditions. Details of the design and summary of the tests are presented in this paper. REQUIREMENT FOR THE TUNER 3.9 GHz Cryomodules (and cavities) designs for the LCLS II Project accumulated its best features from the previous 3.9 GHz cryomodule designed and built by FNAL (for DESY/FLASH) [3] and by INFN for EuXFEL [2]. Parameters of the 3.9 GHz cavity for LCLS II project are presented in the Table 1. The significant difference is in the bandwidth of the cavity. For the LCLS II project half bandwidth must be 90 Hz and peak detuning (with active resonance control) must be less than 30 Hz. TUNER DESIGN Schematics of the tuner design are shown in Figure 1. The coarse tuner is a slim blade tuner (with a 1:20 ratio) with a design that is very close to the design of the tuner developed by the INFN for EuXFEL 3rd harmonics cryomodules [2]. Figure 1: Schematic of the 3.9GHz SRF cavity tuner. Table 1. 3.9 GHz SRF Cavity and Tuner Parameters The major modification of this tuner design introduced by FNAL is adding a fast/fine tuner. Two piezo-stacks have been installed between the slow/blade tuner and the ring welded to the He-vessel. Safety rods have been designed between the cavity end flange and main lever of the tuner. These safety rods protect the cavity during transportation and from non-elastic deformation during cavity/helium vessel system pressure tests. Set-screws and special washers were included to prevent loosening of the assembly screws during warmup and cooldown cycles [4]. The 3.9 GHz SRF cavity tuner will include the same active components (electromechanical actuators and piezostacks) as used on the 1.3 GHz cavity tuner. [4,5]. The electromechanical actuator LVA 52-LCLS II-UHVCX1 built by Phytron per FNAL specification was found to RF frequency 3900 MHz Operating temperature 2 K Total voltage available (16 cavities) 80 MV Average operating gradient 14.5 MV/m Average Q 0 2.0×10 Cavity length (L) 0.346 M R /Q (r /Q ) 750 (2168) Ω (Ω/m) Geometry constant (G ) 275 Ω Longitudinal Cavity Stiffness 5.4 kN/mm Cavity Tuning Sensitivity 2.3MHz mm Coarse (slow) tuner range 750 kHz Fine (fast) tuner range ~1 kHz HOM damped Q value (monopole and dipole) ≤10 6 Lorentz detuning ≤0.6 Hz/(MV/m) Number of cryomodules 2 Number of cavities per CM 8 Cavity alignment requirements (RMS) 0.5 mm Peak detune (with piezo tuner control) 30 Hz Required cavity field amplitude stability 0.01 % (rms) Required cavity field phase stability 0.01 deg (rms) Q ext 2.2×10 Half bandwidth of the cavity 90 Hz Active length of 9 cells 345.96 mm RF beam power per cavity (@300 μ A load) 1.5 kW RF power needed per cavity 1 kW Cavity dynamic load 17 W ____________ *This manuscript has been authorized by Fermi Research Alliance, LLC under Contract N. DE-AC02-07CH11359 with U.S. Department of Energy. #pischaln@fnal.gov MOPOB32 Proceedings of NAPAC2016, Chicago, IL, USA ISBN 978-3-95450-180-9 140 Co py rig ht © 20 16 CC -B Y3. 0 an d by th e re sp ec tiv e au th or s 7: Accelerator Technology Main Systems T07 Superconducting RF be reliable and the rad-hard unit satisfied the requirements of the 3.9 GHz cryomodule for the LCLS II Project [5,6]. The stepper motor will be operated at “full-step” mode and can provide frequency setting resolution at 10-15 Hz/step. The encapsulated piezo P-844K075 will be used for 3.9 GHz fine tuner. It ismade from two 10*10*18mm PICMA butted piezo-stacks built by Physik Instrumente (PI) per FNAL specs for the 1.3 GHz cavity fast tuner. Design details, comprehensive accelerated lifetime tests and radiation hardness tests are described in other papers [5]. Capsulation (and internal preload) of the piezo-stack inside the capsule and the ceramic balls installed between the capsule and tuner frame will minimize shearing forces on the piezo-ceramics and will significantly diminish the problems experienced in the previous version of the slim blade tuner [7]. Two adjustment screws (one for each capsule mounted on the flange welded to He vessel) help to uniformly preload piezo-stacks during assembly. Maximum required slow tuner frequency is 750 kHz (Table 1). The cavity needs to be stretched to ~ 0.3mm. The forces generated by cavity and applied to each piezo-stack will in this case be less than 1kN. With internal preload on the piezo at ~800N, maximum forces on the piezo will not exceed 50% of blocking forces (4kN). Expected stroke from the P-844K075 piezo-capsule at cryogenic temperature is 5-10um. This stroke will provide fine/fast tuning range on the cavity of ~ 12-24 kHz at nominal piezo voltage V=100V. To control the cavity frequency within 30 Hz, detuning voltage applied to the piezo needs to be in the range of 100’s of mV. Figure 2: Prototype Tuner, assembled on the “plastic” model of 3.9 GHz dressed cavity/helium vessel system. For integration purposes the first prototype tuner has been installed on the 3.9 GHz dressed cavity plastic model. (Figure 2). There were no interferences with other components of the cavity. TUNER “WARM” TEST RESULTS To conduct prototype tuner “warm” tests a simple setup “cavity-He vessel mock-up” was built. (Figure 3). On the “mock-up” stand stiffness of the cavity /He-Vessel bellow the system is imitated by a combination of spring washers. Tests were conducted when tuner acted against spring with different stiffness. Figure 3: Prototype Tuner, assembled on the “cavity/He vessel mock-up” stand. Results of the test of the slow/blade tuner are presented in Figure 5. A stepper motor was run in full step mode with 10,000 steps for one spindle rotation (1 mm thread). The slow tuner pushed against a 5.4 kN spring through two piezo-capsules. The displacement of the plate installed between the tuner and the spring replicating cavity were measured by a dial indicator and laser displacement system. (Figure 3). Figure 4: Slow Tuner stroke vs actuator spindle turns. Red – slow/blade tuner operated against spring with k~5.4 kN (expected stiffness of the cavity/He-vessel system). Bluetuner operated against “zero” load. From the two curves presented on the Figure 4 (redwith load (5.4kN spring) and blue with no-load) the stiffness of the blade tuner can be estimated at ktuner~ 27 kN/mm. The coarse (blade) tuner ration changed from ~1:20 (when operated against 5.4 kN load) to ~1:14 when the tuner operated against zero load. From the “5.4 kN load” curve, the slow tuner resolution is estimated to be ~5nm/step or 10-12Hz/step. The results of the piezo tuner test are presented in Figure 5. The piezo test was performed when the tuner was stretched on~150um (three turns on the spindle). At this point the tuner was loaded at ~ 1kN. During the piezo test the voltage on the piezo did not exceed 60V. The maximum stroke (red curve) from the piezo tuner was 12um. Figure 5 also presents the hysteresis curve (blue) of the “free-standing” piezo-capsule. These are ISBN 978-3-95450-180-9 Proceedings of NAPAC2016, Chicago, IL, USA MOPOB32 7: Accelerator Technology Main Systems T07 Superconducting RF 141 Co py rig ht © 20 16 CC -B Y3. 0 an d by th e re sp ec tiv e au th or s measurements of the maximum stroke of the piezo (~18um at V=60V) when stiffness of the blade tuner and “mockup” setup are taken out of consideration. The results of the tuner piezo test presented in Figure 6 allowed estimating efficiency of the tuner system. The piezo efficiency (portion of the piezo stroke that will compress cavity) is near 67%. One of the methods of increasing efficiency is to change the amount of the blades in the region where the piezo is attached to the blade tuner. At the same time the expected piezo-tuner range at cryogenic temperature (even with 67% efficiency) will be near 5um or 10-12 kHz. This is ten times large than the technical requirements of the piezo-tuner (Table 1). Figure 5. Fast (piezo) tuner stroke vs voltage applied to piezo (red). Blue –stroke of “free standing piezo-capsule” vs voltage applied to piezo-stack. CONCLUSION A prototype compact tuner for a LCLS II project 3.9 GHz SRF cavity was designed, built, installed into a “cavity/He vessel mock-up” stand and then tested. The tuner exhibited the expected stiffness, range and resolution for both slow and fast tuning mechanisms. Tuner parameters (as measured warm) met or exceeded technical requirement specifications. The next step is to install the prototype tuner on the dressed cavity for cold tests as a part of an extended dressed cavity design verification program.
The design of the compact tuner for 1.3 GHz 9-cell elliptical cavity will be presented. This compact tuner is designed for future accelerators that will operate in CW and pulsed RF-power modes. The major design features include highly reliable active components (electromechanical actuators and piezo-actuators) and the ability to replace tuner active components through designated ports in the cryomodule vacuum vessel. Results of tuner testing with cold cavity will also be presented.
Fermilab is developing 5-cell elliptical 650 MHz β=0.6 and β=0.9 cavities for project X [1]. A compact fast/slow lever tuner intended for both types of cavities has been developed for final tuning of the resonance frequency of the cavity after cooling down and to compensate the resonance frequency variations of the cavity during operation coming from liquid helium pressure fluctuations. The updated helium vessel (presented at this conference) is equipped with the tuner located at one of the end of the cavity. The tuner design and results of ANSYS analysis of their properties are presented.
The proposed design of the 3 GeV Project X Superconducting Linac employs two families of 650 MHz 5-cell elliptical cavities with different β. The βG =0.61 will cover the 177-467 MeV range. To cover the range between 467-3000 MeV two versions of 650 MHz 5-cell elliptical cavities with βG =0.9 and βG =0.92 has been proposed. The low beam current for CW operation of the Project X requires cavities to operate at a high loaded Q and thus, narrow bandwidth. Therefore it requires the optimal mechanical design of cavities to minimize the sensitivity to microphonics.
The design of 5-cell elliptical 650 MHz β=0.92 cavities to accelerate H- beam of 1 mA average current in the range 467-3000 MeV for the Project X Linac is currently under development at Fermilab [1]. The low beam current enables cavities to operate with high loaded Q’s and low bandwidth, making them very sensitive to microphonics. Mechanical vibrations and the Lorentz force can drive cavities off resonance during operation; therefore the proper design of the tuning system is very important part of cavity mechanical design. In this paper we review the design, performance, operation, reliability and cost of fast and slow tuners for 1.3 GHz elliptical cavities. We also present a design of the slow and fast tuners for 650 MHz β=0.9 cavities based on this experience. The helium vessel (HV) in the new design is equipped with the tuners located at the end of the cavity instead of the initially proposed blade tuner located in the middle. We will present the results of ANSYS analyses of mechanical properties of tuners.
The design of 5-cell elliptical 650 MHz β=0.9 cavities to accelerate 1 mA at an average H-beam current in the range 467-3000 MeV for Project X Linac [1] is under development at Fermilab. The design of a dressed cavity has been mechanically optimized by minimizing df/dP, the sensitivity to microphonics detuning due to fluctuations in helium pressure. We will present the results of COMSOL multiphysics simulations of mechanical resonances and cavity tuneability. We will also present the mechanical design of slow and fast tuners and ANSYS analysis of their properties.
Fermilab is developing 325 MHz SRF spoke cavities for the proposed Project X. A compact fast/slow tuner has been developed for final tuning of the resonance frequency of the cavity after cooling down to operating temperature and to compensate microphonics and Lorentz force detuning [2]. The modified tuner design and results of 4.5K tests of the first prototype are presented. The performance of lever tuners for the SSR1 spoke resonator prototype has been measured during recent CW and pulsed tests in the Fermilab SCTF. The tuner met or exceeded all design goals and has been used to successfully: (1) Bring the cold cavity to the operating frequency; (2) Compensate for dynamic Lorentz force detuning; and (3) Compensate for frequency detuning of the cavity due to changes in the He bath pressure.
As field emission is gradually under control in recent SRF activities, cavity performance is limited by hard quench in most cases. Surface defect has been identified as one of main sources caused cavity quench, scattering cavity accelerating gradient from 12MV/m to 40MV/m. Laser re-melting technique is able to re-shape sharp pit rim to be a flat and smooth surface. In Fermilab, a sophisticated laser repair system has been developed for 1.3GHz quench-limited cavities. A pit in a 1.3GHz singlecell cavity was re-melted by high power laser pulse, after the laser processing the cavity took 30µm light Electropolishing. The cavity gradient achieved 39MV/m in initial run; after another 30µm Electropolishing, it reached 40MV/m. An improved laser repair system, which is able to re-melt surface defects in one meter-long 9-cell SRF cavity, has been developed at Fermilab after the success on single-cell cavity. It successfully re-melted a pit in 9-cell SRF cavity TB9ACC017.
In an exchange of technology agreement, Deutsches Elektron-Synchrotron (DESY) Laboratory in Hamburg Germany has provided a 1.3 GHz cryomodule "kit" to Fermilab. The cryomodule components (qualified dressed cavities, cold mass parts, vacuum vessel, etc.) sent from Germany in pieces were assembled at Fermilab's Cryomodule Assembly Facility (CAF). The cavity string was assembled at CAF-MP9 Class 10 cleanroom and then transported to CAF-ICB cold mass assembly area via a flatbed air ride truck. Finite Element Analysis (FEA) studies were implemented to define location of instrumentation for initial coldmass transport, providing modal frequencies and shapes. Subsequently, the fully assembled cryomodule was transported to the SRF Accelerator Test Facility at New Muon Lab (NML). Internal geophones (velocity sensors) were attached during the coldmass assembly for transport (warm) and operational (cold) measurements. A description of the isolation system that maintained alignment during transport and protected fragile components is provided. Shock and vibration measurement results of each transport and modal analysis are discussed.
The Fermilab High Intensity Neutrino Source (HINS) linac R&D program is building a pulsed 30 MeV superconducting H- linac. The linac incorporates superconducting solenoids, high power RF vector modulators and superconducting spoke-type accelerating cavities starting at 10 MeV. This will be the first application and demonstration of any of these technologies in a low-energy, high-intensity proton/H- linac. The HINS effort is relevant to a high intensity, superconducting H- linac that might serve the next generation of neutrino physics and muon storage ring/collider experiments. In this paper we present the RF design, the mechanical design, the fabrication, the chemistry and testing of the first two SSR1 (Single Spoke Resonator type-1) prototype cavities. These cavities operate at 325 MHz with β=0.21. The design and testing of the input coupler and the tuning mechanism are also discussed.
Fermilab is developing 325 MHz SRF spoke cavities for the proposed High Intensity Neutrino Source. A compact fast/slow tuner has been developed to control Lorentz force detuning and compensate for liquid Helium pressure fluctuations. The tuner design and results of warm tests of the first prototype are presented.
Fermilab is developing 325 MHz SRF spoke cavities for the proposed High Intensity Neutrino Source. A compact fast/slow tuner has been developed to control Lorentz force detuning and compensate for liquid Helium pressure fluctuations. The tuner design and results of warm tests of the first prototype are presented.
The Fermilab High Intensity Neutrino Source (HINS) linac R&D program is building a pulsed 30 MeV superconducting H- linac. The linac incorporates superconducting solenoids, high power RF vector modulators and superconducting spoke-type accelerating cavities starting at 10 MeV. This will be the first application and demonstration of any of these technologies in a low-energy, high-intensity proton/H- linac. The HINS effort is relevant to a high intensity, superconducting H- linac that might serve the next generation of neutrino physics and muon storage ring/collider experiments. In this paper we present the RF design, the mechanical design, the fabrication, the chemistry and testing of the first two SSR1 (Single Spoke Resonator type-1) prototype cavities that were built. These cavities operate at 325 MHz with {beta} = 0.21. The design and testing of the input coupler and the tuning mechanism are also discussed.