Microwave mixers, Second Edition, by Stephen A. Maas, Artech House, Boston, London 1993. 375 pages, 165 pages, 8 tables, 161 references, index. Price: US $76.00 and £59.00 in Europe. ISBN 0-98006-605-1

https://link.aps.org/pdf/10.1103/PhysRevAccelBeams.21.032003

Real-time cavity simulator-based low-level radio-frequency test bench and applications for accelerators

Linear accelerators, like the Free-electron LASer in Hamburg (FLASH) or the European X-Ray Free Electron Laser (E-XFEL) take advantage of the digital Low Level Radio Frequency (LLRF) system to control the phase and amplitude of an electromagnetic field inside superconducting cavities. The real-time control LLRF system, processing data within a few microseconds, has to fulfil performance requirements and provide comprehensive monitoring and diagnostics. The AMC-based controller (DAMC-TCK7) board was developed as a general purpose high-performance low-latency data processing unit designed according to the PICMG MTCA.4 spec. The module provides the processing power, data memory, communication links, reference clock, trigger and interlock signals that are required in modern LLRF control systems. The module was originally designed as a cavity field stabilizing controller for standing-wave linear accelerators. However, the application of the board is much wider because it is a general purpose data processing module suitable for systems requiring low latency and high-speed digital signal processing. According to authors’ knowledge this is the first MTCA.4 module offering 12.5 Gbps links, unified Zone 3 connectivity and advanced Module Management Controller proposed by DESY. The DAMC-TCK7 card was used as a hardware template for the development of the other AMC modules of the XFEL accelerator’s LLRF system. This paper discusses the requirements for the digital real-time data processing module, presents the laboratory performance evaluation and verification in Cryo-Module Test Bench (CMTB) at DESY.

/pdf/high-speed-data-processing-module-for-llrf-4kis9qdj0f.pdf

High-speed data processing module for LLRF

The superconducting accelerator of the European XFEL consists of the injector part and the main linac. The injector includes one 1.3 GHz accelerator module and one 3.9 GHz third-harmonic module, while the main linac will consist of 100 accelerator modules, operating at an average design gradient of 23.6 MV/m. The fabrication and surface treatment by industry as well as RF acceptance tests of the required 808 superconducting 1.3 GHz cavities are close to an end by the time of SRF 2015. The accelerator module assembly, testing and installation in the tunnel is in full swing. First steps of commissioning have been made. The status and results of cavity and module RF tests at 1.3 GHz and 3.9 GHz are presented. INTRODUCTION The 17.5 GeV SRF linac for the European XFEL is currently under construction by a consortium consisting of several European institutes [1]. An enhanced cryomodule production rate of 1.25 eight-cavity-module per week since beginning of 2015 requires an average vertical acceptance testing rate and delivery of at least ten cavities per week. The average cryomodule testing rate has to match this production rate. Testing of both, individual cavities and cryomodules, is performed in a dedicated test facility at DESY (AMTF) [2,3,4,5]. As of July 31, 2015 approximately 740 of the 800 series EUXFEL TESLA-type 1.3 GHz SRF cavities have been produced, and have each undergone at least one vertical acceptance test at AMTF. As of September 10, 2015 ~57 of the 102 EU-XFEL cryomodules (see below) have been tested at AMTF. Vertical and module testing is performed by a team from IFJ-PAN Krakow as an in-kind contribution. The installation of cryomodules and first steps of commission for the injector as well as for the main linac are in full swing. XFEL CAVITIES AND VERTICAL ACCEPTANCE TEST AT AMTF Production Overview Series production of the 1.3 GHz TESLA cavities is equally divided between E. Zanon Spa. (EZ), Italy, and Research Instruments GmbH (RI), Germany. Production includes both mechanical fabrication and surface preparation [6] with the required extensive documentation [7]. Details about the RF measurements for quality assurance during the cavity production and the devices used for frequency measurement and tuning can be found in [8,9,10]. The 800 series cavities required for XFEL (400 per vendor) are delivered complete with a helium tank, ready for vertical testing in AMTF (Fig.1) at DESY. Each vendor also produces an additional 12 cavities without helium tank for the ILC-HiGrade programme [11], which are used as a quality control tool as well as for further R&D. For 8 of these 24 cavities a subsequent assembly of the He-tank will be made. Both vendors must exactly follow well-defined specifications for the mechanical fabrication and surface treatments, but no RF performance guarantee is given. The surface preparation at both vendors starts with a bulk EP followed by 800° annealing, but for the final surface treatment two alternative recipes are in use: EZ applies a final chemical surface removal (“Flash-BCP”); RI applies a final electrochemical surface removal (EP). All cavities are fully equipped with their HOM antennas, pick-up probe and a High-Q input coupler antenna with fixed coupling. The procedures before and after the vertical acceptance test at 2K are described in [12]. At the time of SRF 2015 the mechanical cavity production is nearly finished. The tested ~740 cavities clearly demonstrate that the chosen scheme for mechanical production and surface preparation is successful at both vendors. Due to pores out of DESY specification found in the longitudinal welds of the Helium service pipe made of Titanium about 750 He-tanks had to be modified, partially on He tanks already welded onto cavities. This major effort including the necessary qualification steps has been performed successfully over the last two years and is described in detail in [13]. Vertical Testing Rates In order to achieve the desired testing rate of at least eight to ten cavities per week, the vertical acceptance tests are made using two independent test systems, each consisting of an independent bath cryostat and RF test stand. Each test cryostat accepts an “insert” which supports up to four cavities, greatly increasing the efficiency of cool-down / warm-up cycles. The test infrastructure at DESY (Fig.1) has been in full operation since October 2013 and has achieved a stable average of about 42 vertical tests per month (Fig. 2). All vertical acceptance tests of the 824 cavities will be finished within the current project schedule (end of 2015). Each vertical test is categorized according to well-defined “test reasons”. Depending on the result a categorized “decision” is taken and documented in the cavity and cryomodule managing system [14] of the AMTF as well as in the XFEL cavity data base [15]. Cavities without non-conformities and with acceptable performance usual have only one vertical acceptance test (“as received”) after which they receive the decision “send to string ___________________________________________ #detlef.reschke@desy.de MOAA02 Proceedings of SRF2015, Whistler, BC, Canada ISBN 978-3-95450-178-6 14 C op yr ig ht © 20 15 C C -B Y3. 0 an d by th e re sp ec tiv e au th or s Projects/Facilities progress A03-Funded facilities progress report assembly”. In case of non-conformities (e.g. insufficient cavity performance, RF problem, vacuum leak, mechanical deviation, etc.) the cavity is retested, retreated or sent back to the vendor resulting in additional vertical tests. A detailed analysis of the number and causes of conducted retests and retreatments is given in [16]. Figure 1: Preparation area for vertical testing (top) and vertical test installations (bottom) at AMTF. The vertical acceptance tests follow a standardised procedure, which includes the measurement of the unloaded Q-value (Q0) versus the accelerating gradient Eacc at 2 K, as well as the frequencies of the fundamental modes. For each point of the Q0(Eacc)-curve, X-rays are measured inside the concrete shielding above and below the cryostat. No general administrative gradient limit is applied. The average statistical measurement error is calculated to be ~4 % for Eacc and ~7 % for Q0 [17]. In general the systematic error is about ~10% for Eacc and up to ~20 % for Q0. In few cases the test-to-test comparison showed larger deviations than the above estimated errors. Within the tight production and test schedule an additional retest was performed for a small fraction of cavities for clarification, but often no obvious reason for the observed difference could be identified. In addition to the Q0(Eacc) curves many cavities have had the Higher Order Mode frequencies of the TE111, TM110 and TM011 modes measured [18], depending on the fabrication process. Figure 2: Trend of the vertical test rate. After a successfully completed test, selected key data are transferred to the XFEL Cavity Data Base [15,19,20], which forms the basis of the analyses report here. Definition of “Usable Gradient” and Acceptance Criteria Although all cavities are tested to their maximum achievable gradient (Eacc,max), of greater importance for accelerator operation is the “Usable Gradient” (Eacc,us), which takes Q0 as well as field-emission performance into account. It is defined as the lowest value of:  quench gradient (quench limited);  gradient at which Q0 drops below 10 (Q0 limited);  gradient at which either X-ray detector exceeds the threshold (field-emission limited). For the field-emission limit, the acceptable X-ray thresholds are set to 0.01 mGy/min and 0.12 mGy/min for the top and bottom detector respectively. The threshold 0.01 mGy/min is based on experience from the cavity testing for FLASH. The higher limit for the lower detector is a geometrical effect. At the beginning of production, the criterion for acceptance for module assembly was specified as Eacc,us ≥ 26 MV/m, chosen to give a margin of ~10% compared to the required average design operation gradient (23.6 MV/m at Q0 ≥ 10). Based on an analysis of about 270 cavities tested up to May 2014, including the necessary retreatments and retests, the acceptance criteria was reduced to Eacc,us ≥ 20 MV/m, in order to optimise the number of vertical tests while still maintaining an average module gradient of 23.6 MV/m. Cavities with Eacc,us < 20 MV/m are considered for further processing or re-treatment. The exact nature of the handling of low-performance cavities is judged on a case by case basis. As there are no vendor performance guarantees, retreatments are in general the responsibility of DESY. Nevertheless both vendors have agreed to perform numerous retreatments depending on the case. Proceedings of SRF2015, Whistler, BC, Canada MOAA02 Projects/Facilities progress A03-Funded facilities progress report ISBN 978-3-95450-178-6 15 C op yr ig ht © 20 15 C C -B Y3. 0 an d by th e re sp ec tiv e au th or s VERTICAL TEST RESULTS ‘As received’ from Vendor (1 Acceptance Test) Figure 3 shows histograms and yield curves for the vertical test performance for both maximum and usable gradient, “as received” from the vendors. Compared to [17] the number of vertical acceptance is nearly doubled. The present analysis is based on 671 vertical tests (EZ: 359; RI: 312). Table 1 summarises the average of the distributions shown in Fig. 3. The average usable gradients for both vendors are above the required operational gradient for XFEL. The usable gradient is reduced from the maximum performance by 3.7 MV/m on average, predominantly due to the Q0-value drops below 10. The effect can be seen in Fig. 3 as an increase (top to bottom plot) in the numbers of cavities with performance less than 30 MV/m. For both vendors ~15% of the cavity tests are caused by a necessary retreatment due to field emission, which is about 5% less and a clear improvement compared to the analysis of mid 2014 [17]. There is also a statistically significant difference

Recent Progress with EU-XFEL

This paper presents a detection system for high-accuracy amplitude and phase measurements on RF signals. It is typically used to set up the low-level RF system for particle accelerators. The proposed detector employs the principle of a heterodyne receiver and uses the digital IF architecture with a high oversampling rate to increase the SNR. A discrete Fourier transform-based quadrature demodulator is used to suppress the analog-to-digital converter (ADC) quantization noise and IF harmonics. The system has been evaluated at an IF of 1 MHz and in an RF range of 500-3000 MHz. A graphical user interface is developed to display and record the measured values. The detection accuracy achieved on 500-MHz signals in a 250-kHz bandwidth is better than ±0.0051% for the amplitude and better than ±0.0032 ° for the phase. Experimental results at different frequencies prove that the detection system can meet the demands of state-of-the-art RF control systems for modern particle accelerators.

High-Accuracy Amplitude and Phase Measurements for Low-Level RF Systems

The present configuration of an ILC main LINAC RF station has 26 nine cell cavities driven from one klystron. With the addition of waveguide power coupler monitors, 96 RF signals will be down-converted and processed. A down-converter chassis is being developed that contains 12 eight-channel analog modules and a single up- converter module. This chassis will first be deployed for testing a cryomodule composed of eight cavities located at New Muon Laboratory (NML) - Fermilab. Critical parts of the design for LLRF applications are identified and a detailed description of the circuit with various characteristic measurements is presented. The board is composed of an input band-pass filter centered at 1.3 GHz, followed by a mixer, which down-converts the cavity probe signal to a proposed 13 MHz intermediate frequency. Cables with 8 channels per connector and good isolation between channels are being used to interconnect each down-converter module with a digital board. As mixers, amplifiers and power splitters are the most sensitive parts for noise, nonlinearities and crosstalk issues, special attention is given to these parts in the design of the LO port multiplication and distribution.

/pdf/a-96-channel-receiver-for-the-ilcta-llrf-system-at-fermilab-1e0nmh4b1f.pdf

A 96 channel receiver for the ILCTA LLRF system at fermilab

The proposed RF distribution scheme for the two 15 km long ILC LINACs uses one klystron to feed 26 superconducting RF cavities operating at 1.3 GHz. For a precise control of the vector sum of the signals coming from the SC cavities, the control system needs a high-performance, low-cost, reliable and modular multichannel receiver. At Fermilab we developed a 96-channel, 1.3 GHz analog/digital receiver for the ILC LINAC LLRF control system. In this paper we present a balanced design approach to the specifications of each receiver section, the design choices made to fulfill the goals and a description of the prototyped system. The design is tested by measuring standard performance parameters, such as noise figure, linearity and temperature sensitivity. Measurements show that the design meets the specifications and it is comparable to other similar systems developed at other laboratories, in terms of performance.

/pdf/design-and-evaluation-of-a-low-level-rf-control-system-30vd8pxjg6.pdf

Design and evaluation of a low-level RF control system analog/digital receiver for the ILC main Linacs

The field control of multiple superconducting RF cavities with a single Klystron, such as the proposed RF scheme for the ILC, requires high density (number of RF channels) signal processing hardware so that vector control may be implemented with minimum group delay. The MFC (Multichannel Field Control) module is a 33- channel, FPGA based down-conversion and signal processing board in a single VXI slot, with 4 channels of high speed DAC outputs. A 32-bit, 400MHz floating point DSP provides additional computational and control capability for calibration and implementation of more complex control algorithms. Multiple high speed serial transceivers on the front panel and the backplane bus allow a flexible architecture for inter-module real time data exchanges. An interface CPLD supports the VXI bus protocol for communication to a SlotO CPU, with Ethernet connections for remote in system programming of the FPGA and DSP as well as data acquisition.

/pdf/multichannel-vector-field-control-module-for-llrf-control-of-dxy61xvn13.pdf

Multichannel vector field control module for LLRF control of superconducting cavities

The 3.9 GHz cryomodule and RF system for the XFEL Injector is being assembled and delivered to the underground building in summer 2015, for the injector commissioning in Fall 2015. This contribution outlines the status of the activity and reports the preparation stages of the technical commissioning of the system.

Preparation of the 3.9 GHz System for the European XFEL Injector Commissioning

As part of the research and development towards the International Linear Collider (ILC), several test facilities have been developed at Fermilab This paper presents the latest Low Level RF (LLRF) results obtained with Capture Cavity II (CCII) at the ILC Test Accelerator (ILCTA) test facility The main focus will be on controls and RF operations using the SIMCON based LLRF system developed in DESY Details about hardware upgrades and future work will be discussed

U. Mavric

Papers

A 96 channel receiver for the ILCTA LLRF system at fermilab

Design and evaluation of a low-level RF control system analog/digital receiver for the ILC main Linacs

Multichannel vector field control module for LLRF control of superconducting cavities

Preparation of the 3.9 GHz System for the European XFEL Injector Commissioning

Capture cavity II results at FNAL