Jaroslaw Szewinski

FPGA-based implementation of a cavity field controller for FLASH and X-FEL

The subject of this paper is the design and construction of a new generation of superconducting cavity accelerator measurement and control system. The old system is based on a single digital signal processor (DSP). The new system uses a large programmable array circuit (FPGA) instead, with a multi-gigabit optical link. Both systems now work in parallel in the Free Electron Laser in Hamburg (FLASH). The differences between the systems are shown, based on the measurement results of the working machine. The major advantage of the new system is a bigger area of stability of the machine control loop.

Software layer for FPGA-based TESLA cavity control system

The paper describes design and practical realization of software for laboratory purposes to control FPGA-based photonic and electronic equipment. There is presented a universal solution for all relevant devices with FPGA chips and gigabit optical links. The paper describes architecture of the software layers and program solutions of hardware communication based on Internal Interface (II) technology. Such a solution was used for superconducting Cavity Controller and Simulator (SIMCON) for the TESLA experiment in DESY (Hamburg). A number of practical examples of the software solutions for the SIMCON system were given in this paper.

FPGA-based multichannel optical concentrator SIMCON 4.0 for TESLA cavities LLRF control system

The paper presents an idea, design and realization of a gigabit, optoelectronic synchronous massive data concentrator for the LLRF control system for FLASH and XFEL superconducting accelerators and lasers. The design bases on a central, large, programmable FPGA VirtexIIPro circuit by Xilinx and on eight commercial optoelectronic transceivers. There were implemented peripheral devices for embedded PowerPC block like: memory and Ethernet. The SIMCON 4.0 module was realized as a single, standard EURO-6HE board with VXI/VME-bus. Hardware implementation was described for the most important functional blocks. Construction solutions were presented.

Standardized Solution for Management Controller for MTCA.4

The Micro Telecommunications Computing Architecture (MTCA) standard is a modern platform, that is gaining popularity in the area of High Energy Physics (HEP) experiments. The standard provides extensive management, monitoring and diagnostics functionality. The hardware management is based on the Intelligent Platform Management Interface (IPMI), that was initially developed for management and monitoring of complex computers operation. The original IPMI specification was extended and new functions required for MTCA hardware management, were added. The Module Management Controller (MMC) is required on each Advanced Mezzanine Card installed in MTCA chassis. The Rear Transition Modules (RTMs) require Rear transition module Management Controller (RMC) that is specified in MTCA.4 extension specification. The commercially available implementations of MMC and RMC are expensive and do not provide the whole functionality that is required by specific HEP applications. Therefore, many research centres and commercial companies work on their own implementation of AMC or RTM controllers. The available implementations suffer because of lack of a standard and interoperability problems. The Authors developed a unified solution of management controller fully compliant to AMC and MTCA.4 standards. The MMC v1.00 solution is dedicated for management of AMC and RTM modules. The MMC v1.00 is based on Atmel ATxmega MCU and can be fully customized by user or used as a drop-in-module without any modifications. The paper discusses the functionality of the MMC v1.00 solution. The implementation was verified with developed evaluation kits for AMC and RTM cards.

Implementation of adaptive feed-forward algorithm on embedded PowerPC405 processor for FLASH accelerator

This paper compares algorithms used for control of super-conducting cavities in the FLASH accelerator and laser in DESY Research Center (Hamburg, Germany), and describes features of the adaptive feed-forward algorithm. Described solution was implemented in the PowerPC 405 processor embedded in the Xilinx Virtex II Pro FPGA circuit.

MatLab script to C code converter for embedded processors of FLASH LLRF control system

The low level RF control system (LLRF) of FEL serves for stabilization of the electromagnetic (EM) field in the superconducting niobium, resonant, microwave cavities and for controlling high power (MW) klystron. LLRF system of FLASH accelerator bases on FPGA technology and embedded microprocessors. Basic and auxiliary functions of the systems are listed as well as used algorithms for superconductive cavity parameters identification. These algorithms were prepared originally in Matlab. The main part of the paper presents implementation of the cavity parameters identification algorithm in a PowerPC processor embedded in the FPGA circuit VirtexIIPro. A construction of a very compact Matlab script converter to C code was presented, referred to as M2C. The application is designed specifically for embedded systems of very confined resources. The generated code is optimized for the weight. The code should be transferable between different hardware platforms. The converter generates a code for Linux and for stand-alone applications. Functional structure of the program was described and the way it is acting. FLEX and BIZON tools were used for construction of the converter. The paper concludes with an example of the M2C application to convert a complex identification algorithm for superconductive cavities in FLASH laser.

Multi-cavity complex controller with vector simulator for TESLA technology linear accelerator

A digital control, as the main part of the Low Level RF system, for superconducting cavities of a linear accelerator is presented. The FPGA based controller, supported by MATLAB system, was developed to investigate a novel firmware implementation. The complex control algorithm based on the non-linear system identification is the proposal verified by the preliminary experimental results. The general idea is implemented as the Multi-Cavity Complex Controller (MCC) and is still under development. The FPGA based controller executes procedure according to the prearranged control tables: Feed-Forward, Set-Point and Corrector unit, to fulfill the required cavity performance: driving in the resonance during filling and field stabilization for the flattop range. Adaptive control algorithm is applied for the feed-forward and feedback modes. The vector Simulator table has been introduced for an efficient verification of the FPGA controller structure. Experimental results of the internal simulation, are presented for a cavity representative condition.

Scintillators For High-Temperature Plasma Diagnostics

Gamma-ray measurements during deuterium-tritium cam paigns require detectors characterized by a good energy resolution, a relati v ly high detection efficiency for a few MeV gamma-rays and a fast response time. Scintillat ors proposed for monitoring gammarays in high temperature plasma experiments fulfill most of these requirements, in addition they are rather resistant to neutron damage in comp arison to, e.g., germanium detectors. Intense fast neutron fluxes expected during DT camp aigns reduce considered materials to oxygen-free crystals. CeBr3 scintillators were investigated with the aim to us e them for gamma-ray diagnostics of fusion plasma. Basic properties like ght output, energy resolution, decay time and full energy peak detection efficiency were m asured using a spectrometry photomultiplier. The response of CeBr 3 was studied as a function of crystal volume between 0.5 and 350 cc. A comparison with well-know and used as reference alkali halide scintillators is presented. Measurements were performed at the National Centre for Nuclear Research (NCBJ) using gamma-ray sources with energies up to a few M eV, in particular PuBe and PuC gamma emitters. Natural background radiation spectra allowed to est imate the intrinsic activity of CeBr3 and LaBr 3:Ce scintillators.

Software for development and communication with FPGA based hardware

This document describes aspects of creating software for communication with hardware, especially with FPGA based systems. Features unique for FPGA systems are compared with features of the traditional electronic systems. There are discussed topics like hardware interface definition, or address space description. Connecting client application is shown with Matlab as an example. Solutions with FPGA and embedded processors are discussed.

DOOCS and MATLAB control environment for FPGA based cavity simulator and controller in TESLA experiment

FPGA based cavity simulator and controller is the next generation control system dedicated for high performance, low latency control algorithm development and implementation. The usage of FPGA technology gives users possibility to create many devices on one board and easy exchange, modify or improve VHDL programmed algorithms. In order to provide the full functionality of the system to the user, and meet the requirements of flexibility and extensibility, an appropriate control software is needed. This paper describes the idea and implementation of control environment dedicated for FPGA based devices. As an example of implementation, two control environments have been implemented; the laboratory software based on Matlab, and the application for accelerator operation using DOOCS environment.