Author: Duval, P.
Paper Title Page
MOAPL01 The Control System for the Linear Accelerator at the European XFEL: Status and First Experiences 1
 
  • T. Wilksen, A. Aghababyan, R. Bacher, P.K. Bartkiewicz, C. Behrens, T. Delfs, P. Duval, L. Fröhlich, W. Gerhardt, C. Gindler, O. Hensler, K. Hinsch, J.M. Jäger, R. Kammering, S. Karstensen, H. Kay, H. Kay, V. Kocharyan, A. Labudda, T. Limberg, S.M. Meykopff, A. Petrosyan, G. Petrosyan, L.P. Petrosyan, V. Petrosyan, P. Pototzki, K.R. Rehlich, G. Schlesselmann, E. Sombrowski, M. Staack, J. Szczesny, M. Walla, J. Wilgen, H. Wu
    DESY, Hamburg, Germany
  
  The European XFEL (E-XFEL) is a 3.4 km long X-ray Free-Electron Laser facility and consists of a superconducting, linear accelerator with initially three undulator beam lines. The construction and installation of the E-XFEL is being completed this year and commissioning is well underway. First photon beams are expected to be available for early users in the second half of 2017. This paper will focus on the control system parts for the linear accelerator with its more than 7 million parameters and highlight briefly its design and implementation. Namely the hardware framework based on the MicroTCA.4 standard, testing software concepts and components at real and virtual accelerator facilities and a well-established method for integrating high-level controls into the middle layer through a shot-synchronized data acquisition allowed for a rapid deployment and commissioning of the accelerator. Status and experiences from a technical and an operational point-of-view will be presented.  
slides icon Slides MOAPL01 [6.198 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-ICALEPCS2017-MOAPL01  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPHA189 Automating Operation Statistics at PETRA-3 876
 
  • P. Duval, R. Bacher, H. Ehrlichmann, D. Haupt, M. Lomperski
    DESY, Hamburg, Germany
  • J. Bobnar
    Cosylab, Ljubljana, Slovenia
  
  The quoted machine availability of a particle accelerator over some time range is usually hand-generated by a machine coordinator, who pores over archived operations parameters and logbook entries for the time period in question. When the machine is deemed unavailable for operations, 'blame' is typically assigned to one or more machine sub-systems. With a 'perfect' representation of all possible machine states and all possible fatal alarms it is possible to calculate machine availability and assign blame automatically and thereby remove any bias and uncertainty that might creep in when a human is involved. Any system which attempts to do this must nevertheless recognize the de-facto impossibility of achieving perfection and allow for 'corrections' by a machine coordinator. Such a system for automated availability statistics was recently presented* and we now report on results and improvements following a half year in operation at PETRA-3 and its accelerator chain.
* Duval, Lomperski, Ehrlichmann, and Bobar, "Automated Availability Statistics", Proceedings PCaPAC 2016. 		 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-ICALEPCS2017-TUPHA189  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)