1 Construction of the European X-Ray Free-Electron Laser Facility Integration Challenges & Strategies
The Scientific Introduction 2 The European XFEL is a novel light source for fundamental science a mega microscope Surface physics Material science Geophysics Astrophysics Chemistry Molecular biology Medicine
The European XFEL Facility Introduction 3 Located in Northern Germany: Partly in Hamburg and partly in Schleswig-Holstein Stretching over 3.4 km 10 tunnels 5.6 km About 20 large underground and surface buildings Construction cost: 10 9
5 Main Elements The European XFEL Facility Introduction 1) Buildings & Tunnels > Housing 2) Technical Infrastructure > Supplies 3) Superconducting Linear Accelerator > Electron beam generation 4) Photon Beam Lines & Undulators > X-ray laser light generation 5) Experimental Instruments > Science 4
The Technology Introduction 5 The linear accelerator generates & accelerates electron bunches (17.5 GeV 35000 m e ) The high-energy electrons generate x-ray laser light when passing through undulators
The Science Introduction 6 The ultra short (10-50 fs), highly coherent and brilliant X-ray laser pulses allow: Structure determination Studying exotic states of matter (like in stellar gases) Femto-Chemistry > Movies (The light pulses are about 1000 x shorter & 10 10 x more brilliant than at latest synchrotrons)
The International Context Currently there are worldwide 3 large X-ray FEL projects: Introduction 7 LCLS LCLS (Linac Coherent Light Source), SLAC, USA -> 120 pulses/s (operating since 2010) SCSS (Spring-8 Compact SASE Source), Spring-8, Japan -> 60 pulses/s (operating since 2013) XFEL, DESY, Germany -> 30.000 pulses/s (Superconducting Linac) SCSS XFEL aims for first e - -beam in 2015 and ready for users in 2016 XFEL FLASH (at DESY) operating as user-facility since 2006 > Sets standards for generating VUV and soft X-ray laser light > Sets standards for the scientific exploitation of such radiation > 1/10 of XFEL FLASH
The International Context Currently there are worldwide 3 large X-ray FEL projects: LCLS (Linac Coherent Light Source), SLAC, USA -> 120 pulses/s (operating since 2010) Introduction 8 LCLS SACLA (Spring-8 Angstrom Compact free electron LAser), Spring- 8, Japan -> 60 pulses/s (operating since 2011) XFEL (DESY, Germany) -> 27 000 pulses/s (Superconducting Linac) Goals: first e - -beam in 2015 and ready for users in 2017 SACLA XFEL FLASH (at DESY) operating as User Facility since 2006 > 1/10 of XFEL, i.e. Technology Demonstrator FLASH
The Political Innovation Introduction 9 The primary political idea is to launch a new way in Europe for establishing large-scale science projects (ESFRI roadmap) 12 Countries have decided to build jointly a new international research facility in Europe It is owned by the contributing countries: Limited Liability Company (GmbH) Countries are shareholders Lasting rights & obligations Construction Cost 1G Funding via Cash + In-Kind contributions
Organisational Structure(s) Introduction 10 XFEL Company DESY Partner Institutes Accelerator Consortium Advisory Committees Shareholders Supervisory Authorities 51 Work Packages ( Groups)
Decentralised Project Introduction 11 20 Institutes from 12 Counties contribute in-kind Almost all project activities are joint activities of several institutes Not even the European XFEL Company or DESY There was a lot to learn and to accept by all partners Courtesy of Lars Hagge
The XFEL Construction 12 The Facility Construction Project fascinating and extremely demanding! A (systems) engineering challenge A (technical) management challenge
Organisational Start Situation ( 2006) Facility Construction Plan-approval-procedure and technology development led by DESY groups XFEL GmbH to be formed Collaboration for construction phase to be formed > Responsibilities No central technical coordination / systems integration group No project-wide design and technical management processes (& tools) Technical Technology demonstrated by FLASH ( 1/10 of XFEL ) Collective understandings & individual assumptions based on FLASH, other DESY accelerator facilities, DESY working style... Approved documents: plan-approval documents (incl. safety concept) & technical design report (TDR) No detailed description of Form, Fit and Function (HW & SW) 13
A Fundamental Problem Facility Construction 14 1. Machine layout / physics defines facility design 2. Built buildings constrain machine layout Difficult to solve Buildings & technical infrastructure must be build years before the machine becomes installed Machine layout by far not finalised and many interfaces are not yet defined
The Strategy Facility Construction 15 Organisational Set up a central Technical Coordination for steering the integration Set Up a WP / group for Information & Processes Support (WP40) Doing Set up Steering Means Understand and optimise workflows Manage Requirements & Interfaces Manage Overall Architecture & Changes Develop Installation Sequences Ensure 3C > Completeness, Consistency, Compliance
The XFEL Construction 16 Three Systems Engineering topics for today Steering Means Integration Planning Technical Reviews
Steering Means Facility Construction 17 Institutionalised Steering Means (basically) Quality Management: Structures and provides necessary decision criteria (defines how things shall be done) Reports: Reviews: Audits: Examine and monitor the general progress at regular intervals Examine/ensure readiness for/at crucial project phase transitions (e.g. Design -> Tendering/Fabrication -> Installation) Verify (on-site) compliance and effectiveness of quality management or other processes (target/actual comparison) Change Management: Handles the exceptional issues Issues Tracking Risk Management Schedule Follow-Up Handles pending issues & brings them up at regular intervals Identifies, monitors and helps to limit risks Monitors the schedule compliance and provides early warnings To be understood as main steering instruments
Steering Means Facility Construction 18 Non-Institutionalised Steering Means Personal Interactions: Feel the pulse within the collaboration, give personal acknowledgement and are a / the 'lubricant. Personal interactions are essential. They complement, but do not substitute institutionalised means. The bulk of problems and deviations should be handled via institutionalised processes!
The XFEL Construction 19 Three Systems Engineering topics for today Steering Means Integration Planning Technical Reviews
Integration Office (TC & WP-40) Facility Construction Organising the integration planning on the detailed technical level (incl. steering the integration meeting, steering the infrastructure requirements definition ) Verification & Validation, Refinement and Redaction of Architecture Baselines (e.g. described by Master Model, Inventories, BOMs ) Main vehicles are Master Model & Room Book 20 Processes & Tools developed and serviced by WP40 - "IPS"
Master Model Facility Construction 21 Complete 3D model of entire facility (116347 instancens injun 2011 Can be explored (= walked through) in real time Co-activity of WP40 and TC Courtesy of Lars Hagge
Master Model & Placeholder Concept Facility Construction 22 detailed design WPs create detailed designs as basis for fabrication integration model Placeholder = Maximum part dimensions = Envelope + Space for Tolerances, Tools Placeholders are (de-) coupling activities of WPs and Central Integration WPs must ensure detailed designs fit into placeholders TC creates integration model(s) to ensure that deliverables from different WPs integrate without collisions Courtesy of Stefan Suehl
The 3D-CAD Integration Process Facility Construction 23 Sub-System Designers CAD-QA Team (TC & WP40) CAD-QA Team (TC & WP40) Sub-System Coordinators iterate from here every two weeks create / update subsystem model integrate & release building model perform QA: collision checks agree on resolution strategy RF Ventilation Courtesy of Stefan Suehl
Products in Use and Data Flow iseries / Visview EDMS Facility Construction Solid Edge 24 ADT I-deas RoCAD Courtesy of Stefan Suehl
Exp. Hall Sections & Rooms Concept Facility Construction 25 I1(0-92) Functional View: Beam Line / Facility Sections > Rooms L2 (234-383) L1 (92-169) BC2 (383-467) BC1 (169-234) L3 (467-1689) TL (1901-2059) U1 (2683-2744) T3 (2428-2683) S2 (2171-2428) T1 (2039-2171) U2 (2979-3039) T5 (2744-2979) T6 (2549-3333) T7 (3150-3333) T8 (2824-3333) T5D (3039-3190) I1D (60-68) BC1D (227-232) BC2D (460-467) CL (1689-1901) TLD (2000-2131) T20 (2059-...) T2 (2059-2171) S1 (2171-2373) T4 (2373-2798) S3 (2798-2932) T4D (2932-3107) T9 (2587-3333) T10 (3067-3333)
Requirements & Interfaces Mgm. Facility Construction 26 WPs Excel Templates (easy-to-use) Requirements Database (with web-access) TC Process steered by TC
Infrastructure Coordination Infrastructure Coordination (inter alia) TGA Racks Optical Fibres Radiation Levels & Shielding e.g. Room Book Facility Construction 27 e.g. Racks Inventory e.g. Radiation Levels & Shielding Needs Processes & Tools developed and serviced by WP40 - "IPS"
The Pulsing-Star Concept Facility Construction 28 WPs (trades) provide information to TC (e.g. Req., 3D-Placeholders ) TC imports / assembles > checks > reports problems to WPs WPs (trades) resolve problems & provide updated information back to TC Req., 3D CAD WP/Trade B Req., 3D CAD report provide WP/Trade A report provide Req.-DB Master Model Processes e.g. Req. Consolidation 3D-CAD Integration report provide Req., 3D CAD WP/Trade N Courtesy of Stefan Suehl
Description of SW Systems Facility Construction Example: Low-Level Radio Frequency (LLRF) system > Concept Illustrations but not a thorough modelling 29
Architecture Descriptions Facility Construction 30 Form, Fit & Function Description (HW & SW) Form: Fit: Function: 3D Models, Drawings (various tools) Drawings (various tools) Documents & Presentations (MS Office) Documents & Presentations (MS Office) Physics Simulations (special tools) Became quite thorough for HW dominated systems, less for SW dominated systems
Facility Construction 31 Understand boundaries and required prerequisites (organisational and technical) Optimise the workflows w.r.t. the overall facility integration Parallelise as much as possible Ensure resources and logistics e.g. Installation Workflow in XTL
The XFEL Construction 32 Three Systems Engineering topics for today Steering Means Integration Planning Technical Reviews
Technical Reviews Facility Construction 33 The Challenge: How can TC / PM / WPs ensure project wide 3C, while not being deep experts in most project fields and while for most issues there are also no external experts, which are sufficiently familiar with the particularities) The Solution: > Utilise & Mobilise the project internal deep expertise > Channel it in a systematic process > Focus on steering the process
Technical Reviews - A Core Means Facility Construction 34 The Systematic Process: 1. Identify the Stakeholders and their Fields of Concern 2. Define the High-Level Use-Cases 3. Derive the Requirements & Interfaces 4. Develop the Design 5. Validate & Verify (after each stage) 6. Ensure the outcome is documented, accepted and known by all Stakeholders via their Collective Approval (in the EDMS) Mutual agreement is achieved and documented by means of creating a set of Mandatory Documents, which become (must be) approved by all stakeholders at the process-end
Technical Reviews Facility Construction 35 At XFEL reviews are therefore understood as a Process rather than an event Technical Reviews (components, sub-systems, overall) CDR Conceptual Design Review DR Design Review PRR Production (Tendering) Readiness Review IRR Installation Readiness Review
The XFEL Construction 36 Concluding Assessment
Concluding Assessment Conclusions 37 Very successful: The XFEL Technical Coordination for masterminding and steering the systems integration WP40 for Information Management & Processes Support Master Model, Room Book and other central inventories for describing Form & Fit (of course also all physics simulations!) Conceptual and Production Readiness Reviews for ensuring 3C Project Progress Reviews for following-up the overall progress Not so successful Reliability and safety engineering were rather felt as a burden than as helpful design instruments We didn t manage to set up a project wide Total Quality Management The formal Risk Management remained with the minimum necessary The architecture description of software dominated systems (structure, system borders, interfaces ) remained difficult
Concluding Assessment Where could a more of SE have helped? Conclusions 38 My personal opinion: Consequent Requirements Engineering combined with a thorough computer aided modelling (e.g. sysml) would have created a much more consistent understanding among all contributing people. > Easing and shortening many discussions > Achieving agreements faster among groups, institutes and contractors > Saving time & enhancing planning reliability Also my personal opinion: It is extraordinary what was (is) achieved the way it was (is) done!