 |
CERN Safety Alarm System Supervisory BoardMinutes of the meeting no. 2 held on 02.03.2000 |
| Subjects: | |
| Author: | Uwe EPTING |
| Participants: | Gianpaolo BENINCASA (ATLAS), Paolo CIRIANI (ST), Uwe EPTING (ST) (Secretary),
|
| Distribution: | Participants, Manuel DELFINO (IT), Paul FAUGERAS (AC), Giorgio GOGGI (EP), Dietrich GUSEWELL (EST), Kurt HUBNER (DG), Philippe LEBRUN (LHC), Jurgen MAY (DG), Stephen MYERS (SL), Rui NUNES (ST), Alberto SCARAMELLI (ST), Carlos SOLER (ST) |
The mandate and the minutes of the CERN Safety Alarm system Supervisory Board will be published on a web page after the board members have approved them. The address of the web page is:
http://st-div.web.cern.ch/st-div/mo/CSAM/sup_board/sup_board.htm.
Review of CERN Safety Alarm Monitoring (CSAM) for information of GLIMOS
The CSAM project was presented by Silvia GRAU, Rui NUNES and Carlos SOLER. The presentation is available in PDF format at the address:
http://st-div.web.cern.ch/st-div/mo/CSAM/sup_board/minutes/presentations/csam_20000302.pdf
The project is following the IEC 61508 software safety standard and a presentation for "Safety for Industrial Applications" will be prepared. The scope of the CSAM project does not include the detection equipment, which is under the responsibility of the different experiments. However the whole chain has to be checked because the CSAM system is as reliable as the weakest point. The CSAM team should also make sure that the proposal complies with the INB rules.
Although the CSAM presentation showed only LHC aspects, the project is thought to cover the whole CERN safety infrastructure. All equipment that is actually installed follows the CSAM proposals. It is planned to upgrade old equipment and have a homogenous transmission and monitoring system all over CERN by the start-up of LHC.
Special attention should also be paid to the reduction of false alarms. Therefore only the necessary alarms and reliable equipment should be installed to avoid sending false alarms to the firebrigade. This has to be considered when choosing high sensitive detectors and a large number of detection equipment.
The equipment and the installation have to be approved by the safety responsible for each building and safety zone as well as by TIS. The question, of who is responsible for the surface buildings, was raised. Rui NUNES will write a document that will clarify the needs of all ATLAS and CMS surface buildings including the number of alarms.
The number of the necessary safety alarms for the experiments should also be followed up by the concerned experiment WGs (chaired by the GLIMOS) and checked by the LHC safety co-ordinators (TIS). The CERN Safety Alarm System Supervisory Board has the impression that ATLAS requires too many safety alarms of level 3. The necessity of the safety alarms should be determined together with TIS.
The existing CERN safety regulations have to be considered for the specifications. Most of the basic questions can be answered by using the already existing safety documents and do not have to be re-evaluated another time.
Even if the analysis did not start yet, the interfaces and the technology should be defined before in order to have a coherent safety alarm system.
Status report on implementation of safety alarms in LHC experiments
ATLAS
Gianpaolo BENINCASA presented the safety equipment requirements for ATLAS. The slides are available in paper form. A specification for the surface buildings is available and the number of expected equipment can be calculated from it. For the ATLAS underground areas ~600 alarms for fire detection are foreseen. A working group has been set-up to determine the equipment for the Fire and Gas inside Atlas (FAGIA) project. These numbers will not be available before October 2000.
CMS
Reiner SCHMIDT presented the safety equipment requirements for CMS. The slides are available in paper form. The CMS requirements will be defined in two phases. The first phase will review the risks, necessary signals, reliability and the neighbourhood of the subdetectors and detectors. The second phase will cover the rest of the CMS cavern. The definition of level 2 and level 3 alarms and the corresponding actions are about 90% defined. A first summary gives 66 level 3 alarms for the subdetectors. The report for phase 1 is planned to be finished at end March 2000. Phase 2 will start in June 2000. The first interim report is scheduled for November 2000 and the final report for October 2001.
LHCb
Hans-Jurgen HILKE explained that the LHCb safety requirements will be more or less the same as today at Delphi. In future no flammable gas will be used and no cryogenics installations are foreseen. A detailed analysis will be started not before summer 2000 and results can not be expected before March 2001. At present Delphi has in the order of 100 alarms that are displayed at the local synoptic and one general hardwired alarm is transmitted to the fire brigade.
ALICE
Lars LEISTAM has presented the ALICE requirements. The slides are available in PDF format at the address:
http://st-div.web.cern.ch/st-div/mo/CSAM/sup_board/minutes/presentations/LL_20000302.pdf
ALICE expects from ST/AA that they provide standard detection systems, programming, software, maintenance, repairs and an analysis of the actual existing detectors. They will provide an interface to the slow control, help with the hardware installation and participate in the definition of alarm procedures and graphical user interfaces (GUI). The size and the procedures will be very similar to the existing L3 installation. The actual GSS has 850 signals on a local synoptic display and one hardwired alarm to the firebrigade. The new equipment installation is scheduled for mid 2004, hand held detectors for gas and oxygen should be available for January 2003. ALICE will analyse the existing signals and verify the actual alarms in order to clarify the future needs.
The term "Level 3 alarms" should be clarified and defined in order to avoid confusion in the future. The Level 3 alarms are very well defined in the IS37. All people concerned should refer to this document before starting a discussion about safety alarms. TIS will revise the IS37 document, but that will not change anything at the safety alarm definition. It is clear that not all detector alarms are real level 3 alarms.
The output of the discussion with the four experiments showed, that if more sensitive fire and gas detectors are used, the probability of false alarms would be raised. This should be considered very carefully at the alarm system design and these sensors should only be used where it is absolutely necessary.
Summary and future meetings
The meeting showed that two approaches for the safety equipment are foreseen. ALICE and LHCb will use more or less the existing infrastructure with the support from ST/AA. ATLAS and CMS plan to design and build their own safety system infrastructure. The specific Level 3 alarm safety system for ATLAS and CMS will be designed together with ST and implemented by ST/AA. The global detector safety system including equipment interlocks (DCS = Detector Control System) will be designed and build by ATLAS and CMS themselves. Nevertheless everybody should use the same detection equipment and the same interfaces in order to have a homogenous and coherent safety alarm system. When defining the requirements and the design for the safety system, one should always have in mind:
The Safety Alarm System Supervisory Board is meant to discuss only level 3 alarm related issues. All other safety matters have to be discussed directly with the TIS division.
created at Tue Aug 29 13:35:39 METDST 2000 with the
ST/MO/IN - Minutes Maker