MP3 Meeting

Wednesday 14 Apr 2010, 10:00 → 12:30 Europe/Zurich

874/1-011 (CERN)

Provisional AGENDA: - News from last week. A. Musso. - BLM threshold tests. R. Denz - Circuits viewer and open issues. B. Auchmann. - AOB

Slides

• Minutes_of_MP3_Meeting_of_14_April_2010.pdf
• MP3_14_March.pdf
• MP3_14_March.pptx
• MP3_report_-_week_15.pdf
• MP3_report_-_week_15.pptx

Minutes of MP3 Meeting of 14 April 2010 - RCBCH7.L3B2 et al. circuits problem: Some circuits trips without creating a port-mortem file (for the PC part). Seems to be a FGC problem where a reboot is automatically done. EPC is looking into it. - MB EE delay test: Type test done on Sector 5-6 (see Nuria’s slides). Two different delays added for odd (400 ms) and even (480 ms) point for Main Dipole circuits. Results very positive; no magnets triggered at 10 A/s for run at 2-4-6 kA. Analysis is ongoing to see “how much margin do we have before triggering” (signals shows that for several magnets the threshold on U_QSO and U_res was reached and passed, even if not for an discrimination time sufficient to trigger the QPS). (ACTION: Odd et al.) Plan is to produce the new cards for all the other Sectors and then an extra HWC phase will be negotiated with OP to perform the test on all the sectors. To be defined the test parameters (current level, current ramp rate and derivatives) (ACTION: Reiner, Knud et al.) - RQTD/RQTF problem: A dedicated meeting organized by Andrzej took place last Friday. He reported about it. These circuits are tripping due to the fast feedback (see last meeting). The acceleration applied by the FGC is too high, above the commissioned values. This cannot be changed. The QPS threshold however is critical at low current where a quench in the bus-bars does not propagate fast enough for efficient detection. A proposal to increase the discrimination time for lower currents was retained. This will require just updating a parameter instead of reprogramming and testing the software. Andrzej reminded that the original threshold in 2007 was 20mV. A proposal for a validation time and a current level will come (ACTION: Arjan and Reiner). Nuria informed Andrzej that the case of RCBX correctors in the inner tripled need also to be reviewed as it will be the next feedback needed. - RQTL13.L8B1 : A limitation in maximum current was detected during the squeeze of IP8 (needed 350A instead or 300). The magnet was commissioned up to 550A. The limitation was applied in all RQTL magnets after the discovery that circuits could be unprotected during a QPS loss of communication following a reset. The firmware was already updated in February to cope with this issue and the current limitation is not longer needed. However, it was proposed not increase all magnets but only this one. It was already done by Markus. - Night Service MP3: Original plan scheduled only morning and afternoon shifts as there are not enough experts available. Night shifts are only necessary in case of HWC type tests (EE delays test etc.) or to recover from a technical shutdown. The decision taken in the original plan remains valid, i.e, the person doing the afternoon shifts also does the on-call service during the night. Intervention for QPS problems (PM analysis should in principle be possible from home) should be done by the QPS/MPE Piquet. A clear “check list” for EIC (who call and when) will be prepared by Andrea and discussed ASAP. (ACTION: Andrea) - New tool for Circuit Analysis (history): A tool is needed to have easily all the information about a circuit under analysis (how is composed, major operating parameter, QPS parameter, link to NC at manufacturing and at HWC, etc.) The tool would be a combination of the data already included in the Electrical Database, Circuit Synoptic, MTF, etc. Input will be always choosing a circuit name among a predefined list (to avoid typing errors). Tool must permit to create and attach an “event” to the circuit history. The “event” will automatically create an event also in the MP3 Logbook. To discuss these and other requirements/aspects/suggestions, a dedicated MP3 meeting is called for next Wednesday. - FPA Transient Analysis Tool: It was presented by Odd. Some automatic analysis already defined, but Odd is now waiting for other requests. M. Modena/ N. Catalan.