Choose timezone
Your profile timezone:
2nd Quench Behavior Team meeting
→
Europe/Zurich
Minutes of the 2nd meeting of the LHC quench behavior team
Date: 2015-07-16, 10.30-11.45
Place: 927
Presents:
Gijs de Rijk, Marta Bajko, Per Hagen, Gerard Willering, Ezio Todesco, Sandrine Le Naour , Jean-Philippe Tock
Flattop quenches [G. Willering]
Gerard presents an analysis of the quenches that occurred at flattop during the past three months of operation. We had an average of one event per month, with three events in the period May-July. The rather surprising feature is that all quenches occurred in sector 34 and with 2000 series magnets. Gerard also points out that the sector 34 is the only one that was not kept several hours at flattop during hardware commissioning, but only 30 min.
It is suggested to check the temperature of the sector to verify if cryogenic conditions can account for this anomaly of 34 sector. The possibility of keeping the sector at maximum current for a few hours, or of training it to a slightly higher current value (50 or 100 A more) is discussed. This should be done in the shadow of the LHC operation, during down time due to other causes. It is planned to carry out more analysis, organize a discussion in MP3 and discuss with the management. Then one could present a recommendation to LMC on an action to reduce these quenches and improve availability time.
Test of 3000 series magnets [E. Todesco, and general brainstroming]
Ezio recalls the analysis of Gerard, shown at LMC in April 2015, providing the number of quenches to reach 6.5 TeV plus 100 A margin in the 3000 series during the hardware commissioning. Gerard grouped the magnets in batches of 10 – the plot clearly shows strong variations during the production. The first 130 magnets show a probability of quenching to reach 6.5 TeV plus 100 A margin of 0.1 quench per magnet. Then there is a gradual degradation reaching a peak of 1 quench per magnet around 3180. The situation improves around 3320, and another degradation is present at the end of the production.
We have three spare magnets: 3096 and 3100 belonging to a good part of the production, and 3409 to a bad part. These magnets have been tested in 2010-13, showing training compatible with the 2015 commissioning data. 3409 had a training quench at 11348 A.
It is proposed to test 3409 again in SM18. A possible slot, according to Marta, is October. Marta will prepare a test plan that should include a training and 10h at flattop. Then a thermal cycle could be envisaged. Sandrine points out possible conflict with the dipole test for quench heaters, also to be done in the same period. The priorities will be decided by the group leaders.
Ezio also shows a correlation plot between the number of quenches needed to reach 6.5 TeV plus 100 A margin in the virgin condition and during 2015 hardware commissioning. In virgin data, the degradation of the production appears well before than in the 2015 data, but for the second part of the production there is a good correlation. More analysis will be carried out, also to relate the maximum training current with 2015 performance.
Other actions and future plans [all]
Per confirms that the data of the end shims are available in a DB. Some effort has to be done to reformat them to perform analysis and correlations – it should be done within a couple of weeks and presented in the next meeting.
There are minutes attached to this event.
Show them.
