Tuesday September 25th,
14:00
CCC conference room 874/1-011
Mike
chaired this meeting. There were no comments on the preliminary version of the
minutes of the 30th meeting held two weeks earlier. Massimo reported some news
from the FIDEL meeting of 24 September, which featured a discussion relevant to
the extended magnet measurement programme. The most
important information is that following the transition between software
versions (SLC3 and SLC4) running on lxplus, the
web-based access and analysis of magnetic data may have to be discontinued.
Should this be the case, data analysis will be performed semi-manually. This
triggered several discussions among the participants on the implications for
magnetic measurements in 2008, and in particular the potential increase in the
time needed for data processing (Massimo, Ezio), the
possibility to freeze the existing system (John Miles, Jean-Pierre), and
availability of resources for this work (Mike). In the absence of the persons
responsible, it was decided to follow up this issue (Jean-Pierre).
After the meeting, discussions have taken place with Luca Bottura
and Laurent Deniau, who are responsible for this service,
and who are currently working on the identification of the critical
functionality and on the porting of the analysis tools. A complete re-design of
the data access and analysis system has started, while temporary solutions are
being prepared.
Jan reported the plans for the machine-protection system commissioning (MPSC) and the related studies in the MPSC sub-working group (SWG). Part of his talk, covering a general description and plans, was similar to an earlier presentation which he presented to the LHC MAC on 15 June 2007, the second, addressing the present status, was based on an MPSC presentation by Alick from 12 September. The MPSC SWG comprises 20 members. The SWG web site is linked to the LHCCWG home page.
Core elements of the LHC MPS are beam interlock, beam dumping system, BLM, collimation, quench protection and powering interlocks, and injection system.
Oliver asked whether detector elements like Roman pots are not included. Jan replied that detector components enter only where movable elements are concerned. After the meeting Daniela clarified that the Roman Pots are not part of the machine protection system, i.e. they do not protect the machine, but rather they are users of the protection system via the interlock system and the dedicated BLMs. Jan further commented that, though they have no protective function, the Roman pots can cause great damage to other parts of the machine. For this reason they are in a similar category as the vacuum valves and their impact on the machine protection system will be considered by the MPSC SWG. Ralph pointed out that the Roman Pots are handled through the collimation control system, like all other operationally adjustable aperture limits. They use partially the same hardware as collimators and, in his view, they have a similar machine protection functionality.
A safety level SIL3 needs to be guaranteed for the MPS core part. This is achieved by a good system design, reliable system components, careful follow up of production and installation, and thorough system commissioning. Jan next presented a schematic layout of the LHC MPS and its interfaces.
LHC hardware commissioning includes individual system tests. Experience with MPS equipment was obtained while powering the first LHC sector. Also, the SPS operation in 2006 and 2007 included tests of fast magnet current change monitors (FMCCMs); the use of similarly designed beam interlock systems (BIS) for CNGS, TT40/TT60 and the SPS ring; and the safe beam flag for CNGS operation.
Jan next showed an example of the software interface for the SPS extraction BIS, where maskable and non-maskable inputs are clearly distinguished. Oliver asked for the meaning of a column called “disabled”. Jan answered that channels which are not connected are disabled. He emphasized that of course one should never be able to disable a channel which is operational.
The SPS extraction BIS information contains the same type of energy tracking as will be used for the LHC beam dumping system. Beam extraction is possible only when its energy is OK. For the LHC dump the situation will be slightly different, namely the beam should be dumped before the energy drifts too far away from nominal. The summary display of the interlock system is already developed and it was tested at the SPS.
Jan now presented details of the FMCCMs, whose interlock triggers at a fast current change equal to 1e-4 of nominal value. The performance of part of the MPS system was verified during LHC hardware commissioning. From the sector commissioning, the example of a quench triggering PIC and BIC signals was shown.
In the following, the plans for further MPS commissioning were detailed. Most hours before beam commissioning will be spent on individual system tests and during the LHC cold check-out period. Fairly little time will be needed with beam. The green elements on Jan’s slide 15 can all be tested without beam.
Replying to a question by Daniela, Mike explained that controlled access during the cold checkout will be given when required.
Jan recalled the mandate of the sub working group (SWG) on the commissioning of the MPS, which is to “produce the detailed (=step by step) description of the commissioning of the LHC Machine Protection System and record the test results… The aim is to have a first version available by June / July 2007” , which is going to be tight but this goal had been defined in view of circulating beams in November 2007.
The MPSC SWG is in the process of developing detailed procedures for each system, following a standard format. The procedures are published after approval by working group. They contain the detailed conditions required, as well as the descriptions of all the tests needed.
Each test is filed in the hardware commissioning and test folder (MTF) system, including the test status and test reports. The system is identical to that used for LHC hardware commissioning (HWC). Jan showed an example MTF interface illustrating its structure. The very same system is applied to the MPS.
Jan now commented on different type of procedures and the work of other working groups, such as the hardware commissioning working group and the LHCCWG, pointing out that the MPS Commissioning is complementary to these working groups, and that, unlike them, it covers all commissioning phases, with and without beam, for the individual systems.
In addition to developing procedures and analyzing test results, the MPSC SWG is a forum to discuss MPS related items. Another important function of the MPSC SWG is to establish a team of experts who can be called in during operation, as well as to be a training camp for the EICs.
Jan listed some topics presently under discussion in the MPSC SWG, like the setting of BLM thresholds during operation – e.g., which thresholds to set? when and how to change thresholds? – , the commissioning of the collimation system, and the interlocking strategy for aperture kicker and ac dipole.
Some slides shown by Alick Mc Pherson in the SWG meeting of 12 September summarize the status of the written procedures. The BIS procedure is almost ready in EDMS. First drafts exist for FMCM, LBDS, PIC, vacuum, and WIC. Access, BLMs, collimators, and injection system are less advanced. The development of procedures for other systems, such as MKS&AC dipole, movable objects of the experiments, and safe beam flag, will start soon. The SWG time table foresees to complete drafts of all procedures by 1 October, and to have all procedures included in MTF by mid December. This translates into a time allowance of 1 week per procedure for EDMS, and one week per procedure for the transfer to MTF.
Jan explained that his talk was meant to provide a flavor of what is being discussed in the meetings. He also highlighted that logging is not part of the MPS.
The MPSC commissioning should be modular, with well defined entry conditions for front end systems, and exit conditions to proceed during cold check or with beam. HWC procedures and results in MTF are to be confirmed by the MPSC procedures. The interface with BIS will be tested complementarily to the BIS commissioning.
MPSC validation must ensure that there is no risk to the machine due to operator controls. Also, the dependency on BIS info must be made clear. That is beam info flags should not be used for any critical actions.
Jan concluded that the commissioning of MPS has already started, referring to LHC HWC, SPS extraction and ring BIS, and FMCCM tests. To guarantee a “safe” LHC operation, the commissioning procedures of the different MPS components are being described in detail. A lot work is still to be done. The fruits of the MPSC work should be limited confusion, and that we will never see the destructive potential of the LHC beam.
Mike thanked Jan on behalf of the LHCCWG for the important work of the MPSC sub working group. Daniela asked why the movable detectors were not part of the schematic shown, although their interface had been decided, and just needed to be implemented via the normal procedure. Jan replied that the MPSC SWG wants to verify the MPS conditions are met and check the tests, highlighting that what is not normal at the LHC is the great damage potential, and therefore the commissioning should be done properly. Rudiger added that Bruno had tested the hardware interfaces, but that Jan and MPSC will test all steps systematically, which is done together with Alick. Jean-Jacques remarked that Bruno will test everything he can test, saying that Bruno probably does not have an MTF for all the slots, but that BI has and could provide one. He gave the example of the alignment step in MTF, which his handled and entered in the BI MTF slots by the survey group. Concerning the description of movable devices, Jan observed that it might be that the BI procedure already includes all the steps needed, which would be beautiful, but which will need to be checked. In particular it must be tested that the beam dumping system will be triggered.
Jean-Jacques now commented on Jan’s slide 17. BI has the same intention to use the HWC MTFs. Some information is missing in MTF, however. This point will be discussed in detail at a dedicated meeting. For example, in case of a repetition of steps, the old values are overwritten at the moment, whereas BI would like to preserve the history. Jan will participate in these discussions. Jean-Jacques invited everybody else interested in this subject to contact him.
Thijs remarked that the FMCCMs triggered at fast current changes of order 1e-4, while the power converter control is at the per mil level, i.e. 100 times more precise. Jan and Rudiger both replied that the MPS system should act when the situation becomes dangerous, not earlier. Rudiger also clarified that the exact sensitivity depends on the magnet and individual magnet. There is no reason to have tighter control. The FMCCMs need to detect super-fast failures. The 1e-6 criterion for the power converters is imposed for other reasons (emittance growth etc).
Mike asked about the feedback on Jan’s presentation from the LHC MAC. Jan answered that the MAC found it very interesting. More specifically, the Executive Summary of the 21st LHC MAC Meeting (14-16 June 2007) contains the following paragraph: “The MPS commissioning is well organized and being carried out in a very professional manner. Given the great complexity of the LHC it will be necessary to make this system flexible enough in the beginning of beam commissioning so that it does not prevent that commissioning from proceeding, being so constructed that thresholds for permissive actions can be readily adjusted as needed. It is suggested that during this early period a clear path for evolution of the system to one less flexible and more appropriate to routine operation be established.”
Mike
also inquired the present status of the MPS (not MPSC) WG. Rudiger
announced that an invitation for the following Friday had been sent out by Jorg. Presently under discussion is the level of the beam
flag description. The MPS WG meets at a reduced frequency. All commissioning
questions are now being discussed by MPSC SWG. In case of need the MPS WG could
however meet more often. Lastly, Mike asked whether the beam dump is on the
list of the documenting procedures. Indeed, it is. The beam-dump procedures
will be based on
Tuesday October 9th, 14:00
CCC conference room 874/1-011
Provisional agenda
Minutes of previous meeting
Matters arising
Filling
the LHC:
- Most
robust SPS scheme (Magali)
- Interleaved
filling possibilities (Verena)
- LHC
cycle requirements (Stephane)
AOB
Reported by Frank