LHC BLM threshold modification reports

Meeting Minutes

6rd Meeting

2011-06-30    Mariusz, Ruediger, Eduardo, Barbara

1. Direct Dump Monitors:

• One system which is not working properly will be exchanged coming TS.

• The other 3 systems exhibit a reduced sensitivity above ~4000 counts which will be investigated further.

• A threshold around 4000 - 6000 counts might be appropriate. Ewald will enable a diagnostic channel which is available on this card, that would log the status of the dump signal in TIMBER (while the physical output to the LBDS still remains disabled).

• Eduardo will check whether 2E10 injected beam on TCDQ is sufficient to calibrate direct dump BLMs. A possibility might be during the MD next Sunday 3.6.2011.

2. The cold magnet thresholds in the injection regions, which have not been changed in 2011 will be changed in batches during the next months.

3. The two MF=0.05   (BLMQI.01L2.B1E30_MQXA and  BLMQI.01R2.B2E30_MQXA) to be verified and probably set to MF=0.1

4. for the second half of 2011:

   1. All LHC checked for possible limitations: signal/threshold < 10 and MF > 0.1

   2. Collimator thresholds increase in steps to the nominal (500kW) power deposition value (as established in the MD: LHC-BLM-ECR-0020, CERN-ATS-Note-2011-042 MD ).

5th Meeting

2011-06-08    Markus, Jorg,  Chiara

3rd Meeting

2011-03-16    Mariusz, Markus, Ruediger, Chiara, Eduardo, Barbara, Bernd

1. Injection region BLMs: Decision: do not install RC filters at this TS. Better to disconnect the monitors temporarily from BIS during the scrubbing run (by disabling the connection of the monitor into the MASKABLE/UNMASKABLE BIS input in the DB), in case the injection cleaning does not give the expected reduction in losses on the TDI (giving a high signal on TDI BLMs and a few other monitors behind with less margin to the dump thresholds).
   • TDI monitors: The TDI is robust enough to take a full batch at injection, one TDI monitor is still connected to BIS and the magnets behind are going to dump already at lower losses than when we would start risking to damage the TDI.
   • MBXA position E10: better to disconnect temporarily from BIS during scrubbing, than to install an RC filter (which would act on all RS). The MBXA is still protected by position 2 monitors. The signal in the BLM at injection comes from losses at the TDI - non-local showers.
   • BLMQI.03R8.B1I30_MQXA (only IP8): It is still not understood why the signal is so high in this monitor. Position 3 monitors are redundant to (and less sensitive then) position 2 monitors. With the exception of MQXB, there is no simulated or measured loss scenario in which this is not the case. The signal in the BLM at injection comes from losses at the TDI - non-local showers.

BLMEI.04L2.B1E20_TDI.4L2.B1                                                     O.K. to disconnect permanently from BIS if needed

BLMEI.04L2.B1E10_MBXA           increase 450 GeV thresholds    O.K. to disconnect temp. from BIS if needed

BLMEI.04L2.B1E20_MBXA           increase 450 GeV thresholds

BLMQI.03L2.B2I30_MQXA           increase 450 GeV thresholds

BLMEI.04R8.B2E20_TDI.4R8.B2                                                     O.K. to disconnect permanently from BIS if needed

BLMEI.04R8.B2E10_MBXB           increase 450 GeV thresholds    O.K. to disconnect temp. from BIS if needed

BLMEI.04R8.B2E20_MBXB           increase 450 GeV thresholds

BLMQI.03R8.B1I30_MQXA           increase 450 GeV thresholds    O.K. to disconnect temp. from BIS if needed

2. End-of-run summary (UFO, collimator hierarchy, cleaning losses decomposition) should be coded within the PM application -> Action: Markus to discuss with BLM team (Eduardo, Aurelien, Barbara, Christos, Annika and Tobias)
3. How to display a list of all disabled monitors? Is it possible from the expert threshold application (DB query itself should be straight forward) --> Barbara. A control room application also for non-experts?
4. Dump on TCDQM losses on 8.3.2011: The presentation gives the names of all monitors in the LHC which have a big RC filter (factor 175 reduction in an instantaneous loss) and which are connected to BIS at the same time. They are TCD and MSD in IP6. Correction of the thresholds for this is depending on the loss evolution with time (and therefore not unique). Chiara explains that it was good that the beam had been dumped at this manipulation. The reason for the looses was that the feed-back was not set up for multi-bunch injection and this lead to an orbit excursion in IP6. The dump team prefers to not change the TCD thresholds (but not the MSD thresholds) for the time being. An alternative would be to increase all TCD with big RC filter (factor 175) by a factor of 10.
5. Direct dump BLMs: 4 SEMs in IP6. Eduardo is continuing the analysis. He will check whether there exists a threshold setting which is below the signal during a beam dump (where particles in the abort gap hit the equipment) and at the same time 'safely above' all of these scenarios:
   •  UFO signal on TCSG
   • collimator loss maps
   • collimator halo scraping (especially the one on the 30MJ beam)
   • losses at injection energy, during the ramp and during beam scrubbing

• The proposal is to look at all the last half of 2010 (hardware change), and verify that only during a beam dump the signal is above threshold. For example using TIMBER max. signal over 10 minutes.
• The fact that other sources contribute to the SEM signal (apart from secondary emission in the monitor), is considered not a problem, as long as the total signal is reproducible.
• Data to check in particular: 15.3.2011 at around 19:00 injected beam (around 1E10) shot on IP6 collimators (BLM MPS test).
• Threshold at the moment set to 13kGy/s corresponding to 32768 bits (summing/averaging)

2nd Meeting

2011-03-09    Mariusz, Annika, Ruediger, Jorg, Chiara, Eduardo, Barbara

1. "Cold magnet thresholds, are some long integration time thresholds too low?" Decision taken: During the next TS, increase the last two RS (21s and 84s) to the same value as the 5s RS on: DS and LSS cold magnets (MQ, MQM, MQTLH?, MQX ...). Do not do this change for arc magnets. Reason for increase: Give more margin for luminosity induced losses; The quench test, which lead to the reduction of thresholds in the long RS (LHC-BLM-ECR-0016 ) were done for integration times from 1-5s; at 21s the magnet protection systems catch possible magnet heating as well.
2. MQTLH: 600A magnets currently operated at 150A, not cooled by He. To be checked: Why are the thresholds so high (since when is the MF=1) --> Eduardo/Barbara; How high are the cleaning losses --> Chiara
3. Injection region BLMs to be changed: decision to be taken in one week. In case the injection cleaning is not giving the expected reduction in BLM signal from circulating beam hitting the TDI for the scrubbing run, the losses are expected to be two times higher than in the analysis presented: Foreseen for scrubbing run are up to 1400 bunches in the ring (2010: 900 bunches) with 144 injected at a time.
4. Direct dump SEM monitors to be investigated --> Eduardo

 1st Meeting

2011-02-16    Arjan, Mariusz, Annika, Markus, Ruediger, Jorg, Chiara, Eduardo, Barbara

1. Arjan showed the timing of the QPS versus BLM for Wire scanner induced quench in MBRB (D4) on 1 Nov 2010 at 15h40. The quench occurred about half way through the first of the BLM peaks. To be done: comparison Fluka, BLM, and QP3 based on this timing (Mariusz). Second slide is from Mariusz, showing that the BLM signal times the wire scanner speed is ~ constant for a number of the scans, but different for the very fast (not shown) and for the slowest one. Ruediger suggested to calculate this value for comparison from the PM event from a beam dump due to a wire scan (action Mariusz).

2. Eduardo showed the new proposed thresholds for cold magnets compared to the highest signals for each RS in the last 5 high intensity proton physics fills. It was decided to increase RS1 and RS2 (with respect to what is shown in his slides) by a factor 3. The details of the threshold changes can be found in the LHC-BLM-ECR-0016 (including comparison plots (end of 2010 compared to start-up 2011) for 450 GeV and 3.5 TeV (higher energies have been omitted to increase clarity):

The Table below (from the ECR) gives the typical changes from 2010-start-up to 2010-end-of-run to 2011-start-up for cold magnets.

This web site is owned by the account czam

Site created by CERN Web Services on Wednesday, 12 March 2003 at 11:30