HMPID Shifter's Manual

Table of Content

Introduction

If you need expert help CALL the on-call expert (check ALICE SMS)


smashingcomputer2.gif

  • You should call the on-call expert first if not written otherwise.
  • You can find the name of the on-call shifter on ALICE SMS: https://alicesms.cern.ch/ or you can check one of the big screens in the ACR.
  • For specific questions call the subsystem experts:

subsystems expert office phone mobile phone
HMPID ACR seat current HMPID shifter (not really an expert) 77246  
DAQ, SRC Paolo Martinengo 78434 163757
DAQ/DQM, SRC Levente Molnar 78859 161398
DCS /Liquid System Giacinto de Cataldo 71982 162827
DCS Giacomo Volpe 72926 161834
Gas Antonello Di Mauro 76612 164024

Shifter's tasks and responsibilities

  1. Your goal as an HMPID shifter is to ensure the best quality and largest number of events with the HMPID during data taking! You have to be aware the daily program and the minute-by-minute changes in planning, that are expected in the first days/weeks operation of ALICE + LHC.
  2. Before you arrive to shift (in 2009) check the global ALICE Standing Orders page: http://aliceinfo.cern.ch/Collaboration/Run_Coordination/Run09/shift
  3. Prepare the detector for global data taking in the standalone operation period (approximately 1 hour per day, called "happy hour"):
    1. DCS: – check hardware status & clean trips if necessary and the status of the HMPID top node is READY.
    2. DAQ: – check if all equipments are included, if not, include them, take calibration and standalone run, confirm that the zero suppression is working properly.
  4. In the global data taking period:
    1. Check and ensure safety of the HMPID.
      As of 2009 November 29th the state of the HMPID has to be:
      • Before & during injection, beam dumps, TED & TDI splashes or circulating unstable beams: BEAM_TUNING
      • ALERT! If not: immediately contact DCS operator and ensure BEAM_TUNING state!
      • When stable beam flag is issued by LHC: READY !
      • ALERT! If there is a Happy Hour or Global DAQ test (eg ERP), while there is(are) circulating beam(s) but there is no stable beam flag issued, the HMPID stays in BEAM_TUNING! If the Happy Hour or Global DAQ test is held, while there is no beam, neither TED nor TDI splashes then HMPID can go READY.
    2. Keep the detector in the global data taking all the time (except the happy hour) with the best configuration available.
    3. Make sure the HMPID is kept in the global partition and all HMPID LDCs are included in the global partition – so we actually record data! Yes, it is trivial, but it had happened before and probably you don't want to be the one who missed few hours of the first data taking!
    4. Monitor data quality and report any "strange" behaviour to experts. E.g. noisy pads or group of pads, etc .
    5. Respond to problems, when the shifter action is required. Note: actions can be taken ONLY between runs.
  5. If the action IS necessary, since a new run CANNOT be taken because of the HMPID (e.g.. busy bit, CDH error, HMPID DCS status is not READY) then ask the DAQ shifter & shift leader to wait for the HMPID and ask for how much time do they wait. Quickly diagnose the problem and if your are able to correct the problem try to put back the HMPID to the global partition as soon as possible.
  6. If the action IS NOT necessary, since a new run CAN be taken with HMPID (eg. there is one HV tripped channel) wait for the "happy hour" or a confirmed (by the shift leader/period coordinator) longer break in the data taking.
  7. Write your shift report at the end of your shift at https://twiki.cern.ch/twiki/bin/view/ALICE/HMPIDSHIFTREPORTS2009 including:
    • Start of shift and your name
    • ALICE status: eg. injection/beam through ALICE, etc.
    • global runs taken during the shift (in which the HMPID was included)
    • problem(s) encountered

Overview

What will you use?

To accomplish your tasks, you have to use monitoring and control programs and computers of the HMPID. The main interfaces are the following:
  • Detector Control System (DCS): in this program you monitor the operational status of the HMPID: high and low voltage values, the gas, liquid and cooling systems.
  • Detector Control Agent (DCA): in this program you can start STANDALONE and CALIBRATION runs, monitor the data taking at a low level, check log entries.
  • Local Data Collectors (LDCs): the LDC computers control up to 4 DDLs each, operated through a terminal; you can reset the LDCs and load tresholds here.
  • Local Trigger Unit (LTU): in this computer you can check the status of the DDLs and enable/disable them.
To use these computers and programs you will need to know the passwords. You will receive these passwords during your training.

The HMPID shifter's seat – general information

There are two computers allocated to the HMPID: aldaqacr28 and aldaqacr29, both with two screens. These computers have limited access to the network outside ALICE, so you have to bring your own laptop to able to read emails and check the latest version of this manual. You will also need the appropriate access privileges (ALICE-COF and ALICE-ACR) to be able to take your shifts. Also consult the general rules of shifting at http://aliceinfo.cern.ch/Collaboration/Run_Coordination/Run09/shift/ .

Detector Control System (DCS)

In this part we review how to start and operate the DCS to accomplish various tasks of the HMPID shifter, including changing the state of the detector and monitoring the operational parameters.

Starting the DCS

  1. Login as hmpid user on any of the DAQ workstations in the ACR (aldaqacr28 or 29).
  2. Open a terminal and execute the following commands:
    1. source startHmpid
    2. dcs (dcs is an alias to rdesktop -a16 -g 1280x980 alihmpon001)
  3. Log into alidcson001 using your NICE username and password:
    4hmpmanual002.png
  4. If you see the DCS UI you can skip the next step (see figure).
  5. If not: from the Start menu --> Program --> Startup open HMPID DCS UI and you have to see this:
    DCS UI
  6. Push on the top left corner of the UI the yellow key button and login with your NICE account name and the special password you will receive during your training.
    DCS UI login screen
  7. Double-click the HMPID node in the tree of the left, and the overview of the HMPID will appear. Or alternatively you can right-click it and select view.

Taking and releasing the DCS lock

Only one DCS UI can have control over the DCS at any given time. If you don't have control, you can only monitor the HMPID but can't take any actions. If HMPID participates in a PHYSICS run or ALICE is preparing for BEAM injection the lock is taken by the central DCS operator.

The lock symbol

The lock symbol in the top left corner can have 3 colours:
  • If the lock is red, the DCS control is taken. Moving the mouse on the lock you can check whom took control. (Only one UI at a time can have the lock.)
  • If the lock is green then YOU have the control.
  • If the lock is grey and open no one has the control, you are able to take it.

Taking or releasing the lock

  1. Select the HMPID top node in the FSM tree with a double-click.
  2. Click the big FSM button in the top left corner of the DCS UI, the following the FSM panel of the HMPID will appear.
  3. In the FSM panel click on the top right lock next to the State and "Take" or "Release" the lock. (NOT "Release all")
  4. CLOSE the FSM panel, answer NO for the question "Do you want to release this tree?".

Moving the HMPID to different states

  1. Select the HMPID top node in the FSM tree with a double-click.
  2. Click the big FSM button in the top left corner of the DCS UI, the following the FSM panel of the HMPID will appear.
  3. Click on the button on the top with the current state of the HMPID to see available actions. Always select carefully your action

The states of the HMPID DCS and the procedure to change the HMPID state

The following figure is a visual help to remember how to power on and off the HMPID. The states that appear on the graph are "stable" states of the detector. There are many more states and possible actions, but these are the necessary ones for the shifter.

Powering on and off the HMPID

The very-very short description of stable states:

  • STANDBY: both HV and LV (readout) are turned off. HMPID must be in STANDBY before the magnet current is changing. It is the shifters responsibility to ask the shift leader to notify him well before the magnet is ramped up or down.
  • STANDBY_CONFIGURED: the HV is off, but LV is on.
  • BEAM_TUN: the ALICE detectors are in BEAM TUNING mode when the LHC is injecting beams. LV is on but HV is only at an intermediate level.
  • READY: the HMPID is ready for data taking both LV and HV are on their operational values.

Making the HMPID READY for data taking

These are the necessary steps to make HMPID ready for data taking. You have to have the lock and the issue the command from the FSM panel. If you issue a command you won't automatically move to next stable state but rather into an intermediate state and then to the stable state, nevertheless the manual only mentions the stable states.
  1. Check in the DCS UI if the INFRASTRUCTURE are READY (open the relevant ViewPanel), if not then CALL the On-Call Expert!
  2. Go to the HMP_DCS (the top node of the hierarchy) and check the HMPID Status on the top left corner. If it is READY you can skip the next step.
  3. If the HMPID is not READY:
    1. If the state is STNDBY then go to STBY_CONFIGURED. Wait until the stable state is reached. Follow with step ii.
    2. If the state is STBY_CONFIGURED then go to BEAM_TUN. Wait until the stable state is reached. Follow with step iii.
    3. If the state is BEAM_TUN then go to READY. Wait until the stable state is reached. Follow with step 4.
  4. If the READY state is reached check HV values (click on POWER in the main UI):
    • In CH4 gas the nominal HV is 2050 V.
    • In ArCO2 gas the nominal HV is 1300 V.
    • In pure Ar gas the nominal HV is 300 V.
    • In case of LHC injection tests the voltage can differ. The voltage is always set by an expert.
    • ALERT! WARNING: channel RICH0 HV3 must be at U < 1800 V, channel RICH5 HV1 must be at U < 600 V. If this is not the case call On-Call Expert!!!
    • ALERT! In case of problematic links (stuck busy bits) masked them out and DO NOT move the detector to STANDBY to do a power cycle unless you are instructed to do so by an expert!
  5. If the HV values are correct, the HMPID is ready for data taking.

Monitoring the HMPID from the DCS during data taking

Colour codes in the DCS

The alarm severities in the detector are color coded as follows:
  • green is OK,
  • yellow is warning,
  • red is fatal.
ALERT! If the state is warning / red red then in this state no meaningful data can be taken!

Monitoring HV/LV values

  • On the main DCS panel check if all HV/LV channels are green. If you see an orange HV/LV sector it has tripped. To clean the trip, see in "How to clean a HV/LV tripp".
  • On the main DCS panel (top level node in the FSM tree) click on the box "POWER" of one module, the following panel will appear. From it you can monitor the actual values of voltage and current of all HV and LV channels of that module. ModPower.png

Monitoring the radiator liquid

  1. On the main DCS panel click on the box "C6F14" of one module, the following panel will appear. It shows radiator parameters and its level in the three vessels for that module. ModC4F10.png
  2. Check the Liquid System (C6F14) for ALL MODULES: the three radiators should be completely filled (except RICH6 Radiator 1)! If not CALL the On-Call Expert!

Monitoring the cooling

  1. On the bottom left corner of the main DCS panel click on the box "COOLING", the following panel will appear.
  2. From it you can monitor the state of the HMPID cooling system.
  3. Check if COOLING is ready (green). If not immediately CALL the On-Call Expert! GenCooling.png

Handling ALARMS in the DCS

Once a detector part is faulty (trip channel, over-temperature, gas flow, etc) the relevant UI zone colours change according to the alarm severity. (See the explanation of the colours)

Cleaning a HV/LV trip

  1. If the trip happened during a global run immediately notify the shift leader and/or DAQ shifter and ask him to stop the run and ask for the DCS lock if necessary.
  2. When the global / local run is stopped then start the HV clean procedure:
    1. Navigate in the DCS UI FSM tree to the relevant HV/LV node where the alarm is localized and open View Panel as in picture below:
      hmpalarm001.png
    2. Take note of the module number, channel number and time of the trip to report it to Giacomo Volpe and record it in your shift log (after cleaning the trip).
    3. Click "Click to ARM" to reveal the actual button then click on that button:
      ClearAlarm.jpg and the HV/LV channel should move to OFF state (0 V).
    4. Open the FSM panel and send the command GO_READY for the HV/LV channel, the channel should get back to ready.
HELP Note that if another FSM panel is open first close it otherwise you will not be allowed to navigate.

Data Acquisition (DAQ)

Starting the HMPID DAQ

  1. Login as hmpid user on any of the DAQ workstations in the ACR (aldaqacr28 or 29).
  2. On a virtual desktop start the DCA:
    1. Click on the 6th icon from the left (next to the Firefox button which is called Date HI)
      bar.png to start the Detector Control Agent (DCA).
    2. Then you will see the DCAs listed for ALICE DAQ.
      4hmpmanual0019.png
    3. Click on HMPID and you will see the following two windows:
      4hmpmanual00110.png 4hmpmanual00111.png
  3. Log into the LDCs on an other virtual desktop:
    1. Open 4 terminals (you can execute for i in `seq 1 4`; do xterm & done; to quickly launch four xterms).
    2. In the 4 terminals log into aldaqpc145, aldaqpc146, aldaqpc147 and aldaqpc148 using ssh as user hmpid.
    3. In each terminal execute source startHMPID to activate the shortcuts and HMPID commands.
  4. Log into the LTU machine on the virtual desktop where the DCA is:
    1. Log into alidcscom026 as user hmpid.
    2. Launch 2 terminals (for example with the command for i in `seq 1 2`; do xterm & done;)
    3. In one of the terminals execute vmecrate hmpid (LTU management software) and you should see these two windows:
      4hmpmanual0016.png 4hmpmanual0015.png
    4. In the other terminal execute fs and you have to see this:
      daqhmp.png
    5. If ALL bits under the "E" letters are 0 you are ready to start data taking.

The HMPID DAQ

The 14 HMPID equipments/DDLs are distributed on 4 LDCs and on 7 modules. If there is no ongoing run which includes the HMPID (either GLOBAL or STANDALONE) then the fs command on alidcscom26 should show no busy equipments. If you see busy equipments, you have to follow the appropriate section of the manual. In this section you can read about the the relation of the equipments and the LDCs and on the interpretation of the bit maps.

HMPID Equipment/DDL map

LDC machine name equipments
ldc-HMPID-1-0 aldaqpc145 1536, 1537 (RICH0_left, RICH0_right),
1538, 1539 (RICH1_left, RICH1_right)
ldc-HMPID-2-0 aldaqpc146 1540, 1541 (RICH2_left, RICH2_right),
1542, 1543 (RICH3_left, RICH3_right)
ldc-HMPID-3-0 aldaqpc147 1544, 1545 (RICH4_left, RICH4_right),
1546, 1547 (RICH5_left, RICH5_right)
ldc-HMPID-4-0 aldaqpc148 1548, 1549 (RICH6_left, RICH6_right)

Interpreting the bit maps in the alidcscom026 terminal:

  • When you execute the fs command you get a concise information on the state of the equipments.
  • The most important information comes from two bitmaps: one bitmap shows which equipments are enabled/disabled (0=D=disabled, 1=E=enabled), the second bitmap shows the busy state of the DDLs.
  • The least significant bit corresponds to equipment 1536, the most-significant bit to equipment 1549
  • In the picture of the alidcscom026 terminal above:
    ...10....,...1
    EEEEEEEEEEEEDE
    00000000000000
    means that equipment 1537 is disabled and all others are enabled, and none of them is busy.

Taking STANDALONE and CALIBRATION runs

General preparation for STANDALONE and CALIBRATION runs

  1. Take the lock in the HMPID DCA.
  2. Click on the green DCA button and then click GET_DAQ_RESOURCES
  3. In the DCA check the 4 resources: DCS/RUN_CU , HLT, DAQ_RC, LTU, all 4 of them should be green / ready.
  4. If the HLT is dead unlock it before data taking.
  5. If LTU is dead go to ask the CTP shifter to reset the HMPID LTU.
  6. If DCS is not ready, check why (see DCS part of the manual).
  7. HELP You can start to take data if the HMPID DCS is in READY state as well.
  8. Are you after a power cycle? Take the extra steps in the next section!

Extra steps: when the HMPID is after a LV OFF-ON cycle (power cycle)

  1. In the virtual desktop with the 4 terminals of the LDCs issue the command rst in all of them.
  2. Check the exit codes for every equipment: they should be all zero.
  3. Wait 30 seconds after the end of the last command.
  4. In the VME Crate execute Configuration -> TTCInit
  5. In the alidcscom026 terminal execute fs to check the busy bits.
  6. If you have busy bits, refer to the Resetting busy bits section of the manual.

Taking a CALIBRATION (pedestal) run

  1. Make sure you made the preparation and if necessary, the extra steps.
  2. In the DCA select:
    • "GDC: Event Building OFF ",
    • "LDC: Local Recording ON ".
    • ALERT! Note: if you select something else for GDC, then the CALIBRATION run will terminate with error and the correct pedestals will not be calculated!
  3. From the green DCA button select & issue the command CALIBRATION_RUN
  4. Follow the messages in the infoBrowser window
  5. CALIBRATION Run will stop automatically after ~ 1000 events.
  6. HELP Note for the shifter (COSMIC RUN 2009):
    1. After the CALIBRATION run has ended, you HAVE to upload the pedestals on each LDC executing the command lt !
    2. Check if all links give the Exit Code = 0.
    3. If not try the lt command again a few times, if no success then follow the "extra steps after power cycle" again.
    4. Do lt again.
    5. If still no success then CALL On-Call Expert!
  7. After the CALIBRATION run take a short STANDALONE run (see below the exact steps) and check in the DQM if the "Event Display" is almost empty.
  8. HELP Note: if you see filled lines in the DQM, or parts in the detector:
    1. Stop the STANDALONE run.
    2. Re-apply the lt command in the terminal of the LDC where the filled lines appear (Check the Equipment map).
    3. Start a new STANDALONE_RUN and check the "Event Display" again.
    4. Repeat this procedure until the "Event Display" is clean.
  9. If the "Event Display" in the DQM is clear, HMPID is ready to participate in the global partition.

Taking a STANDALONE (zero suppressed) run

  1. Make sure that you have a good pedestal run and the thresholds are uploaded
  2. In the DCA select:
    • "GDC: No Recording ",
    • "LDC: Local Recording OFF ".
  3. Select and issue the command STANDALONE_RUN from the green DCA button and data taking starts
  4. Check if zero suppression is OK in the HMPID AMORE monitoring.
  5. To stop the STANDALONE run you have to select & issue the STOP_DATA_TAKING from the green DCA button.

Participating in the GLOBAL partition

Giving the HMPID to the GLOBAL partition

  1. If the global CALIBRATION session happens during your shift: take a CALIBRATION run followed by a STANDALONE run.
  2. If zero suppression is ok then in then:
    1. In the HMPID DCA select *RELEASE_RESOURCES*from the green DCA button.
    2. Unlock the HMPID DCA
  3. Go to the DAQ (ECS) shifter and wait until he/she takes the HMPID DCA lock and includes it in the global partition.
  4. Make sure that the ECS shifter includes ALL 4 HMPID LDCs in the global partition. Have it checked with the ECS shifter!
  5. ALERT! NOTE: if the DAQ shifter tells you that the HMPID is not ready: "FERO is NOT Configured" then tell the DAQ shifter to select GET_RESOURCES and after the CHECK_FERO commands on the DAQ control panel (NOT on the HMPID DCA!!!)

Checking if all HMPID LDCs are included in the GLOBAL partition

  1. When you give the control to the DAQ shifters make sure when he/she includes the HMPID in the global partition, he/she checks the status of the HMPID LDCs in the central DAQ configuration.
  2. When the global run starts:
    1. Run the global monitoring (see DQM section) and check if you have data in all enabled equipments.
    2. In the DAQ logBook go to Runs --> Statistics --> click on Run Details (magnifying glass next to run number) --> click on LDCs Statistics tab.
    3. In the LDCs statistics tab check if all 4 HMPID LDCs are included: LDC-HMPID-1-0, LDC-HMPID-2-0, LDC-HMPID-3-0, LDC-HMPID-4-0.
    4. If one or more of the above HMPID LDC is/are missing go to the DAQ shifter to stop the run and include the missing LDC(s)
    5. DO CHECK if the HMPID LDCs are included for ALL new global runs in which the HMPID participates!

Handling errors of the DAQ

Recovering CDH (Common Data Header) errors:

  1. First try: restart the run
  2. Second try: if the restarting of the run fails again with CDH error
    1. In each LDC terminal type rst.
    2. In the alidcscom026 terminal you should see all busy bits: 0 to go to 1.
    3. After ~ 30 seconds, issue the TTCinit command from the VME Crate: Configuration->TTCinit
    4. Try to restart the run
    5. If the run fails again with CDH error, deselect / remove the faulty equipment

Resetting busy bits

When one busy bit is stuck
  1. Identify the busy bit and the corresponding equipment/DDL.
  2. In the corresponding LDC terminal issue rst.
  3. After 30 seconds issue the TTCinit command from the VME Crate: Configuration->TTCinit
  4. All bits in the alidcscom026 window should be 0 now

When one or more busy bits are stuck
Follow these steps and check the busy bits after each of them. If one or more enabled bits are 1 then try the next step.
  1. In each LDC terminal: type rstb and all the enabled busy bits should switch to 0
  2. Select ttcFEEreset from the “Configuration” menu, in the VME crate window.
  3. Select TTCinit from the “Configuration” menu, in the VME crate window.
  4. Try the procedure in Recovering CDH errors.
  5. From the VME Crate panel select Configuration -> ttcFEEcmd then enter 12 as command and click "start"
  6. Deselect the corresponding equipment.

Enabling/disabling and masking equipments

If one of the equipment is busy and you cannot reset it as described above you can try to mask it, but remember: ALERT! one DDL corresponds to a half chamber!

ALERT! Both the masking and the enabling/disabling must be done, you must not skip any of the steps!

  1. Consult the for the HMPID Chamber - Equipment description HMPIDMapping.pdf or check the Equipment map.
  2. In the ALICE DAQ window click on the Select Equipment button.
  3. Tick in/out the equipment(s) you want to select/deselect.
  4. Commit the changes.
  5. Calculate the bit map corresponding to the selected equipments: 1 - selected, 0 - deselected. (HELP note: in the output of fs there is E instead of 1 and D instead of 0). For example it might be 11101101111111: every equipment except 1546 and 1543 are enabled.
  6. Calculate the hexadecimal number corresponding to the bit map. In our example it will be "3b7f".
    HELP There is a script that you can use on aldaqacr28 and aldaqacr29 in the hmpid home directory, to calculate the hexadecimal number:
    • Name: hmpBusyMaskCalculator.sh.
    • Usage: ./hmpBusyMaskCalculator.sh 11101101111111 (of course use the bit map that you need, not this example!).
    • The script will return the hexadecimal value as a string that you have to upload with the fe 0x.... command.
  7. In the alidcscom026 terminal set the mask using the fe command, for example: fe 0x3b7f (ALERT! This is not the right value, you have to calculate it!)
  8. Execute fs and see if you get the right busy bit map.
  9. In the LDC terminals where the equipments belong, issue the rst command.
  10. Wait 30 seconds.
  11. From the HMPID VME Crate window the Configuration->*TTCinit* command.
  12. HELP If you select an equipment which was disabled before:
    • If you have time, take a CALIBRATION run.
    • If you don't have time (e.g. HMPID is in the GLOBAL partition) then in the LDC of the equipment, upload the threshold with the lt command.

If the LTU / VME Crate panel doesn't respond

  1. Try to close the xterm where the panel was started from and open a new one and log into alicedcscom026 and try starting it again.
  2. If it doesn't help, ask help from the CTP (trigger) shifter - probably they have to restart the HMPID LTU proxy!

Data Quality Monitoring (DQM)

Starting the DQM

  1. Log in to aldaqdqm08 as user hmpid. This is the dedicated monitoring machine of the HMPID.
  2. You should see this:
    hmpnewam000.png
  3. Open two xterms from aldaqdqm08.
  4. In one of them start the AMORE agent:
    • for STANDALONE run monitoring execute startHmpNonQaAgentGdcLocal
    • for GLOBAL run monitoring execute startHmpNonQaAgentGdcGlobal
  5. ALERT! Note: there should be only one HMPID AMORE agent running. If you try to start a second HMPID AMORE agent it will crash!
  6. In the other xterm execute startHmpAmoreUi to start the HMPID AMORE User Interface.
  7. You should see this:
    4hmpmanual0010.png
  8. Change the refresh rate to 5 seconds (type 5 in the box in the bottom left corner of the UI with 30 in it and hit enter) and press the Start button.
  9. If the AMORE HMPID agent is running, after every monitoring period, the UI will receive the data and refresh itself, e.g. every 5 seconds.

Monitoring with HMPID AMORE UI

Monitoring during a STANDALONE run after a CALIBRATION run

  1. Check the Event Display if the zero suppression works. You should see an almost empty Event Display like this:
    hmpamore001.png
  2. If the zero suppression is working in the Int PadMap and the Noise Map you have to see a smooth distribution in the chambers, like this:
    • Int Pad Map (Integrated Pad Map)
      hmpamore005.png
    • Noise Map
      hmpamore007.png
  3. Check the number of pads distribution on the "nppe" tab. For a chamber (2 DDLs) the average has to be around 50:
    hmpamore009.png

Monitoring during a Cosmic Run

  1. Check the displays: Event Display, Int PadMap, Noise Map and also the Number of Pads ("nppe") as in the STANDALONE runs.
  2. Check the charge distribution:
    hmpamore008.png
  3. Since in the cosmic runs we expect few charged particles, the distributions have to peak around 3-4 ADC and there are almost no entries in the high ADC region.

Detecting problems with the HMPID AMORE UI

Common problems

What you see? Possible cause
The Event Display has lots of pads fired. The zero suppression is not set (or you are running CALIBRATION run).
The Event Display shows some filled boxes or lines like in picture below. Thresholds are not loaded properly. Load them again!
In cosmics run, the charge distribution (PadQ tab) has a broad distribution. Noisy channels are not properly masked.

Reporting problems

  1. In each monitoring tab on the right panel you can find the Save button.
  2. Click on the Save button to save the ROOT canvas of the actual monitoring tab.
  3. Select file type "eps".
  4. Save the display to the hmpid home directory as: year_month_day_RunNumber_FirstName_LastName.eps
  5. Make a note in your shift report, shortly describing the problem and giving the name of the eps file(s).

Remarks on HMPID AMORE UI

  • Since the HMPID AMORE UI is under development, the Pedestal tabs don't work yet. If you click on them, you will crash the UI. Kill it and restart it.

Appendix

List of acronyms

CTU Central Trigger Unit
DAQ Data AQusition
DCA Detector Control Agent
DCS Detector Control System
DDL Detector Data Link
DQM Data Quality Monitoring
FSM Finite State Machine
Edit | Attach | Watch | Print version | History: r8 < r7 < r6 < r5 < r4 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r8 - 2009-12-03 - AndrasAgocs
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright &© 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback