https://atlasop.cern.ch/partlogs/

Topo FW reload

• https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneTopoTrigger#How_to_load_a_FW_version_in_P1

• Reverting Topo FW Procedure if run is ongoing:
  • stop the atlas run
  • kill L1calo partition
  • reload fw
  • Restart L1Calo (IGUI option just above Restart TTC partition)
  • Put L1Calo back IN (the Kill automatically takes it OUT)
  • Start a new run

Surface Test Facility STF

• 0004-S-054, quiet room: 0004-S-049
• https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneCaloCernSurfaceTest
• atlas-tdaq-lvl1-calo-stf
• JIRAS: https://its.cern.ch/jira/issues/?jql=project%20%3D%20ATLLONECAL%20AND%20component%20%3D%20STF
• As in Oct. 2020 Murrough did setup in the current login scripts if run on the TDAQ testbed to sets L1CaloSTF as the default partition if you login to a machine whose name starts with pcl1c or sbcl1c and uses CernTest otherwise.

Oncall shifts

• https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloExpertResponsibilities

Power down instructions:

• https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneCaloCernSurfaceTest#STF_Power_down_Actions

Power up

• https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneCaloCernSurfaceTest#STF_Power_up_Actions
• when inserting handles for the modules they will be powered up when the shelf is up. Need to disable them via shelf manager: clia boards check what is in the shelf, clia deactivate board X to power a board down. X is the physical slot.
• ALTI setup. Login the sbc for the alti, type the lines to set it up. go on a stf computer and run the partiton. One time it got a bus error.

https://atlasop.cern.ch/elisa/display/439217

Instructions from Rhys: see mail called "ALTI reset"

• Paul's monitoring plots: after a pwer cut the crono jobs are stopped. Need to restart them: scritps are here /afs/cern.ch/atlas/project/tdaq/level1/calo/www/html/dcsmon

Hardware

• PPM # 22 in slot # 8
• Wiener Crate VME64x http://file.wiener-d.com/documentation/Crates_6000/WIENER_Series_6000_Manual_A2.pdf
  • Control of the Power Supply 6021 via Fan tray -> Hold the POWER and MODE switches up simultaneously until the display shows„Config: Wait....“ and „Config: Ready !“. Then release both switches.

TREX tests at STF and 104

• connect to pcatkip07 (desktop below the PPM crate)
• login into the sbc:

      pc-tbed-pub
      sbcl1c-stf-00
     

Bit Files

• Tigran's area: /eos/user/t/tmkrtchy/phase1/bitfiles or https://cernbox.cern.ch/index.php/apps/files/?dir=/__myshares/bitfiles%20(id%3A223125)&
• To be taken from KIP and saved in lxplus (for pprotest upload via sw) or in the desktop (for vivado upload on the flash memory)

  ssh -p 122 franchin@portal.kip.uni-heidelberg.de 
  rooster
  cd /data/barn01/PPrArea/bitfiles/
  scp ... franchin@lxplus.cern.ch::PPROTESTxxx/bitFiles/.
  

PPROTEST

• download from git KIP the needed pprotest version ( Feb 19: https://git.kip.uni-heidelberg.de/Atlas-Trigger-Software/PPROTEST, Nov19: https://git.kip.uni-heidelberg.de/Atlas-Trigger-Software/PPROTEST_v1/tree/develop), (firmware repository here: https://gitlab.cern.ch/atlas-l1calo-ppr-firmware)
• look at dependencies in the documentation, which bit file goes for which pprotest version: ( OLD, sping 2019 https://git.kip.uni-heidelberg.de/Atlas-Trigger-Documentation/FirmwareDocs/blob/master/Notes/firmware_dependencies_phase_1.md) (Nov 19: https://git.kip.uni-heidelberg.de/Atlas-Trigger-Documentation/FirmwareDocs/tree/master/Releases_Trex_v1.0)
• setup_pprotest_TR (or other functions, look in .bashrc) and make
• change module configuration in PPROTEST in dbFiles/crateConfiguration/cern_crate.cfg (in Feb 19 there is only one PPM and TREX in slot 8).
• use ./PROG-FPGAs to program ExREM, PREDATOR and DINO

Monitor Temperature, Voltages

 ./MONIT_TVIalarms  

•  ./MONIT_TVItoFile  

  creates a file with T, V I

•  ./HIST_TVIfromFile 

  (better to do it from lxplus and not from sbc) reads the log file and creates root file with plots

VIVADO, program flash memories

• installed in my laptop Vivado lab edition and in pcatkip07 Vivado web server.
• one can connect through the laptop via ethernet to the Vivado web server in pcatkip07 to program flash memories
  • first configure via pprotest the REM, DINO and PREDATOR with any FW.
  • Check if USB cable is connected to TREX and enable one of the two FPGAs, either PREDATOR or DINO with ./PROGR_TrexFlashMems and chose USB-JTAG loading (2) for DINO, (5) for predator
  • connect as root in pcatkip07
  • cd
  • type Victor's script:

     ./start_vivado_hwserver.sh 

    -> this opens the port 4241, need to be doe only once, if the PC is not switched OFF it will stay open
  • Open Vivado HW Manger in the laptop -> Connect to remote server -> hostname: pcatkip07, port 4241 -> should see the FPGA enabled with the PPROTEST
• see the FPGA in the list, right click -> Add Configuration Memory Device -> put parameters seen in these screesnhots: http://franchin.web.cern.ch/franchin/L1calo/STF/TREX_flash_program/

Test Rig

• Official twiki: https://twiki.cern.ch/twiki/bin/view/Atlas/LevelOneCaloCernTestRig

•  ssh -X pc-tbed-pub

  , setup:

  source .bashrc,   setup_l1calo_TR

• database configuration (to be copied from the official one, usually written by Bruce and that can be found here:

   /tbed/oks/tdaq-06-01-01/l1calo/partitions/CernTest.data.xml 

  and to be copied in my se4tup here:

   /tbed/user/franchin/oks/tdaq-06-01-01/l1calo

•  runme 

  starts the Gui graphic interface with the L1 info taken from database defined in the setup_l1calo_TB

•  ssh -X pcl1c-ros-01  /dsk1/l1c/rosData/  (rosData/data  old) 

  where test data are saved
• To look at raw data:

   dumpl1calo -P -d /dsk1/l1c/rosData/data_test.1467018834.calibration_.daq.RAW._lb0000._ROS-01._0001.data   

• Test Rig in Building 4: https://twiki.cern.ch/twiki/bin/viewauth/Atlas/Lab4Testbed

Shutdown Test Rig (Bruce's instructions)

• Turn off all 4 VME crates, and the NIM-bin. Topo is already off.
• sudo /sbin/shutdown -h pcl1c-pm-01. That is the top server in the rack on the left.
• sudo /sbin/shutdown -h pcl1c-ros-01. That is the server in the middle of the right rack.
• shutdown sandbox DCS machine, (server middle of the left rack: push and hold button.)
• shutdown two box PCs at the bottom of the left rack: push and hold front-panel buttons. (Yuri's jtag server, right. DCS for legacy test-rig, left)
• turn off all the breakers at the back wall behind the racks.
• turn off the air-conditioner (front panel - widget at bottom of display.)

In Goldfish bowl:

• shutdown two DELL mini-pcs. I did already, actually, but if Uli is around he might reboot: CTRL-ALT-F1 or F5. Then CTL-ALT-DEL
• if box PC is on, please shut-down (push hold power button).
• I usually pull out the power from the wall - window side left.

ACE

example of output example of output

[franchin@pcl1c-ros-01 ~]$ dumpl1calo -P -d /dsk1/l1c/rosData/data_test.1467018834.calibration_.daq.RAW._lb0000._ROS-01._0001.data | less
Reading from file: /dsk1/l1c/rosData/data_test.1467018834.calibration_.daq.RAW._lb0000._ROS-01._0001.data
Event: 0  16777216
RDO Object Type: PpmFadc  Crate: 0  Module: 8  Eta: 0  Phi: 0  Layer: 0
      Values:  10 f f f f       Flags:  0 0 0 0 0 
RDO Object Type: PpmFadc  Crate: 0  Module: 8  Eta: 0  Phi: 4  Layer: 0
      Values:  14 13 14 14 14       Flags:  0 0 0 0 0 
.................................    same for all 7000 channels, 5 ADC slices and then the LUT...
RDO Object Type: PpmLut  Crate: 0  Module: 8  Eta: 0  Phi: 0  Layer: 0
      Values:  3030000       Flags:  100 
      LutCP: 0 LutJEP: 0 PedCorr: -253 
RDO Object Type: PpmLut  Crate: 0  Module: 8  Eta: 0  Phi: 4  Layer: 0
..........................................

CleanUp a Partition

•  
   pmg_kill_partition -p CernTest
   rmgr_free_partition_resources -p CernTest
  

Monitor PPM Temperature Voltage

example of output for PpmTempMonitoring example of output

Monitoring

• Asymmetric TOB distribution, more occupancy on C side than on A side.. That's because sliding window goes on the left, more probable to have an hit on the left than on the right

Mistimed stream monitoring

• root files stored in EOS: /eos/atlas/atlastier0/rucio/data18_13TeV/physics_Mistimed/00348354/data18_13TeV.00348354.physics_Mistimed.merge.HIST.x554_h295/data18_13TeV.00348354.physics_Mistimed.merge.HIST.x554_h295._0001.1
• Martin's twiki with PDFs: https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneCaloMistimedMonitoring
• Martin's explanation: The mistimed stream is composed of several HLT triggers which run on each event. Out of these HLT triggers, we are only interested in one, which is monj400. This particular HLT trigger selects all events in which J400 fired one BC before or after the actual event, by looking at the CTP information. In Tier0 we have an athena monitoring sub-task which processes the mistimed stream events, and analysis all events taken by monj400 in more detail. Merged histograms are produced. Then a cron job checks only if these histograms are available and makes plots and posts to the twiki.

CTP Per bunch monitoring: /eos/atlas/atlascerngroupdisk/det-ctp/PerbunchMonitoring/2021

Mistimed Events in 2015

• Twiki: https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneCalo2015MistimedEvents
• Matin's talk: https://indico.cern.ch/event/503447/contributions/1180841/attachments/1236401/1815645/l1c_160301_openEB.pdf
• Trigger presentation: https://indico.cern.ch/event/536523/contributions/2183913/attachments/1281968/1905109/TriggerOps_ATLASWeekly_AStrubig.pdf
• Rainer presentation: https://indico.cern.ch/event/434823/contributions/1080859/attachments/1132872/1619897/150728_ATLASWeekly_L1Calo_stamen.pdf
• Alan: https://indico.cern.ch/event/406235/contributions/972072/attachments/1208661/1762114/L1Calo-Status-Report.pdf

Look at Raw Data

l1calomap

• Twiki page with informations here: https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloMappingTool

• Calibration raw data. From the trg pc type the alias

   cal17  

  (alias for pc-tdq-calib-17) and go to look for the raw data in

   /data  

  (from 2017 they are also copied in a common area visible from all P1 pc:

   /net/pc-tdq-lfs-acr/clients_data/scratch/l1calo/data/   

•  dumpl1calo -d data_test.1447413464.calibration_L1CaloTest.daq.RAW._lb0000._ROS-00._0001.data -P | less 

  where the data are the one taken by me with a test in the testrig (playback option, with value 80 in each channel, 1000 events). here I see all the value of the 5 adc of raw data for each event. at the end of the file there are also LUT data 9in this case empty)..

•  event2mapxml -d data_test.1447413464.calibration_L1CaloTest.daq.RAW._lb0000._ROS-00._0001.data  

  create an xml file for the raw data file

•  event2mapxml -k dcm  

  TO dump events from a current running run. Once inside the event2mapxml I can ship N events with -N or add N events in the same file with +N
• event2mapxml -p ATLAS -N 62 -k dcm -t 0x84 -e -E -s 100 -w 0 -R0

•  l1calomap.sh event.map.xml & 

  open with the graphical interface the file event.map.xml that has been created with the event2mapxml command.

  Dataset-->Local-->Open (eventmapxml  file) 

  . And for opening the histograms--> Chose raw, columns and Update histo and Local/event/fadc. (in the map i can look at one particular fadc slice with: Dataset-->Local-->event--> and Fadc0-15 (to swithc between one and the other I can just type the ">" and "<" with the keyboard)

Show ACE screenshot Show ACE screenshot

• rmHistogrammer: to check plot of rate (LUT_CP) of a particular run for particular L1channels (seen with a script who tells you particular noisy channels) rmHistogrammer (to be run on mon06 - ssh pc-l1c-mon-06 (was 04 in 2016))

  rmHistogrammer -1 '15/07/2015 19:15' -2 '15/07/2015 20:15' -i c0100101 -i 0ba04000

rmHistogrammer -1 '13/03/2016 16:00:00' -2 '14/03/2016 10:00:00' -i 06120803 -d /l1c/rates -l -f 'png'   

• for more than one histo in the same plot, just add -i and the TT:

rmHistogrammer -1 '31/05/2016 07:00:00' -2 '31/05/2016 16:00:00' -i 01130f00  -i 01130f01 -i 01130f02 -i 01130f03 -i 01130e02  -d /l1c/rates -l -f 'png'

• Once the plot has been created copy it in lxplus to display it. Go in lxplus and then:

scp atlasgw:/path/exact name .

• Once can use the same command also to retieve rate plots. To know which item correnspond to which ID, go on the oncall page, SysRates oncall page tells you the ID to use for all the L1Calo rates. Just hover the mouse over the rate and the ID shows in the top right corner. (see attachment)

Example Example

* ID rates:

calculate LUT result from raw data

• First etract the 5 ADC values

•  ./ppmchansim.py -d "$L1CALO_DB_CONNECT" -f Physics -c 0x011e0b01 -m HighMu -p 1 35 53 76 67 57  

Dead Channels not in the twiki:

• 1100000: https://atlasop.cern.ch/elisa/display/274889
• lista di canali che nell'online non sono presenti: https://atlasop.cern.ch/elisa/display/259945
• EBC in emergency mode ??? https://atlasop.cern.ch/elisa/display/268883
• killed Tile tower: 06100d03 https://atlasop.cern.ch/elisa/display/278043

Debugging session with Calorimeter Pulses

• Ask to calo people to start a run with pulses
• Open the L1CaloStandalone or L1CaloCalibration partition (with setpart) and set the following values: DefaultDSSTrigger, PhysicsRunPars, Receiver Gain 1, No Generator. The need to open the calibration partition is because receivers are not included in the standalone one, if last operation done with the calibration partition was a DAC scan it has as last step the Gain 0 option and then it is remembered and we will not see anythign coming from the calorimeter, so we need to open the calibration partition and put Gain 1.
• While the L1Calo partition is running, open l1calomap and click on: IS--> Partition--> Select the L1Calo partition that is running, then IS--> PpmRates-->Tower_C or Tower_J and I see the pulses rates. If I want the histograms I can open them

L1CAlo Special RUns

• job option for our derivations: https://svnweb.cern.ch/trac/atlasoff/browser/PhysicsAnalysis/DerivationFramework/DerivationFrameworkL1Calo/trunk/share/L1CALO1.py
• (in 2020) https://gitlab.cern.ch/atlas/athena/-/tree/master/PhysicsAnalysis/DerivationFramework/DerivationFrameworkL1Calo/share
• L1CALO1: applies cuts (>36 ADC counts) and trows away empty towers (smaller files, broken in 2017). Used e.g. for HI 7+1 runs (maybe for special runs where we want physics_Main?) (estimated dimensions (MartinK): 50-200 kb/ev
• L1CALO2: every event taken for each TT, Mistimed stream and L1CaloCailb stream (when run). used for special run for filters and noise cuts (estimated dimensions (MartinK): 300-400kb/ev)
• L1CALO3: DESD_EGAMMA
• L1CALO4: DESD_CALJET
• L1CALO5: DESD_PHOJET
• L1CALO6: DESD_SGLEL

• Some useful special runs for filter coefficients (Sten) data17_13TeV.00335302.calibration_L1CaloCalib.merge.DAOD_L1CALO2.c1144_m1873 56 188452017415 2019-09-13 16:55:44,

Some Software Stuff

From Martin's mail....

Splash Events

• 2018RC ATLAS twiki: https://atlasop.cern.ch/twiki/bin/view/Main/BeamSplashSetup2018
• L1Calo expert manual: https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloOnCallManual#Configuring_L1Calo_for_Beam_Spla
• 2018 configuration: https://atlasop.cern.ch/elisa/display/370668

Calibration runs

• Runs taken by shifters, esplanatin and instructions here: https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloCalibrationShifter
• Offline analysis explained in L1CaloCalibrationSilvia twiki page
• Tile PMT scan automatic output here: https://pcata007.cern.ch/tile/l1calo/
• Ramp Maker results CAF2 in web page: http://atlas-l1calo.web.cern.ch/atlas-l1calo/calib/Caf2Test.php
• Don't run the takecalb script from atlas-pub pc but from the physically located in the CR pcs otherwise we get error in publishing online analysis results on web, as Phos4 scan or dac-ped

Raw data saved in eos: eosmount foldername, later do unmount

 /eos/atlas/atlastier0/rucio/data16_calib/calibration_L1CaloEnergyScan/00288331  .

  /eos/atlas/atlastier0/rucio/data21_calib/physics_Main/00395154/   .

 xrdcp -r root://eosatlas//eos/atlas/atlastier0/rucio/data16_calib/calibration_L1CaloPprPhos4ScanPars/00301316/data16_calib.00301316.calibration_L1CaloPprPhos4ScanPars.daq.RAW /tmp/franchin/. .

Calibration account

• in lxplus, user: l1ccalib (where also all CAF2 results are stored is in: w0/DaemonData/jobs/CAF2/OutputCondor

Standard calibration (DAQ, PED) expert mode

Calibration Panel Expert Mode Calibration Panel Expert Mode

• Screeshot calibration panel expert mode:

DAQ Panel Pedestal Run DAQ Panel Pedestal Run

• Screeshot pedestal run:

cd /det/l1calo/scripts/takecalib/   
./takeCalib.sh

/det/l1calo/takeCalib/tdaq7

It openss the calibration Panel. If you want to validate the scans, be sure to go in L1Calo Expert mode and start from there the DAC and PED scan (it takes ~ 70 min for both scans). If you validate the runs be sure that the calorimeters are quite and are not doing the calibration as well, otherwise they might send some unwanted pulses. you only have 5 minutes of time between the pdf opens and you validate the run (and authomatically updates the DB). Otherwise it need to be done by hand. When the DB is updated with the new DAC scan the pedestal run should already take the new DAC values in order to set the pedestal to 32 ADC counts. If there is no time in between interfills you can validate PED only, but if you validate DAC be sure there is enough time to validate also PED. DAC scan has 59 steps (the number of steps can be seen in the run panel, L1Calo panel, RodMon)

Saturated Tile CIS scan

Saturated LAr pulse scan

/det/lar/project/shiftarea/710.41.00/LAR_PARTITION/v12r0/Calibration/L1Calo

/barrel/L1MultiDac-001/parameters.dat
9
8 16 40 80 120 160 270 350 420
1
0
1
ffffffff ffffffff ffffffff ffffffff

Saturated files are (See what files are used in /det/l1calo/takeCalib/L1CaloRunTypes.py

more L1MultiDacSat-Taylor/parameters.dat
10
900 1100 1300 1500 2000 3000 4000 5000 6000 7000
1
0
1
ffffffff ffffffff ffffffff ffffffff

Phos4 scan

• Run1 Phos4 chip: https://cds.cern.ch/record/405081/files/LHCC-98-36_307.pdf *LAr (Tile) phos4scan need LAr (Tile) partition to be used: LArgL1CaloCombined. Calibraion panel --> L1Calo--> Expert mode--> LAr Phos4scan (~ 50 minutes)
• Used for timing calibration of the calibration chain (pulser run). Send a pulser signal and move in in time untile to find the right setting to sample the peak with the cental adc slice.
• It will take four (historically) so-called PHOS4 run, which scans the L1Calo input timing per trigger tower in approximately 1 ns resolution, on LAr side pulsing only one single layer at a time.
• Every time the analysis is perfomed a file with output calibrations is produced but not updated in database (need to do it by hand) but not so often, pulser timing is pretty stable
• For timing of physics pulses need physics data and Claire's analysis...

Screeshot DAQ panel for Phos4 scan Screeshot DAQ panel for Phos4 scan

• Screeshot calibration panel expert mode:

• Martin's eplanation of the scan: The nMCM has 24 delays to cover one BC i.e. 25 ns. The fine timing scan starts with a delay of 0 for all towers, and takes a certain number of events for this step. This ‘certain number’ is either defined via a time (assuming fixed rate) or number of events (assuming low enough rate to be able to poll fast enough). Which one is used here we need to check in the COOL run parameter folder. With a fixed delay (step) we digitise the incoming analogue pulse at fixed locations every 25 ns (okay, 12.5 for the 80 MHz but let’s leave that away), i.e. always at N*25ns + step * 1.04ns (where 1.04 is 25/24). With the next step, you digitise the analogue pulse at 2.08 ns etc, until with step 25 you are back to the full BC. I.e. in this way we scan the pulse in steps of 1.04 ns. What the analysis does, it calculates the average of the ADC values and puts them into the histograms shown in the PDF: the x axis shows here the step number modulo 24 so to say (or maybe it shows the transformation to ns, can’t remember).

Layer Phos 4 scans

• checkScan.py: The script needs to be told where to find the data files. For some historical reason the default is not where the calibration partition copies them to (its probably where they are on the event builder PC). So what you need is something like:

  cd /det/l1calo/scripts
  for run in 372234 372235 372236; do ./checkScan.py -s /net/pc-tdq-lfs-acr/clients_data/scratch/l1calo/data -n -r ${run}; done

• Script to update the run summaries produced by takeCalib with the analysis results information

  > /det/l1calo/takeCalib/updateRunSummary.py -r 372234,372235,372236
  

• copy the root file that has been created from P1: /www_ALL_data/l1calo/calib/pho/ to lxplus
• analysis the TBB delays with Martin's script:

http://franchin.web.cern.ch/franchin/L1calo/Phos4/analysis_P4LayerScansMartin.py

LAr pulses, trigger monitoring tool

setpart L1CaloCombined
setpart LArgL1CaloCombined
runme

Otherwise one can also monitor the PPM rates in the L1CaloStandalone partition but settings the right parameters in L1CAlo panel (parameters copied from the LArgL1CaloCombined partition:

L1Calo Run Type = Default
Run Parameter Set = PhysicsRunPars
Readout Configuration = Default
LTPI = SlaveToLAr
Trigger Menu= trigconf-default
Calibration Data = calib-oracle-atonr-trigger-r2
Conditions Data = calib-oracle-atonr-trigger-r2
receivers gain strategy = CalibGainsEt

Run analysis

xrdcp -r root://eosatlas//eos/atlas/atlastier0/rucio/data16_calib/calibration_L1CaloPprPhos4ScanPars/00301317/data16_calib.00301317.calibration_L1CaloPprPhos4ScanPars.daq.RAW/data16_calib.00301317.calibration_L1CaloPprPhos4ScanPars.daq.RAW._lb0000._SFO-1._0001.data /tmp/franchin/.

setup_l1calo
source source .localOKS_bashrc 
 analysePpmRun -d /tmp/franchin/data16_calib.00301317.calibration_L1CaloPprPhos4ScanPars.daq.RAW._lb0000._SFO-1._0001.data -r 301317 -a PprPhos4ScanPars -o /tmp/franchin/ -f Calib2
plotCalib -f /tmp/franchin/PprPhos4ScanPars.Calib2.00301317.root -a PprPhos4ScanPars -n 15 -i Calib2

Setup OKS database in lxplus

May 2020: OKS tdaq9 database: https://atlasop.cern.ch/cvs/viewvc.cgi/tdaq-09-00-00/combined/partitions/L1CaloCombined.data.xml?view=markup

• OKS GIT instructions: https://gitlab.cern.ch/atlas-tdaq-software/oks/-/blob/master/doc/RELEASE_NOTES.md

Instructions from Martin: in order to set up a (fake but correct) oks database outside P1, the easiest is to download a real P1 oks database from the archive. For those all single oks files are merged into one big one. Murrough has a note with the link on a twiki. It is reachable from the L1Calo twiki under ‘databases’. There you want to follow the ‘web interface’ link and the instructions by Murrough. https://twiki.cern.ch/twiki/bin/view/Atlas/LevelOneCaloDatabases#OKS_Archives Regarding what archive to download, it is not really important as long as all PPMs were included in the run. DAC scan, pedestal run and PHOS4 scan analysis do only care about PPM readout. So usually I choose a recent L1CaloCalibration archive to download. For L1CaloStandalone you never know what was going on. After downloading, you have to copy and unpack the tar file to your personal oks folder. If we followed my setup it might be at ~/l1calo/dbPoint1/oks/. So in this particular case it would then be ~/l1calo/dbPoint1/oks/tdaq-07-01-00/l1calo/partitions. You can look up that path also in your .localOKS_bashrc (or whatever we named the file you have to source before running the analysis on lxplus). In your new database folder you have to create a soft link with name L1CaloCalibration.data.xml pointing to your downloaded L1CaloCalibration..data.xml file. Finally you have to adjust the database connection string which differs inside and outside P1. For that I think it should be okay if you simply replace all occurrences of ATONR_COOL with ATLAS_COOLPROD.

Hope that will work … Cheers, Martin

Calibration processing code

https://gitlab.cern.ch/atlas-l1calo/CalibrationProcessing

CPM Input scan

• taken from the calibration panel (few minutes) If there are errors about the CMX doesn't matter (becaue the CPM changes the clock). Results can be found in the crate (e.g. ccc02 in /logs/tdaq-06-01-01/L1CaloCalibration/results)

CP Run2

JeM Input scan

• again taken from calib pael. Same as befre fr CMX errors. Results from the L1Calo twiki, Calib --> Status --> and i can find the log file or the pdf with results. Red is the band forbidden. Blue bands are missing chanels

Calibration

• Silvia's twiki page with calibration stuffs: https://twiki.cern.ch/twiki/bin/view/Main/L1CaloCalibrationSilvia

Tile CIS

• Tile calibration updates link
• Tile calibration System (Arely Nov 17): https://indico.cern.ch/event/678290/contributions/2784913/attachments/1569370/2474819/TileCalibration_arelycg.pdf
• Tile calbration energy reconstruction: with RampMaker we use UPD4
• Tile laser scan results: http://pcata007.cern.ch/commissioning/level1/
• Mapping PMT to Trigger tower: http://hep.uchicago.edu/atlas/tilecal/level1/geometry.html

Sanya's instructions:

asetup 21.0.71,Athena
ReadCalibFromCool.py --folder=/TILE/OFL02/CALIB/LAS/LIN --module=LBA10
--chan=20 --gain=0 --pmt --begin=343000

TileCalibTools : INFO     Resolved globalTag 'CONDBR2-BLKPA-2018-06' to
folderTag 'TileOfl02CalibLasLin-RUN2-UPD4-16'

(341222,0)  LBA10 pm 21 ch 20 LG    0.944506  646.900024
(345808,0)  LBA10 pm 21 ch 20 LG    1.000000  -1.000000
(347606,0)  LBA10 pm 21 ch 20 LG    1.017112  646.900024
(348534,0)  LBA10 pm 21 ch 20 LG    1.013614  646.900024
(349137,0)  LBA10 pm 21 ch 20 LG    0.989601  646.900024
(349712,0)  LBA10 pm 21 ch 20 LG    0.968799  646.900024
(350144,0)  LBA10 pm 21 ch 20 LG    0.957218  646.900024

ReadCalibFromCool.py --folder=/TILE/OFL02/CALIB/LAS/LIN
--tag=CONDBR2-BLKPA-2018-03 --module=LBA10 --chan=20 --gain=0 --pmt
--begin=343000

TileCalibTools : INFO     Resolved globalTag 'CONDBR2-BLKPA-2018-03' to
folderTag 'TileOfl02CalibLasLin-RUN2-UPD4-16'

--folder=/TILE/OFL02/CALIB/CIS/LIN
--folder=/TILE/OFL02/CALIB/CES

--tag=UPD1   or --tag=UPD4

Check masked Tile channels

ReadBchFromCool.py --tag=UPD4 --module=LBC56 --chan=10 --pmt --begin=300000

ReadBchFromCool.py --module=LBC56 --pmt --begin=300000 | grep -v good

http://zenis.dnp.fmph.uniba.sk/tile.html

Receivers

• Twiki page receivers https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloReceiversExpert
• Receiver control is via legacy USB 1.1, not supported by modern servers. In 2019 installed Raspberry pi controllers to provide smart USB hub services acting as the bridge from ethernet to receiver USB.

HV correction

https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloOnCallManual#Updating_Receiver_gains_and_LAr

https://indico.cern.ch/event/638444/contributions/2639266/attachments/1490522/2316697/bracinik10jul17_calibStatus.pdf

https://cds.cern.ch/record/830849/files/ATL-COM-LARG-2005-003.pdf?version=1

Change receivers gains

• with the new sqlite file from juraj: (instruction here: /det/l1calo/doc) *Juraj provides files (update: from analysis authomatic, force: values put by hand because problematic channels). to be copied in

   /det/l1calo/coolData/gains

   cd /det/l1calo/coolData/gains

   ../../scripts/updateGains.sh updateLAr_mar16_v1.sqlite

• To revert or change some channel, go in P1, setup l1calo, ace, open DB, search for receiver folder and look for the channel (name of receiver channels, not TT).

Show ACE screenshot Show ACE screenshot

Monitor receivers inputs

https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloReceiversExpert#How_to_configure_the_Monitoring

Change Firmware

PPROTEST OLD VERSION

There is a version already installed in P1 in

 /det/l1calo/ppmTools/pprotest 

CALIPPR FW

• Get the new version of PPROTESTatCERN from Jan (if it has been changed recently), copy in lxplus and compile it (after setup_pprotest)
• compress the folder, copy in P1 and decompress

 tar -zcvf archive.tar.gz directory/   

 tar -zxvf archive.tar.gz  

• Jan's twiki here: https://wiki.kip.uni-heidelberg.de/KIPwiki/index.php/Atlas_Privat:JJ_HowTo_P1FWupdate
• bit files are (the most recent) in the folder:

   /afs/cern.ch/work/j/jjongman/public/bitFiles 

  . Take the needed verion and put in the pprotest package in

   bitFiles/ 

• Check in the file:

    ppmServices/src/PpmFpga.cxx  

  at the very end, that the FW version is the wanted one (Currently used End 2016: RemFPGA_6.a.0.bit, Calippr_2.4.5.bit
• If need to come back to another FW version: reput in the cxx file the new bit files, recompile, make tar, move to P1, untar, upload in the sbcs
• Check the currect PPM slot mask in the file

   PPROTESTatCERN/ppmServices/apps/EepromProgramming.cxx 

  
 // Board SlotNR Mask         5  6  7  8  9 10 11 12  13 14 15 16 17 18 19 20
      unsigned char ppmMask[16] = {1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1}; // P1
//    unsigned char ppmMask[16] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 0};    // ZDC P1
//   unsigned char ppmMask[16] = {0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0};    // TestRig
//    unsigned char ppmMask[16] = {0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0};    // TestRig - L1CaloTest Partition only
   unsigned char mcmMask[16] = {1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1};

• enter in P1 network. Create 8 xterm and in each of them log in the sbcs:

   sbc-l1c-pcc-0(0-7) 

  (or use the alias

   pcc0n 

• In each of them:

   setup_pprotest 

  and cd

  PPROTESTatCERN 

•  ./EepromProgramming 

  . It come out a menu like this one:
• to load CALIPP FW:
  • select (1) and give 16, all PPM.
  • select (6) load CALIPP to EPROM, give nMCm number; 16 (all)
  • Insert the magic key (58).
• It starts to program and take ~ 45 min. Once it is done (check it did 16 PPM otherwise the mask was wrong!).
• Once it is done:
  • Select (7): reload CALIPP from EPROM to FPGA.
  • After that one can check that everything went fine:
  • Select again all PPM and do (9). One should read the correct version in each of the 16 PPMs
• Never do the JTAG programming! it takes forever because it does in series ever MCM. (once I did by mistake, need to stop it, load the FW in the FPGA (only in the first PPM ) with 7 and then restart with 6 and then 7
• I I cannot communicate with the VME crate maybe is becuse the crate has been shutdown, I need to reload the REM first (it does automatically when I run the standalone partition, but If I don't do it, I can do with the Eeprogramming)
  • Select (2) : Load REM firmware
  • Load from FLASH memory of from softare? (f/s)? Select s
  • then I can update the Calipp or read the info from the crate

REM Firmware

./FlashProgramming

http://franchin.web.cern.ch/franchin/L1calo/Tests_FW_Jan/REM_ipload/REM_upload_options

Chose the options. The first time you chose all PMTs, then need to first do option (3) that loads the FW in one of the 6 slots. Select option 0: REM and select the wanted slot. (End 2017 in use is slot 5 with REM verson 6.b.0 (in 2016: 6.a.0 in slot 4)). Is OKS who choses which slot to use for the run, the others once uploaded with a version can keep the version for next use. If I do th option (4) I move the FW in the FPGA but this will be reset at the next configire of the run because will read from OKS which oe to use. If I do option (6) I read what is in the FPGA, so if I upload the FW in the FLASH but without moving to the FPGA it doesn't read it. The OKS file can be seen in git here (search for flash-blockN, whith N from 0 to 5). N.B in the pprotest the counting is between 1 to 6 so pay attention of shifting by one what is in OKS wrt to the REM slot! LCD = lvds cable driver, it does some precompensation and fan-out of the lvds signals coming from the nMCMs going out to the CP & JEP, it is the bottom right daughterboard on the ppm

If there are VME errors, try loading the ReM FPGA using the ./EepromProgramming tool and the 'via software' option, then using the VME reset. This brings the modules into a cleanly configured state where you have access to all the memory modules.

NEW PPROTEST.GIT

• Change the crate configuration and mapping in this file:

   dbFiles/crateConfiguration/cern_crate.cfg 

• There are some executable files, the ones that start as INFO to display info about one PPM (configured in the previously described file), two called PROGRAM (PROGR_McmEeproms for programming CALIPPR FW and PROGR_PpmFlash for programming he flash with the REM firmware). Follow the instructions, see instruction example below. Before reading info is better do a VME_Reset first.

File contents File contents

PROGR_MmcmEeproms PROGR_MmcmEeproms

PROGR_PpmFlash PROGR_PpmFlash

./PROGR_McmEeproms

http://franchin.web.cern.ch/franchin/L1calo/Tests_FW_Jan/LoadFW/PROGRAM_McmEeproms

http://franchin.web.cern.ch/franchin/L1calo/Tests_FW_Jan/LoadFW/PROGRAM_McmEeproms

• REM:

/.PROGR_FPGAs

PPM

PPM Mapping

• EM
• HEC
• Tile
• FCAL: 4 PPMs: (crates 4 & 5, PPMs 0 & 8)
• SPARES: 4 PPMs:

PPM Calibration

• DAC scan The pedestal is adjusted and measured on regular basis by the combination of DAC scans and pedestal runs. First the linear dependency between the DAC setting (0-255) and the ADC value is measured by scanning the DAC range and measuring the ADC output for each point. An analysis determines for each tower the slope and offset of the linear function and stores them in the database. Using these parameters, the DAC setting is determined and loaded at CONFIGURE in order to result in a pedestal of 32. The DAC calibration is needed to compensate for differences of the ADCs.
• PED scan Since the precision of the DAC is coarse than the precision of the ADC (approximately factor of 2) we cannot rely on the DAC calibration alone in order to ‘predict’ the pedestal precisely enough to be used as zero-line in the LUT. Hence in second step after the DAC calibration we take a pedestal run which simple measures the pedestal, histograms it and determines mean and rms.

Hardware

• KIP database: http://www.kip.uni-heidelberg.de/atlas/db/DbPPr/signin2.php

Power cycle single PPM

ppmOff #

L1Calo Readout modes

• Default: 5+1 (7+1 used for filter coefficient calculation, 15+1 80MHz used for timing purposes, special runs)
• COOL datbase with parameters elog
• Deadtime settings for each readout mode: l1whiteboard
• Parameters: NumSamples40, ADC_latency, NumAdcSamples elog
• Event size: (Steve): With the standard five slices each PPM slink has a data size of about 220 words - see for example this:

https://atlasdqm.cern.ch/webdisplay/tier0/1/physics_L1Calo/run_364214/run/L1Calo/ROD/rod_1d_PpPayload

Change Low-High mu settings: AC (high mu), matched filters (low mu)

oncall manual

https://atlasop.cern.ch/oncall/l1calo/L1CaloModStat.php

(Instructions from Martin for 2016 run, how to do manually)

• Loading low-mu matched filters, LUT slopes and noise cuts:
  • (a) First step is usually to upload the three coolinit files to the corresponding results folders. In this case I have done this already so you can directly proceed with the validation in (b). However, when reverting to high-mu settings you will have to do this step with proper high-mu files of course, and instructions are given below. Nevertheless, for completeness and emergencies, the files for low-mu

/det/l1calo/coolData/acfirsglb_physics/single_bunch  

- PprFirFilterResults_Physics_single_bunch.coolinit
- PprLutValuesResults_Physics_single_bunch.coolinit
- PprNoiseCutResults_Physics_single_bunch.coolinit

• • (b) Validation of results. In this step the matching attributes of the specified results folders are copied to the validated folder PprChanCalib (or PprChanExtra, respectively). The latter folders are used to configure the system, so this step will have impact on the performance. Hence, before running the validation it might also be a good idea to dump the previous status of PprChanCalib (PprChanExtra) which could make reversion in case of accidents easier. For this I usually create a 'work' folder. Below an example sequence of commands.

> cd /det/l1calo/coolData/acfirsglb_physics/single_bunch
> mkdir work_201609xx
> cd work_201609xx
> dumpfolder.py -f Physics -i PprChanCalib
> mv PprChanCalib_Physics.coolinit PprChanCalib_Physics_201609xx_before.coolinit

> useCalib -d "$L1CALO_DB_CONNECT" -f Physics -v PprChanCalib -r PprFirFilterResults
> useCalib -d "$L1CALO_DB_CONNECT" -f Physics -v PprChanCalib -r PprLutValuesResults
> useCalib -d "$L1CALO_DB_CONNECT" -f Physics -v PprChanCalib -r PprNoiseCutResults

• II. Loading high-mu AC filters, LUT slopes and noise cuts:

(a) Upload the three coolinit files to the corresponding results folders. The results folders are used for storing calibration results without actually using them. So no changes to the performance of the system are done in this step. The coolinit files with the AC filters can be found in /det/l1calo/coolData/acfir25ns_physics/20160726/acMu29.

> cd /det/l1calo/coolData/acfir25ns_physics/20160726/acMu29
> loadfolder.py -d "$L1CALO_DB_CONNECT" -f Physics PprFirFilterResults_Physics_20160721_ACmu29.coolinit
> loadfolder.py -d "$L1CALO_DB_CONNECT" -f Physics PprLutValuesResults_Physics_20160721_ACmu29.coolinit
> loadfolder.py -d "$L1CALO_DB_CONNECT" -f Physics PprNoiseCutResults_Physics_20160721_ACmu29.coolinit

• (b) Validate the results. In this step the matching attributes of the specified results folders are copied to the validated folder PprChanCalib (or PprChanExtra, respectively). The latter folders are used to configure the system, so this step will have impact on the performance. Hence, before running the validation it might also be a good idea to dump the previous status of PprChanCalib (PprChanExtra) which could make reversion in case of accidents easier. For this I usually create a 'work' folder. Below an example sequence of commands.

> cd /det/l1calo/coolData/acfir25ns_physics/20160721/
> mkdir work_201609xx
> cd work_201609xx
> dumpfolder.py -f Physics -i PprChanCalib
> mv PprChanCalib_Physics.coolinit PprChanCalib_Physics_201609xx_before.coolinit

> useCalib -d "$L1CALO_DB_CONNECT" -f Physics -v PprChanCalib -r PprFirFilterResults
> useCalib -d "$L1CALO_DB_CONNECT" -f Physics -v PprChanCalib -r PprLutValuesResults
> useCalib -d "$L1CALO_DB_CONNECT" -f Physics -v PprChanCalib -r PprNoiseCutResults

PPM COOL ID

PPM test, software patches

• Usually Martin put the patches in P1 here:

    /det/l1calo/ppmTools/patches  

• Using su l1calo copy the interesting libray (renaming it without the extension, in

    /det/l1calo/releases/tdaq-06-01-01/l1calo/pro/installed/patches/x86_64-slc6-gcc49-opt/lib/  

• log files are in each sbc crates (sbc-l1c-pcc-0n) here:

  /logs/tdaq<version>/<partition>  

HDMC (Hardware Diagnostic Monitoring and Control system)

• From the sbc open it with

   dbhdmc 

  and open the correct ppm crate (depending from which sbc it is lounched)
• The igui will open, File --> LoadDBcrate --> Select the ppm crate, ppm, mcm and channel. Open the interested register and click on Read

Show Screeshots Show Screeshots

Make some hot towers via LUT

• login in the correct sbc in P1 while the test run in ongoing (ssh -Y sbc-l1c-pcc-0c)

•  dbhdmc 

  after having setup_l1calo
• chose the right MCM, and channel, and chose either the lutCP or lutJEP. Open. Read values, There is a list of them. take the last value with zeroes and fill with ffff. The point

• It should be suddenly effective and no need to put back. once the PPM is configured in the next run it reads from the DB the correct value
• The tower should become hot in the mapping tool.

Show Screeshots Show Screeshots

Pedestal Corrections

  TRIGGER/L1Calo/V2/Configuration/PprChanDefaults/CR43_PedCorrEnable  

CPM-JEP

Overflows

Event with 6 jets Event with 6 jets

CMX Ovrflows

https://atlasdaq.cern.ch/info/mda/db/ATLAS/365304/Histogramming.l1calo-athenaHLT./L1Calo/CPM_CMX/Input/cmx_2d_tob_Overflow

L1Topo

CTP rates Topo CTP rates Topo

• Validation Topo plots, run from AOD, can be configurble on the fly (from Sara Alderweireld): https://salderwe.web.cern.ch/salderwe/tmp/topotest/

Topo Caldo

connect to pc-l1c-topo-00

either go in /atlas-home/0/simoniel/executables/DANGEROUSTOOLS
or copy /atlas-home/0/simoniel/executables/DANGEROUSTOOLS/topoCaldo.py in your home

for test with LAr no need of any special keys or any special configuration (they already set up ATLAS as they need)

run: python topoCaldo.py
from the menu “Partition” select “ATLAS”
from the menu “Pattern” select “TwoEm90” - when you select the pattern the hot tower will start *automatically* to stop it you need to select “None” from the “Pattern" menu. This process will cause a lot of errors messages to the run control but it is exactly what LAr people want ;)

Not necessary but if you are curious: the time stamp on when you activate any pattern can be monitor from or L1Calo on-call page in the ERS messages filtering for “Undefined” for the application ID

Errors while running

L1Calo webpages links

• Official l1calo known problem page: https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloKnownProblems
• Hardware interventions twikii; https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloHardwareInterventions

some encountered errors

• Nov 17, Stopless removal of ROS-TDQ-CALPP-01, https://atlasop.cern.ch/elisa/display/367662

•  ERROR l1calo-check-status-app cs::BadModuleStatus Bad module status seen for l1calo-cp3-cam

  https://atlasop.cern.ch/elisa/display/281077 problem with relative timing of clocks arriving at input of different boards in the crate. Error appeared in 2016: 19.03.16, 28.04.16
• 100% Busy topo. https://atlasop.cern.ch/elisa/display/301457
• 100% topo busy start of run, TTC restart should help https://atlasop.cern.ch/elisa/display/301713
• TWIKI https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloMrsMessages
• Error cs:BadModStatus seen for l1calo-roc0-rod0. See also in the error plot in tier 0 and online. If the error repeats the shifter should call. Look online ine in the OncallPage

• PP0 not responding and giving stoppless removal. Problem with the sbc. reboot sbc worked. https://atlasop.cern.ch/elisa/display/308361 https://atlasop.cern.ch/elisa/display/308340
• If a monitoring pc dies need to be replaced with another one (there are 4 in total, two are used, two spares) but we need to change in OKS the address. The same if we replace an sbc. Need to learn how to modify oks
• TRP not publishing rate anymore because aplication died. Need to be restarted by run control: https://atlasop.cern.ch/elisa/display/317134
• Calibration panel not starting. Close, exist restart X. https://atlasop.cern.ch/elisa/display/316878
• Noise burst at adjust phase. tipically they will be quet during stable beam. Noise Killer might intervene and kill noisy cells https://atlasop.cern.ch/elisa/display/317096
• Topo busy for ~ 1min due to clock swithches (usually outside stable beam) elog
• Psare PPM causing busy L1CAlo https://atlasop.cern.ch/elisa/display/316026
• jep0-jem03, occasional Parity errors on LVDS link (October 2016) https://atlasop.cern.ch/elisa/display/325741
• check-status on cp0-cam (October 2016) https://atlasop.cern.ch/elisa/display/325739
• Problems with L1Mu2Topo_RCD app: in it living on a TGC PC (pc-tgc-mon-rod-01), but is also related to MUCTPi and L1Topo. If TGC PC is down it gives errors on our side..
• JEM simulation error (1/min). Oct 17. Solved reloading the JEM FW after end f the run. Tolerable defect affecting Ex calculation. https://its.cern.ch/jira/browse/ATLLONECAL-109
• crsh monitor application at end run, https://atlasop.cern.ch/elisa/display/364236

Topo (2017)

• FATAL due to IPbus transition failure (error at CONNECT): 1/2 time per week. to go down and up again with the ATLAS partition.
• l1calo-topo-l1topo1 error (Access attempted on non-validated memory). --> This is not an issue regardless the status of the partition. It likely happend during a lock switch. During clock switch there is a sequence of resets to re-synchronize the various part of L1Calo and L1topo is last, and subjected to a rather long resets: if the monitoring application try to read number (registers), the IPbus would not be responding due the on-going reset. Therefore the error. Topo input/output mismatches with CMX/CTP
• Topo busy and stopless removal at ramp down: https://atlasop.cern.ch/elisa/display/357928
• Everytime Topo RoIB is busy, TTC restart doesn't help, need to reload topo fw (and go to shutdown with the run before)

L1Calo Busy

• HLT backpressure physics related or generated by a very hot tower (look for hot towers TOBs)
  • HLT backpressure should be visible by the Busy Panel RoIB XOFFS and also from pbeast: https://atlasop.cern.ch/tdaq/pbeast_dashboard/#/dashboard/file/Dashboard_ATLAS_Complete

• Sometimes topo get busy at beginning of the run. Solved with a TTC restart

Operation

Dead Times

• CTP tiki with dead times: https://twiki.cern.ch/twiki/bin/view/Atlas/LevelOneCentralTriggerDeadtimeSettings

https://indico.cern.ch/event/708589/contributions/2909405/subcontributions/249519/attachments/1638435/2615082/L1CT.pdf

CTP deadtime and project tag settings:

• New CTP webpage with deadtimes per run: https://atlas-l1ct.cern.ch/pbeast_plots/

For STABLE BEAMS (data16_13TeV): Simple 7, Complex: b0=6/811 (L1Calo ); b1=42/381 (TRT); b2=9/351 (LAr); b3=2/400 (PIX/IBL)

For beam commissioning (data16_comm): Simple 16, Complex: b0=15/370 (L1Calo ); b1=42/381 (TRT); b2=9/351 (LAr); b3=1/400 (PIX/IBL)

For cosmics running (data16_cos): Simple 16, Complex: b0=15/370 (L1Calo ); b1=42/381 (TRT); b2=9/351 (LAr); b3=1/400 (PIX/IBL)

For beam splashes (data16_1beam): Simple 2500 (LAr in 32 samples mode), Complex: b0=6/811 (L1Calo ); b1=42/381 (TRT); b2=9/351 (LAr); b3=1/400 (PIX/IBL)

For high-rate running (data16_comm): Simple 6, Complex: b0=15/370 (L1Calo ); b1=42/381 (TRT); b2=9/351 (LAr); b3=7/260 (PIX/IBL) NB: WITHOUT Pixels

For high-rate (100 kHz) running (data16_comm): Simple 4, Complex: b0=15/370 (L1Calo ); b1=42/381 (TRT); b2=9/351 (LAr); b3=7/260 (PIX/IBL) NB: WITHOUT Pixels

For low-rate ID cosmics running (data16_comm): Simple 16, Complex: b0=15/370, b1=42/381, b2=9/351, b3=1/400; SMK: 2297, L1PSK : 6495, HLTPSK: 4830, BGK: 1003; NB: ONLY WITH Pixels, SCT and TRT (using FastOR trigger); include also other sub-systems

DATABASES

L1Calo Databases https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneCaloDatabases

OKS (with all hardware configurations)

• Rhys command (Oct 2020): from P1 PC:

 dbe -f /atlas/oks/tdaq-09-02-01/l1calo/partitions/L1CaloStandalone.data.xml

To include/exclude any crate (e.g. if one PP crate dies and we need to exclude from the partition, eiher do unning the partition and commit and reload or do with dbe, navigate to the right partition, double click and type the name of excluded crates in "Disabled" and save (you need to know the exxact name you want to disable). Vice versa, to enable it, remove it from the disabled list (right click).

 dbe -f /atlas/oks/tdaq-09-02-01/combined/partitions/L1CaloCombined.data.xml

Show Examples of dbe Show Screeshots

e.g. see here the combined partition list: e.g. see here the names of our crates: e.g. see here the results for ATLAS partition:

here inside there are the combined partitions and LArgL1CaloCombined and TileL1CaloCombined

 dbe -f /atlas/oks/tdaq-09-02-01/combined/partitions/ATLAS.data.xml 

* https://atlasop.cern.ch/twiki/bin/view/Main/L1CaloOksDatabase

• https://twiki.cern.ch/twiki/bin/viewauth/Atlas/DaqHltOks

oks_data_editor filename.xml 

Martin, Bruce modify manually the xmls file. To do it: in P1 create a tmp folder (and then you will erase after having done, to dont do mistakes next time, or remember to update it next time, before modifying with

oks-update.sh 

• For example to change the address of the REM formware version :

oks-checkout.sh /atlas/oks/tdaq-06-01-01/l1calo/sw/l1calo_firmware.data.xml 

  "PpmFpgaProgram" "PpmRemFPGA-flash-block3" 

example example

  

<obj class="PpmFpgaProgram" id="PpmRemFPGA-flash-block2">
 <attr name="BinaryName" type="string">"FlashMemory"</attr>
 <attr name="Description" type="string">""</attr>
 <attr name="Authors" type="string" num="0"></attr>
 <attr name="HelpURL" type="string">"http://"</attr>
 <attr name="VersionID" type="u32">0x60907</attr>
 <attr name="CadProject" type="string">""</attr>
 <attr name="CheckString" type="string">""</attr>
 <attr name="Checksum" type="u32">0</attr>
 <attr name="ChipType" type="string">"XCV1000E"</attr>
 <attr name="FlashRamBlock" type="u32">0x2</attr>
 <attr name="SourceURL" type="string">""</attr>
 <attr name="ProgramType" type="enum">"RemFpgaDefault"</attr>
 <attr name="DeviceName" type="enum">"RemFpga"</attr>
 <rel name="Needs" num="0"></rel>
 <rel name="BelongsTo">"FpgaConfiguration" "ppm-rev6-flash"</rel>
 <rel name="Uses" num="0"></rel>
</obj>

<obj class="PpmFpgaProgram" id="PpmRemFPGA-flash-block3">
 <attr name="BinaryName" type="string">"FlashMemory"</attr>
 <attr name="Description" type="string">""</attr>
 <attr name="Authors" type="string" num="0"></attr>
 <attr name="HelpURL" type="string">"http://"</attr>
 <attr name="VersionID" type="u32">0x60a00</attr>
 <attr name="CadProject" type="string">""</attr>
 <attr name="CheckString" type="string">""</attr>
 <attr name="Checksum" type="u32">0</attr>
 <attr name="ChipType" type="string">"XCV1000E"</attr>
 <attr name="FlashRamBlock" type="u32">0x3</attr>
 <attr name="SourceURL" type="string">""</attr>
 <attr name="ProgramType" type="enum">"RemFpgaDefault"</attr>
 <attr name="DeviceName" type="enum">"RemFpga"</attr>
 <rel name="Needs" num="0"></rel>
 <rel name="BelongsTo">"FpgaConfiguration" "ppm-rev6-flash"</rel>
 <rel name="Uses" num="0"></rel>
</obj>

If we want the new one we will upload it with Jan's code in slot 4, and then wen we want to use it we have to change the fw version in the slot and the line here

  PpmFpgaProgram" "PpmRemFPGA-flash-block4"  

oks-commit.sh -u l1calo_firmware.data.xml   -m "comments that goes in cvs"

• To disable a PPM , checkout the following file:

 /atlas/oks/tdaq-06-01-02/hw/l1calo_crates.data.xml

COOL (with calibration stuff)

ACE config in P1

• How to configure the database opening:

Show Screeshots ACE Show Screeshots ACE

• Full L1Calo chain cools database example :

Screeshot Screeshot

• Example on how to open it : If on wants to see the db changing vs time:

Screeshot Screeshot

• Full PPrChanCalib:

Screeshot Screeshot

Other stuff

• Setting proxy to edit P1 twiki pages:

https://atlasop.cern.ch/twiki/bin/view/Main/EditTwikiFromAnywhere#ProxySetup

• sysadmins twiki: https://atlasop.cern.ch/twiki/bin/view/SysAdmins/GatewayProxySetup

Mainteinance

• Tile parallel elog: https://pcata007.cern.ch/elog/Maintenance/

HLT releases and merge requests

https://gitlab.cern.ch/atlas/athena/merge_requests/12888

Online Software

• Tile towers warnings: /atlas/moncfg/tdaq-06-01-01/combined/triggerMonitor/BeamStatesToIgnore.dat

Online Software

• Oct16 Murrough's presentation CMake, seful links: slides, twiki with instructions

Isolation and Noise Cuts

• Queen Mary Git: https://gitlab.cern.ch/QMUL/TrigEff

LAr recommissioning

• LAr twiki pages: https://twiki.cern.ch/twiki/bin/viewauth/LAr/LS2InstallationAndCommissioning
• shared cernbox folder with TBB results: https://cernbox.cern.ch/index.php/apps/files/?dir=/__myshares/NewTBBdelayCalc%20(id%3A214246)&
• During technical runs we need to ask to leave out our rosses for Lar support: pc-tdq-ros-calpp-00, pc-tdq-ros-calpp-01, pc-tdq-calib-17
• LAr calendar https://espace.cern.ch/LAr-SW/_layouts/15/start.aspx#/Lists/LAR_PARTITION/calendar.aspx

Phase2

• Tile demonstrator installed in 2019 in LBA14

External Documentation

FELIX

• FELIX manual: https://atlas-project-felix.web.cern.ch/atlas-project-felix/user/felix-user-manual/versions/4.0.6/
• ATLAS FELIX PAGE: https://atlas-project-felix.web.cern.ch/atlas-project-felix/
• felix busy handling: https://atlas-project-felix.web.cern.ch/atlas-project-felix/user/docs/BusyXonSpecs.pdf

ALTI

L1Calo Results

• L1Calo Public Results https://twiki.cern.ch/twiki/bin/view/AtlasPublic/L1CaloTriggerPublicResults
• Performance Trigger 2015 https://cds.cern.ch/record/2140103/files/

Interesting talks

• L1Calo Joint meeting twiki; link
• Jan:
  • Feb 2016 Joint: PPM HW/FW and Sat BCID: link
  • Sept 2015 Joint: PPm HW/FW: link
  • June 2014: Joint: HW/FW status: link
  • 2016 high RR school, summary PPM: link
  • 2016 KIP HW: Finding Sat BCID Thresholds: link
  • 2016 KIP Analysis meeting: BCID L1calo: link
  • 2014 Pedestal Corrections: link
• Claire
  • Feb 2016: PPM performance: link
  • Sept 15: BCID 80: link
  • Sept 15: Timing update (40 MHz runs)
  • Nov 16, Sat80 performances: https://indico.cern.ch/event/588576/contributions/2372909/attachments/1373169/2083748/analysiscontribution_17Nov2016.pdf

• Sebastian Feb 16: Mistimed BCID analysis: link
• Stanislav
  • Feb 16: improvedLUT jets: link
  • Oct 15: Miss Et, Jet calib link
  • Nov 15 TGM: Et, jet calib: link

• Timing talks
  • Fabrizio JM July 17: talk
  • Thomas TBB calibration, 2020: talkKIP

• Calibration:
  • https://indico.cern.ch/event/510610/contributions/2143238/attachments/1263029/1868011/250416_summaryCalib_Silvia.pdf
  • MartinK OFC LAr: https://indico.cern.ch/event/772504/contributions/3210032/attachments/1788150/2912115/OFCs_Calibration_EMEC.pdf https://indico.cern.ch/event/777509/contributions/3280846/attachments/1786280/2908374/New_Physics-to-Pulser-Corrections_in_EMEC.pdf

• • 2019 tests with calorimeters: https://indico.cern.ch/event/793457/

• KIP meetings, PPM related talks
  • Francesco BCID pulser runs: link

• Ben ATLAS weekly, L1 general (may 2016): link
• Martin ATLAS weekly Sat BCID (March 2016) link
• Trigger Operation, ATLAS weekly (May 2016) link
• Rainer ATLAS weekly mistriggered events in 2015: https://indico.cern.ch/event/434823/contributions/1080859/attachments/1132872/1619897/150728_ATLASWeekly_L1Calo_stamen.pdf

• Trigger talks
  • https://indico.cern.ch/event/541088/contributions/2201697/attachments/1290872/1922542/20160614_trigger_menu.pdf
  • https://indico.cern.ch/event/517101/contributions/2030723/attachments/1256623/1855315/Menus_20160412.pdf

• General L1Calo Talks
  • Oct 2015 M. Polka: https://cds.cern.ch/record/2059328/files/ATL-DAQ-SLIDE-2015-780.pdf
  • Nov 15. Kate TDAQ week link

• Topo Talks
  • Eduard ATLAS weekly Sept16: link
  • Imma Oct16: link

• LHC Status Summary
  • Martin 5Oct2017: Martin Talk Gruffalo

• Mistriggered events
• Steve: 7June16, 25Nov15

• Filters
• Poster Andrew: http://epweb2.ph.bham.ac.uk/user/daniells/Talks/L1Calo/RALPoster.pdf
• Andrew 2013 talk1, talk2 talk 2012 25 vs 50 ns (2012)

• *Noise Cuts optimization * 2015 Steve https://indico.cern.ch/event/446823/contributions/1110657/attachments/1157457/1664633/lowmute150921.pdf

• FCAL Geometry https://indico.cern.ch/event/165077/contributions/1417357/attachments/198964/279273/l1c_111207_lar.pdf

• MBTS noise https://indico.cern.ch/event/131301/contributions/1350605/attachments/95271/136141/mbtsxtalk110607.pdf

• Mistimed Streams: talk1 , talk2

ATLAS Software tutorial

• Computing tutorial link
• Software tutorial link
• Python twiki link
• Sw basis https://twiki.cern.ch/twiki/bin/view/AtlasComputing/SoftwareTutorialSoftwareBasics#Grid_Certificates

L1Calo Code

• Online weblage ModStat https://gitlab.cern.ch/atlas-l1calo-online/www-atlasop-oncall/-/blob/master/l1calo/L1CaloModStat.js
• https://gitlab.cern.ch/atlas-l1calo-online/iguiL1Calo/-/blob/master/jsrc/l1calo/igui/CpmStatusInfoPanel.java#L168)

is_ls -p ATLAS -n L1CaloStatus -R '.*cp3.*cpm06.*' -v -N

PileUp

LHC Stuff

• LHC orbit = 3564 bunch crossings BC
• LHC bunch crossing Clock (sent via TTC): 40.08 MHz
• BCR= bunch crossing counter reset
• ECR= Event Counter REset
• 8b4e schema (from October 2017): to cope with 16L2 problem, instead of a continuous train of bunches spaced by 25 nanoseconds, this beam consists of mini-trains of eight bunches spaced by 25 nanoseconds and four empty bunch slots. This irregular beam pattern suppresses the formation of electron clouds compared to the standard beam. While the LHC operated with up to 2556 bunches in July, operation with "8b4e" limits the number of bunches to around 1920.
• Reduction of Crossing angle: http://home.cern/cern-people/updates/2017/09/lhc-report-colliding-angle
• Filling Schema Web Site (user lpc) http://lpc.web.cern.ch/SchemeEditor.htm
• LPC Evian Dec17: https://indico.cern.ch/event/663598/contributions/2782518/attachments/1573604/2487449/Evian2017_ExperimentsInput2018_final.pdf
• TTCrx manual http://ttc.web.cern.ch/TTCrx_manual3.12.pdf

Emittance scans

• 22 March Ale: https://indico.cern.ch/event/708578/contributions/2909361/subcontributions/252147/attachments/1621316/2579771/2018-03-22-dailyrun.pdf
• 27 March, ATLAS weekly, Nicoletta: https://indico.cern.ch/event/716501/contributions/2945291/attachments/1623629/2584759/Weekly_27032018.pdf

Heavy Ions

• Spread in TE: https://atlasop.cern.ch/elisa/display/361073

Pulse shape Oscilloscope

• https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneCaloPulseShapes

L1Calo documents

• Readout https://edms.cern.ch/ui/file/761046/5/PPM_S-Link_data_formats_v2.1.pdf
• Hardware specifications https://atlas-l1calo.web.cern.ch/atlas-l1calo/html/orgweb/Modules/Modules.html
• Hardware specification MSU page: https://web.pa.msu.edu/hep/atlas/l1calo/

Simulation

Overlay Simulation

• https://indico.cern.ch/event/590779/contributions/2383766/attachments/1376860/2091072/L1Calib_20161124.pdf
• https://indico.cern.ch/event/279530/contributions/634995/attachments/511924/706533/Haas_ATLAS_PileupOverlay_3-19-2014.pdf

Data Quality

• Murrough's fixes of Ped Correction Histos: https://atlasop.cern.ch/elisa/display/367375

L1Calo pictures

• KIP server: http://www.kip.uni-heidelberg.de/atlas/docs/l1calo/plots
• STF https://twiki.cern.ch/twiki/bin/view/Atlas/LevelOneCaloSTFPhotos

Phase I

Power consumption

• PPM: (Martin) citing the PPM paper: "The power consumption was measured using a single, fully populated PP crate in a test lab setup that was operating all PPMs in the typical data processing mode, resulting in 175 A on the +3.3 V supply and 150 A on the +5.0 V supply. This corresponds to 84 W per PPM, which is well below the worst case estimate of 100 W.” So I guess from there, multiplying with 16 modules per crate, it would be about 1.4 kW per crate.
• eFEX: 8KW per shelf

L1Calo Upgrade documentation

• Twiki Reviews: https://twiki.cern.ch/twiki/bin/view/Atlas/LevelOneCaloUpgradeModules
• FOX https://cds.cern.ch/record/2286379/files/ATL-COM-UPGRADE-2017-014.pdf?version=4
• Latency talk Antoine: https://indico.cern.ch/event/963953/contributions/4070004/attachments/2126184/3579831/latency_run3.pdf

TREX

• https://edms.cern.ch/ui/file/1533071/0.2/TREX_TechSpecs_v0.2.pdf,
• https://cds.cern.ch/record/2646164/files/ATL-COM-DAQ-2018-175.1.pdf?version=3
• Zynq: https://indico.cern.ch/event/799275/contributions/3413781/attachments/1861984/3060431/vandrei_trex_zynqmpsoc_20190613.pdf
• EDMS https://edms.cern.ch/ui/#!master/navigator/project?P:1008597770:1008597770:subDocs
• Wiener fantray https://file.wiener-d.com/documentation/Crates_6000/WIENER_Series_6000_Manual_A2.pdf
• STF Wiener fantray web interface: http://stf-trex-crate.cern.ch/

CTP

• Multi partitions for commissioning: https://twiki.cern.ch/twiki/bin/viewauth/Atlas/LevelOneCentralTriggerMultipartitions

Other Links

• Jan's twiki at KIP: https://wiki.kip.uni-heidelberg.de/KIPwiki/index.php/Atlas_Privat:JJ_Base
• Sysadmins open a redmine tiket: https://atlasop.cern.ch/twiki/bin/view/SysAdmins/ContactSysAdmins#Redmine_open_a_ticket

Run 3 stuff

• May 2019 Trigger workshop: https://cds.cern.ch/record/2679559/files/ATL-COM-DAQ-2019-098.pdf? , https://indico.cern.ch/event/772409/timetable/

Talks

• TDAQ speaker commetee https://twiki.cern.ch/twiki/bin/viewauth/Atlas/TDAQSpeakersCommittee

Trouble shooting

• start calibration run and having error to start.daq -> try to start calbration partiton manually

-- SilviaFranchino - 2015-11-16

• USA15 L1Calo Phase1 racks: