Detector and Trigger Plans - Overview

1
Detector and Trigger Plans - Overview

• Accelerator
• Plans
• October shutdown
• D0 Control Room shifts
• Experiment technical support
• Sub-detectors plans
• Trigger
• D0 operating efficiency
• Commissioning and data collection
• Data quality monitoring
• Summary
• Thanks for the input from many individuals!

2
Accelerator Operation

• Tevatron peak luminosity is around 2.1031 over
  past two months
• Improvements in peak luminosity are expected soon
  due to installation of cooling tanks in
  antiproton accumulator during June shutdown
• only about 30 of antiprotons are reaching final
  collisions for now
• transverse beam sizes are more then expected
• Detailed plan of increase of Tevatron luminosity
  is developed
• studies periods each week
• Beams Division plans to split weekly 2 days of
  studies into smaller intervals
• periodic shutdowns

3
Accelerator Performance

• There is slow improvement in per week delivered
  luminosity
• higher initial luminosity value
• better operating efficiency
• 70 of stores end gracefully
• Run Ia goal is 15pb-1 per week
• January 2002 plan is to deliver 300 pb-1 to each
  of the experiments by the end of 2002
• as of today delivered luminosity about twice
  lower then planned
• revision of the plan is expected soon
• How can D0 help
• utilize luminosity more efficiently
• reduce number of interruptions in accelerator
  operations
• there is no 10 minutes access!

4
October Shutdown

• It is a bad plan which admits of no
  modifications
• Original plan was 6 weeks shutdown starting
  September 30th 2002
• Single major reason for the October shutdown
• Improvement in vacuum and other elements of
  recycler
• Considerations
• Last(!) prolonged shutdown for the recycler
• Are 6 weeks enough?
• Far from perfect accelerator performance even
  without recycler
• Prolonged interval needed to get accelerator to
  pre-shutdown conditions
• Experiments desire to collect decent data samples
  for Winter 2003 conferences
• Current plan
• Not earlier then October 2002
• Not less then 6 weeks

5
Accelerator Backgrounds

• D0 benefits greatly from extensive shielding
  installed during Run IIa upgrade
• Much lower backgrounds on muon detectors
• less then 0.1 occupancy
• Reduced beam halo
• D0 beam halo is stable and affects 1-2 of
  crossings
• Beam related backgrounds are affecting primary
  forward proton detector
• Task force (with CDF) is formed
• Late March CDF believed to experience high
  (1Mrad/sec) dose rate irradiation which caused 6
  silicon ladders to fail
• Abort of partly de-bunched beam
• Not confirmed by booster studies
• CDF is located close to the Tevatron abort
• D0
• Located favorably
• D0 silicon High Voltage did not even tripped
• Typical integrated dose during store shot setup
  for D0 is 5-10 rads
• Low in comparison with silicon radiation hardness

6
D0 Control Room Shifts

• Over last 6 months achieved reasonably stable
  shifts sign up
• 5.103 shifts per 6 months are needed
• This year requires peak number of shifters
• Still many systems are in commissioning stage
• Require extensive trainings and active work
  during data taking
• Introducing extra shifters as systems become
  available
• L2
• Global monitor (planned)
• Starting combining shifts for different
  sub-detectors
• Forward Proton Detector (FPD) and Calorimeter
• Based on relative success of current shift duties
  rules only minor modifications are adopted for
  July-December 2002 period
• 8 shifts for each non-resident on Masthead per
  6 months
• 10 shifts for each resident (gt 6 months at
  Fermilab) on Masthead per 6 months
• Thanks to all shifters who keep the experiment
  running it is fun, experience, understanding of
  your detector performance, and a must to get our
  data sample

7
D0 Experiment Technical Support Groups

• Two teams of electrical and mechanical
  professionals are reliably supporting operations
  of the D0 experiment
• Mechanical Russ Rucinski group
• Electrical Rick Hance group
• Perform extensive amount of 24/7 duties
• All cryo control and monitoring
• Power supplies and magnets
• Heating and cooling
• Emergency responses
• Detector opening and closing
• Can you move 5000 tons with 0.5mm precision?
• We plan to continue 24/7 presence of Ops shifter
  for the duration of Run II
• Let D0 control room crew concentrate on high
  quality data taking
• Will provide support for commissioning/installatio
  n/operation jobs
• Requests should be made in advance

8
To do for the D0 Detectors

• All major detectors are in operation
• Second order commissioning and improvements
• Luminosity detector
• Operates stable over many months
• Still with Run I electronics
• Improvements in calibration constant
• On-line numbers for integrated luminosity
• Is CDF luminosity really 6 higher?
• D0/CDF task force
• Silicon detector
• Reliability of the cathedral power supplies and
  cooling
• Understanding of grassy noise issue
• High voltage leak currents
• Monitoring of the radiation dose

4 F Disks
6 Barrels
12 F Disks
9
To do for the Detectors

• Fiber tracker and preshowers
• Great progress since AFE boards became available
• Resolve AFE boards failures/problems
• Understand pedestal variations to reduce hits
  occupancy
• Monitor light output, VLPS, and AFE responses
• Provide calibrated thresholds/signals for CTT
  trigger
• Integrate forward preshower (and FPD)
• Calorimeter
• Fix noise on Level 1 trigger signals
• Stable pulser calibration
• Operation with 1.5s vs 2.5s zero suppression
  threshold
• COMICS based reliable parameters download and
  verification
• 25 BLS low voltage power supplies have to be
  upgraded

10
To do for the Detectors

• Inter Cryostat Detector (ICD)
• Replace bad PMTs (few electronics channels)
• Updated ICD addressing and test using LED pulser
• MIP calibration with data for every individual
  tile
• Solenoid
• Long term stability monitoring
• Corrections to field map
• Geometry, tracks fit based corrections
• Muon system
• Improve stability of readout (synhronization and
  other errors)
• Level 2 trigger signals issues
• Individual thresholds for central A layer
  counters
• T0s and time-to-distance for PDTs (momentum
  resolution)

11
To do for the D0 Detectors

• Forward Proton Detector (FPD)
• Connect dipole spectrometer to AFE boards
• Phase I (10 detectors/5 spectrometers) to be
  included in readout by October
• FPD Trigger Manager commissioning will proceed in
  parallel with DFE commissioning
• Commission luminosity monitor TDC boards to
  include trigger scintillator information

12
D0 Trigger

• Trigger framework
• Existing prescaler logic could report
  mis-measured luminosity for prescaled triggers
• Plan on how to resolve this problem is developed
• Automated SCL inits
• To reduce downtime (see below)
• L1 trigger
• Fiber tracker (CTT) trigger
• Implementation is delayed due to higher then
  specified trigger decision time
• Different options of resolving this issue are
  discussed by experts
• Estimated delay is 6weeks
• Calorimeter trigger
• Debug recently achieved eta coverage up to 2.4
• Dead and noisy towers, calibration
• Missing Et
• Extend eta coverage
• Muon trigger
• Optimize new trigger terms based on scintillator
  and wire detectors
• Re-establish data vs trigger simulator comparison
• Attempt to shave 132ns from FPGA logic

13
D0 Trigger

• Level 2 trigger
• Increase processing bandwidth from current
  0.5kHz
• Code speedup/max_opt 1-2 kHz
• Multiple Alphas/Betas in crate gt2 kHz?
• Silicon track trigger
• In progress of installation
• Expect beam commissioning later this year (fiber
  tracker trigger needed)
• Fiber tracker
• Waiting for inputs expected in September
• Calorimeter trigger
• Running on-line (/eta/lt0.8 for now)
• Requires better understanding of efficiency,
  cuts, rejection
• Muon trigger
• Well developed single muon, di-muon
• Resolving problems with synhronization of signals
  coming from hundreds of front-ends
• Detailed monitoring of the inputs is needed
• Algorithms optimization (geometry for Pt cut,
  etc.) is in progress

14
D0 Trigger

• Level 3 trigger
• Running on-line for quite a while
• Development of fast and efficient algorithms with
  good rejection is driving the schedule
• Tracking Level 3 is coming on-line next week
• Improve electron and muon rejection
• Vertex for missing Et resolution improvement
• Calorimeter is one of the most advanced systems
• Efficiently filtering jets and em objects
• Coming in p12 missing Et, non-linearity
  corrections, hot cell killer
• Muon
• Filtering on limited sub-set of muon triggers
• Technical issues with muon geometry, efficiency
  to be better understood
• Level 3 trigger basic philosophy
• Develop as much as possible inclusive single e,
  mu, tau, jet filters
• Combine tools into di-muon, muonjet, etc.
  filters
• For all trigger levels
• After trigger is running on-line and commissioned
  
• ID and physics groups have to take the lead to
  optimize cuts, study efficiencies, and select
  optimal trigger terms for their analyses
• Trigger board will coordinate these efforts

15
DAQ

• Currently could utilize new DAQ system rate
  capabilities in a limited way due to complex
  system of distributed readout and front-end
  crates
• Multi-buffering operation in tracking crates just
  arrived
• Correct tick and turn operation to be verified
• With estimated DAQ rate capability in the range
  of 1kHz we are currently limited to 140Hz for
  reliable on-line operation
• Current most serious problem is loss of
  synhronization in muon scintillator crates
• Every 10 minutes at 140Hz
• There are other concerns on the horizon
• Tracking crates front-end busy fraction at 140Hz
  is 4
• DAQ rate task force has been organized (chaired
  by Ron Lipton)
• Very well defined goal
• Bring experts from different sub-systems together
  to resolve problems preventing us from running at
  full DAQ capabilities
• Support of the DAQ system operation will continue
  to be responsibility of the DAQ team and
  appropriate detector groups
• DAQ team has developed list of improvements
  related to core of the DAQ system
• Software bugs fixes (small annoyances)
• Complete Linux Port
• Supervisor, Monitor Server are still on Windows
• Improve shifter monitoring and documentation

16
On-line System and Controls

• Key system for data handling/transfer and on-line
  data quality monitoring
• Run II design rate of writing data to tapes at
  50Hz has been achieved
• Operations
• get serious with Significant Event System
  (alarms)
• important issue for high quality data collection
• active participation from detector/trigger groups
  is required
• migrate detector groups to use COMICS downloads
• migrate detector groups to use Hardware Database
• improve support of Examines
• Applications
• deploy run configuration/quality database
• introduce global DAQ monitor application
• Software issues
• install, test Python v2.2 (update to t02.23 soon)
• support of software releases
• System issues
• solve nfs and static route problems
• add more Examine nodes
• control room monitor upgrades

17
D0 Operating Efficiency

• Definition
• of collisions used
• Global runs (recorded)
• Special runs
• Commissioning
• Goals
• From Run I experience
• 190pb-1 delivered 120pb-1 recorded (60, Main
  ring vetoes, etc.)
• Should do better in Run II gt 90
• Moving target
• Major reasons of luminosity loss are changing as
  more systems come on-line
• What was a concern a few months/weeks ago is
  replaced by new enemies
• Steady increase in operating efficiency during
  February/March steady state operation
• Recent numbers
• 60 for the last 2 months (lower before)
• 75 last week (after implementation of
  mutli-buffer readout)

18
D0 Operating Efficiency

• Major components of the efficiency
• Dead time
• Sub-system is setting busy signal due to low
  bandwidth L1, L2, L3, DAQ
• Adjustable and can be set at almost any value by
  adjusting trigger rates
• Due to lack of multi-buffering in tracking crates
  had to accept relatively large L1 front-end-busy
  fraction (20) to collect decent data sample
  before June shutdown
• Currently running at about 4 front-end-busy
  fraction (140Hz readout rate)
• Downtime
• Data collection is stopped (or could not be
  started) due to failure of a specific component
  (or components)
• Currently issues are
• Muon scintillator front-end crates errors
• High Voltage trips
• Redownloading of problematic crates
• Begin/end runs
• Operator errors
• Commissioning activities (prescales selection,
  trigger studies,)
• SCL inits
• Used to synhronize trigger/DAQ system
• Over last 3 months we average 3.103 SCL inits
  per month during global runs
• About 15 of luminosity blocks (but not events)
  is wiped out

19
Commissioning and Data Collection

• Many important Run IIa systems are coming late
  (some are still in pipeline)
• Seriously complicates data collection
• Commissioning requires substantial amount of
  devoted beam time
• Requires often changes to operating conditions
  (trigger list, etc.) and documentation
• We operate balancing between physics data
  collection and commissioning activities
• Data are needed for Reco development and physics
  analysis
• We have decent luminosity sample which is
  providing interesting results
• Our detector is much better then in Run I
• Even with available on tapes luminosity data
  samples in some channels (J/y -gt mm, etc.) are
  orders of magnitude above Run I
• Long term we need full D0 detector capabilities
  we planned for
• Since the beginning of the Run II we perform
  above mentioned activities in parallel
• All control room activities are coordinated by
  the Run Coordinator
• Priorities are based on Collaboration goals
  defined together with D0 Spokes
• Goal is to minimize global data taking
  disruptions (do whatever could be done without
  beam) providing necessary time for sub-systems
  commissioning

20
Data Quality Monitoring

• This issue was discussed at different meetings
  during this Workshop
• Basic outcome is that we need multilayer data
  quality checks
• Significant Event Server (alarms), different
  GUIs (HV, readout, ), check lists,
• Examines and event Display
• Detector, trigger, physics (Global Monitor)
• Bad run lists
• Off-line physics based monitoring
• Checks everything from detectors, triggers,
  luminosity to reco, cuts, and databases
• Run quality database which will combine (most) of
  the above results is under development
• The sooner in the above list the problem is
  discovered the better
• Fix problem and mark data set with appropriate
  flag
• Sub-system experts are working hard optimizing
  monitoring without overloading shifters and
  physics analysis groups
• Size of minimal unit for monitoring was
  discussed event? Luminosity block number? Run?
• Complexity is drastically increasing with
  reduction in segmentation
• Some physics studies may not require every system
  working perfectly
• Stability of data collection is a key issue
• In many cases it is easier to fix the problem
  then develop monitoring tools
• Success on all levels (including analysis)
  strongly depends upon stability
• Last time forward muon parameters were changed in
  January 2002

21
Specs, Reliability, and Schedule

• In such large experiment as D0 with many
  independent groups involved following written
  specifications is absolutely critical
• CTT trigger decision time appeared to be above
  specs (most recent example)
• Delayed CTT trigger implementation
• Could cause considerable amount of work for all
  sub-systems for re-timing the experiment
• Considerable amount of the experiment downtime is
  due to low reliability of different systems
• We plan to run for many years with limited number
  of accesses
• Reliability should be taken into account at the
  design and construction/testing stage
• Recent examples (serious in terms of downtime and
  potential damage)
• Muon VME power supplies
• Silicon low voltage power supplies
• Interface boards water heat exchangers
• Schedule is not what you might do, it is what
  you will do
• Lab and funding agencies approach is changing
  (Run IIb)
• Mixture of different cultures in the experiment
• Wrong time estimates are causing considerable
  manpower losses and reduce prioritizing
  efficiency

In cathedral area
22
Summary

• Incredible progress in the D0 detector
  capabilities over last year
• All detectors are working and down to second
  order improvements
• No show stoppers!
• All trigger levels are in operation with rapidly
  improving capabilities
• New DAQ is performing well
• Data collection monitoring is advancing
• Data collection efficiency is reasonable
• We have to finish installation of a few critical
  systems and commission recently installed
  equipment
• Will provide necessary support for these
  activities
• Steady high efficiency operation with in depth
  data quality monitoring is becoming highest
  priority for most of the sub-groups
• Accelerator performance is critical for success
  of the D0 physics program
• We should provide feedback and minimize
  disruptions to accelerator commissioning and
  operations
• Current plan is to deliver 300pb-1 by the end of
  the year 2002
• Strong support of the control room shifts and
  detector/trigger/on-line activities by the
  Collaboration is critical in getting interesting
  physics results soon