Run 18 Calibration Datasets

 Below are the calibration dataset considerations for isobar needs in Run 18:

• EMC
  • emc-check runs once per fill (at the start)
    • For the purpose of EMC pedestals and status tables
    • 50k events
• TOF
  
  • VPD
    • Resolutions needs to be calibrated (for trigger) and confirmed to be similar for both species
• TPC
  • Laser runs
    • For the purpose of calibrating TPC drift velocities
    • Every few hours (as usual)
    • Either dedicated laser runs, or included in physics runs
    • ~5000 laser events
  • GMT inclusion in a fraction of events
    • For the purpose of understanding fluctuations in TPC SpaceCharge
    • Trigger choice is not very important, as long as it has TPX and during most physics running
    • Something around 100-200 Hz is sufficient (too much will cause dead time issues for the trigger)
  • Vernier scans
    • For the purpose of understanding backgrounds in the TPC that may be different between species
    • Once for each species under typical operating conditions
    • 4 incremental steps of collision luminosity, each step ~1 minute long and on the order of ~50k events (total = ~4 minutes)
    • TPX, BEMC, BTOF must be included
    • Minimum bias trigger with no (or wide) vertex z cut
    • You do not have access to view this node
  • Low luminosity fill IF a TPC shorted field cage ring occurs
    • Only one species needed (unimportant which)
    • Minimum bias trigger with no (or wide) vertex z cut
    • ~1M events
    • ZDC coincidence rates below 3 kHz
  • Old ALTRO thresholds run
    • For the purpose of TPC dE/dx understanding
    • Only one species needed (unimportant which)
    • ~2M events
    • Could be at the end of a fill, or any other convenient time during typical operations
  • Magnetic field flipped to A ("forward") polarity before AuAu27 data
    • For the purpose of acquiring sufficient cosmic ray data with both magnetic fields to understand alignment of new iTPC sector

-Gene

Run preparation Run VI (2006)

This page is meant to organize information about the calibration preparations and steps required for each subdetector for upcoming runs. Items with an asterisk (*) need to be completed in advance of data.

For the run in winter 2006, the plan is to take primarily pp data. This may lead to different requirements than in the past.

TPC

• * Code development for 1 Hz scalers.
• * Testing of T0/Twist code with pp data.
• * Survey alignment information into DB.
• Drift velocity from lasers into DB (automated).
• T0 calibration as soon as data starts.
• Twist calibration as soon as data starts (for each field).
• SpaceCharge and GridLeak (SpaceCharge and GridLeak Calibration How-To Guide) as soon as T0/Twist calibrated (for each field).
• Would like a field flip to identify origins of offsets in SpaceCharge vs. scalers.
• dEdx: final calibration after run ends by sampling over the whole run period, but initial calibration can be done once TPC momentum is well-calibrated (after distortion corrections).

FTPC

• HV calibration performed at start of data-taking.
• Temperatures into DB (automated)
• Rotation alignment of each FTPC done for each field (needs calibrated vertex from TPC[+SVT][+SSD]). There was concern about doing this calibration with the pp data - status???

SVT

• Steps for pp should be the same as heavy ion runs, but more data necessary.
• Self-alignment (to be completed January 2006)
• Requires well-calibrated TPC.
• Requires a few million events (a week into the run?)
• Would like field flip to discriminate between SVT/SSD alignment vs. TPC distortions.

SSD

• * Code development in progress (status???)
• Requires well-calibrated TPC.

EEMC

• EEMC initial setting & monitoring during data taking relies on prompt and fully automatic muDst EzTree production for all minias-only fast-only runs. Assume fast offline muDst exist on disk for 1 week.
• Initial settings for HV: 500k minbias fast-triggered events will give slopes necessary to adjust relative gans. Same 60 GeV E_T maximum scale as in previous years.
• Pedestals from 5k minbias events, once per fill.
• Stability of towers from one 200k minbias run per fill.
• Highly prescaled minbias and zerobias events in every run for "general monitoring" (e.g. correlated pedestal shift)
• Offline calibrations unnecessary for production (can be done on the MuDst level).
• "Basic" offline calibration from MIP response in 5M minbias fast events (taken a few weeks into the run)
• "Final" offline calibration from pi0 (or other TBD) signal requires "huge" statistics of EHT and EJP triggers to do tower-by-tower (will need the full final dataset).
• * Calibration codes exist, but are scattered. Need to be put in CVS with consistent paths and filenames.

BEMC

• * LED runs to set HV for 60 GeV E_T maximum, changed from previous years (status???).
• Online HV calibration from 300k minbias events (eta ring by eta ring) - requires "full TPC reconstruction".
• MuDsts from calibration runs feedback to run settings (status???).
• Pedestals from low multiplicity events from the event pool every 24 hours.
• Offline calibrations unnecessary for production (can be done on the MuDst level).
• "Final" offline tower-by-tower calibration from MIPs and electrons using several million events.

TOF

• upVPD upgrade (coming in March 2006) for better efficiency and start resolution
• Need TPC momentum calibrated first.
• Requires several million events, less with upVPD (wait for upVPD???)

Run preparation Run VII

There is some question as to whether certain tasks need to be done this year because the detector was not moved during the shutdown period. Omitting such tasks should be justified before skipping!

TPC

• Survey alignment information into DB (appears to be no survey data for TPC this this year)
• High stats check of laser drift velocity calibration once there's gas in the TPC: 30k laser events with and without B field.
• Check of reversed GG cable on sector 8 (lasers) once there's gas in the TPC: same laser run as above
• Drift velocity from laser runs (laser runs taken once every 3-4 hours, ~2-3k events, entire run) into DB (automated); check that it's working
• T0 calibration, new method from laser runs (same laser runs as above).
• Twist calibration as soon as data starts: ~100k events, preferrably central/high multiplicity, near start of running for each field
• SpaceCharge and GridLeak (SpaceCharge and GridLeak Calibration How-To Guide) as soon as T0/Twist calibrated: ~100k events from various luminosities, for each field.
• BBC scaler study for correlation with azimuthally asymmetric backgrounds in TPC: needs several days of generic data.
• Zerobias run with fast detectors to study short time scale fluctuations in luminosity (relevant for ionization distortions): needs a couple minutes of sequential, high rate events taken at any time.
• Need a field flip to identify origins of offsets in SpaceCharge vs. scalers as well as disentangling TPC distortions from silicon alignment.
• dEdx: final calibration after run ends by sampling over the whole run period, but initial calibration can be done once TPC momentum is well-calibrated (after distortion corrections).

FTPC

• HV calibration performed at start of data-taking (special HV runs).
• Temperatures into DB (automated)
• Rotation alignment of each FTPC done for each field (needs calibrated vertex from TPC[+SVT][+SSD]): generic collision data

SSD

• Pulser runs (for initial gains and alive status?)
• Initial alignment can be done using roughly-calibrated TPC: ~100k minbias events.
• P/N Gain-matching (any special run requirements?)
• Alignment, needs fully-calibrated TPC: 250k minbias events from one low luminosity (low TPC distortions/background/occupancy/pile-up) fill, for each field, +/-30cm vertex constraint; collision rate preferrably below 1kHz.

SVT

• Temp oscillation check with lasers: generic data once triggers are set.
• Initial alignment can be done using roughly-calibrated TPC+SSD: ~100k minbias events.
• Alignment, needs fully-calibrated TPC: 250k minbias events from one low luminosity fill (see SSD).
• End-point T0 + drift velocity, needs fully-calibrated SSD+TPC: same low luminosity runs for initial values, watched during rest of run.
• Gains: same low luminosity runs.

EEMC

• Timing scan of all crates: few hours of beam time, ~6 minb-fast runs (5 minutes each) for TCD phase of all towers crates, another 6 minbias runs for the timing of the MAPMT crates, 2 days analysis
• EEMC initial setting & monitoring during data taking: requests to process specific data will be made as needed during the run.
• Initial settings for HV: 200k minbias fast-triggered events will give slopes necessary to adjust relative gans, 2 days analysis
• Pedestals (for offline DB) from 5k minbias events, once per 5 hours
• Stability of towers from one 200k minbias-fast run per fill
• "General monitoring" (e.g. correlated pedestal shift) from highly prescaled minbias and zerobias events in every run.
• Beam background monitoring from highly prescaled EEMC-triggered events with TPC for at the beginning of each fill.
• Expect commissioning of triggers using EMC after one week of collisions
• Offline calibrations unnecessary for production (can be done on the MuDst level).
• "Basic" offline calibration from MIP response in 5M minbias fast events taken a few weeks into the run
• "Final" offline calibration from pi0 (or other TBD) signal requires "huge" statistics of EHT and EJP triggers to do tower-by-tower (still undone for previous runs).
• Calibration codes exist, but are scattered. Need to be put in CVS with consistent paths and filenames (status?)

BEMC

• Timing scan of all crates
• Online HV calibration of towers - do for outliers and new PMTs/crates; needed for EMC triggering. Needs~5 minutes of minbias fast-triggered events (eta ring by eta ring) at beginning of running (once a day for a few days) - same runs as for EEMC.
• Online HV calibration of preshower - matching slopes. Not done before, will piggyback off other datasets.
• Pedestals from low multiplicity events from the event pool every 24 hours.
• Offline calibrations unnecessary for production (can be done on the MuDst level).
• "Final" offline tower-by-tower calibration from MIPs and electrons using several million events

upVPD/TOF

• upVPD (calibration?)
• No TOF this year.

preparation Run VIII (2008)

This page is meant to organize information about the calibration preparations and steps required for each subdetector for upcoming runs.

Previous runs:

• Run preparation Run VI (2006)
• Run preparation Run VII

 

Red means that nothing has been done for this (yet), or that this needs to continue through the run.
Blue means that the data has been taken, but the calibration task is not completed yet.
Black indicates nothing (more) remaining to be done for this task.

 

TPC

• Survey alignment information into DB (appears to be no survey data for TPC this this year)
• Drift velocity from laser runs (laser runs taken once every 3-4 hours, ~2-3k events, entire run) into DB (automated); check that it's working
• T0 calibration, using vertex-matching (~500k events, preferably high multiplicity, once triggers are in place).
• Twist calibration as soon as data starts: same data
• SpaceCharge and GridLeak (SpaceCharge and GridLeak Calibration How-To Guide) as soon as T0/Twist calibrated: ~500k events from various luminosities, for each field.
• BBC scaler study for correlation with azimuthally asymmetric backgrounds in TPC: needs several days of generic data.
• Zerobias run with fast detectors to study short time scale fluctuations in luminosity (relevant for ionization distortions): needs a couple minutes of sequential, high rate events taken at any time.
• dEdx: final calibration after run ends by sampling over the whole run period, but initial calibration can be done once TPC momentum is well-calibrated (after distortion corrections).

TPX

???

FTPC

• HV calibration performed at start of data-taking (special HV runs).
• Temperatures into DB (automated)
• Rotation alignment of each FTPC done for each field (needs calibrated vertex from TPC[+SVT][+SSD]): generic collision data

EEMC

• Timing scan of all crates: few hours of beam time, ~6 fast runs (5 minutes each) for TCD phase of all towers crates
• EEMC initial setting & monitoring during data taking: requests to process specific data will be made as needed during the run.
• Initial settings for HV: 500k minbias fast-triggered events will give slopes necessary to adjust relative gans, 2 days analysis
• Pedestals (for offline DB) from 5k minbias events, once per fill
• "General monitoring" (e.g. correlated pedestal shift) from highly prescaled minbias and zerobias events in every run.
• Beam background monitoring from highly prescaled EEMC-triggered events with TPC for at the beginning of each fill.
• Expect commissioning of triggers using EMC after one week of collisions
• Offline calibrations unnecessary for production (can be done on the MuDst level).
• "Basic" offline calibration from MIP response in 5M minbias fast events taken a few weeks into the run
• "Final" offline calibration from pi0 (or other TBD) signal requires "huge" statistics of EHT and EJP triggers to do tower-by-tower.
• Calibration codes exist, but are scattered. Need to be put in CVS with consistent paths and filenames (status?)

BEMC

• Timing scan of all crates
• Online HV calibration of towers - do for outliers and new PMTs/crates; needed for EMC triggering. Needs~5 minutes of minbias fast-triggered events (eta ring by eta ring) at beginning of running (once a day for a few days) - same runs as for EEMC.
• Online HV calibration of preshower - matching slopes. (same data).
• Pedestals from low multiplicity events from the event pool every 24 hours.
• Offline calibrations unnecessary for production (can be done on the MuDst level).
• "Final" offline tower-by-tower calibration from MIPs and electrons using several million events

PMD

• Hot Cells (~100k generic events from every few days?)
• Cell-by-cell gains (same data)
• SM-by-SM gains (same data)

VPD/TOF

• T-TOT (~3M TOF-triggered events)
• T-Z (same data)

Run 12 CuAu200

Regarding the P14ia Preview Production of Run 12 CuAu200 from early 2014:

A check-list of observables and points to consider to help understand alayses' sensitivity to non-final calibrations

To unambiguously see issues due to mis-calibration of the TPC, stringent determination of triggered-event tracks is necessary. Pile-up tracks are expected to be incorrect in many ways, and they constitute a larger and larger fraction of TPC tracks as luminosity grows, so their inclusion can lead to luminosity-dependencies of what appear to be mis-calibrations but are not.

1. TPC dE/dx PID (not calibrated)
   1. differences between real data dE/dx peaks' means and width vs. nsigma provided in the MuDst
   2. variability of these differences with time
2. TPC alignment (old one used)
   1. sector-by-sector variations in charge-separated signed DCAs and momentum spectra (e.g h-/h+ pT spectra) that are time- and luminosity-independent
   2. differences in charge-separated invariant masses from expectations that are time- and luminosity-independent
   3. any momentum effects (including invariant masses) grow with momentum: delta(pT) is proprotional to q*pT^2
      1. alternatively, and perhaps more directly, delta(q/pT) effects are constant, and one could look at charge-separated 1/pT from sector-to-sector
3. TPC SpaceCharge & GridLeak (preliminary calibration)
   1. sector-by-sector variations in charge-separated DCAs and momentum spectra that are luminosity-dependent
   2. possible track splitting between TPC pad rows 13 and 14 + possible track splitting at z=0, in the radial and/or azimuthal direction
   3. differences in charge-separated invariant masses from expectations that are luminosity-dependent
   4. any momentum effects (including invariant masses) grow with momentum: delta(pT) is proprotional to q*pT^2
      1. alternatively, and perhaps more directly, delta(q/pT) effects are constant, and one could look at charge-separated 1/pT from sector-to-sector
4. TPC T0 & drift velocities (preliminary calibration)
   1. track splitting at z=0, in the z direction
   2. splitting of primary vertices into two close vertices (and subsequent irregularities in primary track event-wise distributions)
5. TOF PID (VPD not calibrated, BTOF calibration from Run 12 UU used)
   [particularly difficult to disentangle from TPC calibration issues]
   1. not expected to be much of an issue, as the "startless" mode of using BTOF was forced (no VPD) and the calibration used for BTOF is expected to be reasonable
   2. broadening of, and differences in mass^2 peak positions from expectations are more likely due to TPC issues (particularly if charge-dependent, as BTOF mis-calibrations should see no charge sign dependence)
   3. while TOF results may not be the best place to identify TPC issues, it is worth noting that BTOF-matching is beneficial to removing pile-up tracks from studies

Intrinsic resolution in a tracking element

Foreword: This has probably been worked out in a textbook somewhere, but I wanted to write it down for my own sake. This is a re-write (hopefully more clear, with slightly better notation) of Appendix A of my PhD thesis (I don't think it was well-written there)...

-Gene

_______________________

Let's establish a few quantities:

• E_intr : error on the measurement by the element in question
  
  • σ_intr² = <E_intr²> : intrinsic resolution of the element, and its relation to an ensemble of errors in measurement
• E_proj : error on the track projection to that element (excluding the element from the track fit)
  • σ_proj² = <E_proj²> : resolution of the track projection to an element, and its relation to an ensemble of errors in track projections
• E_track : error on track fit at an element including the element in the fit
• R_incl = E_intr - E_track : residual difference between the measurement and the inclusive track fit
  • σ_incl² = <(E_intr - E_track)²> : resolution from the inclusive residuals
• R_excl = E_intr - E_proj : residual difference between the measurement and the exclusive track fit
  
  • σ_excl² = <(E_intr - E_proj)²> : resolution from the exclusive residuals

Let us further assume that the projection from the track fit excluding the element E_proj is uncorrelated with the intrinsic error of the measurement from the element: <E_proj E_intr> = 0. This implies that we can write:


Our goal is to determine σ_intr given that we can only observe σ_incl and σ_excl.

To that end, we utilize a guess, σ'_intr, and write down a reasonable estimation of E_track using a weighted average of E_intr and E_proj, where the weights are w_proj = 1/σ_proj², and w_intr = 1/σ'_intr²:


Substituing this, we find...


Dropping terms of <E_intr E_proj>, replacing terms of <E_proj²> and <E_intr²> with σ_proj² and σ_intr² respectively, and multiplying through such that all terms on the right-hand-side of the equation have the denominator (σ'_intr² + σ_proj²)², we find


We can substitute for σ_proj² using σ_excl² = σ_intr² + σ_proj²:


And solving for σ_intr² we find:


This is an estimator of σ_intr. Ideally, σ_intr and σ'_intr should be the same. One can iterate a few times starting with a good guess for σ'_intr and then replacing it in later iterations with the σ_intr found from the previous iteration until the two are approximately equal.

_______

-Gene

SVT Self-Alignment

Using tracks fit with SVT points plus a primary vertex alone, we can self-align the SVT using residuals to the fits. This document explains how this can be done, but only works for the situation in which the SVT is already rather well aligned, and only small scale alignment calibration remains. The technique explained herein also allows for calibration of the hybrid drift velocities.

Miscellaneous TPC calibration notes

The log file for automated drift velocity calculations is at ~starreco/AutoCalib.log.

Log files for fast offline production are at /star/rcf/prodlog/dev/log/daq.

The CVS area for TPC calibration related scripts, macros, etc., is StRoot/macros/calib.

Padrow 13 and Padrow 40 static distortions

These two distortion corrections focus on static non-uniformities in the electric field in the gating grid region of the TPC, radially between the inner and outer sectors. This region has an absence of gating grid wires where the subsector structures meet, allowing some bending of the equipotential lines, creating radial electric field components. In both cases, the region of distorted fields is quite localized near the endcap and only over a small radial extent, but this then distorts all drifting electrons (essentially from all drift distances equally) in that small radial region, affecting signal only for a few padrows on each side of the inner/outer boundary.

Padrow 13 Distortion
For the original TPC structure, the was simply nothing in the gap between the inner and outer sectors. More about the gap structure can be learned by looking some of the GridLeak documentation (which is different in that it is dynamic with the ionization in the TPC. This static distortion was observed in the early operation of STAR when luminosities were low, well before the GridLeak distortion was ever observed. It was modeled as an offset potential on a small strip of the TPC drift field boundary, with the offset equal to a scale factor times the gating grid voltage, where the scale factor was calibrated from fits to the data. Below are the related distortion correction maps.

Padrow 40 Distortion
With the iTPC sectors, a wall was placed at the outer edge of the new inner sectors with conductive stripes on it to express potentials primarily to suppress the GridLeak distortion. It was understood that this would modify the static distortion, and in the maps below it is apparent that the static distortions became a few times larger with the wall, but in both cases only at the level of ~100 μm or less at the outermost padrow of the inner sectors. It is also worth noting that the sign of the distortion flipped with respect to Padrow 13.

This distortion correction is determined from the sum of 3 maps that represent the distortion contributions from (1) the wall's grounded structure, (2) the potential at the wall's tip [nominally -115 V], and (3) the potential on the wall's outer side [nominally -450 V]. Plots on this page were made using the nominal potentials.

Distortion correction maps:
The maps are essentially uniform in z, so the maps shown below focus only on the radial and azimuthal dependencies. Further, there is no distortion more than a few centimeters away from the inner/outer boundaries, and there is essentially a 12-fold symmetry azimuthally (though this is not strictly true if different wall voltages are used on different iTPC sectors), so the maps zoom in on a limited region of r and φ to offer a better view of the fine details. Also, the inner and outer edges of the nearest padrows are shown as thin black lines on some plots to help understand their proximity to the distortions.

Open each map in a new web browser tab or window to see higher resolution. Units are [cm] and [radian].

	Δ(r-φ) vs. r and φ	Δ(r) vs. r and φ
Padrow 13
Padrow 13 lines indicating locations of rows 13 and 14
Padrow 40
Padrow 40 lines indicating locations of rows 40 and 41

-Gene

RunXI dE/dx calibration recipe

This is a recipe of RunXI dEdx calibration by Yi Guo.

TPC Hit Errors

• Instructions
• Calibration results
• Hit errors in zero field

TPC T0s

Global T0:

• Instructions (as of 2012, calibrated along with Twist)
• Run 10 calibration
• Run 12 calibration

Twist (ExB) Distortion

 Run 12 calibration