Difference between revisions of "Mcsmear updates"

From GlueXWiki
Jump to: navigation, search
(→‎FDC)
 
(15 intermediate revisions by the same user not shown)
Line 1: Line 1:
 +
 +
= Tasks =
 +
 +
* Dead channel handling - Sean
 +
* Smearing updates - Detector groups
 +
** See below
 +
* Run-dependent values from RCDB - Sean & Dmitry
 +
* [[Mixing Simulated Events]]
 +
 +
== Per-Channel Efficiency ==
 +
 +
Efficiencies for simulated data should match those of real data when possible.  To start with, we will compare hit-level efficiencies.  To match simulated and real data efficiencies, we will calculate correction factors.  These factors will have values between 0 and 1, and we will randomly drop hits at a rate of (1-factor).
 +
 +
The following principles should be followed when determining these factors:
 +
 +
# Efficiencies should not be set to zero unless the channel was actually disabled during the runs
 +
# Efficiencies should be determined for each channel.
 +
# Efficiencies should be determined either on a run-by-run basis or as an average over a group of runs
 +
# Efficiencies should be determined in way that is unbiased by the trigger
 +
 +
=== New Tables ===
 +
 +
These tables contain factors to modulate the efficiency of each channel in order to match the data.
 +
 +
* FCAL/block_mc_quality
 +
* CDC/channel_mc_quality
 +
* START_COUNTER/paddle_mc_quality
 +
* BCAL/channel_mc_quality
 +
* FDC/package[1-4]/wire_mc_quality
 +
* FDC/package[1-4]/strip_mc_quality
 +
* TOF/paddle_mc_quality
  
 
= Current Settings =
 
= Current Settings =
 +
 +
Will set table fields to "green" when updated.
 +
 +
Should load parameters from CCDB where possible.
  
 
== General Comments ==
 
== General Comments ==
Line 7: Line 42:
  
 
TRIGGER_LOOKBACK_TIME = -100 ns
 
TRIGGER_LOOKBACK_TIME = -100 ns
 +
 +
Should we properly model readout windows?
  
 
== CDC ==
 
== CDC ==
Line 22: Line 59:
 
|}
 
|}
  
−
[more in hdgeant]
+
HDGeant calculations that need revisiting:
 +
* longitudinal drift
 +
* gain
  
−
Noise parameterization from 2007.
+
=== CDC Threshold Calculation ===
 +
 
 +
threshold = CDC_THRESHOLD_FACTOR * CDC_PEDESTAL_SIGMA
  
 
== FDC ==
 
== FDC ==
Line 42: Line 83:
 
|}
 
|}
  
−
[more in hdgeant]
+
HDGeant calculations that need revisiting:
 +
* wire drift calculation?
 +
* lorentz effects?
  
−
Noise parameterization from 2007 [only wires].
+
=== FDC Threshold Calculation ===
 +
 
 +
FDC_PED_NOISE = -0.0938 + 0.0485*FDC_CATHODE_SIGMA
 +
 
 +
threshold = FDC_THRESHOLD_FACTOR * FDC_PED_NOISE
  
 
== FCAL ==
 
== FCAL ==
Line 62: Line 109:
 
== BCAL ==
 
== BCAL ==
  
−
TBD
+
{| class="wikitable"
 +
! name !! value !! comment
 +
|-
 +
| BCAL_TWO_HIT_RESOL || 50. ||
 +
|-
 +
| BCAL_ADC_THRESHOLD_MEV || 1. ||
 +
|-
 +
| BCAL_FADC_TIME_RESOLUTION || 0.3 || (comment: to be updated)
 +
|-
 +
| BCAL_TDC_TIME_RESOLUTION || 0.3 || (comment: to be updated)
 +
|-
 +
| BCAL_SAMPLINGCOEFA || 0.042 ||
 +
|-
 +
| BCAL_SAMPLINGCOEFB || 0.013 ||
 +
|-
 +
| BCAL_MEV_PER_ADC_COUNT || 0.029 || 
 +
|}
  
 
== TOF ==  
 
== TOF ==  
Line 93: Line 156:
  
 
== TAGM ==
 
== TAGM ==
−
 
−
fixed in mcsmear
 
  
 
{| class="wikitable"
 
{| class="wikitable"
Line 107: Line 168:
  
 
== TAGH ==
 
== TAGH ==
−
 
−
fixed in mcsmear
 
  
 
{| class="wikitable"
 
{| class="wikitable"
Line 121: Line 180:
  
 
== PS ==
 
== PS ==
−
 
−
fixed in mcsmear
 
  
 
{| class="wikitable"
 
{| class="wikitable"
Line 133: Line 190:
  
 
== PSC ==
 
== PSC ==
−
 
−
fixed in mcsmear
 
  
 
{| class="wikitable"
 
{| class="wikitable"

Latest revision as of 12:37, 20 July 2016

Tasks

  • Dead channel handling - Sean
  • Smearing updates - Detector groups
    • See below
  • Run-dependent values from RCDB - Sean & Dmitry
  • Mixing Simulated Events

Per-Channel Efficiency

Efficiencies for simulated data should match those of real data when possible. To start with, we will compare hit-level efficiencies. To match simulated and real data efficiencies, we will calculate correction factors. These factors will have values between 0 and 1, and we will randomly drop hits at a rate of (1-factor).

The following principles should be followed when determining these factors:

  1. Efficiencies should not be set to zero unless the channel was actually disabled during the runs
  2. Efficiencies should be determined for each channel.
  3. Efficiencies should be determined either on a run-by-run basis or as an average over a group of runs
  4. Efficiencies should be determined in way that is unbiased by the trigger

New Tables

These tables contain factors to modulate the efficiency of each channel in order to match the data.

  • FCAL/block_mc_quality
  • CDC/channel_mc_quality
  • START_COUNTER/paddle_mc_quality
  • BCAL/channel_mc_quality
  • FDC/package[1-4]/wire_mc_quality
  • FDC/package[1-4]/strip_mc_quality
  • TOF/paddle_mc_quality

Current Settings

Will set table fields to "green" when updated.

Should load parameters from CCDB where possible.

General Comments

Smearing is performed by adding a Gaussian function, unless otherwise described.

TRIGGER_LOOKBACK_TIME = -100 ns

Should we properly model readout windows?

CDC

name value comment
CDC_TIME_WINDOW 1000. readout window (ns)
CDC_PEDESTAL_SIGMA 50.0 charge smearing factor
CDC_THRESHOLD_FACTOR 4.0 sparsification threshold = CDC_THRESHOLD_FACTOR* CDC_PEDESTAL_SIGMA
CDC_TDRIFT_SIGMA 2.8181818181818e-9 drift time smearing width*1.0e-9 (ns)

HDGeant calculations that need revisiting:

  • longitudinal drift
  • gain

CDC Threshold Calculation

threshold = CDC_THRESHOLD_FACTOR * CDC_PEDESTAL_SIGMA

FDC

name value comment
FDC_TIME_WINDOW 1000. readout window (ns)
FDC_CATHODE_SIGMA 150.0 cathode charge smearing factor
FDC_PED_NOISE -> charge smearing factor = -0.0938 + 0.0485*FDC_CATHODE_SIGMA
FDC_THRESHOLD_FACTOR 4.0 sparsification threshold = FDC_THRESHOLD_FACTOR * FDC_PED_NOISE
FDC_TDRIFT_SIGMA 1.0-9 drift time smearing width*1.0e-9 (ns) [both wire and cathode]

HDGeant calculations that need revisiting:

  • wire drift calculation?
  • lorentz effects?

FDC Threshold Calculation

FDC_PED_NOISE = -0.0938 + 0.0485*FDC_CATHODE_SIGMA

threshold = FDC_THRESHOLD_FACTOR * FDC_PED_NOISE

FCAL

name value comment
FCAL_BLOCK_THRESHOLD 0.02 (GeV)
time smearing 200. fixed in mcsmear (ps)
FCAL_PHOT_STAT_COEF 0.035
energy smearing -> 1.0 + Gaussian( FCAL_PHOT_STAT_COEF / sqrt(E) )

BCAL

name value comment
BCAL_TWO_HIT_RESOL 50.
BCAL_ADC_THRESHOLD_MEV 1.
BCAL_FADC_TIME_RESOLUTION 0.3 (comment: to be updated)
BCAL_TDC_TIME_RESOLUTION 0.3 (comment: to be updated)
BCAL_SAMPLINGCOEFA 0.042
BCAL_SAMPLINGCOEFB 0.013
BCAL_MEV_PER_ADC_COUNT 0.029

TOF

name value comment
TOF_SIGMA 0.155563 time smearing factor
TOF_PHOTONS_PERMEV 8000. conversion factor
dE smearing -> Gaussian(sqrt(npe))
TOF_THRESH_MEV 0 threshold on smeared dE

START COUNTER

name value comment
START_SIGMA 0.3 time smearing factor
START_PHOTONS_PERMEV 200. conversion factor
dE smearing -> Gaussian(sqrt(npe))
STC_THRESH_MEV 0 threshold on smeared dE

TAGM

name value comment
TAGM_TSIGMA 0.200 TDC smearing factor
TAGM_FADC_TSIGMA 0.350 fADC smearing factor
TAGM_NPIX_PER_GEV 1.e5 poisson smearing

TAGH

name value comment
TAGH_TSIGMA 0.350 TDC smearing factor
TAGH_FADC_TSIGMA 0.450 fADC smearing factor
TAGH_NPIX_PER_GEV 5.e5 poisson smearing

PS

name value comment
PS_SIGMA 0.200 time smearing factor
PS_NPIX_PER_GEV 1.e5 poisson smearing

PSC

name value comment
PSC_SIGMA 0.200 time smearing factor
PSC_PHOTONS_PERMEV 5.e5 Gaussian smearing
Retrieved from "https://halldweb.jlab.org/wiki/index.php?title=Mcsmear_updates&oldid=76455"