Minimal Upgrade Layout

It consists of the minimal set of improvements to the existing LHCb detector without introducing major changes.

This is not necessarily a realistic upgrade model, but it is intended to help us understand limitations of the layout presently employed in LHCb and motivate new ones.

It is also used as a stepping stone in development of the LHCb software to allow for upgrade simulations to co-exist with the simulation code of the present detector.

Versions v1.0 and v1.1

The first large upgarde MC data sets were generated in this model. Later we also generated MC with more recent v2.1 (see below).

Geometry (Gauss)

These versions implement following changes in the detector geometry ( see the presentation at the July 15, 2009 Upgrade Meeting ):

  • Thinner RF foil (xml code supplied by Thomas Latham). Without changing the foil geometry, the foil material is diluted to cut down its radiation thickness by half (preliminary studies by Ray Mountain indicate that this may be feasible).

  • T-stations (xml code supplied by Jan Amoraal). Z-positions of IT and OT planes in each T-station are swapped. Moving IT behind OT should help OT occupancy problem.

  • Aerogel radiator is removed in version v1.0. Version v1.1 has aerogel in. This is the only difference between these two versions.

Digitization (Boole)

In addition to the changes in geometry (affecting Gauss step), there are two spill-over suppression measures implemented in Boole:

  • OT readout gate is reduced from 75ns to 50ns

  • Spill-over suppression assuming 40 MHz readout is implemented in all silicon detectors (VELO,TT,IT). The implementation by Tomasz Skwarnicki & Marcin Kucharczyk is described in Sep.09, 2009 presentation.. In this implementation only subtraction of the spillover from "Previous" to the "Current" bunch crossing is implemented. After the data had been produced, we realized that the spillover from "Next" to the "Current" in VELO detector is equally important (see Hans Dijkstra's Nov.11,2009 presentation.). Therefore, the produced MC data samples overestimate the spillover background in VELO.

Produced data in LHCb database

All productions were done "with MC truth" info in Boole and Brunel steps. So far the following data were produced:

   Event type                         Statistics (at each luminosity)

   13104011 Bs_phiphi=DecProdCut     ~500,000 
   30000000 minbias                ~2,000,000 
   10000000 incl_b                    ~25,000 in xdst format (contains MCHits 
                                                                  usually available only in .sim files)

 

The data can be found in the LHCb bookkeeping data base under "MC -> Upgrade" tab and "Beam7TeV-UpgradeML1.1-MagDown-LumiX-25ns" beam/detector conditions where x=2,5,10,20 stands for luminosity in units of 10^32 cm^-2 s^-1.

There is also a single data set without aerogel in RICH1 of Bs_phiphi=DecProdCut (500,000 events) at Lumi2 under "Beam7TeV-UpgradeML1.0-MagDown-Lumi2-25ns"

For comparison with existing LHCb detector, we produced one sample of Bs_phiphi=DecProdCut (500,000 events) at Lumi2 with "MC09" geometry which can be found under "Beam7TeV-VeloClosed-MagDown-Lumi2-25ns". Unlike other MC09 MC data sets this one includes spill-over simulation. In the bookkeeping database it is placed under "Upgrade" folder.

Hans Dijkstra run official Bs->phiphi offline selection on Bs_phiphi events (only a fraction of full statistics) and stored the selected events at: /castor/cern.ch/user/d/dijkstra/Selections-Upgrade/LHCbPhysics.Bs2PhiPhi-upgrade-minimal-lX.dst, where X=2, 5,10 and 20.

Versions v2.1

Upgrade MC appearing in bookkeeping data base with "Beam7TeV-UpgradeML2.1-MagDown-LumiXX-25ns" (with aerogel) were generated with Gauss v38r5, Boole v21r5, Brunel v37r4. The database tages can be found at: $APPCONFIGOPTS/Conditions/Upgrade-20100617-md100.py:

LHCbApp().DataType  = "Upgrade"
LHCbApp().DDDBtag   = "mul-20100617"
LHCbApp().CondDBtag = "sim-20100510-vc-md100"
The geometry modifications compared to the standard LHCb detector are the same as outlined above. However, these changes where implemented on LHCb DDDB tag "head-20100510" (which is newer than the previous MUL geometries).

Improvements in spill-over suppression in Velo (suppression of feed-across from the "Next" bunch crossing) and in TT (work by Adam Webber) were implemented.

The following data sets were produced at each of 4 luminosity points:

   Event type                         Statistics (at each luminosity)

   13104011 Bs_phiphi=DecProdCut     ~50,000 
   30000000 minbias                 ~200,000 
   10000000 incl_b                   ~50,000 
   11114001 Bd_Kstmumu=DecProdCut    ~50,000
   11124001 Bd_Kstee=DecProdCut      ~50,000
   13102201 Bs_phigamma=DecProdCut   ~50,000

 

How to generate your own MC data

Instructions below apply to v1 geometry; for more recent v2 geometry please use release numbers and database tags shown above.

Gauss v37r5

Example how to generate events with Minimal Upgrade Layout v1.0 (without aerogel), beam energy of 7 TeV and luminosity 2 x 10^33 ("Lumi20"):

     gaudirun.py $APPCONFIGOPTS/Gauss/Upgrade.py \
                 $APPCONFIGOPTS/Gauss/MC09-b7TeV-md100-nu6,8-25ns.py \
                 $APPCONFIGOPTS/Conditions/Upgrade-20090917-md100-gauss.py \
                 $DECFILESROOT/options/eventType.opts \
                 $LBPYTHIAROOT/options/Pythia.opts \
                 Gauss-Job.py

(you can also replace the first 3 files with $GAUSSOPTS/Gauss-Upgrade.py which references them).

See $GAUSSOPTS/Gauss-Job.py for example of the last input file (this is where you define name and location of the output file and how many events to analyze).

Use following $APPCONFIGOPTS/Gauss/ files for different luminosities:

   * Lumi2:   2 x 10^32   MC09-b7TeV-md100-nu0,7-25ns.py  
   * Lumi5:   5 x 10^32   MC09-b7TeV-md100-nu1,7-25ns.py  
   * Lumi10: 10 x 10^32   MC09-b7TeV-md100-nu3,4-25ns.py  
   * Lumi20: 20 x 10^32   MC09-b7TeV-md100-nu6,8-25ns.py
To generate with aerogel in place use $APPCONFIGOPTS/Conditions/Upgrade-20090917-md100.py (instead of Upgrade-20090917-md100-gauss.py). Please notice that both versions must be processed with Upgrade-20090917-md100.py in Boole and Brunel steps (which is a default tag for the upgrade simulations).

Boole v19r6

     gaudirun.py $APPCONFIGOPTS/Boole/Upgrade-WithTruth.py \ 
                 $APPCONFIGOPTS/Conditions/Upgrade-20090917-md100.py \ 
                 Upgrade-Files.py

you can also do:

     gaudirun.py $BOOLEOPTS/Boole-Upgrade-WithTruth.py \
                 Upgrade-Files.py

which relies on default tags set for the upgrade simulations.

See $BOOLEOPTS/MC09-Files.py for example of how Upgrade-Files.py should look like (defines input and output files).

You can also use "-NoTruth" versions if you don't care about MC truth info being saved.

Brunel v35r6p1

     gaudirun.py $APPCONFIGOPTS/Brunel/Upgrade-WithTruth.py \
                 $APPCONFIGOPTS/Conditions/Upgrade-20090917-md100.py \ 
                 Upgrade-Files.py

See $BRUNELOPTS/MC09-Files.py for example of how Upgrade-Files.py should look like (defines input and output files).

You can also use "-NoTruth" versions if you don't care about MC truth info being saved.

Panoramix (instructions below tested with v17r5)

     python $myPanoramix -D MC09 -c upgradePanoramix.py <other options e.g. "-f none" for no input file> 

Where upgradePanoramix.py file contains a single line:

     User_opts = upgradeDDDB.py
and the upgradeDDDB.py file redirects Panormaix to the upgrade geometry DB:
from DetCond.Configuration import *
CondDB().PartitionConnectionString = {"DDDB":"sqlite_file:$SQLITEDBPATH/DDDB_upgrade.db/DDDB"}
from GaudiConf.Configuration import LHCbApp
#   Min.Upgrade.Layout v1.1 with aerogel 
#  (for v1.0 geometry with no aerogel use "mul-gauss-20090917"; cannot be used for any reco in Panoramix)
LHCbApp().DDDBtag   = "mul-20090917"
# same as for MC09
LHCbApp().CondDBtag = "sim-20090402-vc-md100"

-- TomaszSkwarnicki - 09-Oct-2009

Edit | Attach | Watch | Print version | History: r6 < r5 < r4 < r3 < r2 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r6 - 2010-10-28 - TomaszSkwarnicki
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    LHCb All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback