Difference between revisions of "Tuesday, July 10, 2012"

From Hall A Wiki
Jump to: navigation, search
(→‎EM Physics I)
 
(8 intermediate revisions by the same user not shown)
Line 1: Line 1:
−
==Physics List 1 Notes ==
 
−
*Slide 12
 
−
** To aid in ConstructProcess() a function ConstructEM() may be defined to construct all the ElectroMagnetic Processes.  In this function, it is '''recommended''' that you use the G4PhysicsListHelper to take care of initializing the process in the proper order!!! This is new starting with 9.5
 
−
 
−
 
−
Tuesday, July 10th, 2012  - see talks online as well
 
−
 
 
== Morning Talks ==
 
== Morning Talks ==
  
Line 56: Line 49:
 
=== Physics II ===
 
=== Physics II ===
  
−
*
+
:: Not much is said that is not on the slides
−
*
+
 
 +
* EM
 +
** standard
 +
** low energy - 100 eV and below
 +
** optical photon
 +
<br>
 +
* Hadronic
 +
** Pure hadronic
 +
** Strong radioactive decay
 +
** Photonuclear
 +
** Leptonuclear
 +
<br>
 +
* Decay and Parameterization
 +
** weak decay
 +
** EM decay
 +
** Parameterized - EM showers
 +
<br>
 +
* Processes
 +
** final state - all the changes you are going to apply to incident particle
 +
** energy loss, momentum change, generation of secondaries
 +
** mixed processes are a mixture of discrete, continuous and At Rest
 +
<br>
 +
* threshhold is a distance, not an energy - 0.7*mm in PhysicsList
 +
** different energy production threshholds depending on energy
 +
** transition between discrete and non-discrete mode - not actually stopped
  
 
=== Geometry II ===
 
=== Geometry II ===
 +
* Copy number starts from zero and is incremented
 +
* G4PVParameterized
 +
** GEANT4 doesn't know how to parameterize your volume unless you tell it
 +
** can't use BREPs solids
 +
** Can't get copy number of the mother volume
 +
** using nested parameterizations - can change position but not size using replicas
 +
 +
<br>
 +
* G4 Replica
 +
** slide 18 - Radial axis is called - kRho, not kRaxis
 +
 +
* Parameterized and Replica are different
 +
** Replica is faster - if they are all the same
 +
** but if you need gaps in between need to use parameterized
 +
 +
* Note that mother volumes can be tubes with nothing in the center, for example
 +
* will still fill the volume - even if you give an offset
 +
 +
<br>
 +
* G4 Division
 +
** may have gaps in this case at "top" and "bottom"
 +
** side surfaces should be touching the mother
 +
 +
* G4 Replicated Slice
 +
** can have gaps between daughters
 +
<br>
 +
* Nested parameterization
 +
** slide 28 - no actual gaps, just for illustration
 +
** this is the exception to putting parameterized volume inside a replica
 +
*** not changing the size, just the placement
 +
<br>
 +
* Assembly volume
 +
** acts as an envelop for the daughter volumes!
 +
<br>
 +
* Touchable
 +
<br><br>
  
 
=== Visualization II ===
 
=== Visualization II ===
 +
:: Could use some more notes here
  
 +
* /vis/open
 +
** x, y sizes of window and position on screen relatively to top and right hand side
 +
** Autorefresh false good for getting configuration correct
 +
* slide 10 - simplest is open window, draw detectors
 +
* axes are x, y, z -  red, green, blue (I think)
 +
* Draw hits draws all hits in hits collection(s)
 +
* need to use smooth trajectory to get the control points to do the circular arc
 +
* /vis/verbose errors <-- only tell me if I did something stupid
 +
* /vis/verbose warnings <-- tell me you could draw something, even if it is okay for tracking
 +
* OpenGL postscripts good enough for talks; pub quality - use DAWN
 +
* some commands call multiple lower level commands; error messages will come from lower level
  
 
== Afternoon Talks ==
 
== Afternoon Talks ==
 
=== Primary Particle ===
 
=== Primary Particle ===
 +
:: Could probably use some more notes here.
 +
* UserPrimaryGeneratorAction should NOT generat particles, but just control it
 +
** this method is evoked at each event (no UI commands)
 +
 +
<br><br>
 +
 
=== Physics III ===
 
=== Physics III ===
−
=== EM Physsics I ===
+
* no absolute cutoff - everything tracked to zero energy
 +
* cut 450 keV is fine - but wastes CPU time with delta rays in Pb
 +
 
 +
<br><br>
 +
 
 +
=== EM Physics I ===
 +
:: not much is said that is not on the slides
 +
 
 +
* you have to check the settings for step limits
 +
* example - slide 33
 +
 
 +
<br><br>
 +
 
 
=== Hands On II ===
 
=== Hands On II ===
 
<br><br> Return to [[JLAB GEANT4 Workshop, July 9-13, 2012]]
 
<br><br> Return to [[JLAB GEANT4 Workshop, July 9-13, 2012]]
 
<br> Return to [[MOLLER GEANT4 Simulations]]
 
<br> Return to [[MOLLER GEANT4 Simulations]]
 
<br> Return to [[MOLLER at 11 GeV E09-005]]
 
<br> Return to [[MOLLER at 11 GeV E09-005]]

Latest revision as of 15:19, 10 July 2012

Morning Talks

Physics I

  • slide 10 - can explicitly list particles
  • Need to tell the particles that they need to move
  • can group EM physics and all other physics if you would like
  • slide 12 - while writing methods in Physics list - use physics list helper (new 4.9.5)
    • helps with ordering
    • list processes, it will put them in correct order
    • you do need to instantiate a particle iterator
  • ConstructGeneral is "fictitous" - you can make it anything you like
    • need to loop over all particles (not just those that can decay)
    • check to make sure decay process is applicable for that particle
  • Cuts
    • final mandatory method you have to implement
    • 0.7*mm recommended for most applications
    • needed for all 4 (gamma, e-, e+, proton) and no others (even neutrons)
    • can change cut for individual particles (don't have to be the same)
    • if you use a standard package, it wll be set


  • Modular physics lists
    • some nice features, like certain things automatic
    • basically a bookkeeping device for EM vs hadronic processes
    • for example proton, electron, decay physics all separate
    • still have to do setCuts, but it is simpler


  • Prepackaged physics lists - the "easy" way to do it
  • adding hadronic physics is more involved than EM physics
    • need to use models
    • for example - for neutrons - there is a basic model, then there are HP_neutron model
    • there are several physics lists which have been validated by experiment
  • geant4.cern.ch/support/proc_mod_catalog/physics_lists/physicsLists.html
  • really are a "best guess" of the physics needed in a given use case
  • may not do exactly what you want
  • possible to use prepackaged physics lists and remove one process and replace it


  • Reference Physics Lists
    • subset of prepackaged physics lists


  • Questions
    • Hadronic - Bertini vs. binary cascade - you have to decide - need to validate
      • Bertini handles more particles
      • Binary better for lower energies
    • photonuclear - EM or Hadronic - both! - see future talk




Physics II

Not much is said that is not on the slides
  • EM
    • standard
    • low energy - 100 eV and below
    • optical photon


  • Hadronic
    • Pure hadronic
    • Strong radioactive decay
    • Photonuclear
    • Leptonuclear


  • Decay and Parameterization
    • weak decay
    • EM decay
    • Parameterized - EM showers


  • Processes
    • final state - all the changes you are going to apply to incident particle
    • energy loss, momentum change, generation of secondaries
    • mixed processes are a mixture of discrete, continuous and At Rest


  • threshhold is a distance, not an energy - 0.7*mm in PhysicsList
    • different energy production threshholds depending on energy
    • transition between discrete and non-discrete mode - not actually stopped

Geometry II

  • Copy number starts from zero and is incremented
  • G4PVParameterized
    • GEANT4 doesn't know how to parameterize your volume unless you tell it
    • can't use BREPs solids
    • Can't get copy number of the mother volume
    • using nested parameterizations - can change position but not size using replicas


  • G4 Replica
    • slide 18 - Radial axis is called - kRho, not kRaxis
  • Parameterized and Replica are different
    • Replica is faster - if they are all the same
    • but if you need gaps in between need to use parameterized
  • Note that mother volumes can be tubes with nothing in the center, for example
  • will still fill the volume - even if you give an offset


  • G4 Division
    • may have gaps in this case at "top" and "bottom"
    • side surfaces should be touching the mother
  • G4 Replicated Slice
    • can have gaps between daughters


  • Nested parameterization
    • slide 28 - no actual gaps, just for illustration
    • this is the exception to putting parameterized volume inside a replica
      • not changing the size, just the placement


  • Assembly volume
    • acts as an envelop for the daughter volumes!


  • Touchable



Visualization II

Could use some more notes here
  • /vis/open
    • x, y sizes of window and position on screen relatively to top and right hand side
    • Autorefresh false good for getting configuration correct
  • slide 10 - simplest is open window, draw detectors
  • axes are x, y, z - red, green, blue (I think)
  • Draw hits draws all hits in hits collection(s)
  • need to use smooth trajectory to get the control points to do the circular arc
  • /vis/verbose errors <-- only tell me if I did something stupid
  • /vis/verbose warnings <-- tell me you could draw something, even if it is okay for tracking
  • OpenGL postscripts good enough for talks; pub quality - use DAWN
  • some commands call multiple lower level commands; error messages will come from lower level

Afternoon Talks

Primary Particle

Could probably use some more notes here.
  • UserPrimaryGeneratorAction should NOT generat particles, but just control it
    • this method is evoked at each event (no UI commands)



Physics III

  • no absolute cutoff - everything tracked to zero energy
  • cut 450 keV is fine - but wastes CPU time with delta rays in Pb



EM Physics I

not much is said that is not on the slides
  • you have to check the settings for step limits
  • example - slide 33



Hands On II



Return to JLAB GEANT4 Workshop, July 9-13, 2012
Return to MOLLER GEANT4 Simulations
Return to MOLLER at 11 GeV E09-005

Retrieved from "https://hallaweb.jlab.org/wiki/index.php?title=Tuesday,_July_10,_2012&oldid=16014"