Progress Report of Region3


1) Clean room status

Starting this year we assembled during one week the clean room in the detector lab #217. The clean room has an area of 336 sft and is divided in 3 sections for
wire stringing, HV plane streching and assembly.
We are still waiting for the electrical hookup of the clean room because this has become a project of W&M  facilities planning and construction division  =8-(

2) Garfield Simulation

Graduate Student Carissa Capuano and I are working on the (de)correlation of the vertical drift position and drift time (known as xt-plot). We now have switched from Maple to ROOT
for the fit and analysis of the Garfield simulation results.  Main goal is to get an overal parameterization of the xt-plot for any track angle. This would allow a fast and simple inversion
of the xt-plot into a tx-plot, which is needed for the measured drift time to drift distance extraction.

See Elog entry : http://dilbert.physics.wm.edu/elog/Software/67?hide=0,

Current problem is a stable extraction of  two cut positions which divide the xt-plot into 3 drift section  (low cut around x =0.2cm, high cut around 0.4cm )
where different fits (linear or quadratic) are applied.

3) Geant4 simulation progress

- Juliette and I discovered independently that the Geant3 output ntuple contains strange event values. I still depend on the Geant3 output from which I extract
   the primary event values (e.g.. energy, momenta and vertex) for Geant4. Meanwhile Root had problems to convert the PAW ntuple correctly into a Root tree
   which caused more confusion on my side. Juliette provided me last week a clean Geant3 ntuple and with the newest Root version everything is working again.
   However I'm still working on an independent elastic ep event generator.

  http://dilbert.physics.wm.edu/elog/Software/51  (strange PAW ntuple)
  http://dilbert.physics.wm.edu/elog/Software/71  (clean ntuple, however strange peak in Q2 distribution)
  http://dilbert.physics.wm.edu/elog/Software/65  (Root based elastic ep event generator, no radiation correction and target energy loss included)

- Geant4 visualization is now running under Linux (Coin3D used for rendering realtime graphics: www.coin3d.org), no need for a second computer running under Windows for visualization.

- included now almost all major elements in the Geant4 simulation (http://dilbert.physics.wm.edu/elog/Software/72)



- at present: running the simulation for different VDC drift cell setting  for best tracking efficiency
    scanning for  wire angle: 30 deg   (completed, was before 45 deg)
    scanning  for cell height : 26 mm   (completed, not changed)
    scanning for wire distance : stepping 1/64" between 25/64" and 30/64"  (in progress)

- example of the hit pattern seen by the front VDC (http://dilbert.physics.wm.edu/elog/Software/77)

- Cerenkov detector simulation

In the Geant4 code the V-shaped cerenkov detector is read out with a mockup photomultiplier on each end (http://dilbert.physics.wm.edu/elog/Software/44) .
The cerenkov bar and the PMT have realistic optical and physical properties based on Neven's reports and other resources, e.g. SiO2 refraction index , PMT quantum efficieny
and the refraction index of the PMT window. You could also include more complexe photocathode properties like reflection and angle dependency.

Just for fun I turned on the cerenkov effect for the detector where Geant4 tracks the *polarized* cerenkov photons. The cerenkov bar in the simulation is naked, no wrapping material
was applied. For a better visualization the absorption for optical light in the air was set to extremly high values, otherwise the picture would be flooded with escaped cerenkov photons.
See http://dilbert.physics.wm.edu/elog/Software/45 for a wireframe visualization of the cerenkov detector with cerenkov photons inside.
In general Geant4 has a powerful handling of optical photons and surfaces (see Geant4 basic overview about  OpticalPhoton.pdf).

==> bad news: Here are my current preliminary result: http://dilbert.physics.wm.edu/elog/Software/48 , basic message: detection efficiency drops exponentially and not lineary.
                        There is clearly a minimum at the center of the cerenkov bar, the detection sum wil

==> good news: Postdoc Michael Gericke is getting started to take a deeper look into this important matter  =8-)


That's all folks !