Project

General

Profile

"lbne35t4apa_v3" -- NOT PUBLIC YET

Author:

----------------------------------------------------
----------------------------------------------------

Passes GeometryTest, free of overlaps

One last issue before committing -- after all the changes, wrong number of channels again. Will be fixed by friday. If you would like to beta-test this with your code before it is public, go to your lbnecode/lbne/Geometry/gdml directory and:

cp /lbne/app/users/talion/Dev35/srcs/lbnecode/lbne/Geometry/gdml/lbne35t4apa_v3.gdml .
cp /lbne/app/users/talion/Dev35/srcs/lbnecode/lbne/Geometry/gdml/lbne35t4apa_v3_nowires.gdml .
cd $MRB_BUILDDIR
make install

Then add the following line at the end of your job script:

services.user.Geometry.GDML: "lbne35t4apa_v3.gdml" 

----------------------------------------------------
----------------------------------------------------

lbne35t4apa_v3 has changed significantly from lbne35t4apa_v2. There are new volumes throughout the cryostat, finally bringing its painfully asymmetrical quality to life. More importantly, there is significantly more detail (up to <<1mm accuracy around the APAs themselves), making this a good geometry for conducting studies on effects of particles crossing through an APA or into other APAs. Cold dimensions are used.

Origin / Coordinate System

The reasoning behind the placement of the origin is historical, following MicroBooNE's lead, as well as being constrained by LArSoft requirements. The APAs are in the same yz plane, with y pointing vertically and z horizontally. The x-axis is orthogonal to this plane, with the short drift side in the negative x and the long drift side in the positive x. The x=0 center is at at the face of the first wire plane an electron would encounter (U) drifting on the long drift side. The y=0 center is exactly in between the two middle, medium-sized APAs. The z=0 center is on the very "front" face of the active region of liquid argon.


Below is a diagram (not to scale) depicting the x=0 center in detail. The perspective is looking down in the negative y direction from above. It is very hard to render this with the real GDML.

Volumes

APAs

  • Frame Tubes -- stainless steel, hollow, liquid argon on the inside.
  • G10 Plastic Boards -- Physically used to wrap wires and impose plane pitch. May have a rare but interesting effect on through-going particles
  • Wire Planes -- Different active widths/heights for each (insightful figures to come)
Missing:
  • G10 grid board at readout end of APAs, which overlaps the wire planes and creates a dead region of wires. This region is not even included in the length of the generated wires since the endpoints of the active region are the desired endpoints in LArSoft.
  • Grid plane outside of the U plane, which is used to control the field. This has too many holes to behave well in Geant4-GDML tracking.

Cryostat (default material LAr)

  • Neck -- filled with gaseous argon (technically this is in volDetEnclosure so Cryostat remains a box)
  • Concrete and Foam padding, Steel Shell -- also in volDetEnclosure
  • CPAs -- broken down into tube framing and very thin, biased sheet.
Missing:
  • Field cage. This has too many holes to behave well in Geant4-GDML tracking, so this cage should probably be parameterized in the source code.
  • Any purification instrumentation in and below the neck.

Photon Detectors

-- to come --

Detector Enclosure / World (default material Air)

  • Neck
  • Steel Shell, Foam Padding, Concrete exterior
Missing:
  • Concrete walls of the hall
  • Dirt on the sides and below
  • Hall roof?
  • Building immediately above

Caveats

For CRY simulations:
  • No detector enclosure volumes such as concrete walls (except of course the ones around the cryostat)
  • No world volumes such as the dirt along the sides of the hall and below, or the air above.
  • Default is air, though, so the cryostat is effectively just suspended in air
For optical simulation:
  • No field cage (or PDs yet?)
  • No grid plane on APAs
Other:
  • CPAs slightly more removed from the active drift region than in reality. This was to avoid a particularly tricky overlap for now, but will not risk the correctness of what is drifted.