FEDRA reconstruction and csv conversion
Specific notes for our configuration, keep track how what I did wrong and right
Intro
Always keep track of parameters. Always check if parameters change between bricks/walls/tests/other.
Reconstruction procedure
Making scanset
makescanset -set=21.0.0.0 -from_plate=60 -to_plate=1 -dz=-1315 -suff=cp.root
Linking notes for SND@LHC
Due to high track density, trying to perform unique linking leads to failure.
Therefore, two linking prodecure is used:
Find shrinkage correction, requiring Theta > 0.05. Also, in a small central region, 2.5 cm x 2.5 cm. Shrinkage and angular corrections are computed
Use shrinkage correction from previous computation. Shrinkage and angular correction computations are disabled (DoCorrectShrinkage and DoCorrectAngles are set to 0). All surface, all angles
Note: even if the region is the same, it appears that there are more couples in firstlinking. However, checking the "good couples" (eCHI2P<2.4&&s.eW>20&&eN1<=1&&eN2<=1&&s1.eFlag>=0&&s2.eFlag>=0), numbers become compatible, as tested with background check films linking.
Also, global linking in nusrv1 required some ad-hoc changes:
DoubleFilterOut with cell 6, instead of 1 (not recommended in normal operation);
Increase maximum TTree size (safer change)
However, testing in RUN0 standard linking did not lead to appreciable changes in track quality (b000431 test).
Track and vertex reconstruction
emtra -set=21.0.0.0 -new -v=2
Remember to change the x and y ranges in track.rootrc according to the target unit we are considering.
A file *.trk.root will be created. Link it to linked_tracks.root
ln -s b000021.0.0.trk.root linked_tracks.root
Then we are ready for actual vertex reconstruction:
root -l vertexing.C
A file vertextree.root will be created
Csv conversion
python csvconversion.py
Two files will be created:
brick21.csv, containing all the information about the couples
brick21vertices.csv, containing the vertices and tracks ID from recosntruction
Summary of all bash scripts commands to do
Bash scripts should be found in macro_snd GitHub, copy them into your main conversion folder. Check that paths are set properly.
The main idea is that the previous operations should be performed for all bricks, via .sh commands:
Create folders b0000{11..14, 21..24,31..34,41..44,51..54}/p001...p060
copy 2FEDRA.rootrc and check the settings;
source preparebricks.sh
source doreco.sh
source csvconversion.sh
source addvertexinfo.sh
hadd vertextree_allbricks.root */vertextrree.root
Last updated