DF Logbook
2018
May
Week of the 6th
I acquired and installed GEMC, COATJAVA, and CED on Ubuntu 16.04. Installation instructions can be found here.
GEMC Install Notes
GEMC and the common environment are found here
There was some trouble in installing the common environment (ce) due to an awk program (mawk, in this case) segfaulting. I was able to get around this by downloading a different awk program, gawk. After downloading the ce I had some trouble with the default version of Geant4; updating the Geant4 version numbers in ce/settings.xcconfig and ce/versions.env to 4.10.04.p01 resolved the issue.
After the GEMC install proper, it was also necessary to install the clas12Tags repository, where scons took issue with an undefined variable in physcisList.cc. Uncommenting the line which defined theParticleIterator allowed the tags to be compiled.
COATJAVA Install Notes
Release versions can be found here. Otherwise, one must clone either the development version or a fork of it.
Attempting to run any of the coatjava scripts with ./ChupaThingy
throws a source not found error, use bash ChupaThingy
instead.
CED Install Notes
Downloads can be found here.
Week of the 13th
Finished Troubleshooting installs for the most part.
Wasn't seeing any meaningful events from GEMC. Took the issue up with Dr. Forest who showed that it was likely a combination of bad starting vertex placement and small sample size. He suggested using Lund files instead. I currently have one that produces 26 usable events. In time I'm going to have to learn to create my own.
Spent some time learning how to format the wiki.
Updating the development code to the most recent master branch from JLab seems to have fixed the magfield map problem. The environmental variables still have to be set, but they are the generic names again. Instructions for setting up the development code to be updated from the JLab code can be found here.
Week of the 20th
Spent Monday and most of Tuesday updating this wiki and tackling the database problem.
The version of clas12Tags I'm for using for GEMC (4a.2.3) doesn't seem to use the database at all. Will try with 4a.2.2 and see if that uses the database at all. As for the IDE, I've set up the environmental variables CCDB_CONNECTION=sqlite:///../../DB/clas12.sqlite.05162018
and CCDB_DATABASE=../..DB/clas12.sqlite.05162018
. The previous version of clas12.sqlite I was using apparently lacked some fields that coatjava wanted, but this one appears to work.
Spent the rest of the week updating the wiki and defining the reconstruction elements.
Requested Pictures
Electron scattering event via CED at 27 degrees and 1.2 GeV, phi=-4.0,
Screenshots of IDE output window with and without non-zero vertex.
Week of the 27th
I have come to the conclusion that my current method of going through the code and exhaustively defining all of the variables and functions is a fool's errand. The current plan is to find tools to build UML diagrams from the code itself and find an easy way to build flowcharts.
UML
Currently attempting to figure out how to use the easyuml plugin for NetBeans. While I can make uml diagrams for single classes, I can't seem to get even an inheritance tree working.
FlowCharts
I've downloaded the packages graphviz sudo apt-get install graphviz
and used pip to install pygraphviz sudo pip install pygraphviz
. Documentation on how to use the libraries can be found here and here.The script RecoFlowChart.py has been written such that it pulls in RecoFlowChart.gv and spits out RecoFlowChart-yy-mm-dd.png.
He3 Tube Project
I believe the CAEN Mod. N842 CFD that I"m using is not working to spec. After looking at the documentation, It should be triggering at voltages well below -5mV, but it isn't.
Procured an Ortec 672 Spec Amp from the IAC.
Clas 12 Collab Call
Had significant problems with the audio and missed much of the call.
June
Week of the 3rd
Thesis
Dug into the code some more and created flowcharts for an Overview, DCHBEngine.processEvent, and DCTBEngine.processEvent.
He3 Tube
Created a flowchart to describe the connections between the electronic components of the setup.
Collected data with the threshold set in increments of 10mV between 10-250mV, inclusive. Didn't find the end of the plateau... will consult with Forest.
Added a CAEN N110 Dual attenuator between the spec amp and LED. Set to 20 dB should result in a 90% reduction of voltage.
Attenuator Problems:
When set to not attenuate, I'm still getting attenuation ~90% at 20mv. When set to 20dB, it appears to be attenuating more than that.
Week of the 10th
Thesis
Discussed with Forest what exactly I'm doing as I've lost the thread a little. My current goal is to dig into the code in the direction of the Swimmer so that I can understand what is occurring there and modify it.
He3 Tube
Characterized attenuator with pulse generator, It appears to be working more or less exactly as it should now that Forest has corrected my screw-ups. Took data points with the attenuator now mixed in: 1-15mV in 1mV increments, 15-45 in 5mV increments, and 45-55 in 1 mV increments.
Added the attenuator to the Hardware Flowchart.
Learned how to use Grace (graphing tool).
Week of the 17th
Thesis
He3 Tube
Finished Taking data for the no source run. Produced graphs using that data. Will wait for Forest's approval of those before taking data with the source.