Richard
Richard
reported on general items:
Ashok (who was delayed for this meeting) made tremendous progress:
He got Athena running on the fates!
He produced an ntuple with 10 events. Tayfun had a look at it and
found that all the entries were there, but he hasn't had time to
check them out in detail.
Warren Shaw, who joined our meeting for the first time, is getting
set up with a computing account at CERN.
For next week, we should all send an email to Michel about our travel plans
for the foreseeable future, so we'll have some idea of who is going to
be where and when.
Margret
reported that she found out the the setup for the material runs
is already implemented in LArG4TB. Using a visualization tool, she
made the additional iron plates visible. The large walls in
front pot the cryostat (Fe, Snit., Pb) are also present in the
code, but their `holes' have been coded as long vertical slits.
Margret sent an emails to Peter and Pavla to find out the
proper size of the holes and just before the meeting got back an
email from Andrei Grunion with his best knowledge of the
hole sizes.
Margret also looked at leakage in terms of particle leaving the
detector. She booked some histograms and plotted the xyz positions
of the point in space where a particle is leaving the HEC,
the energy of the particle as it is leaving the HEC, and the
type of particle leaving the detector.
Plotting the x vs y coordinates to the detector exit point of
the leaked tracks leads to an outline of the detector boundary.
This plot revealed that it is important to specify the option
`hec-type=sa' when running the code. (Omitting this option
implements the entire HEC wheel into the testbeam!)
Michel and Richard pointed out that Margret should concentrate on
finding out how to get at the energies deposited in the cells
by electromagnetic processes and by hadronic processes separately.
She should also think about whether it would be a good idea to invite
a LArG4TB expert to UVic or to go to Munich for a meeting, and to
take action on that soon.
Tayfun
had more updates on his phi modulation response. He showed the
results of his event-by event phi-modulation-corrections of the x-position
measured by the calorimeter and finds that the energy fluctuations in
phi go down, but are eliminated. Michel pointed out that the position
reconstruction in the calorimeter actually introduces a bias, which
may be why the correction does not have a more dramatic effect..
Tayfun plotted the peak-to-peak height of the (energy response vs phi) plot
against the beam energy and found that it has a linear dependence.
He also looked at the global time and finds that he has to do a
very minor correction only - Rob and Naoko's treatment of the pulse-heights
really improved things.
After applying the corrections Tayfun started to look at the systematics.
He plotted sigma(x) vs. x and found a very distinct structure, with
larger sigma(x) at the outer x-positions of the calorimeter.
Michel thought that this might be due to leakage.
Tayfun also plotted (x-mwpc - x-calo) vs. x-mwpc and finds
a linear dependence.
The general feeling that this would likely be due to geometric effects.
Tayfun will now continue his systematic studies.
Tamara
looked at the details of the HV correction and will have results next week.
Ashok
was tied up in another meeting and joined the testbeam meeting toward the end.
He reported that he has been experimenting with Athena releases 7.5, 7.8 and 8.0
and found that all of them are working on Mercury.
He produced an ntuple the night before the meeting. The only problem he encountered
was afs timing out on Mercury, but that problem is not present on the fates, and besides,
data can always be obtained via ftp.
The next thing he'll try doing is to checkout a package and modify the source code.
Ashok also thinks that things would be a lot more stable if it
we could bring the conditions database here.
Michel
has explored running Athena within the ask (`Athena-Startup-Kit') environment,
and gave us a very nice rundown on how to use it and how it compares to
running Athena `the old way'.
A somewhat updated version on the ask procedure that worked is given here:
How to run Athena from scratch, modify some code, using ask
% cd $HOME
% mkdir work
% cd work
Launch the athena startup kit
% ask
Checkout LArCalorimeter package
>>> checkout('LArCalorimeter', '8.0.0')
Set up the running environment
LArCalorimeter> setup()
For example, Checkout LArHECTBAna package
LArCalorimeter> checkout('LArCalorimeter/LArTestBeam/LArHECTBAna', '8.0.0')
Now you can edit a file, for example the file
${HOME}/work/LArCalorimeter/LArTestBeam/LArHECTBAna/LArHECTBAna-02-04-02/src/LArTBSignalBuilder.cxx
If you have edited a file, you can compile the LArHECTBAna package. The first time
this can take some time since all files need to be compiled
LArHECTBAna> make()
You can find out what are the available jobOption files
LArHECTBAna> LArHECTBAna.scripts
Run Athena. Results in ${HOME}/work/LArCalorimeter/LArTestBeam/LArHECTBAna/LArHECTBAna-02-04-02/run
LArHECTBAna> run('share/LArHECTBAna_jobOptions_combined.txt')
You can then get out of ask with ctr^D
Next time you login, you do
% cd ${HOME}/work
% ask
>>> select('LArCalorimeter')
LArCalorimeter> setup()
LArCalorimeter> select('LArHECTBAna')
LArHECTBAna> run('share/LArHECTBAna_jobOptions_combined.txt')