EMC-EMT Noise Tower Calibration
Taking the Data
There are several steps to this and I will attempt to go through them as
thoroughly as possible.
-
Run cratestart from any machine in the Control Room. (This is located in
$BFROOT/detector/trg/emt)
-
Log into the three windows that appear.
-
In the window labeled EmtCrate type "xyplex of-emt" This
will enable you to watch the ROM running through the test.
-
In the window labeled Control Level change directory into a release
and srtpath. Then type "RunCalManager OrcCyles". You will then be
asked to enter a crate mask. For the EMT this should be 400. You
will then be asked to enter the configuration key value, WAIT
-
In the window labeled Event Level also change dir into "/nfs/babar/calib/emt".
From here you should then run "$BFROOT/detector/trg/emt/EmtTCALNoise_logger".
This will automatically start up correctly opening another window.
-
You may now enter the correct configuration key. For this calibration the
value is 82.
-
The system will now run through 7 major cycles each containing 100 Minor/L1Accepts.
As each Major cycle completes you will see a message in the OEP window
stating that there has beenchanged event damage of 0x2000. This is completely
normal.
-
Once this has finished running you will see a message in the Control Level
window asking you to either enter another config key value or 0 to exit.
Type 0 <CR>.
-
All extra windows that appeared will no dissappear and the Control &
Event level windows will return to their prompts. You will now need to
rename the resulting file that has been made. In the Event level window
do an "ls -al babar-emc-emt*". This will give ou a list of the previously
done calibration xtc files & one with the same first part to the name
but with the normal 56bit ending. This should be renamed to the next regular
number in the babar-emc-emt-tcalXXXXXXXX.xtc list (eg. babar-emc-emt-tcal00000001.xtc).
-
You are now ready to analyse this data.
Analysing the Data
After renaming the file to something consecutive with the other taken Tower
Calibrations you can now using OEP look at the data you have just taken.
It is best to do this on a machine such as Shire or Percheron to speed
up this process since the detector will be waiting for you to finish.
-
[shire01]> cd $BFROOT/detector/trg/2.8.0 (or
any recent online release)
-
[shire01]> srtpath
-
[shire01]> ir2boot
-
[shire01]> cd ../emt
-
[shire01]> ./noisetower <run-name>
This will produce an hbook file in the dir ./hbooks.
You may look at this now with your choice of paw variants
Therefore in paw:
PAW>h/fil 1 hbooks/babar-emc-emt-tcalXXXXXXXX.hbook
PAW>exec tcalmon
This will produce as output a ps file in the dir ./psfiles. It
is best if this is now printed and looked at.
Now the following procedure must be followed.
[shire01]> cd /nfs/bbr-srv02/dataflow/constants/emt
[shire01]> cp aaaalink.mask run<next_run_number>.mask
[shire01]> chmod 744 run<next_run_number>.mask
[shire01]> emacs run<next_run_number>.mask& This will
bring up a file with the same format as below up for editing
Start off by looking at the plot labeled Phi
Energy V Time Theta 0; As can be seen from this example the following
toers are hot for theta 0. You must remeber that the Phi numbering runing
up the Y axis start from bin 0!
Hot towers = 1, 13, 16, 18, 21, 27, 32, 33, 34 & 35
The file run<next_run_number>.mask must now be altered to
reflect these towers that must be masked off. A 0 represents a masked tower.
All the other towers in the fiule below have been change to reflect
the results shown by the following 6 plots.Towers 1,
2, 3, 4,
5 & 6
Example of a mask file below;
The last thing that you now must do is in the /nfs/bbr-srv02/constants/emt
directory;
-
[shire01]> rm aaaalink.mask
-
[shire01]> ln -s run<next_run_number>.mask aaaalink.mask
You are now ready to hand the system back over to the shift crew.
NOTE: Editing the mask file by hand is a temporary measure which will
soon become automated.
EMC-EMT Serial Number Calibration
Taking the Data
The basics of taking this calibration are the same as the above Noise calibration
though a different logger is used & a different crate mask. In the
Event Leval window type EmtTCALSerial_logger instead of EmtTCALNoise_logger.
All this does is to use a differnt crate mask for running the calibration.
The crate mask for this is:
Crate Mask = 7ff
Since this is a different type of calibration then a different Calibration
key must therefore be used. In this case it is 81.
The usage of RunCalManager and the xyplex is exactly the same as above.
Analysing the Data
Here you also want to be logger onto a machine such as shire. The startup
procedure invloving srtpath & ir2boot is the same and also eventually
you also want to be in the $BFROOT/detector/trg/emt directory.
From here the executable is:
serialtower babar-emc-emt-tcalXXXXXXXX
Since this application produces alot of output to the screen it would
probably be est if you pipe this to a file so that it can be looked at
later. One of the EMT group would popbably be best to have a look through
this file so it isn't necessary for you to. The hbook file is put in the
same hbooks directory as above
Open paw and execute towercheck.kumac on it. This also prints output
to a ps file and it wouls also be best if this was printed out and put
in the Trigger folder. An EMT person would probably be best to look at
this so if necessary phone one of them or wait until daytime!
