F2/EMC analysis how-to for shift takers
Instructions for online analysis of F2/EMC Experiments (E12-10-002/E12-10-008)
1. We would like to keep the deadtime below 20 % so make sure to prescale accordingly before you start a run. That basically means to keep the DAQ target rate below 2000 Hz (look at the prescale gui that's in the DAQ VNC session.)
2. Once you accumulated 50 K CODA events (50000 CODA events):
for SHMS detector checks: ./run_shms.sh <run_number>
for HMS detector checks: ./run_hms.sh <run_number>
--> This will bring up the online GUI for you to do detector sanity checks. Look carefully at all histograms and compare to the golden run which you can find in the binder.
--> It will also bring up an emacs window where you can check: current, trigger type, trigger and tracking efficiencies
As an example, this is what you would find in this emacs window:
Run #: 2081
BCM1 Current: 34.886 uA
BCM2 Current: 35.742 uA
BCM4a Current: 34.168 uA
Pre-Scale SHMS 3/4 : -1
Pre-Scale SHMS EL-Real : -1 --> this should be different than -1 (all the rest -1) when we take neg polarity runs
Pre-Scale SHMS EL-clean : 5 --> this should be different than -1 (all the rest -1) when we take pos polarity runs
Pre-Scaled Ps1 SHMS Computer Live Time : -0.6279 %
Pre-Scaled Ps2 SHMS Computer Live Time : -3.2613 % --> this should be larger than 80% if we take neg polarity runs
Pre-Scaled Ps3 SHMS Computer Live Time : 98.5311 % --> this should be larger than 80% if we take pos polarity runs
SING FID TRACK EFFIC : 0.8380 +- 0.0097
E SING FID TRACK EFFIC : 0.8417 +- 0.0100 --> this should be larger than 90% typically
Plane 1: 0.997916
Plane 2: 0.996795
Plane 3: 0.991199
Plane 4: 0.976989
3_of_4 EFF : --> this should be larger than 99.5% typically
--> Once you close the emacs window another script will run without you doing anything. Watch what's printed on the screen at the end: this should give you an idea if the PID legs pedestals have drifted compared to the golden run. If they did, ** shout **. This would need to be fixed ASAP as it affects the efficiency of our PID-dependent trigger (remember: for production we run with ELREAL or ELCLEAN).
As an example, this is how it would look:
!!!!!!!!!!!!!!!!!Attention Needed IT AFFECTS TRIGGER EFFICIENCY!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
chan 2 bar 3 pNGCER Mean golden run: 134.5001 Mean current run: 116.7607 Mean diff: 17.7394
- =:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=
3. Once you got 100 k CODA events (100000 CODA events) please do the following:
./run_shms_f2_stat.sh <run_num> <statistics goal for SHMS electrons **in K**>
./run_hms_f2_stat.sh <run_num> <statistics goal for HMS electrons **in K**>
This will tell you how many CODA events you need to take in SHMS and HMS in order to reach our statistics goal for electrons per kinematic setting. The statistics goal per kinematics (in K) is read from the tables you have in front of you (also posted at: ); please, please pay attention when you type.
As an example: ./run_shms_f2_stat.sh 3333 200
(here the run number is 3333 and the statistics goal is 200000 electrons or 200 K electrons, the shells take as argument the number of electrons in K)
Continue with the current SHMS and HMS runs until you accumulated the number of CODA events you got by running the shells above. We would like to keep the runs under one hour so stop/start a run if you need more than 1 hour to reach the statistics goal for electrons.
4. Once you ended a run do the replay of the whole run so you can get the full electron statistics in that run by doing the following:
./run_shms_f2_stat_all.sh <run_num> <statistics goal for SHMS electrons **in K**>
./run_hms_f2_stat_all.sh <run_num> <statistics goal for HMS electrons **in K**>
This will tell you what percentage of the whole statistics you got in that particular run for that particular kinematic setting. Write that number of electrons (in units of K and in percentage) in the printed table you have right in front of you.
We are done with this run!
