[d2n-analysis-talk] T3 Trigger Efficiency Revisited

[David Flay]

Hi Brad,

I've revisited the T3 trigger efficiency in light of some comments Diana
had concerning my write up:

http://www.jlab.org/~flay/thesis/tech_notes/trig_note.pdf

and then I double-checked your email concerning how to calculate it (from
July 12, titled LHRS Trigger Efficiency Study) :

"So, this expression:

eff_3 = bit3/(bit3 + bit4)

 bitN = trigger N latch bit, after prescaling.  This is set
         if and only if:
         - a TN trigger is seen at the TS within 10ns of whatever
           generated the L1A, .AND.
         - it passes the prescale condition for trigger N"

My previous calculations (seen in the note currently) did this but using
the DL.bitN variable -- which looks at the number of TN (N = 3,4) triggers
that set the bit pattern, and is not the same as the latch pattern, which
is what you were referring to above (I believe).

I re-ran the code using the DL.LTN (latch pattern) variables, and I obtain
the graph attached, which shows the trigger efficiency as a function of
run number.  The overall average is ~98%.  Two runs are as low as 92%. 
Only 3 show exactly 100%.

Here are the averages binned by kinematic setting in the LHRS:

======================= Results =======================
T3 trigger efficiency average (for all runs): 98.1014
For each kinematic:
p = 1.2300, E = 1.2300: 99.9560
p = 0.6000, E = 4.7300: 98.0200
p = 0.6000, E = 5.8900: 98.4012
p = 0.8000, E = 4.7300: 97.8476
p = 0.9000, E = 5.8900: 98.0216
p = 1.1300, E = 5.8900: 98.2423
p = 1.2000, E = 5.8900: 98.2806
p = 1.2700, E = 5.8900: 97.5517
p = 1.4200, E = 4.7300: 98.6211
p = 1.4200, E = 5.8900: 97.5404
p = 1.5100, E = 4.7300: 97.6364
p = 1.5100, E = 5.8900: 98.6002
p = 1.6000, E = 4.7300: 97.8905
p = 1.6000, E = 5.8900: 98.4202
p = 1.7000, E = 5.8900: 98.4451

These numbers are quite different from the ~99.95% efficiencies I obtained
utilizing the DL.bitN variable.

It is nice that there are far fewer runs that show the 100% efficiency
value.  I'm just not sure if this is absolutely correct.

Should we also include the DL.bitN variable as well (since they are good
T3s too) -- that is,

N_T3 = DL.bit3 + DL.LT3?

What do you think?

Thanks,

Dave


-------------------------------------------------
David Flay
Physics Department
Temple University
Philadelphia, PA 19122

office: Barton Hall, BA319
phone: (215) 204-1331

e-mail: flay at jlab.org
            flay at temple.edu

website: http://www.jlab.org/~flay
              http://quarks.temple.edu
-------------------------------------------------
-------------- next part --------------
A non-text attachment was scrubbed...
Name: T3_trig_eff_8_16_10.png
Type: image/png
Size: 15514 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/d2n-analysis-talk/attachments/20100816/967e5754/attachment.png