[Clas_cascades] Fwd: Re: analysisnote_Lambda1520_v3

Thu Jan 27 13:27:17 EST 2011

Replies:

1) I don't understand what you mean about the jump appearing for eg3.
2) It's phi-dependent and there's two PMTs.  For example, you could  
have a track go through the scintillator right near a good PMT and get  
your trigger, but a low-gain PMT on the far edge will have a low ADC.   
The trigger bit doesn't distinguish which side of the paddle it came  
from.  Also, the color scales are drastically different between  
triggered/un-triggered.
3) ADC = 600 is just a safe target for the high-voltage calibration,  
it's not the threshold for triggering and it's not used in the  
calculation of the efficiency.  It's just a guide to show that if the  
ADC is significantly lower, you may have a problem triggering.
4) It depends on the trigger bits for eg3.  For g13, we had a two- 
sector coincidence (ST X TOF, no pre-scale, no tagger), and trigger  
bits 1-6 were set if trigger signals arrived from sectors 1-6.  I  
dunno about eg3, but the time windows were large for g13 (120 ns for  
SC, 28 ns for ST), so they did not influence it anyway (except for  
maybe slow spectator protons, but they never reached the TOF anyway).   
Keep in mind that if your Lambda decays outside of the start counter,  
it will not have a trigger signal).

  - Paul

On Jan 27, 2011, at 12:57 PM, Haiyun Lu wrote:

I have quite a few questions on the plots.
1. In the plots of triggered ADCs, the lower edges of red ( high  
occupancy) bars are threshold of each paddle. It's exactly like the  
case EG3 triggers on the photon energy so that there is a jump  
appearing. If this is true, the inefficiency is not so terrible  
( maybe except the large-angle paddles)
2. The ADCs of no trigger are greater than ADCs with trigger. Why? For  
example, channel 23, 24 in sector 1; 25, 27 in sector 2...
3. In the calculation of efficiency, the threshold is assumed at 600.  
However, look at the plots, large samples of events triggered are  
below 600. It's also not consistent with efficiency shown. Any typo?
4. I wonder How to determine if it's triggered or not. Is it possible  
that it requires two-track trigger and signals from other two sectors  
already satisfy the trigger system and this one missed it due to timing?

Haiyun Lu
WEH 8402
Department of Physics
Carnegie Mellon University
5000 Forbes Avenue
Pittsburgh, PA  15213
412-268-1714

On 01/27/2011 11:11 AM, Zhiwen Zhao wrote:
>
> -------- Original Message --------
> Subject: Re: [Clas_cascades] analysisnote_Lambda1520_v3
> Date: Wed, 26 Jan 2011 15:16:39 -0500
> From: Paul Mattione<pmatt at jlab.org>
> To: Zhiwen Zhao<zwzhao at jlab.org>
>
> No, I mean triggering efficiency.  The thresholds for SC detection are
> 20mV, but for triggering are 100mV.  Therefore you can have paddles
> that reconstruct pi-'s just fine but that won't send trigger signals.
> A really quick way to check is to look at the ADCs of reconstructed
> pi-'s and see if they look as terrible as mine do (Trigger
> efficiencies were set assuming the pedestal subtracted ADC is at 600):
>
> http://clasweb.jlab.org/rungroups/g13/wiki_secure/index.php/PiMinusTriggerEfficiencyStudy#ADC_Study
>
> eg3 may have been fine, I'm just wondering if you've looked into it.
> The effect this has depends on your event topology: if you have
> several pi-'s in your final state and some of them are hitting these
> low-gain paddles then you may not get triggers.  It may have a larger
> effect for eg3 than other experiments since you guys had the three
> sector trigger.
>
>   - Paul
>
> On Jan 26, 2011, at 3:09 PM, Zhiwen Zhao wrote:
>
> hi,Paul
>
> When you are talking SC trigger efficiency, do you actually mean SC
> efficiency?
> I didn't really do SC eff study, but I did cut away some paddle with
> low occupancy away as showed in the note.
> Once I have looked g10's SC eff study, but it's takes a very different
> approach from what you did and is more similar to a DC hit eff.
> For your approach, I will need to study all kaon/pion/proton SC eff.
> That's gonna be a lot of work.
> With the large trigger eff correction of eg3, I hope SC eff will be
> buried in the other big error bars.
>
> Did you put your result into gpp? How about your pion and proton SC  
> eff?
>
> Zhiwen
>
>
> On 01/26/2011 02:51 PM, Paul Mattione wrote:
>> Hey Zhiwen, did you try to do a trigger efficiency study on the SC
>> paddles? For g13, I found a lot of SC paddles that had really low
>> triggering efficiency because they had low gain (old PMTs at max
>> voltage
>> but weren't replaced yet). For me, this mostly had an effect on
>> pions at
>> back angles.
>>
>> Old but "close enough" study plots (trigger efficiency of pi-'s with
>> momentum>  400 MeV/c):
>> http://clasweb.jlab.org/rungroups/g13/wiki_secure/index.php/PiMinusTriggerEfficiencyStudy#P_.3E_400_MeV
>>
>>
>> - Paul
>>
>> On Jan 20, 2011, at 2:35 PM, Zhiwen Zhao wrote:
>>
>> Hi, folks
>>
>> I put up things related to the note at the page
>> http://clasweb.jlab.org/rungroups/eg3/wiki/index.php/Analysisnote_Lambda1520
>>
>> which is linked from the main wiki page
>>
>> I has got comment from Kijun and Elton. More comments are welcome.
>> Thanks a lot
>>
>> Zhiwen
>>
>>
>> On 12/14/2010 02:20 PM, Zhiwen Zhao wrote:
>>> Hi, folks
>>>
>>> I posted the Lambda1520 analysis note v3 on wiki and you can also
>>> get it
>>> directly from
>>> http://www.jlab.org/Hall-B/secure/eg3/zwzhao
>>> /analysisnote_Lambda1520_v3.pdf
>>> This is for reviewing within eg3
>>> Please send me your comment and I really appreciate it.
>>>
>>> Zhiwen
>>> _______________________________________________
>>> Clas_cascades mailing list
>>> Clas_cascades at jlab.org
>>> https://mailman.jlab.org/mailman/listinfo/clas_cascades
>> _______________________________________________
>> Clas_cascades mailing list
>> Clas_cascades at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/clas_cascades
> _______________________________________________
> Clas_cascades mailing list
> Clas_cascades at jlab.org
> https://mailman.jlab.org/mailman/listinfo/clas_cascades