[Halld-tagger] microscope beam test

[Alexander Somov]

Richard,

I think that the only useful information we have obtained 
from the beam test was the amount of light we can get from
the scint. fibers - the plots you have showed during the
collaboration meeting.

Measuring any timing resolution of photodetectors
(photonics sipms), that cannot be operated at high rate,
and, that, consequently, will not be used, is of some 
limited interest to the microscope detector.

Obtaining timing information from the fadc spectrum using
a linear filter is a methodologically interesting topic, but
I doubt that the method  could surpass the tdc resolution
(though it can probably provide complementary timing
  measurements)

One of the problems we had with the microscope detector was 
that it had not been calibrated; I was not able to convert the adc
amplitude to the number of sipm pixels, specifically for the
summed channels.

Anyway, I think that we have learned many things from the beam test.


Cheers,
        Alex



On Wed, 2 Nov 2011, Richard Jones wrote:

> Alex,
>
> Yes, we are in essential agreement about the time resolution, I think.  If 
> you look at your 2D time walk plots, you can see that the regions where 
> you get less than 300ps resolution are out in the tails of the amplitude 
> distribution where there are hardly any events.  We should not consider 
> those regions very relevant for measuring performance properties.
>
> https://halldweb1.jlab.org/wiki/index.php/BLTWG_Meeting_10/31/2011
>
> The only one of the 3 summed channels that has a clear enough amplitude 
> peak to see it clearly separated from the threshold cutoff is the "sum 2" 
> (your first one).  In that one, the peak amplitude is around 900 adc 
> channels.  There you are getting about the same time resolution as I am, 
> around 350ps.  In the other two sum channels you show, unfortunately the 
> gains seem to be somewhat lower, and the threshold cutoff actually cuts 
> into the peak distribution, and trying to pick off a time resolution at 
> the intensity peak is impossible.  Going way out into the tail where there 
> are a few dozen events out of 50,000 (like looking at sum channel 3, adc 
> 1300 where you get less than 300ps rms) doesn't really work, I would say.
>
> Thanks for showing this.  I do not doubt any of the results that you show. 
> I just want to make sure that any claims of resolution that we make based 
> on these data are solidly supported by the bulk of the data, and not 
> cherry picking results in a corner of parameter space that look 
> artificially good.  I think we agree about this.
>
> -rtj
>
>
>
>
> On 11/2/2011 10:57 AM, Alexander Somov wrote:
>> Hello Richard,
>> 
>> Just following up our discussion during the beamline meeting
>> regarding the microscope beam test, I have linked a few time
>> resolution plots to the appendix of our last meeting
>> https://halldweb1.jlab.org/wiki/index.php/BLTWG_Meeting_10/31/2011
>> (I made these plots quite time ago)
>> 
>> The time resolutions  were computed with respect to the RF signal;
>> I also applied time walk corrections.
>> 
>> In general, the resolutions are consistent with that you have
>> presented.
>> 
>> Cheers,
>>          Alex
>
>
>