[Halld-cal] Fwd: Spectra from SiPM Gate Scan

Yi Qiang yqiang at jlab.org
Tue Mar 20 17:40:57 EDT 2012

-------- Original Message --------
Subject: 	Spectra from SiPM Gate Scan
Date: 	Tue, 20 Mar 2012 13:42:13 -0400
From: 	Yi Qiang <yqiang at jlab.org>
To: 	William Brooks <william.brooks at usm.cl>, Elton Smith <elton at jlab.org>

Hi Will,

All our SiPM ADC spectra can be found in the online directory:
and the following sub-directories hold the gate scan histograms for SiPM #16:

Sub_Directory			Filter	Temp	Gate	V_25	V_Set	V_off	Current	Pos	HV
20120316_163956/SiPM_0016	36	17.605	150	71.537	71.123	0.000	1.050	1	1
20120316_164426/SiPM_0016	36	17.645	200	71.537	71.122	0.000	1.045	1	1
20120316_164730/SiPM_0016	36	17.615	250	71.537	71.122	0.000	1.044	1	1
20120316_165027/SiPM_0016	36	17.595	300	71.537	71.122	0.000	1.045	1	1
20120316_165336/SiPM_0016	36	17.64	350	71.537	71.127	0.000	1.056	1	1
20120316_165637/SiPM_0016	36	17.59	400	71.537	71.124	0.000	1.051	1	1
20120316_165952/SiPM_0016	36	17.655	500	71.537	71.123	0.000	1.047	1	1
20120316_170336/SiPM_0016	36	17.615	600	71.537	71.126	0.000	1.054	1	1
20120316_170650/SiPM_0016	36	17.605	800	71.537	71.123	0.000	1.047	1	1
20120316_171019/SiPM_0016	36	17.665	1000	71.537	71.128	0.000	1.061	1	1

In each folder there is a png file called Spectra.png and it will show your the spectra of individual cells with that gate setting.
gives you the spectra with 1000 ns gate.

Here is the link to the wiki page where I posted my early study of after pulses using three different methods:
The second method was using the average waveforms from oscilloscope with different cuts.
In this test, I used a very fast pulse generator, ~ 1ns, to drive a red laser diode to produce a narrow light pulse,
and I also used a filter to reduce the average number of pixels fired to close to 1 (1.185 as you see in the fit).
In order to reduce the dark noise, I cooled the device down to -7 degC.
I first tried to extract a single pixel waveform which is purely triggered by the light source, I did that by cutting on the single photo electron peak in a narrow window coincident with the light input and on the pedestal of the ADC spectrum outside the window.
The resulting pulse shape is an average of all waveforms passing this cut.
Then, I compare it to the average waveform without any cut. Here since the dark pulses are completely random to the gate, the average of them should just be a flat background.
I normalize the single pixel waveform to the average waveform by comparing the leading edge and the difference between these two waveforms gives me the residue mainly caused by after pulses.
Since the average number of pixels fired is very close to 1, even if there is any time distribution in the input light source, the subtraction should greatly cancel its effect.
What I got is about 12% for after pulses by fitting the residue's tale using an exponential shape, which is principle consistent with other methods dealing with the tale only.
What I also saw is that there is about 10% fast component left in the residue as well, which may be attributed to the cross-talk.

Since we have a very fast laser in JLab now, we can carry out this test using that setup again.


*Yi Qiang*   Ph.D. /Staff Scientist/
*Email*: yqiang at jlab.org <mailto:yqiang at jlab.org>
*Tel*: (757) 269-7237 *Fax*: (757) 269-6331
*Web*: https://userweb.jlab.org/~yqiang 
12000 Jefferson Ave F351, Newport News, VA 23606
Jefferson Lab <http://www.jlab.org>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/halld-cal/attachments/20120320/2f22ec29/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 4844 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/halld-cal/attachments/20120320/2f22ec29/attachment.jpe 

More information about the Halld-cal mailing list