Elton and all,
I have created a page on the wiki (under main page, link "Silicon
Photomultiplier Performance") to record this conversation, because I
think things are starting to become clearer. Please look at my new
section added at the end, in particular my final paragraph quoted below.
If one wants to compare the Sensl and
Photonique devices, it is only
fair to compare them under the same operating conditions: 1.5V over
threshold. From Carl's measurements, the actual efficiency of the Sensl
C20 device is about 5% [he measured 4.4% at 1.2V overbias, so I round
up to 5% at 1.5V] compared with 24% for the 2 mm Photonique device
under the same conditions. The factor of 5 in PDE between the
Sensl and Photonique devices can only be explained by the supposition
that Sensl never intended their C20 device to be considered for
scintillator readout. Nor should 12% @ 500 nm be considered
acceptable,
given that commercial devices from Photonique exist with a factor 2
larger PDE. We have heard that CPTA technology is primitive, and that
may be so, but then it should not be difficult to reproduce their
performance in these new devices from Sensl.
Richard Jones
Elton Smith wrote:
Understanding PDE information
-----------------------------
Elke, Carl and I went over the information we have on the SiPM photon detection efficiency. We believe now that we understand the various pieces of information and they are consistent. Let me start with the first two rows of Zisis talk at the collaboration meeting (p 5, 1x1 mm2 SiPM parameters)
Design # Pixels DR @ 4V Fill Factor Dark Current Peak PDE
(MHz) (%) (muA) @4V (%)
C20 620 2.9 17 -2 8.5
A20L 920 5.9 34 -3 17.0
>From Carl's talk (p 37, QE and PDE), you can determine the ratio of the PDE@4V to PDE@1.2V, which was the chosen operating point for Carl's measurements. The ratio is approximately 2. Therefore, the expected PDE@1.2V for the C20 is about 8.5/2=4.2. Carl measured it to be 4.4+/-0.5.
-> Conclusion: At the operating point of about 1.2 V above breakdown, the PDE for the C20 (1mm2, 17% fill factor) SiPM is about 4.5%.
So far, our experience tells us that the practical operating point for these devices is about 1V over voltage. As the voltage above the breakdown increases further signal to noise decreases and even pulse shape become distorted.
If we use the A20L instead of the C20, the PDE will double, but so will the dark rate. We believe that the estimated 110MHz rate for the 4x4 array is for the 1V over voltage, thereby consistent with a PDE of 4.5%.
Comments are welcome. In particular, it would be useful for Richard to remind us the conditions that were used for his measurements of the PDE for the 1x1 and 2x2 mm2 SiPMs.
Cheers, Elton.
Elton Smith
Jefferson Lab MS 12H5
12000 Jefferson Ave
Suite # 16
Newport News, VA 23606
elton@jlab.org
(757) 269-7625
|