Photomultiplier Single Electron (Photon) Response

Methodology

If a PMT is pulsed with light, it will produce a signal. By controlling the light level that reaches the PMT, the amplitude of the signal will vary proportionately. When the amount of photons reaching the PMT is very low, it becomes possible to distinguish the number of photons through the amplitude. If the number of photons is extremely low, it can be assumed that the number of photons that reach the PMT follow a poisson distribution.

If the light level is such that the PMT only generates a signal in 1 out of 10 original light pulses, or less, this means that the mean of the poisson distribution of the number of photons reaching the PMT is below 1. In this situation, the probability of having more than one photon reach the PMT becomes vanishingly small. As a result, there are effectively only two situations, either no photon reaches the PMT or only one photon reaches the PMT. Consequently, the response of the PMT to single photons can be observed by ignoring the events where there is no PMT pulse.

Experimental Setup

In order to reach the necessary conditions above, the PMT is placed in a PMT base, the magnetic shielding is attached to the PMT but the end of the PMT is left open. This assembly is then placed inside a dark box. At the other end of the dark box a controllable light source is placed (a light source that can be pulsed), for this purpose a LED is a good candidate. The voltage applied to the light source as well as the duration of the pulse are controlled in such a way that the PMT produces a pulse in less than 10 pulses supplied to the light source. The pulse supplied to the light source is used as a trigger. The output of the PMT is recorded.

Measurements 2020

Dark Box

For the dark box, the one built for the experiments in 2019 was used. A LEMO passthrough had to be added to the other end of the box in order to help avoid noise from the pulse supplied to the LED to propagate to the signal coming from the PMT. This noise is likely related to the lack of termination on the LED side as well as the exposed nature of the LED selection PCB board (see description below for full details).

When the PMT and LED are placed inside this box, the LED is approximately 33 cm from the front face of the PMT.

The box has approximate dimensions of 90cm in length, 25cm high and 13cm wide. The inside of the box is very reflective since it is made out of metal, so a dark cloth was used to line the box and eliminate reflections. The cloth is significantly larger than the box, so a lot of the cloth was outside the box, but this ended up being useful to wrap around the lid and help even more in maintaining the inside light-free.

  • Side view of the dark box:
    DarkBox.jpg

Light Source

For the light source, a triple light source was built in 2018 or 2019. This light source has 3 LEDs:

  • Red
  • Green
  • Orange

The LEDs are mounted on a first prototyping board which is connected to a second prototyping board through a ribbon cable. The second prototyping board has a LEMO connector, for placing the input signal, i.e. the pulse. The second prototyping board also has a jumper cable and an expansion header, in this way it can be chosen to which LED the pulse is supplied to. However, a consequence of this setup is that it is likely to produce a bit of unwanted interference.

Note that there is no 50 Ohm termination and the LEDs are directly connected to the pulse source, this leads to some level of reflection on this line, which also contributes to the noise mentioned above.

TODO: Add photo of light source

Pulse Source

Is a Tektronix AFG 3252 (2 GS/s; 240 MHz) Minimum pulse width is limited by frequency, we want very small width, which meant we had to go with higher frequency. This is a bit undesirable, not the best source for us. Next time use the other one TODO: Figure out the pulser that we normally use and update the info here

The pulse parameters that were used:

  • Frequency: 1 kHz
  • Pulse width: 10 ns
  • Leading: 2.5 ns
  • Trailing: 2.5 ns (Are these included in the pulse width or added?)
  • Low: 0 mV
  • High: depends on the LED, start at 600 mV

TODO: Add photo of pulser

TODO: Add picture of ideal pulse

TODO: Add picture of actual pulse

Measurement Device

LeCroy Waverunner 8104 (1 GHz; 20 GS/s)

LED pulse signal on Ch1, 200 mV/div with -600 mV offset (so that the base line is close to the bottom of screen and the pulse is well seen)

PMT output was connected to Ch2, 20 mV/div with 60 mV offset (so it is close to the top and we see the pulse well on the bottom)

Trigger on LED pulse signal. Positive edge, 300 mV threshold 20 ns/div, -100 ns offset so that the original pulse is to the left The gate for all functions should be: 3.7 div to 9 div Make a table have as first entry the minimum of the pulse The second entry the area of the pulse Make a histogram of the second entry, and only display when needed. The histogram with a max of 3000 events. Disable the "auto find", set bins to 100, center to -150, width to 50 The third entry of the table should be the histogram integral of the part that corresponds to the baseline, we can estimate the fraction of pulses this way. Set the gate to 7.2 to 10 At a later point, added a second histogram of the first table column. Changed the max of both histograms to 10000 events. Bins 100, center -40 mV, width 10 mV. Changed first histogram to center -300e-12 width 100e-12

TODO: Add photo of Oscilloscope

Also used a NIM crate with a discriminator and a scaler for measuring the noise of the PMTs. The descriminator had its threshold set to the lowest possible value, should be around 20 mV. The scaler was set to count in units of ms and the counter was set to 300000, corresponding to 300 s, which is 5 minutes.

TODO: Check minimum value

Setup

Note the boxes on top of the dark box, they contain screws bolts and other heavy knick-knacks to help keep the lid shut and avoid any light getting in

TODO: Add photo of full setup

Start by measuring the noise of the PMT with voltage applied, but no pulses to the LED. This should only require the NIM crate. Then, the PMT is connected to the Osciloscope and the voltage of the LED is adjusted until we see pulses 1 out of 10. Finally, several acquisitions can be made at different PMT voltages.

Can also attempt to see double photon peak by very slowly increasing the LED voltage

Results

Old Measurements (from 2018)

The methodology for this old data does not necessarily follow the same methodology as above since the methodology was not written down. However, the methodology is very similar and there should be no harm in assuming a similar setup.

Code

All the code performing the data analysis can be found on the daquser account. The code was written in jupyter under SWAN (swan.cern.ch) using cernbox as the storage. All files were placed under the swan directory: ~/labdata/2018/lab11/PMTs/, in particular note the jupyter notebook called ProduceHistos.ipynb (Also the file ParseRaw.ipynb or the python version of this same file ParseRaw.py). The cernbox space corresponds to an eos mounted space which can easily be accessed from lxplus. The cernbox/swan home directory corresponds to the path: /eos/user/d/daquser. Consequently, the directory above can be accessed through: /eos/user/d/daquser/labdata/2018/lab11/PMTs/ and the jupyter notebook at /eos/user/d/daquser/labdata/2018/lab11/PMTs/ProduceHistos.ipynb. However, it is best to access the jupyter notebooks through SWAN or CERNBOX in order to adequately see the contents.

Measurements

Measurements of the Single Electron Response (SER) of the photomultipliers. For a full list of the available photomultipliers, see PhotomultiplierInventory2018.

ID LED HV Timebase Scale Int Min Int MaxSorted ascending Result Comments
7980 1.420 2250 10 ns/div 20 mV/div 1 4 PM7980_2250.png Best separaton so far in our photomultipliers
7980 1.420 2200 10 ns/div 20 mV/div 1.5 5 PM7980_2200.png Best separaton so far in our photomultipliers
6810 1.390 2350 10 ns/div 20 mV/div 1.5 5 PM6810_2350.png Not the best separation and operating close to maximum voltage
17248 1.420 2200 10 ns/div 20 mV/div 2 5.5 PM17248_2200.png OK
27821 1.460 2275 10 ns/div 20 mV/div 1 5.5 PM27821_2275.png  
27821 1.460 2300 10 ns/div 20 mV/div 1 5.5 PM27821_2300.png  
15697 1.500 2250 10 ns/div 20 mV/div 2.5 6 PM15697_2250.png  
Sune 1.470 2150 10 ns/div 20 mV/div 2 7 PMSune_2150.png  
Sune 1.470 2198 10 ns/div 20 mV/div 2 7 PMSune_2198.png  
Sune 1.470 2250 10 ns/div 20 mV/div 2 7 PMSune_2250.png  
8919 1.445 2490 10 ns/div 20 mV/div 3.5 7 PM8919_2490.png No separation
8141 1.430 2350 10 ns/div 20 mV/div 2 7 PM8141_2350.png Comment
17208 1.530 2375 10 ns/div 20 mV/div 2 7 PM17208_2375.png Separation practically missing
9050 1.430 2475 10 ns/div 20 mV/div 2 7 PM9050_2475.png Separation practically missing
7469 1.430 2490 10 ns/div 20 mV/div 2 7 PM7469_2490.png Separation practically missing
7405 1.470 2400 10 ns/div 20 mV/div 3 7 PM7405_2400.png
  • Separation practically missing, maybe to be redone at higher voltage
  • With repeat measurement, it seems to discharge

8657 1.480 2490 10 ns/div 20 mV/div 2.5 7 PM8657_2490.png  
7337 1.420 2300 10 ns/div 20 mV/div 5 7.5 PM7337_2300.png Similar to PM 7497
7481 1.435 2496 10 ns/div 20 mV/div 3.5 8 PM7481_2496.png No separation
7497 1.420 2300 10 ns/div 20 mV/div 4 8 PM7497_2300.png Weird results:
  • We do see a peak moving with the HV, but there is some kind of underlying structure. What is it?
  • The larger pulses also seem to always come accompanied by a second smaller pulse.
15598 1.480 2300 10 ns/div 20 mV/div 4 8 PM15598_2300.png Comment
11518 1.680 2300 10 ns/div 20 mV/div 3.5 8 PM11518_2300.png  
11261 1.600 2300 10 ns/div 20 mV/div 3.5 8 PM11261_2300.png  
8885 1.420 2200 10 ns/div 20 mV/div 4.5 9 PM8885_2200.png Peak was broad and not well defined
6094 1.380 1625 10 ns/div 20 mV/div     PM6094_1625.png  
6094 1.380 1650 10 ns/div 20 mV/div     PM6094_1650.png  

-- CristovaoDaCruzESilva - 2018-07-31

Topic attachments
I Attachment History Action Size Date Who Comment
JPEGjpg DarkBox.jpg r1 manage 164.5 K 2020-06-17 - 03:46 CristovaoDaCruzESilva Side view of the dark box
PNGpng PM11261_2300.png r1 manage 10.2 K 2018-06-21 - 17:59 GianfrancoMorello Single electron response for PM 11261 @ 2300V
PNGpng PM11518_2300.png r1 manage 10.3 K 2018-06-21 - 17:58 GianfrancoMorello Single electron response for PM 11518 @ 2300V
PNGpng PM15598_2300.png r1 manage 10.8 K 2018-06-13 - 11:11 CristovaoDaCruzESilva Single electron response for PM 15598 @ 2300V
PNGpng PM15697_2250.png r1 manage 10.6 K 2018-06-21 - 18:37 GianfrancoMorello Single electron response for PM 15697 @ 2250V
PNGpng PM17208_2375.png r1 manage 10.5 K 2018-06-15 - 08:52 GianfrancoMorello Single electron response for PM 17208 @ 2375V
PNGpng PM17248_2200.png r1 manage 10.1 K 2018-06-13 - 11:07 CristovaoDaCruzESilva Single electron response for PM 17248 @ 2200V
PNGpng PM27821_2275.png r1 manage 10.4 K 2018-06-21 - 18:37 GianfrancoMorello Single electron response for PM 27821 @ 2275V
PNGpng PM27821_2300.png r1 manage 10.4 K 2018-06-21 - 18:38 GianfrancoMorello Single electron response for PM 27821 @ 2300V
PNGpng PM6094_1625.png r1 manage 10.1 K 2018-07-31 - 11:26 CristovaoDaCruzESilva Single electron response for PM 6094 @ 1625V
PNGpng PM6094_1650.png r1 manage 10.2 K 2018-07-31 - 11:27 CristovaoDaCruzESilva Single electron response for PM 6094 @ 1650V
PNGpng PM6810_2350.png r1 manage 10.2 K 2018-06-13 - 10:23 CristovaoDaCruzESilva Single electron response for PM 6810 @ 2350V
PNGpng PM7337_2300.png r1 manage 10.7 K 2018-06-13 - 10:24 CristovaoDaCruzESilva Single electron response for PM 7337 @ 2300V
PNGpng PM7405_2400.png r1 manage 10.4 K 2018-06-18 - 11:13 GianfrancoMorello Single electron response for PM7405 @ 2400V
PNGpng PM7469_2490.png r1 manage 10.2 K 2018-06-15 - 11:51 GianfrancoMorello Single electron response for PM 7469 @ 2490V
PNGpng PM7481_2496.png r1 manage 10.7 K 2018-06-12 - 17:36 CristovaoDaCruzESilva Single electron response for PM 7481 @ 2496V
PNGpng PM7497_2300.png r1 manage 10.9 K 2018-06-12 - 17:39 CristovaoDaCruzESilva Single electron response for PM 7497 @ 2300V
PNGpng PM7980_2200.png r1 manage 10.2 K 2018-06-13 - 10:21 CristovaoDaCruzESilva Single electron response for PM 7980 @ 2200V
PNGpng PM7980_2250.png r1 manage 10.5 K 2018-06-13 - 10:21 CristovaoDaCruzESilva Single electron response for PM 7980 @ 2250V
PNGpng PM8141_2350.png r1 manage 10.4 K 2018-06-14 - 15:09 GianfrancoMorello Single electron response for PM 8141 @ 2350V
PNGpng PM8657_2490.png r1 manage 10.4 K 2018-06-21 - 17:57 GianfrancoMorello Single electron response for PM 8657 @ 2490V
PNGpng PM8885_2200.png r1 manage 10.5 K 2018-06-13 - 11:09 CristovaoDaCruzESilva Single electron response for PM 8885 @ 2200V
PNGpng PM8919_2490.png r1 manage 10.7 K 2018-06-12 - 17:23 CristovaoDaCruzESilva Single electron response for PM 8919 @ 2490V
PNGpng PM9050_2475.png r1 manage 10.4 K 2018-06-15 - 10:29 GianfrancoMorello Single electron response for PM 9050 @ 2475V
PNGpng PMSune_2150.png r1 manage 9.9 K 2018-06-12 - 17:12 CristovaoDaCruzESilva Single electron response for PM Sune @ 2150V
PNGpng PMSune_2198.png r1 manage 9.7 K 2018-06-12 - 17:18 CristovaoDaCruzESilva Single electron response for PM Sune @ 2198V
PNGpng PMSune_2250.png r1 manage 9.8 K 2018-06-12 - 17:18 CristovaoDaCruzESilva Single electron response for PM Sune @ 2250V
Edit | Attach | Watch | Print version | History: r21 < r20 < r19 < r18 < r17 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r21 - 2020-06-25 - CristovaoDaCruzESilva
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    BL4S All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback