Counting photons with an MPPC

[Jan Panteltje]

On a sunny day (Tue, 27 Aug 2019 06:23:42 -0700 (PDT)) it happened George
Herold <gherold@teachspin.com> wrote in
<aac65532-5337-4f6c-9cfa-76d7dde37084@googlegroups.com>:

On Tuesday, August 27, 2019 at 9:12:49 AM UTC-4, pcdh...@gmail.com wrote:
For some of the SPAD's one photon causes the whole capacitance
to discharge (by the over-voltage)... another photon before it
recharges is likely missed

That's not a huge issue with these ones, at least not in analogue mode. Their
pixels are only 50 um square, so they have lots (3400-odd iirc) and they
recharge again in about 20 ns. Each one dumps about 2 million electrons,
so the saturation current is about 55 mA. At 53 V bias, the poor thing would
be dissipating almost three watts, which would turn it to lava.

And ruin your day. :^) I assume you have some current limit in the supply line
that
will stop that from happening... when some tech open's the unit up to room
light.

Years agao I looked for SPADs and got 'we do not sell this to your country' crap.

After reading some of this here, I decided to look on ebey if they EXIST[1] now.

No S.P.A.D
No single photon avalanche detector,
but looked at some other entries,
this caucht my attention, surprized:
https://www.ebay.com/itm/262592051982
VL53L0X World smallest Time-of-Flight (ToF) ranging sensor NEW
I though maybe of use for my drone's anti collision system, so started reading the desciption:
....
The VL53L0X integrates a leading-edge SPAD array (Single Photon Avalanche Diodes) and embeds ST\u2019s second generation FlightSenseTM patented technology.
The VL53L0X's 940nm VCSEL emitter (Vertical Cavity Surface-Emitting Laser), is totally invisible to the human eye, coupled with internal physical infrared filters, it enables longer ranging distance, higher immunity to ambient light and better robustness to cover-glass optical cross-talk.
Key Features
Fully integrated miniature module
940nm Laser VCSEL
VCSEL driver
.....

LOL, and there is the SPAD
Any comments?
I have ordered one just now 8$26 with free shipping.
??
There is even an arduino library
https://github.com/ScruffR/VL53L0X


[1] Not on ebay means does not exist.

 

[Phil Hobbs]

On 8/26/19 6:57 PM, Dimiter_Popoff wrote:

On 8/22/2019 18:06, Phil Hobbs wrote:
On 8/22/19 10:08 AM, jjhudak4@gmail.com wrote:
On Wednesday, August 21, 2019 at 6:19:01 PM UTC-4, Phil Hobbs wrote:
Well, the MPPC demo system is done.  It has four of our small
proprietary boards (controller/SMPS, TEC driver, voltage-controlled
amplifier, and APD bias) plus a handwired pHEMT-boostrapped front end
and a box of voltage regulators.  Works from single photons up to about
4 mA in one range.

This video shows it counting photons, which I still think is a cool
thing to be able to do.

https://electrooptical.net/www/sed/MPPCphotonCounting.mts

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

Please excuse this naive question as this is not my background, but....
What kind of detector is being used?

It's a Hamamatsu S13362-3050.  The product will use a tiled array.

how do you know if you missed a photon? have two or more coincident
ones? or miscounted? e.g. false positives and false negatives.

You miss most of the photons.  The probability of detection is only
about 40% at most, and only 74% of the area is active.  It's roughly
competitive with a PMT.

If my recollection from college physics is correct, a photon has a
certain energy level.  Do all photons have the same energy level or
do they vary?

They vary.  Energy is hc/lambda.

I would imagine if they can contain different energy levels then
discriminating >between n>1 hits would be very tricky.

The pulse height doesn't depend on the photon energy.


That is surprising, semiconductor gamma-ray detectors (e.g. HPGe) have
the best energy resolution to date. Obviously this being a small device
its energy range would not go high enough so there is likely no
practical application for that.

The physics is completely different. Ge(Li) and IG detectors get their
high resolution from the primary particle producing N carriers, where N
is proportional to its energy.

MPPCs are avalanche devices, where a detection event causes the whole
pixel to discharge until the avalanche stops.

On a side note, what throughput did you manage? (counts/second after
which the counts which make it through start to decline).
I recently did photon counting using a good old PMT (well not old
really, a fairly new Hamamatsu model), easily got > 20M cps which was
plenty for the application (a TLD reader, yet to be announced, don't
look for it yet on the website, perhaps next week .

Because of the highish capacitance, the output pulses aren't as
symmetrical as a PMT's--the fall time is about 20x longer than the rise.
So to get PMTish photon counting rates you need a more intelligent
discriminator circuit that can reliably see events that arrive on the
tails of other events.

Cheers

Phil Hobbs

(coming to you from beautiful Sydney BC waterfront)

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Phil Hobbs]

On 8/27/19 9:23 AM, George Herold wrote:

On Tuesday, August 27, 2019 at 9:12:49 AM UTC-4, pcdh...@gmail.com
wrote:
For some of the SPAD's one photon causes the whole capacitance
to discharge (by the over-voltage)... another photon before it
recharges is likely missed

That's not a huge issue with these ones, at least not in analogue
mode. Their pixels are only 50 um square, so they have lots
(3400-odd iirc) and they recharge again in about 20 ns. Each one
dumps about 2 million electrons, so the saturation current is
about 55 mA. At 53 V bias, the poor thing would be dissipating
almost three watts, which would turn it to lava.


And ruin your day. :^) I assume you have some current limit in the
supply line that will stop that from happening... when some tech
opens the unit up to room light.

Yup, a couple of parallelled LND150s inside the voltage feedback loop.

There's also an I2C pressure sensor to detect the chamber venting, so
that we can turn off the bias and warm the device back up to 20C before
we get ice all over it.

Cheers

Phil Hobbs







--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com