Using an APD/Photon Counter

[joaocandre]

I'm trying to implement a biasing and readout circuit to interface with an APD from Hamamatsu (http://www.hamamatsu.com/resources/pdf/ssd/s12572-025_etc_kapd1043e.pdf). The problem is to how properly bias the APD (need around ~65V to get into geiger mode). The readout would be handled by a non-inverting charge/integrating amplifier, but the biasing is a bit trickier because I'm woring with an Arduino and have only a single 5V supply at my disposal.

I've been looking at the LT3482 IC, but I have some doubts about how exactly I should design the circuit: since I've already have a non-inverting charge/integrator amplifier in mind, I was wondering if it would be possible to ignore the current monitor funcionality of this IC - or do I need the current monitor powered to get a proper bias?

Also, I've noticed that there are three output pins: VOUT1, VOUT2 and APD. In the examples supplied in the datasheet, the APD is connected to the last, but in that case are the others for?

Any input is welcome! Thanks in advance!

 

[davenn]

I'm trying to implement a biasing and readout circuit to interface with an APD from Hamamatsu (http://www.hamamatsu.com/resources/pdf/ssd/s12572-025_etc_kapd1043e.pdf). The problem is to how properly bias the APD (need around ~65V to get into geiger mode).

I don't see a reference to that ??

 

[hevans1944]

You would be well-advised to purchase the Hamamatsu MPPC Module C11205-350 which contains the biasing and charge amplifier circuitry you need, operating from a ±5 V DC supply.

 

[joaocandre]

I don't see a reference to that ??

I need to bias the APD above the breakdown voltage, which sits around 65V +/- 10, as can be seen in the products datasheet, in the link I supplied.

 

[davenn]

I need to bias the APD above the breakdown voltage, which sits around 65V +/- 10, as can be seen in the products datasheet, in the link I supplied.

Again I don't see the reference to to the -65V

 

[hevans1944]

Again I don't see the reference to to the -65V

@davenn : an avalanche photo diode, or APD, is a quantum-mechanical photon detector typically constructed as a PIN junction device operated with reverse bias. At low reverse bias voltages it behaves as a simple photo diode with reverse current proportional to light intensity. At a critical value of reverse bias, the break-down voltage, the APD will enter an impact ionization or avalanche mode where a single photon event causes multiple electron-hole pairs to be created. In the absence of current limiting, the avalanche condition will lead to the destruction of the device. A series resistor, called a quench resistor, will cause the PIN reverse bias to rapidly drop below a level sufficient to sustain the avalanche condition.

More than you probably need to know about APDs and Geiger mode operation can be found in the PDF I have uploaded that describes a modern device.

This was all state-of-the-art stuff in the 1960s era. Biasing was extremely tricky and much care was required to approach the break-down voltage, or slightly exceed it for Geiger mode operation, while limiting the avalanche current to a safe amount and quenching after each photon event. The single PIN APD was most often used as a photon counter to replace photo-multiplier tubes (PMTs) driving pulse height discriminators configured as a multi-channel analyzer. This was pretty impressive stuff, counting single photons, but multichannel analyzers were expensive.

Today, we have arrays of APDs, each with its own quench resistor, all wired in parallel so any one pixel in the array can generate a photon event. Thanks to modern wafer processing techniques, the Hamamatsu arrays no longer suffer (as much) from after-pulse effects while the increased area over a single APD offers advantages in increased responsivity. I think they are still tricky to bias and instrument. The Linear Technology LT3482 series helps to solve the biasing problem, but biasing is still temperature sensitive and this must be corrected by either TE cooling the array or holding it at a constant temperature or by measuring the array temperature and applying a bias correction. This is a job well-suited for a modern micro-controller such as the Arduino.