0
\$\begingroup\$

I'm experiencing a few problems with responsivity of APD (model: First Sensor AD500-8-S1).

enter image description here

This is the circuit I tried to make.

Just simply apply voltage (positive polarity) to the cathode for reverse bias voltage (used high voltage power supply ps325 of Standford Research System) and connect the anode to an oscilloscope to calculate the photocurrent by measuring the voltage.

First, I tried to measure the typical Si photodiode sensitivity without power supply.

I used a He-Ne laser(633nm) with 3.5µW and the Si sensitivity with 0V at 633nm is about 0.5A/W.

However, the voltage signal difference on the oscilloscope when applying the He-Ne laser was 50mV.

Since I used 1MΩ coupling of oscilloscope, so the responsivity I got is (50mV)/(1MΩ X 3.5µW) = 0.014 A/W, which is way smaller than the reference.

I have no idea why this problem happens, since the circuit is pretty simple.

enter image description here

Another problem is, according to the multiplication graph, that when the reverse bias exceeds 90V the multiplication increases extremely.

However, the voltage signal difference on oscilloscope by applying the He-Ne laser was 12V for reverse bias 91V, 15V for reverse bias 101V, 18V for reverse bias 111V and 23V for revere bias 121V.

It didn't increased dramatically and the responsivity at each reverse bias was also way smaller (about 10 times) than the reference (responsivity of 633nm at 90V (M=100) is about 35A/W according to the graph below).

enter image description here

Can somebody give me some advice on this?

\$\endgroup\$
2
  • \$\begingroup\$ Ask a focussed and targeted question; don't presume that what you feel is a problem is that obvious to anyone reading this question. This is a question and answer site and, questions are what you bring to the table. \$\endgroup\$ Commented Nov 23, 2021 at 11:00
  • \$\begingroup\$ Are you measuring the bias voltage across the APD or across the power supply terminals? With a 1 Meg resistor those will be very different values. \$\endgroup\$ Commented Nov 24, 2021 at 0:01

2 Answers 2

Reset to default
1
\$\begingroup\$

Since I used 1MΩ coupling of oscilloscope, so the responsivity I got is (50mV)/(1MΩ X 3.5µW) = 0.014 A/W, which is way smaller than the reference.

I have no idea why this problem happens, since the circuit is pretty simple.

I agree with the other answer. Most likely you are not getting all of the beam onto the active area of the detector.

  • Have you optimized the position of the detector in the beam?

  • Is the beam incident on the detector at a large angle to normal to the detector surface?

  • Has the beam has expanded when it reaches the detector to be larger than the detector active area?

  • Is something reflecting part of the beam away? Does your detector or one of the optics in your system have an AR coating designed for a wavelength other than 632 nm?

  • Is the 3.5 uW power of the laser a maximum power rating or an actual measured value? Are you operating the laser at full power?

Other possibilities

  • Is the power supply to the APD limiting the output current?
\$\endgroup\$
0
\$\begingroup\$

You need to check the coupling from the laser to the APD. I guess the laser you are using is a free space output with a small beam size. But as the APD has only 0.5mm diameter, you need careful alignment and maybe some lenses to get best coupling efficiency. Maybe a G-lens just near the glass window of APD and it helps a lot. The gain vs voltage and temperature changes with individual device. you need to read the test report of the APD you have, not the graph from the datasheet. The silicon APDs have very large optimized bias voltage, even from same supplier. I have used some si APDs with Vbr from ~100V to almost ~400V.

\$\endgroup\$
1
  • \$\begingroup\$ I checked the beam size of the laser with beam profiler and the FWHM was 40um, full width at 10% of maximum peak was 100um. So, it seems like this is not the reason. From the datasheet of this apd, this apd's breakdown voltage was 80~120V, that was the reason I didn't increase the voltage over 120V. \$\endgroup\$ Commented Nov 23, 2021 at 10:38

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.