Calibration of the Linear Regime of Photodiode Detector


Huijie Miao

SUNY at Stony Brook

Optics Rotation Project 2

Advisor: Thomas Weinacht



¡¤ Introduction and Motivation

¡¤ Photodiode Detector

¡¤ Experimental Setup and Results

¡¤ Acknowledgements



Introduction and Motivation


The following diagram(fig.1) shows the construction of the photodiode which we used for the detector.

fig. 1

The silicon photodiode is constructed from single crystal silicon wafers. It requires high purity silicon. In fact, it is just a PN junction device. The ¡°p¡± layer is very thin, which is formed by thermal diffusion or ion implantation of the appropriate doping material (usually boron). The front contact is the anode and the back contact is the cathode. The active area is covered by some antireflection coating which is optimized for particular irradiation wavelengths.


According to the characteristics of semi-conductor, we know, at normal temperature,  the thermal energy produces a ¡°depletion region¡± around the PN junction. The width of the depletion region can be changed by applying a voltage across the photodiode. If a positive voltage is in contact with the N type end of the diode while a negative voltage is in contact with the P type end of the diode, which means we reverse bias the diode, the depletion region will be enlarged. The majority carriers in P region (holes) are attracted by the negative voltage, which draws them away from the depletion region. And the majority carriers in N region (electrons) are drew away from the depletion region by the positive voltage. The attractive forces result an enlargement of the depletion region, consequently, the energy gap between the two regions.


The reverse-biasing results in a large sensitivity for detecting radiation. And the output voltage of the photodiode is extremely linear with respect to the power applied to the photodiode junction. However, a too high input power may cause the photodiode saturated. It means when the input is higher the some certain value, the output voltage doesn¡¯t change obviously with it.


Since the price of a photodiode detector(phototransistor) is more then $100, and the price of a photodiode is just around $20, we want to make the detector with the photodiode ourselves. That¡¯s our main motivation for the project. Further, we calibrated the linear regime of the detector. This project is pretty practicable and attractive.



Photodiode Detector


Fig.2 shows the circuit which we followed to make the detector.

fig. 2

The photodiode we used is FDS010-Photodiode, SI. The rising time is £1ns (Measured with 50W load and 12V bias), the active area is 0.8mm2 (Æ1.0mm) and the spectral range is 200~1100nm.





Here are some values we picked up for the circuit:

    Capacitance:  0.1µF

Resistance:  1kW

         VBIAS =  9V (reverse)



Fig.3 shows the detector which we made in our laboratory.


fig. 3




Experimental Setup and Results


Here is the experimental setup. The light source is Ti : Sapphire laser pulses, whose period is about 1 ps. An changeable aperture is used to control the input power. A fast lens is placed between the light source and the photodiode to make sure that all the light will be focused onto the detector. Thus the calibration will be independent on the size of the light spot. To avoid to burned the photodiode, a ND filter is used to lower the intensity of the input light. The filter we use is with ND=3, which mean it will lower the input power by 1000 times. Fig.4 and fig.5 show the setup in our lab.


fig. 4


fig. 5


We measured 24 dots to calibrate the linear regime of the detector. Fig.6 gives the result of our calibration.

fig. 6


Y = A + B * X


Parameter Value     Error


A     4.03         0.61

B     5.93         0.04

R                     N

0.99949             24




The linear correlation coefficient R is 0.999, so the linearity of the response is quite good. And we got the relation between the input power and the output voltage. Thus we calibrated the linear regime for the detector successfully. It works just like a power meter. We can use it to tell the laser power easily. However, because the spectral response of the photodiode (shown in fig.7), if we use it for other wavelength laser, we need to calibrate it again. But for the Ti : Sapphire laser in our lab, it is effective.

fig. 7





Professor Weinacht helped me a lot with my project. He helped me to pick up such an interesting and practicable topic, introduce me the most basic facilities in the lab, encouraged me to use them and gave me many useful advises. Under his guidance, I became familiar with the atmosphere in the lab. I am so grateful to his generous help. Also, I want to thank my lab mates¡ªPatrick and David. They always lend me a hand when I encountered problems.