When the photodiode is illuminated with CW laser radiation, there is a steady rate of carrier generation (electrons and holes). The electric field that exists inside the photodiode causes the electrons and holes to drift in separate directions towards the anode and cathode, giving rise to the photocurrent which is measured by the display unit. A small percentage of these carriers recombine before they leave the photodiode and thus do not contribute to the measured photocurrent. When the photocurrent reaches about 1mA, the voltage on the load resistance of the display unit starts to rival the potential between the separated photo-generated charges. The photodiode’s internal potential is insufficient to push additional current through the circuit and the photodiode response becomes saturated. The maximum power specifications of our photodiode sensors are derived from this consideration.
When measuring average power of laser pulses, the measurement will be accurate if effectively all of the photogenerated charge exits the photodiode and contributes to the measurable current without undergoing recombination in the device. The discharge time of the photodiode is more relevant than the actual laser pulse width, which is often much shorter. Saturation occurs when the photo-generated charge density approaches a certain device-specific value and begins to increase the recombination rate. The laser pulse energy at which saturation occurs depends mostly on the response time of the photodiode, as it dictates how fast the charge carriers can leave the photodiode before they recombine with each other. Another important factor is bias voltage. A negative bias voltage applied to the photodiode will increase the effective potential of the separated photo-generated charges and cause the charge carriers to exit the photodiode faster. This increases both the photodiode response time and saturation pulse energy.
PD300-UV and PD300-IR were tested with Spectra Physics VGEN pulsed lasers at 1064 nm and 532 nm. The results are presented in the following table. For each pulse energy, we compared the readings with and without a fixed 50% filter. The criteria for saturation was a decrease of 2% in the measured power without the filter relative to the reading expected based on the measurement made with the filter.