Light pulse measurement with photodiodes

[jstamour802]

I am trying to measure very high light pulses by using photodiodes. Light levels are expected to be upwards of 2-3W of irradiance.

Pulse-widths are from 20uSec to 1mSec. Example light source is a highly over-driven high brightness LED, typically 200-300mW in power (during recommended current levels)

My basic design is to have a reverse-biased photodiode fed into a voltage->current converting circuit. The pulsed light will excite the diode and the circuit should effectively convert voltage output from the diode to a current output so I can estimate the power output from the pulsed light by following a responsivity curve.

The problem is, I can't find a way to measure such high light levels with these devices. This is a very low-speed application, 10-20Hz. All the photodiodes I can find saturate at very low levels and I can only read low levels of power.

Anyone have any ideas?

 

[berkeman]

Can you attenuate the light signal by some known amount? Use crossed polarizers at an angle that you set to a calibrated attenuation amount (using DC light to calibrate the attenuation amount).

 

[n0_3sc]

The pulsed light will excite the diode and the circuit should effectively convert voltage output from the diode to a current output...

From my understanding it is not entirely the "pulse" that excites the diode but rather the CW (continuous wave) component of your input that's exciting the diode. Thats why your diode is saturating. Most semiconductor diodes are rated up to several mW. The polarizer idea by berkeman is a good idea - utilizing something like "Malus' Law" :
See bottom of page: http://en.wikipedia.org/wiki/Polarizer"

Since your source is fairly broadband what about getting a Thermal Power Meter? They usually handle up to 3-4W. (Again measuring total average power)

I am currently trying to measure peak powers and have a thread:
https://www.physicsforums.com/showthread.php?t=221909
if that's of any interest keep watching, hopefully someone should have a good idea.

Indeed your pulses are fairly long compared to the carrier lifetime of a semiconductor diode, in which case you could probably come up with a (complicated) circuit using an input trigger from your pulse generator and only have the diode "looking" at the instants that a pulse hits the diode...I don't know how well that would work though.