I am having issues with calculating some parameters for some novel though low efficiency (E-10 A current generated under illumination, responsivity E-9 A W-1) photoconductor type photodetectors.
I have primarily had issues with noise based values such as NEP and signal: noise
I have obtained a time dependent photoresponse curve at +40 V bias under darkness and solar lamp illumination. with responsivity, rise/ fall times etc obtained.
I then tried to measure noise using a lock in amplifier (taken under darkness + 40 V)
The R current value obtained from noise measurement also appears to be higher than the current obtained under time dependent photoresponse across most frequencies.
For noise data the frequency range was 10^2 to 10^6 Hz, noise data was taken with units A/√Hz
current was also measured. The device were placed under 40 V bias during testing and tested across a 1000 ohm resistor, I converted the noise data from A/√Hz to W/Hz.
By integrating under the power spectral density frequency graph a power of 2.8E-7 W is obtained, however, from here I can't seem to be able to calculate sensible values for NEP and signal: noise.
For signal:noise if the noise data is 2.8E-7 the signal power (40V x current) does not seem to provide a sensible signal noise
From time dependent photoresponse it seems I should be getting a snr around single digits or low double digits
How should I go about calculating NEP and signal to noise from the data obtained? Any comments could be much appreciated. Is the noise data sufficient or is it possible to obtain noise from the time dependent photoresponse?
I have attached the noise equivalent power and time dependent photoresponse curve. I also have large noise peaks at certain frequencies but am not sure of the cause
I think you would be better off driving your system with wide band noise of precisely known power. Then you just measure total output power with and without signal. However, you are measuring very low power levels - are you taking all the necessary precautions to shield your devices from external electromagnetic interference? E-10 A in 1000 ohms is 0.1 microvolts. Almost all radio stations will give signals greater than this, and confound your readings. Best wishes with your research!
- Badges
- Science topic