As part of a Laboratory Electronics class, my lab-mate and I had to design and build a lock-in amplifier which would measure a modulated LED light signal. We had 3 lab sessions to work on it (3 hours each), and there's certainly some improvements that we're aware we could've made given more time, but we were pretty happy with our results.
A PDF of the schematic can be found here , and the corresponding LTSpice file can be found here. We got a photo-current sensitivity of 400 pA, which was actually somewhat high compared to some other groups in the class, but we had about the second best range (ie how far we could separate the LED source and the photo-diode receiver and still see a SNR > 1) at 2 feet.
Here's a picture of the front-end. The front-end stage was powered by batteries and housed in a metal box to reduce noise and interference. The front-end consisted of a Hamamatsu silicon photo-diode, followed by two gain stages built with OP-27's. The lock-in part of the circuit used the AD 630 Balanced Modulator IC, followed by a two-pole low-pass filter at 1 Hz.
We observed that we were picking up 120 Hz noise from the room lights on the back-end board (contained AD 630 circuit). We weren't sure if this was due to the wiring on the board or the scope probe acting as an antenna (if we left the scope probe floating, we'd see the same noise we saw on our circuit output). Some other things we could've improved on include: increasing second-stage gain, trying to filter the 120 Hz noise, adding DC gain after the lock-in low-pass filter, and experimenting with different op-amp (one with less bias current) and different photo-diode biasing schemes (possibly using a photo-transistor, although our photo-transistors in the lab had much smaller windows). These improvements would likely increase our photo-current sensitivity to the 1 pA scale and increase our range as well.
No comments:
Post a Comment