• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 46 subscribers
  • Views 471 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

OPT301 Settling issue and design question

Nate Oborny
Prodigy 20 points
Other Parts Discussed in Thread: OPT301

I am using an OPT301 combination photodiode transimpedance amp to pick up the presence of a fluorescent chemical in a small (<40um) fluid filled channel. The light level from this tiny amount of the chemical is very small and consequently I have been using the OPT301 with an external 10G resistor for gain. This does work although I'd like to push it maybe 10x further while minimizing the noise. Here in lies my problem or, more to the point, my several problems:

1. 10G is as large a surface mount resistor as I can easily obtain. I was planning therefore to use a T-Network to accomplish a higher overall gain with smaller value resistors. I have read however that this will increase my 1/f noise more than a single resistor would and was hoping for any advice the forum might have in limiting this effect.

2. As I have increased the gain on the OPT301, I have seen an increasingly long settling time in response to the signal. My signal of interest is <10Hz as no more than 10 bands of the fluorescent chemical will be able to pass the detector in under a second. The rise time of the detector appears more than adequate for this rate. However, once a signal is detected, the drop off back to zero can take a considerable amount of time. Am I correct in thinking this is a direct result of the extreme gain? What is the best way to eliminate this effect?

Thanks,

Nate Oborny

  • 0
    Guru 21975 points

    Nate,

    The noise of the feedback resistor contributes directly to the output noise of a TIA. The noise is proportional to the square-root of the resistance whereas the desired signal increases proportionally with resistance. Thus signal-to-resistor_noise ratio improves with increasing resistor value. A T-network is much like adding another gain stage after a lower-gain TIA. Noise and signal are both amplified equally so the S/N ratio is that of the lower-gain TIA.  I don't know a way to beat this problem. Higher resistor values are available but probably not in surface mount. You could, of course, use several in series.

    I don't see that a first-order photodiode model and circuit theory can answer your second question. Few users have ventured into your hyper-ohm gain range so this effect may have never been seen before. I suspect that this is a subtle effect involving slow migration of residual charge in the photodiode. Unfortunately, the designer of this product passed away early this year and we've lost much of his historical knowledge. (RIP, Wally.) The solution may be a different photodiode.

    Regards, Bruce.