EEVblog Electronics Community Forum

Electronics => Projects, Designs, and Technical Stuff => Topic started by: arivalagan13 on April 20, 2021, 05:19:17 am

Title: Transimpedance amplifier noise calculation
Post by: arivalagan13 on April 20, 2021, 05:19:17 am
Hi all
I want to calculate the output noise voltage at the output of an op amp-based transimpedance amplifier. Here I've attached a file where I did an example calculation. Can somebody please tell me whether my calculations are correct???


Regards
M. A :( :( :( :(
Title: Re: Transimpedance amplifier noise calculation
Post by: Kleinstein on April 20, 2021, 07:16:46 am
The 20*log (1+32p/1p) part is not correct. This is somewhat mixing with the dB scale. However the numerical result for 20*log .. is not far off the corret factor without the 20*log part. So by lucky coincidence the mistake does not change much.

The calculation assumes that the ouput is sampled not much faster than some 50 kHz, so it does not include much higher frequency noise. Chances are this is OK for the application.

There would be a small contribution from low frequency 1/f noise of the OP. So kind of add the 0.1-10 Hz RMS noise to the final result. This would be negligible in this case with high TIA gain and high BW, but could be an issue in other cases.
Title: Re: Transimpedance amplifier noise calculation
Post by: awallin on April 27, 2021, 03:16:16 pm
FWIW I wrote some python scripts that predict dark-noise and frequency response for simple one op-amp transimpedance designs.
there's also a few comparisons to real-world measurements in the repo:
https://github.com/aewallin/TIASim

if there's something not clear please ask or submit a github issue and I may improve the documentation.

HTH
Title: Re: Transimpedance amplifier noise calculation
Post by: graybeard on April 29, 2021, 06:14:05 am
The temperature for the resistor is Kelvin, not Centigrade

p.s. I still use mhos for conductance as well!
Title: Re: Transimpedance amplifier noise calculation
Post by: harerod on April 29, 2021, 07:56:11 am
The temperature ... is Kevin, not Centigrade
...
You Americans surely have a hard time with units not statute, ey? YMMD.  :-DD