• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 47 subscribers
  • Views 415 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.

OPA2376 Noise Density Part-to-Part Variations

Charlie Sweet
Prodigy 140 points
Other Parts Discussed in Thread: OPA2376

Hi Folks,

I have selected the OPA2376 as a pre-amp for an EEG filtering application based on its low noise for low supply voltage and current characteristics.

How can I get information on voltage noise density and current noise density variations from part-to-part (statistical or min / max info) specifically including the frequency range 1Hz to 100Hz (signal bandwidth 2- 47Hz)?

Thanks,

Charlie

  • 0
    TI__Guru 68171 points

    Charlie,

    The typical graph of OPA2376 input voltage noise spectral density is given on the front page of the PDS while the input current noise spectral density I simulated below - it increases in its value around 100kHz due to common-mode and differential input capacitance causing an increase in the input bias current- IB is the main factor in the input current noise equation: Inoise=sq-rt[2*q*IB].

    The voltage noise does not change significantly from part to part for the units coming out from the same wafer lot or another lot manufactured within days of each other (they vary just few percent) BUT the long term wafer process variation (lots manufactured several month or a year apart) may cause much more profound noise variation.

    Thus, the broadband input voltage noise spectral density (typical 7.5nV/rt-Hz) may vary by +/-33% (six-sigma) so for wafer lots manufactured a year apart this may mean a noise value anywhere between 5nV/rt-Hz to 10nV/rt-Hz.  

    In case of low frequency voltage noise (see grpah below), for the parts coming from wafer lots many months apart, 1/f noise can vary even more significantly - from 0.5x to 2x - due to the variation in 1/f corner frequency that may occur at half or double frequency (again six sigma variation).  Therefore, the OPA2376 integrated input voltage noise, from 0.1Hz to 10Hz, (typical 0.8uVpp) may vary anywhere between 0.4uVpp to 1.6uVpp.

    The input current noise is a function of the input bias current, Inoise=sq-rt(2*q*IB), thus it varies as a square function of IB variations.  Since IB increases with temperature (doubling every 10 deg C), as the temperature increases from 25C to 125C, IB doubles 10 times, (125C-25C)/10C, so IB increases 1024x (2^10), resulting in the input current noise increase of ~32 (1024^0.5).  Therefore, the typical OPA2376 input current noise spectral density at 125C should be around 8fA/rt-Hz - sq-rt(2*1.6E-19*0.2E-12*2^10)