• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 50 subscribers
  • Views 1055 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

OPA657: TIA noise analysis

Lu Tan
Prodigy 10 points
Part Number: OPA657
Other Parts Discussed in Thread: THS4631, TINA-TI

I have a few questions regarding TIA noise analysis (specifically for OPA657):

1. SNOA942A, Table 1-> All noise terms are shown in units of pA/RtHz. My understanding is that this is because the noise was modeled as an input current source. Since I need to establish a noise floor on the output in Volts, do I multiply a noise term with the feedback resistor gain Rf to obtain the output noise.

For example, in Table 1, for the THS4631, the noise voltage term is 8.7 pA/RtHz and Rf = 810 Ohms. Is the output voltage noise 7047pV/RtHz (=8.7pA/RtHz * 810 Ohms)?

2. SBOA060 -> I follow the voltage noise calculation, but I do not follow the current noise and resistor noise calculations. Can someone elaborate on those calculations?

3. Are there guides on SPICE simulation for TIA noise? 

  • 0
    Guru 140200 points
    Hi Lu,

    1. Yes, you must multiply this noise with the feedback resistor. Concretely spoken the given noise current is a spectral density. So, you still must multiply it with the SQRT of bandwidth. In frequency regions where the noise current density varies you must choose small portions of bandwidth and add all these terms geometrically.

    2. In the document SBOA060 are some typos. You might want to have a look at this thread:

    e2e.ti.com/.../2539745

    Kai
  • 0 in reply to kai klaas69
    TI__Genius 14146 points

    Hi Lu,

    TI Precision Labs has some detailed videos explaining different noise sources. But in summary, voltage noise must be calculated piecemeal from the noise gain bode plot, current noise must be multiplied by source resistance and then calculated same as above, and resistor noise is generated by each resistor, and each must be added seperately based on their location in the signal path. Usually the most significant is the resistance between the V+ pin and ground.

    I don't know of any SPICE noise simulation documentation, but the above videos will show you how to simulate total noise in TINA-TI.

    Best regards,

    Sean