• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 48 subscribers
  • Views 1108 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.

System Noise Analysis

Hari T O0
Genius 3820 points


Hi Team,

Please find the attached image.This describes how to calculate noise in a signal chain.It consists of an Amplifier,Sallen key filter,ADC Driver and an ADC.

O/P noise of each stage is given and it is converted to rms noise.

For converting the O/Pnoise of sallen key filter to RMS they used the amplifier bandwidth,Could you please tell me why?

May I know how the rms noise of ADC driver is calculated which bandwidth is used for the same.

Could you please tell me how ADC noise is obtained as 22.3uRMS

Regards

Hari

  • 0
    Guru 140200 points
    Hi Hari,

    makes no sense to me. Where do you have this scheme from?

    Kai
  • 0
    TI__Guru 69431 points

    Hari,

    In the cascaded configuration of three stages as shown above, the noise of the first stage(Gain Amp) will totally dominate the total noise because the noise of the later stages (Filter Amp and ADC Driver) get divided by the gain of the first stage (G=50) while being referred to the input (RTI).

    It is not clear to me if the 39nV/rt-Hz and 4nV/rt-Hz shown above is a noise of second and third stage RFI or RTO but regardless the calculation above is wrong as it does not take into account the noise bandwidth of each stage.

    See link below for detailed explenation fo the noise calculation:

  • 0 in reply to Marek Lis
    Genius 3820 points
    Hi Mark,

    Thank you very much
    39nV/rt-Hz and 4nV/rt are the RTO given. For converting 39nV/rt-Hz to RMS they used NEB as 50Khz,
    Vrms=(39nV/rt-Hz) *Sqrt(5000)=8.7uVrms.
    I need a clarification that for finding the RMS noise of second stage(Sallen Key Filter) why the bandwidth(50Khz) of previous stage is taken.

    Regards
    Hari
  • 0 in reply to kai klaas69
    Genius 3820 points
    Hi Kai,

    I obtained this from Avnet website.

    Regards
    Hari
  • 0 in reply to Hari T O0
    TI__Guru 69431 points

    Hari,

    The calculation of 39nV/rt-Hz RTO spectral noise across the 50kHz signal bandwidth should be:  Vrms=(39nV/rt-Hz) *Sqrt(50,000*1.57)=10.9uVrms while the correct value should be based on 1kHz third-order SK LPF: Vrms=(39nV/rt-Hz) *Sqrt(1000*1.16)=1.33uVrms - see below.

  • 0
    Guru 140200 points

    Hi Hari,

    I have found the article with this scheme in the www. This scheme is no proper noise calculation at all, but shall only demonstrate how important it is to also take the noise bandwidth into consideration and to consider the RMS noise value instead of only looking at the noise density. The scheme is only a fabricated example.

    Kai

  • 0 in reply to Marek Lis
    Genius 3820 points
    Hi Marek,

    Thank you very much.

    Regards
    Hari