• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 47 subscribers
  • Views 281 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.

TSW1200 rms noise definition

Mikael Landing
Prodigy 10 points
Other Parts Discussed in Thread: ADS4249

Hi,

I am using the TSW1200 2.5 to evaluate the ADS4249. I now wonder how the rms noise presented in the GUI is calculated and what relationship it has with SNR?

I expected it to be SNR=-(RMS_noise+10log10(NOP/2)) where NOB is number of points in FFT, but the numbers do not work out.

From the screenshot below: -(-106.13+10log10(4096/2))= 73.0 which is not eqal to the presented 69.44 db in the GUI.

There is no info in the help file and it seems to be have added recently because in some helpfile GUI screenshots it is not there.

 Thanks for your support!

  • 0
    TI__Mastermind 29220 points

    Hi,

    i think the rms calculation is off by a few dB in the TSW1200 GUI.  I think the TSW1200 GUI simply took the points in the noise floor of the FFT and calculated the rms value of the values as seen in the plot.  The correct way would have been to take the values back to linear scale, take the rms of the values, and then convert that back to dB down from full scale.  When we did the GUI for the more recent TSW1400 we revisited this issue and got it right (i hope) for the TSW1400, and I believe the numbers work out if you are looking at the reorted SNR and rms of the noise floor on that GUI.  And the rms of the noise floor didn't correlate between the TSW1200 GUI and the newer GUI for the TSW1400.

    In the TSW1200 GUI, however, the noise floor value is just for display purposes.  it is *not* used by the GUI to calculate anything else such as SNR.  The SNR value is calculated directly from the FFT array data, not from the calculated rms value. 

    Regards,

    Richard P.