• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 46 subscribers
  • Views 1301 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.

ENOB determination with DC current

Benjamin Nordman73346
Prodigy 10 points

Hi everybody,

I have an issue for measuring the ENOB the MSC1200 with a DC.

In order to this, I took some information from TI's website. I found this course from Bonnie Baker but I don't understand her calculation:

http://e2e.ti.com/videos/m/analog/97246.aspx

According to her, the equation for a DC will be

ENOB = N - log2(sigma)

where N is the resolution and sigma the standard deviation considered as the rms of the noise.

I took some measures and I found my standard deviation was between 0 and 1V. Consequently, le log2(sigma) become negative and I found an ENOB value bigger than the resolution, which is totally absurd.

Do you understand where is my mistake?

Do you reckon the equation for the AC ENOB = SINAD - 1,76/6,02 is alright if ever I swap the SINAD with the SNR - which is equal to SINAD when there is no distortion issue.

Thanks for you responses!

  • 0
    TI__Guru** 113765 points

    Benjamin,

    The problem you are having is with your value for standard deviation.  The number to use is for the noise of your converter with shorted inputs.  You also need to bias the shorted inputs to within the common mode input range of the MSC1200, if you are using the buffer, otherwise you can short them to ground.

    If you look at the footnote under Table 1 of the datasheet for the MSC1200, you will see the same formula for ENOB as you have described but broken down a bit further.

    So, you have a 24-bit converter, which is your maximum resolution, and you subtract out the number of bits that are noise inherent to the device/system.  The standard deviation is the rms noise, and not the input range of your signal, as the inputs are shorted.

    Using an industry standard factor, Vpp = 6.6 * Vrms, or Vrms = Vpp/6.6.  Determine the standard deviation from a data set collected, and plug that value into your formula.

    Best regards,

    Bob B