Post Go back to editing

SNR Measurements in datasheet

Category: Datasheet/Specs
Product Number: AD7608

Hey,

Figure 19 in the datasheet for the AD7608 is showing results of signal to noise ratios that can be expected with different oversampling rates.

Is it possible to get any information on how this is measured?
I have the development board EVAL-AD7608SDZ, I assume that there has to be fitted a resistor shorting the pins of the SMB connector (R5 in schematic screenshot below).
If this is the case, do I short with 1Mohm resistor to match the input impedance of of the ADC, or to short it directly with 0 ohms.

Appended:

- Figure 19 from AD7068 datasheet
- Screenshot of V1 SMB connector from EVAL-AD7608SDZ schematics

  • Hi,

    I'll look into this and get back to you.

    Regards,
    Jo

  • do I short with 1Mohm resistor to match the input impedance of of the ADC, or to short it directly with 0 ohms

    You can short it with 0 ohms. But this gives too good results and not realistic ones.
    Keep in mind that your actual source impedance has inherent noise level too, that adds to your ADC SN.

  • Will I then expect better results than showed in the datasheet?
    What setup do I need to reproduce the results that are showed in the datasheet?

    I have done some measurements and I'm wondering if they are correct.

    The signals are generated by a RSDG805 Waveform generator, and the noise referenced is measured by shorting R5 with a 0 ohm resistor.

    I am getting a difference in results based on if I remove DC offset of the noise signal in post-processing of the data
    Why is it that I am measuring a DC offset when shorting the differential outputs to ground?

    Appended:

    Screenshot of SNR calculated with and without DC offset removal