• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 49 subscribers
  • Views 651 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.

DAC80004: Noise specs?

David F. Chan
Expert 5500 points

Part Number: DAC80004

Hi all,

I am wondering if there is a discrepancy with the rated noise figure in the datasheet, as seen below.

60nV√Hz * 100kHz^0.5 = 19uVRMS at 100kHz BW, which does not seem to match what is shown in Figure 49 if this math is correct. Can you please verify?

  • 0
    TI__Genius 11750 points
    Howdy David,

    This DAC includes an operational amplifier, so it's not just limited by resistor thermal noise, but also 1/f noise from the amp. You'll notice this from figure 49. To calculate output noise you would need to integrate (area under curve) of the plot shown in Figure 47. This accounts for the higher value presented in the datasheet.

    Best Regards,
    Matt
  • 0
    TI__Prodigy 420 points

    Hi David,

    I have two comments

    1) For the broadband calculations, you are assuming brick wall filter - we typically use 1 pole filter therefore you need to multiply the BW (100KHz by 1.57)

    2) You are not taking 1/f noise consideration into your calculation. The 1/f noise for this process is quite large. In plot 49 we only go upto 100Hz (because of limitation of the instruments). I can imagine that if we go close to 1Hz, the noise density could be quite high (few 10s of uV/rtHz). The RMS noise that we specify in the datasheet is not calculated but measured, therefore i would trust this value.

    Thanks,

    Rahul Prakash

  • 0 in reply to Rahul Prakash
    TI__Expert 5500 points
    Hi Matt and Rhaul,

    Thanks for the speedy response. However, if the measured 100uVRMS figure is correct, wouldn't the normalized noise be closer to 200nV/√Hz when accounting for a 1pole filter?
  • 0 in reply to David F. Chan
    TI__Genius 11750 points

    Howdy David,

    I think I understand the question, but please correct me if I'm incorrect.  The value presented in the datasheet does not seem to be a normalized value, but a datapoint that corresponds to the noise spectral density plot at 1kHz and 10kHz.  This datapoint gives the user an overall understanding of the thermal noise of the DAC, as the thermal noise is a uniform noise source.  The other independent noise source, is the 1/f noise that Rahul mentions -- at lower frequencies, this noise dominates.  For a approximate understanding of the noise in Vrms, the area under the curve is usually calculated, with respect to the desired bandwidth.  Please refer to the "integrating noise" section of this application report:

    http://www.ti.com/lit/an/slva043b/slva043b.pdf

    Best Regards,

    Matt