• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 46 subscribers
  • Views 241 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.

Original question:

DAC82001: Output noise (0.1 - 10Hz) graph

DAC82001: Noise performance and figure 6-21

William J
Prodigy 10 points
Part Number: DAC82001
Other Parts Discussed in Thread: DAC82002

Tool/software:

Hi TI,

I have 2 questions that I hope you can help with.

1. The Datasheet for DAC82001 (and DAC82002) lists the noise to be 10nV/sqrt(Hz) at 10kHz, but figure 6-21 shows the noise to be ~30nV/sqrt(Hz). In another thread (https://e2e.ti.com/support/data-converters-group/data-converters/f/data-converters-forum/1361750/dac82001-output-noise-0-1---10hz-graph) the figure is mentioned as being inaccurate of DAC82001's true noise performance. My question is if the figure is correct in the 1kHz to 1MHz range or if there is another figure to use?

2. Figure 6-23 shows the clock feedthrough. Is this for both outputs (DAC82002) or does the channels differ in digital feedthrough?

Best regards.

  • +1
    TI__Mastermind 42273 points

    Hi William, 

    The only noise contribution in the 1kHz to 1MHz frequence range you're asking about is resistor thermal noise from the 6.25kΩ resistor string which is constant over frequency. The real plot would look like a flat line at 10nV/sqrt(Hz). the plot is showing close to a flat 30-40nV/sqrt(Hz) over frequency. When I checked with the validation engineer who took this plot when I was answering the questions in the post you linked, he told me this is basically the noise floor of the instrumentation that was used to take the plot. We don't have another plot. The dominating factor in application would be reference noise from whatever external reference you use, and noise from the output buffer you choose. 

    Both channels would have the same feedthrough response as far as the general magnitude and timing of the spikes. 

    Best,

    Katlynne Jones