• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 47 subscribers
  • Views 689 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.

DAC60508: Sample/Update rate (Msps) and settling time

Marco Failla
Prodigy 85 points

Part Number: DAC60508

Tool/software:

Hello,

i'm finding the right solution for my application where I need at least 800 ksps out of a DAC.

I know the empirical rule that out sample rate is just the inverse of the settling time, as a maximum value, real things are worst.

While looking at TI catalog, under <10 MSPS DAC section, I see many DAC with 1 MSPS and settling time bigger than 1 us.

For example the model I'm looking for is DAC60508 with a settling time of 5 us but a given sample rate of 1 MSPS from what I can see in the product detail.

It should be 1/ 5us = 200 ksps

Datasheet is never referring to sample rate as a parameter.

So, where is the true? 

Thanks in advance.

  • 0
    TI__Mastermind 42253 points

    Hi Marco, 

    What is your application? Are you trying to create some waveform like a sine wave?

    Best,

    Katlynne Jones

  • 0 in reply to Katlynne Jones
    Prodigy 85 points

    Hello,

    yes, I need to create a 40 kHz sinevawe but with at least 10 samples in a period, for signal processing reasons.

    So I need an update rate of 400 ksps with a bit of margin.

    I don't understand how the sample rate in the product details of TI DAC is calculated, because I only know there is a relationship with the settling time.

    Thanks

  • 0 in reply to Marco Failla
    TI__Mastermind 42253 points

    Hi Marco, 

    Are all 8 channels going to need a 40kHz sine wave?

    The update rate/sample rate are determined by the SPI frequency, and the settling time. The settling time is less important when you are making small code changes to form a sine wave. The 5us settling time is for a +/-2LSB settling band, and I doubt that is a requirement for your sine wave. For a sine wave, we can just look at the contribution from the SPI frequency. If you're using a 50MHz SPI clock, 10 codes per period, and 24 clocks required per update, your update rate is closer to 50MHz/(24*10) = 208kSPS. 

    If you really need all channels to be updating simultaneously, divide that number by 8 to get 26kSPS. 

    Best,

    Katlynne Jones