Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 46 subscribers
  • Views 465 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DAC8831 practical speed?

Mohsen Jamalabdollahi
Prodigy 80 points
Other Parts Discussed in Thread: DAC8831

Has anyone used the DAC8831 before? what is the maximum signal frequency that it can build on 2.5 MSPS?  Is it a good choice for generating a  2.5M/16  = 156KHz  sine wave?

  • 0
    TI__Mastermind 46290 points

    Mohsen,

    The maximum rate that you could possibly write data to the DAC is limited by the SPI timing requirements for a write sequence. Specifically this is tLead, tsck, tLag, and ttd. These values sum to 370ns or ~2.7MHz. The DAC output takes 100ns for the output to settle from zero-scale to full-scale after the word has been latched by the rising edge of CS in the SPI interface. In this case the maximum throughput is limited by the SPI interface, rather than the settling time of the DAC, and the maximum sampling rate of the DAC is ~2.7MHz. This means you could provide ~17 updates per period for a 156kHz sinusoid or 16 updates per period using the 2.5MSPS figure.

    Resistor based DAC topologies, both string based and R-2R ladder based, are optimized for DC performance specifications. Accordingly you'll predominantly find DC specifications for these devices in their datasheets. The feasibility of using a DC precision DAC for an AC application really depends on your signal integrity requirements and what sort of filtering you think you can implement to clean the signal. An AC waveform from the DAC8831 is going to see frequency content from the update rate of the DAC, digital feedthrough of the SPI signals through to the analog output, and another AC phenomenon called glitch (which the DAC8831 specifies at 35nV-s, a relatively high figure but something we should expect to see from an R-2R ladder based DAC).

    Delta-sigma or current-steering DACs are much more commonly used in AC signal generation applications. These devices are frequently specified for AC performance rather than DC performance and showcase a number of attractive features inherent in their architectures that are conducive to AC signal generation.

    Unfortunately the signal band you've proposed is sort of in a void of coverage between audio DACs and high-speed DACs that limit your available choices.

    In short the direct answer to your question is the feasibility of this approach really depends on your signal integrity (THD, SNR, etc) goals as well as how much filtering you're able to integrate in the back-end. A more typical approach for you to consider may be using a VCO using a precision DAC to drive the VCO voltage and set the frequency.

    Can you share more details about your application?

  • 0 in reply to Kevin Duke
    Prodigy 80 points

    Dear Kevin,

    Thanks a lot for your help, I am trying to build a underwater modem for sensor nodes communication. The SPI interface comes from the C6713 DSK which can provide the 50MHz CLK SPI interface. Considering this interface do you think that this ADC can follow the 16 samples of a sine wave up to 156MHz without and distortion? Considering the 25 v/microSec as slew rate of 8831,it suppose to follow  up to10V at 1/2.5M = 0.4 microSec for the next sample, however, the maximum slew rate of a 156KHz sine wave is less than 1 volt for .4 microSec for VREF = 3V.

    Please also consider that we need this DAC at transmitter where the samples are noise less and limited to the desired bandwidth. As an example of desired transmitted signal you can consider the chirp signal where the frequency of sine wave increases linearly up to 156KHz. 

    Now, based on this information do you think that the DAC8831 is a proper option for us or not?

    Thanks again,

  • 0 in reply to Mohsen Jamalabdollahi
    TI__Mastermind 46290 points

    Mohsen,

    The C6713 has a sufficiently fast SCLK to realize 16 samples per period at 156kHz. I'm not intimately familiar with the serial peripheral on the device though so I cannot speak to whether it can meet the full set of timing requirements of the DAC8831. I would suggest you post on the C6000 forum to get someones advice on that matter.

    It is difficult for me to predict via calculation or simulation the quality of the signal you would create using the DAC8831 in this manner. Parameters like digital-feedthrough and glitch are dependent upon PCB layout parasitics and the specific word being written to the DAC. We would likely see some impact of the DC error parameters of the DAC as well (i.e. linearity, offset, and gain). Furthermore the integrity of this signal would be dependent on the specific points used in the look-up table to generate these sinusoids.

    It is plausible to use the DAC8831 in your application. Again, it would come down to the exact definition of your performance criteria when you say "noise-less" or "without distortion". There is an evaluation module available for the DAC8831 that you may consider using to prototype the system.

    Being that precision DACs are generally meant for DC applications, I would not say that this is the most appropriate application for this device. As I mentioned though you may find that there are not many options to service the specific band you are interested in so there may not be a directly appropriate device for this application.