• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 9 subscribers
  • Views 469 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.

Carrier and Sideband Suppression for the TRF3703

Robert Murano
Intellectual 1275 points
Other Parts Discussed in Thread: DAC5687, DAC5686, DAC5682Z, DAC5688

What is the carrier and sideband suppression for the TRF3703? If it is calibrated, how well does the calibration hold?

  • 0
    TI__Intellectual 1275 points

    Robert Murano said:

    What is the carrier and sideband suppression for the TRF3703? If it is calibrated, how well does the calibration hold?

    Sideband and LO Suppression are pretty interesting topics of discussion. For a signal much less than the interface BW, there are two cases, depending on whether the LO and sideband are adjusted or unadjusted. Unadjusted means no attempt at compensation.
     
    Consider for example the TRF3703-17. The attachment to this post has figures taken from the TRF3703-17 datasheet found on the website
    are 4 plots with carrier and sideband suppression, unadjusted and adjusted. The unadjust carrier (LO) is shown in the top left graph, and ranges from ~ -38 dBm below ~ w GHz to ~ -35 dBm for > 2 GHz. Carrier suppression typically is not affected by signal power, which could range from > 5 dBm for a CW signal around -5 dBm for a modulated waveform with a PAR/backoff of 10 dB. So in dBc, the carrier suppression would range from -43 dBc to -33 dBc below 2 GHz and -40 to -35 dBc above 2 GHz, depending on PAR and backoff of the signal.
     
     
    Most of our dual interpolating DACs such as the DAC5686, DAC5687, DAC5688, DAC5682z have an offset mechanism that can be used for suppression of the LO feedthrough. Alternatively, a 12-14 bit low speed DAC could be used to sum in an offset voltage. Typically this can reduce the LO feedthrough to better than 80 dBc. However, over time/temperature/voltage/frequency variation, for the example the case where the system is calibrated 1 time in manufacturing, then this typically would increase to ~> 50 dBc.
     
    An example is shown in bottom left figure, where the initial calibration is better than 80 dBm, and then maintains ~ 45 dBm over temperature and frequency.
     
    The unadjusted sideband suppression is shown in top right plot. It is > 40 dBc over most of the 400 Mhz to 4 GHz frequency range and is worst case - 35 dBc above 3 GHz. Sideband suppression tracks linearly with carrier power, so the suppression is the same for a CW signal and modulated signal (excluding wide bandwidth effects).
     
    Some of our interpolating DACs like the DAC5688 and DAC5687 have an sideband compensation mechanism (what we can quadrature modulator correction - QMC. However, the logic needed to adjust the phase and gain is straight forward and could be included in an FPGA.
     
    If compensated, the adjusted sideband suppression can reach > 70 dBc, and maintain > 50 dBc across time/temperature/voltage/frequency for a couple 100 MHz RF frequency band. An example is shown over temperature and frequency (200 MHz band) in the lower left plot.
     
     Across a wider RF frequency band, the phase would need to be recalibrated to maintain > 50 dBc, but could probably maintain 45 dBc across 0.5 - 1 GHz in the 0.7 - 3.5 GHz range. Above or below that the phase changes faster with RF frequency.
     
    Note that with a real upcoversion system, the LO cannot be adjusted and the sideband will be full power.

    Regards,

    Robert

     

    TRF3703_carrier_sideband_suppression.pdf