• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 9 subscribers
  • Views 433 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.

TRF3765, VTUNE ACCURACY IN AUTOMATIC CALIBRATION

Raul Gracia
Prodigy 20 points

Hello.

In my application, I need to obtain a Vtune the nearest possible to 900mV, therefore I configure the parameter VCO_CAL_REF to 0.9V, and in order to obtain the nearest possible value I configure the parameter CAL_ACC with a value of 3.

However, the obtained value VCO_TRIM, is different for each calibration varying from 56 to 60, and therefore obtaining in each calibration different values of Vtune (from 0.46 to 1.62V), and therefore different phase noises in each calibration. I observe that these results are independent of the value used in CAL_ACC, obtaining similar results for values from 0 to 3.

- Is it possible to obtain more accuracy in the Vtune calibration?

- Does the accuracy depends of the configuration of the charge pump?

- How can I increase the accuracy of the Vtune calibration?

Thanks in advance and best regards.

Raul Gracia

  • 0
    TI__Expert 3300 points

    The most accurate setting for calibration accuracy is CAL_ACC = 0. The least accurate is CAL_ACC = 3. Your result distribution (VCO_TRIM = 56 to 60) is higher than normal, apparently reflecting this loose setting.

    The VCO + cap search algorithm distribution will be also tighter with slow cal_clk_freq. The next biggest thing you can do to reduce VTune distribution is use a calibration clock frequency closer to 100kHz than the higher speed limits by adjusting CAL_CLK_SEL according to your chosen pfd frequency.

    Charge pump current is not a factor.

    If you are using a single frequency instead of multiple frequencies, it is possible to calibrate once, record the resulting setting, and force re-use of that setting. This would eliminate yoru distribution altogether but only applies in a special case.

    Note that VTune will, by design, shift as temperature changes. Make distribution measurements at the same temperature level. Along the same lines, you certainly can reduce the nominal VTune target to 900mV from the default 1.04V, but be certain to verify that it retains lock at your temperature extremes.