• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 483 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: TRF3765

Eric Cui13
Intellectual 920 points
Part Number: TRF3765

Hi,

We are using the TRF3765 as an RF LO in the 2-4 GHz range, with a 10 MHz external reference.

At room temperature we have no issues, but at ~5 C the chip fails to lock detect when programmed with a new frequency.

Any ideas on what could be going on, or tests I could run to troubleshoot?

Thank you.

  • 0
    TI__Mastermind 18065 points
    Based on your description it sounds like the original programmed frequency is still locked at the cold temperature but when programmed with a new frequency the device fails to lock.

    Here are a few things to monitor to troubleshoot. Monitor (through readback of the SPI) the programmed VCO selection and the programmed cap array setting (VCO TRIM). Also monitor the Vtune voltage on the board with a voltmeter.

    I assume that you are enabling calibration upon the frequency change. Calibration searches for the optimum VCO (there are four) and the optimum VCO trim (0-63) to keep the Vtune voltage at the optimum level (~1.5V). If the device is calibrated at room temp and then brought to cold, there is sufficient range to keep the device locked. You will see the Vtune voltage shift accordingly. If you change frequencies without calibrating, the device may not be in the proper trim value to achieve lock.

    My first guess is that a VCO calibration is not being done. If so, the simple resolution is to re-calibrate to get it to lock. Conceivable, you can pre-determine the VCO and trim that is optimized for the new frequency and just program that in directly without re-calibrating.

    If that is not the issue, then my second guess is that the new frequency is right on the borderline between two VCOs. Monitoring the VCO and trim level will tell you if you are on the edge. If VCO trim is near 0 or near 63 then it is on the edge. Change Cal_Clk_Sel to /128 to slow down the calibration to allow the algorithm to converge to a more optimum place.

    --RJH