• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 21 subscribers
  • Views 436 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.

LMX2595: Changing charge pump current results in unwanted transient

Dylan g
Prodigy 170 points
Part Number: LMX2595


Hello, we have recently encountered a problem when stepping frequency that we are unsure why it happens

When tuning to a new frequency in Full Assist, we are opening the charge pump from 3mA to 6mA, writing out the calibrated data/numerator/etc., then moving the charge pump back to 3mA around 8µs after the last frequency related data is written out. When this happens, we see on our Signal Source Analyzer a transient ranging from 64kHz to 270kHz, before going back to the desired frequency.

We were wondering why this was happening, as the part is locked when we move the charge pump current back to 3mA

  • 0
    TI__Guru** 110810 points

    Hi Dylan,

    Did you see similar transient if, you are already at a certain frequency, and then you program the charge pump to a different current setting? 

  • 0 in reply to Noel Fung
    Prodigy 170 points

    Yes that is what we are encountering

    The output frequency is already settled within <25kHz of desired, and we even extend the amount of time we wait (around 25µs after settling) before switching the charge pump. The deviation of the transient surprised us since we were already phase locked

  • 0 in reply to Noel Fung
    Prodigy 170 points

    Hey Noel, would you happen to know the amount of time the LMX takes to react to a change in the charge pump from an SPI command, and how long it takes to change it?

  • 0 in reply to Dylan g
    TI__Guru** 110810 points

    Hi Dylan,

    Charge pump is part of the PLL loop, whenever its characteristic is change, it will affect the loop performance. A glitch is expected if you change its value as the change will affect loop filter bandwidth directly. In addition, charge pump turns on every fpd cycle, if your change was made when the charge pump is turned on, charge pump current gets changed immediately and as a result disturb the loop. I am not surprise to see glitches when we change the charge pump current after the PLL is locked.