• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 21 subscribers
  • Views 564 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.

WEBENCH® Tools/LMK04832: Loop Filter settings

Venu Reddy
Intellectual 320 points

Part Number: LMK04832

Tool/software: WEBENCH® Design Tools

Hi

I want to re-design the loop filter for LMK04832 for both PLLs. However the webbench tools doesnot support it. I tried using platinum tool, with the same user guide values from LMK04832, it shows different loop filters results.

Here are the Inputs, For PLL1, Input frequency 10MHz, fpd =1MHz, Loop BW = 14Hz, Phase margin = 50, Kpd = 0.45mA, Kvco = 2.5KHz/V, Output freq = 100MHz, VCXO =100MHz.

For PLL2, Input frequency = 100MHz, Fpd = 100MHz, Loop BW = 220KHz, Phase margin =70, Kpd = 3.2mA, Kvco = 13.6MHz/V, Output freq, 10MHz, 100MHz, 50MHz. So I am using VCO0.

I am optimizing for Lowest possible Jitter. Can you provide me the optimal loop filter settings and values ?

Regards

Venu

  • 0
    TI__Mastermind 28031 points

    Venu,

    The instructions have insufficient information to uniquely define the loop filter.  Assuming Kvco, Kpd, and N fixed as well as R3, R4, C3, and C4 integrated, we have 3 unknown components.  We therefore need 3 paramenters. Loop bandwidth and phase margin are two.  The third constraint is the gamma optimization parameter.  In PLLatinum Sim, go to advanced mode and it will show this.

    If you copy the information from the EVM instructions into PLLatinum Sim, it gives the following parameters in advanced mode:

    Loop Bandiwdth:  429.9441 kHz

    Phase Margin:  76.2855 deg

    Gamma:  1.0771

    Note that for design purposes T3/T1 and T4/T3 are ignored because R3, R4, C3 and C4 are fixed.  Also, I see that R3 on the design screen is wrong and should be changed to 0.2 kOhm.  Then if I set the capacitor and resistor step to "Ideal"  I get the correct values.  (note that using them at 10% creates a small round-off error).  See plot below

    So I see that the loop bandwidth is a little off and so is the phase margin.  Bottom line is that I trust PLLatinum Sim over the EVM instructions.  I see that these instructions are dataed 2018, but LMK04832 was added to PLLatinum sim in 2018.  The calculations for loop filter values for PLLatium sim are robust, provided all the inputs (including gamma) are given.  Tinkering with the settings, it looks to me that perhaps the values given in the EVM instructions are making different assumptions about R3, R4, C3, and C4 than are stated.  If I make these zero, it's closer to what hte instructions say.

    Regards,

    Dean

    For your settings, it seems that you are using your own external VCO.  If so, import the VCO phase noise and PLLatinum Sim will recommend a loop bandwidth based on optimal jitter.