• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 1903 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.

SN74LV4046A: Loop bandwidth design

Kawai
Guru 19785 points
Part Number: SN74LV4046A

Hi Team,

Could you please help us how to design Loop Bandwidth using SN74LV4046A ?

I know that there are R and C as a loop filter, however, we would like to know how to calculate the loop bandwidth with the used R/C value.

I found App Note : SCHA003B, however, since the document is old, LV4046A information was not included. Is there any material that states the design procedure in specific ?

SCHA003B down load link
http://www.ti.com/litv/pdf/scha003b

Best Regards,

Kawai

  • 0
    TI__Guru* 97046 points

    Hi Kawai-san,

    The corner frequency, also called the 3dB bandwidth, for a first order low pass filter (RC circuit) is:

    fc = 1/(2πRC)

  • 0 in reply to Emrys Maier
    Guru 19785 points
    Hi Emrys-san,

    Thank you for your reply.

    So it is just a normal RC filter. If you want to tighten the loop bandwidth user could enlarge R and C value. Is there any other condition to be met for the PLL loop to be stable ?

    I was wondering because normally charge pump current in PLL affects loop bandwidth and filter value affects damping factor.

    Best Regards,
    Kawai
  • 0 in reply to Kawai
    TI__Guru* 97046 points
    My apologies Kawai-san, I misunderstood the original question.

    I am not familiar with loop bandwidth calculation for PLL circuits, however it's my understanding that the SN74LV4046A was designed as a 'drop-in' replacement for the HC4046A, which the document you linked to specifically discusses. The same equations should apply.
  • 0 in reply to Emrys Maier
    Guru 19785 points
    Hi Emrys-san,

    Thank you for the information.

    What we are facing is that when drop-in replaced "MC74HC4046A" (being EOL) to SN74LV4046A, the system output jitter had degradation (out of compliance test spec). We had tried changing the RC to a larger value which probably designs loop bandwidth lower , however, could not see any improvement. Regarding the specification of the device, it seems SN74LV4046A has better performance.

    Please advise us what we could do.

    Best Regards,
    Kawai
  • 0 in reply to Kawai
    TI__Guru* 97046 points
    I'm afraid a difference in performance is to be expected when switching from one manufacturer to another. In this case it seems that our device doesn't have as good of jitter, and I unfortunately don't have a solution to offer.

    If you can share the design surrounding the LV4046A, we could try to duplicate the circuit in our lab and perform tests to see if there's any way to improve the issue.
  • 0 in reply to Emrys Maier
    Guru 19785 points
    Hi Emrys-san,

    Thank you for your reply and your support.

    I am closing this post as our customer came up with a solution. It seems some additional passives were required for their application compared with the competitor device.

    Best Regards,
    Kawai
  • 0 in reply to Kawai
    TI__Guru* 97046 points
    Kawai-san,
    Thanks for letting me know. I'm glad they found a solution.

    If you can share the final solution (if it is not proprietary) it would be very helpful for us in the future.
  • 0 in reply to Emrys Maier
    Guru 19785 points
    Hi Emrys-san,

    Customer had added 1M Ohm pull down at the PC2out pin.
    They are using external VCXO to obtain better phase noise.

    We have below questions from customer.

    Q1). Is it acceptable to add 1M Ohm pull down to GND at PC2out pin ? It is connected to loop filter and VCXO.

    Q2). Why did the performance improved by adding 1M Ohm resistor ?

    Q3). What is the output impedance of PCxout pins ?

    Best Regards,
    Kawai
  • 0 in reply to Kawai
    TI__Guru* 97046 points
    Can you provide the full circuit they are using? I can't say why a single resistor improves their circuit if I don't know what it is.
  • 0 in reply to Emrys Maier
    Guru 19785 points

    Hi Emrys-san,

    Thank you so much for your continuous support.

    Could you please send me an email to below email address ?

    I can send you the schematic data.

    Best Regards,
    Kawai