I am new to PLLs and am trying to use one to generate a frequency ramp between 5.725 GHz and 5.875 GHz. I have found tools online that help design loop filters for PLLs, and all of these tools require that the user know the desired loop filter bandwidth and phase margin. I have not been able to find suggestions for choosing filter bandwidth. Also, I understand that filter bandwidth and phase margin are related, but I don't know how to choose appropriate values for them for this application.

  • \$\begingroup\$ What experiences and skill sets do you bring to the table, regarding nearly \$6\:\text{GHz}\$ signals? (You may want to include some discussion of the reason why you need this and some of your thoughts.) \$\endgroup\$
    – jonk
    Commented Aug 30, 2019 at 1:54
  • \$\begingroup\$ Can you define slew rate MHz/ms tracking rate or capture time or LPF BW of clock noise or BW output of signal and BW of noise, then any performance features you expect? Anything? \$\endgroup\$ Commented Aug 30, 2019 at 2:00
  • \$\begingroup\$ I am a total beginner with PLLs, so no. PLL is for FMCW radar. Specifically, I am trying to understand a project I found online. Unsure of how designer made decisions like this one. In most sources I have found nebulous recommendations for choosing bandwidth like "increase loop bandwidth for more rapid loop response." \$\endgroup\$
    – Saunders
    Commented Aug 30, 2019 at 2:24
  • \$\begingroup\$ I understand you need to read more to ask a better question. Use the key words like this. scholar.google.ca/…. Then when U hit a paywall, search the title.. try again... researchgate.net/profile/Frank_Herzel/publication/… \$\endgroup\$ Commented Aug 30, 2019 at 4:56
  • \$\begingroup\$ to achieve a clean phase trajectory, you may need a higher-order loop \$\endgroup\$ Commented Aug 30, 2019 at 14:43

1 Answer 1


A PLL uses integration to eliminate steady state phase error in a stable input. Whereas with a ramp f input, there is a gain-dependant fixed phase-error ( as well as other variables)

Not having done FMCW RADAR before, may I refer to an excellent article.

It defines all the variables that affect BW , settling time and fixed phase error which is an indicator of Spectral Density noise BW that contributes (error and ) has both 1st and 2nd order sensitivity.


If the link does not work , search..

Phase Noise Modeling for Integrated PLLs in FMCW Radar

Frank Herzel, Arzu Ergintav, and Yaoming Sun


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.