When I was deriving the transfer function of a differential active loop filter provided by ADIsimPLL, I had some problems. The circuit diagram is shown below. The differential active loop filter that I calculated

First, for a differential active loop filter, is it right that calculating the transfer function only requires half of calculation of the circuit? And could please provide the result H(s)=Vo(s)/Vi(s). Second, the phase-frequency detectors(PFD) I use output voltage signals which is different from those that output current signals. So I should calculate voltage to voltage transfer function instead of current to voltage transfer function of the loop filter. Did I understand correctly? The result of my calculation is H(s)=Vo(s)/Vi(s)=-(1+SR2C2)/(SR1C2*(1+SR3C3)).

Looking forward to your answer.

  • \$\begingroup\$ It depends on the state of the NU and ND which is a switch with a pullup . Hunting high s gain depends on R Ratios while tracking gain integrates with RC ratios with 200R source. \$\endgroup\$ Dec 2, 2021 at 4:35
  • \$\begingroup\$ Thank you! I am still confused about how to calculate the transfer function of the active loop filter in the picture. Could you please explain it in more detail? \$\endgroup\$
    – shang xu
    Dec 2, 2021 at 6:43
  • 1
    \$\begingroup\$ They are the same. The noninverting input impedance matches the inverting input's. Or simulate it. Can you spot a difference (hint: the traces completely overlap)? \$\endgroup\$ Dec 2, 2021 at 8:00
  • \$\begingroup\$ Thanks a lot! This approach of yours provides me with a good way of thinking. There is one thing I still can't understand. How to calculate the transfer function of the loop filter? \$\endgroup\$
    – shang xu
    Dec 2, 2021 at 8:51
  • \$\begingroup\$ They are not voltage sources, rather open collector when off and current sink when on. When not locked, the PD differential output is a 2Vpp sawtooth from 3 to 5V of the difference frequency. But when locked it will pulse up and down in alternate cycles. So take the Thevenin equivalent of RCR as the source impedance, then the differential gain is Zfb/Zs then add LPF, so it is a 3rd order equation with 3 caps. \$\endgroup\$ Dec 2, 2021 at 14:54


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.