1
$\begingroup$

I want to build and simulate a 6 GHz, 4th order lumped-element Chebyshev bandpass filter with a maximum passband ripple of 0.1 dB and a bandwidth of 1 GHz, and as I understand it, there should be 3 lobes in the S11 response, but instead my circuit has a huge drop at the center frequency. I'm not certain what could be causing this or if this is even common, but I'm assuming it has to do with my implementation. The equations I used to get the component values from the normalized lowpass filter coefficients are as follows and the schematic from MWO is also attached;

enter image description here

enter image description here

The elements in the circuit are (from left to right); [C1 = 3.53078138643953e-06 uF, L1 = 0.199281477338676 nH, L2 = 10.3941804502207 nH, C2 = 6.76935843301265e-08 uF, C3 = 5.63670104271011e-06 uF, L3 = 0.124828215212789 nH, L4 = 6.51082585059285 nH, C4 = 1.08069137002873e-07 uF].If anyone can see where I'm going wrong, and would be kind enough to point it out to me that would be greatly appreciated.

$\endgroup$
1
  • $\begingroup$ Is the figure what you get from your simulations? In that case everything should be okay because $S_{11}$ is the reflection coefficient, which should be small inside the passband, and large outside the passband. $\endgroup$ – Matt L. Nov 13 '20 at 12:09
0
$\begingroup$

You are confused about which parameters are what. The $S_{11}$ parameter is the power that's reflected back from the network; the $S_{21}$ parameter is the desirable bit that's passed forward.

Just by conservation of energy, if the filter is lossless then $\left | S_{11} \right | = 1 - \left | S_{21} \right |$. So that horrible deep null in $S_{11}$ actually just means that $S_{21}$ is at unity, which is desired.

$\endgroup$

Your Answer

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

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