Very nice and educative. I have seen commercial LPF quoting 0.1dB insertion loss for 5 pole or 7 pole filters, is this figure achievable or it's a false claim?
Interesting how the attenuation turns up, I've seen something similar and wondered why, your explanation seems reasonable, I wonder if other explanations will surface.
the transmission zero (notch) ALWAYS occurs because of some cross coupling between two resonators or between the input and output. So it is either a magnetic coupling bween the two toroids that is causing it or most likely your board layout is causing a (weak) coupling between the input and output.
@@levelupeelab I think in this filter it is more likely due to a weak coupling between the toroidal inductors rather than between the input and output because in the layout nothing stands out to me as being the culprit. You can put a metal shield around the coils or something to separate them and see if that eliminates the notch. Simulation should be able to show that too
Very nice and educative. I have seen commercial LPF quoting 0.1dB insertion loss for 5 pole or 7 pole filters, is this figure achievable or it's a false claim?
Thanks for the update! I never tried ultraspherical filters before. Should give it a try in one of my future projects! 🙏
Interesting how the attenuation turns up, I've seen something similar and wondered why, your explanation seems reasonable, I wonder if other explanations will surface.
the transmission zero (notch) ALWAYS occurs because of some cross coupling between two resonators or between the input and output. So it is either a magnetic coupling bween the two toroids that is causing it or most likely your board layout is causing a (weak) coupling between the input and output.
Interesting. I didn't apply any thought to coupling when I laid out the board.
@@levelupeelab I think in this filter it is more likely due to a weak coupling between the toroidal inductors rather than between the input and output because in the layout nothing stands out to me as being the culprit. You can put a metal shield around the coils or something to separate them and see if that eliminates the notch. Simulation should be able to show that too
S21, and s11 and s22 should be measured too in filters
Interesting.