rejection in low pass filter design

[buyan]

A Low Pass Filter Design

Hello,

For a LPF design, I designed it using lump components first and then transferred the lump components to microstrip lines. The simulation result is a bit different at the cutoff frequency.

Could anyone tell me the possible reasons? Thanks a lot.

rgds
buyan

 

[gecky]

Re: A Low Pass Filter Design

Hi,

what you see is normal and expected. This is due to the nature of Transmission Lines (TL). TL provide the admittance/reactance only at the centre freq that you design it at. But generally the performance of TL and lumped filters are similar in the passband.

Also note that TL filters are periodic, unlike lumped, since L=lamda @ f1 becomes L=2lamda @2*f1....

Hope this answers you initial doubt. Find out more in filter books like "Lancaster", the ebook is found somewhere is this forum.


Cheers!

 

[amicloud]

Re: A Low Pass Filter Design

Hello,

For a LPF design, I designed it using lump components first and then transferred the lump components to microstrip lines. The simulation result is a bit different at the cutoff frequency.

Could anyone tell me the possible reasons? Thanks a lot.

rgds
buyan

Maybe it doesnt take the discontinuity affection into account.you should use a
full-wave analyzer to simulate it.and the optimise the sizes to your desirable design.

 

[J1Lee]

Re: A Low Pass Filter Design

The reason is that you use approximations when transforming from lumped to microstrip elements. For closed approximations, the high impedance line (about 80 or 90 ohms) should be used for series inductor and low impedance line (about 20 or 30 ohms) should be used for shunt capacitors. For each shunt capacitor, you can also use a radial microstrip line or two open-circuited stubs instead of one.

 

[buyan]

Re: A Low Pass Filter Design

Hi,

what you see is normal and expected. This is due to the nature of Transmission Lines (TL). TL provide the admittance/reactance only at the centre freq that you design it at. But generally the performance of TL and lumped filters are similar in the passband.

Also note that TL filters are periodic, unlike lumped, since L=lamda @ f1 becomes L=2lamda @2*f1....

Hope this answers you initial doubt. Find out more in filter books like "Lancaster", the ebook is found somewhere is this forum.


Cheers!

Hi gecky,
Thank you so much.
BTW, what does the center frequency mean? I did the transformation from lump components to microstrip lines using the cutoff frequency of the low pass filter.

For the periodic of microstrip filter, does it mean the filter has the same response at f1, 2f1, 3f1....?

rgds
buyan