Microwave lowpass filter help

[ife101]

Hello,

I have been investigating both chebyshev and butterworth lowpass filter s11 and s21 parameters in ADS using different types of filters. The responses are generally siilar but I was unclear as to the main differencess between these responses in:

i. lumped element
ii. transmission line (varying widths and lengths)
iii. microstrip lines

so far my understanding is that microstrip lines will have less losses than lumped ciecuit elements but I'm not really sure why.

The cut off frequency is 4GHz (at least -20dB attenuation at 6GHz).
Any help would be appreciated!

(I am new to microwaves and this forum:lol

 

[Phytech]

For the same design, the loss is related to resonator unloaded quality factor.

 

[ife101]

thanks for the help. This is the first I've heard about quality factor. From the bit of research i've done i will have more than 1 q factor (both filters are 5th order).

The noticeable differences between the butterworth nd chebyshev is that there are spikes in the return loss (s11) which i understand is from the poles in the filter. i'm not sure why a chebyshev of the same order would have poles.

For each filter type (butterworth and chebyshev) - lumped, transmission lines, microstrip circuit implementation seem to have decreasing return loss in that order.

i've looked everywhere for what could be causing this,,,:-?

 

[Phytech]

thanks for the help. This is the first I've heard about quality factor. From the bit of research i've done i will have more than 1 q factor (both filters are 5th order).

The noticeable differences between the butterworth nd chebyshev is that there are spikes in the return loss (s11) which i understand is from the poles in the filter. i'm not sure why a chebyshev of the same order would have poles.

For each filter type (butterworth and chebyshev) - lumped, transmission lines, microstrip circuit implementation seem to have decreasing return loss in that order.

i've looked everywhere for what could be causing this,,,:-?

Chebyshev filter has ripple (equal ripple) in the passband, therefore, you see the spikes in the return loss. While butterworth (also called maximum flat) has maximum flatness in the center (for passband) or zero (for lowpass). So you see maximum return loss in the center or at zero.

It depends on the design not on the type of filter. Lumped can have worse return loss than microstrip or transmission line.

 

[ife101]

Thanks alot! This definately makes sense now. Really appreciate it.:grin: