multiple order passive RC filter

passive RC filter

hi,

How would I go on designing 2nd or third order passive filter(please no actives) using just RC and without LCs?
Would I just cascade the RCs?

Re: passive RC filter

Would I just cascade the RCs?

Click to expand...

**broken link removed**

Rgds,
IanP

Re: passive RC filter

Passive RC filters are restricted to real poles and thus offer a very limited choice of filter characteristics. None of the "classical"
2nd or higher order filters can be realized. What's your specification for the filter?

Re: passive RC filter

I have a 20MHz bandwidth signal. It could be centered in any frequency up to 100MHz.
ex) 0~20MHz, 20~40MHz, 40~60MHz, 60~100MHz.
Basically I will have a single(one and only one) 20MHz channel within in the 0~100MHz frequency range. I need to filter a signal at 4GHz. preferally the more the better. Definitely more than 20dB.

I also have a question in Phase shift for the in-band signal.
If I set my RC cutoff frequency to 200MHz, there is a good amount of phase shift happening in the in-band for a 80~100MHz signal. Would this phase shift not matter if we assume the phase is shifting linearly in that frequency range, since then the group delay is just constant? In the beginning I thought that I might have to set fc to 1GHz to avoid the inband phase shift, but this value of fc is too high!

passive RC filter

A Bessel filter would achieve linear phase in the passband. Unfortunately it requires complex poles, respectively a LC or active RC design.

You have to decide about acceptable phase deviations, if you are restricted to passive RC.