Surface Mount RF Filter Guidelines?

I've used the "solver" optimizer in MS Excel to do similar "engineering." I suppose with Mason's rule you could get the transfer function and enter it into a spreadsheet.

I made an arithmetically symmetric bandpass using the MS optimizer, but that was just moving the poles and zeros around.

That's a great overview. I'm not sure how you'd find it if you didn't know it was there! I googled the companion AN mentioned in that PDF and didn't find it until I just used the link in the document.

--
Ben Jackson AD7GD

http://www.ben.com/

Hi all,

I was looking with a practical eye to the figure7 at page 5 from the previous application note. And what did I saw? Three graphs showing spurious images below -70dB (the first and the second filters) near the badwith of the signal under interest or -60dB images far away from the bandwith. In practical life for such a filter (110Mhz the fundamental and up to

330MHz the third harmonic), does not count too much if you're using a, b or c filter topology. More that that, in real life you haven't such big room for filter components as suggested on drawings. And all topologies have important spurios signal closely to the fundamental. I've play a lot with AD DDS and unfortunately they can't gave a true clean output signal only with too much struggle, not talking a bout the ugly sinx/x anvelope which require a VGA for correction.

greetings, Vasile

Short of designing a filter using 20 mil MgO substrate with TBCCO deposited on both sides and cooled to 77 K. The filter's 0402 parts listed on the BOM don't take that much board space. Yes the spurs are close to the fundamental signal that's why they used an elliptical topology.