ceramic filter cavity
Lupin said:
Can anyone tell me what factors in a cavity filter can determine a spurious signal?
Thanks in advance.
Lupin
Read filter design book, possibly find couple of them here in edaboard.
And old books but still usable is Zverevs 'Handbook of filter syntesis' and L.Matthaei, Lee Young and E.M.T jones 'Microwave filters impedance-matching networks and coupling structures' (Artech house)
This both books written in 'handbook design era' as 60-70s without computer support and programs and many (all??) filter books written in modern time have this books as reference...
Matthaeis book is possible to find as pdf in P2P grey area on Internet, but Zverevs is not scanned and PDF:ed yet (needs 1200 dpi to make usable quality of charts... is _very_ high detalied - and very nice typgraphic settings and quality is miles higher compare to days ugly Word document) - and very, very expensive to buy...
but...
Interdigital and digicom bandpass filter have more or less spurios on 3f,5f,7f of f passband
and also possible make unwanted different wave modes inside filter cavity for other spurr frequency depend of mecanical design.
If you make 'mecanical' cavity filters - you need good mecanical workshop and try, try, try couple of designs - is not possible to simulate everything and need measure with NVA on real things
for example hard to simulate is Q-value and loss depend of rod-shape in mounting point and contact pressure and resistance in sealings, loss in end-cap and tuning cap (difference between brass and silver-plated tuning screw in endcap can give 0.3 dB difference in transmission loss...) etc. If high power filters - passive IM, voltage breakdown or plasma flashing inside, different temperature on different place in high power mode can change frequency and tunings (going from low to higer transmissions loss and more heat... and bad standing wave) etc.
(for example in 100 MHz combining filter with 40 degree rod, can have 14 ampere current (peak) and 400 Volt (peak) in top capacitance only for 3 Watt power... - think 2 kW power....)
common problem is not possible to simulate is leaking RF in sealings between cavity etc. - is practical spent very much time to hunting leaking RF in wrong place (more screws) and hard to make wanted leaking on other places (zeros), build good tuning point in right place in mechanic for high yeld manufacturing, easy tuning process, temperature stability with exotic material (invar) etc.
Simulate to find right length of rod and dimension of cavity and find right coupling factor between cavity is smallest and most easy part of filter design, tricky part is make working filter in real world, easy tuning and stable over time and temperature, and high yeld manufacture to low price.