I need some help selecting the best variable coil for an FM narrowband receiver 455 kHz (quadrature detector) . The one I am using now has a Q of about 30 and BW of 15 kHz. Seems like a higher Q/narrower bandwidth would give me better sensitivity.

 

I suppose you can always use 455 KHz for the IF freq in an FM receiver. Are you proposing to wind your own coil? One suspects the characteristics of the entire IF strip have more to do with signal bandwidth than the detector coil, which only works with the signal applied to it.

 

Mossman, are you proposing to modify so called wideband 2-way radios (25 kHz) to narrowband (12.5 kHz) by reducing the receive bandwidth and transmit deviation? Some of the manufacturers of the radios offer kits to do that. The FCC says you are allowed to do that provided the modified radio is re-type accepted as narrowband. Once type accepted, all radios you modify are covered by the type acceptance.

 

I'm simply asking that if I use an adjustable coil with a sharper Q and more narrow bandwidth, can I achieve better sensitivity? The coil I am looking at has a bandwidth of 15 kHz and a Q of 30, which doesn't sound ideal for narrowband applications with deviations of 5 kHz or so.

 

I would think that the front end design had more to do with sensitivity than the demodulator. The better the signal to noise ratio in the front end, the more gain you can apply in the IF, and so you have more signal to demodulate.

 

Ok, but wouldn't a higher Q in the 2nd IF stage result in sharper transitions/deviations thereby improving the sensitivity of the quadrature detector/discriminator?

 

What about using a ceramic filter such as 734-455-DU found at http://www.oscilent.com/spec_pages/Filters/F-734.html

 

Think I answered my own question. Since an FM quadrature detector detects phase changes, you would want a useable amplitude across the entire band for reliable detection. In other words, if the received signal deviation was +/- 5 kHz, you would want the 3 dB bandwidth of the quadrature coil to be at least 10 kHz. Am I right?

 

Yes.

http://www.eetimes.com/design/autom...s-Pt-4-sensitivity-noise-front-end-amplifiers