On Sun, 29 Nov 2009 19:08:16 -0500, "Michael A. Terrell"
wrote:

Looking at the data sheet of a modern AM/FM front end chip, the AM
section appears to be Hi-Z input:
http://www.atmel.com/dyn/resources/prod_documents/doc4913.pdf

Exactly! use an extra stage to hide a problem, rather than deal with
it properly.

I beg to differ somewhat. The problem with anything below about 7 MHz
is that atmospheric noise far exceeds the field strength of any
possible weak signal worth receiving. It makes no sense to have an
ultra-low-noise receiver front end on a lower frequency receiver.

However, note that this chip was also made for receiving short wave
DRM and HD Radio, which can be up to 30MHz. The level of complexity
necessary to build a tracking filter, with automagical impedance
matching is a bit much for consumer electronics. The result of using
a simple amplifier is lousy image rejection, overload problems,
sub-harmonic problems, intermod problems, and lousy NF which has an
effect on 7 Mhz reception.

An untuned amplified front end for impedance matching might seem like
an inferior solution, but it's certainly the cheapest, which drives
the consumer car radio market.

You know that a broadband, untuned input circuit has lower gain and
more noise that a tuned circuit designed to pass little more than a
couple channels at a time.

Current or voltage gain? In this case, it's current gain, which seems
quite high (35dB). It's prime purpose is to transform the rather high
antenna impedance down to something the chip can digest (50 ohms???).
Certainly an untuned front end has a worse NF than one that is tuned
and matched. However, the atmospheric noise at BCB frequencies will
swamp out any alleged improvement produced by the better sensitivity.

The older, tuned input design reduced the
design by at least one gain stage, along with the extra tracking
problems. Better designs used moving slug tuners with three or four
coils.

These daze, the slug tuners would cost more than the extra stages.
Transistors are one of the cheapest parts in the radio.

Lets remember that most people don't give a damn about listening
to AM radio anymore, and certainly not on long drives. They might
listen to the local news, or a ball game, but everything else is FM, Cd
or MP3.

See Ibiquity, DRM and HD Radio:
http://en.wikipedia.org/wiki/Digital_Radio_Mondiale
http://en.wikipedia.org/wiki/HD_Radio
http://www.ibiquity.com/hd_radio
BCB and SW stereo is kinda strange, but does work in strong signal
areas.

The best Am receivers I've owned were a solid stated ARN-6 DRF, and a
couple frequency selective voltmeters with a tuned antenna and preamp.

http://jproc.ca/rrp/rrp3/argus_arn6.html
My guess(tm) is that the reason they're "better" is not because of
improved sensitivity, but because of improved overload handling or
dynamic range. Much of the garbage and trashy sound is nothing more
than overload or intermod from strong nearby stations.

You've talked about designing two way radios, and I've worked with
telemetry designs on multiple bands, along with C-band & Ku band sat TV
equipment. No one in their right mind would use an untuned antenna
system and expect good performance. The noise floor, intermod and
co-channel interference would be horrible.

True. I suppose someone could conjure an proper BCB receiver. Most
of the better ham radio HF transceivers would certainly quality, such
as Elecraft K3:
http://www.w1vd.com/ElecraftK3.html
Well, maybe not. AM is certainly low on the priority list. Still,
the test numbers are far better than the typical AM/FM receiver.

However, even if the receiver were superior, the real problem is
hiding in the antenna system. One big improvement for BCB only would
be to install a big fat loading coil at the base of the antenna. It
wouldn't improve the received signal strength, but it would provide a
lower impedance to the coax and receiver front end, eliminating the
front end impedance matching stage.

The bottom line is always cost. The next incremental improvement in
performance is going to require substantial hardware in the receiver
front end. It's not a trivial change and will be expensive. I don't
think that Joe Sixpack will pay the price.


--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558