Michael
Amen brother but you have to give Jon credit for his pursuit as he will
learn much and he'll be able to tell us why Marantz or McIntosh didn't make
a High Quality TRF tuner. Or maybe we'll all be drooling over the Noring
High Quality TRF tuner. Or maybe he'll morf from Noring to Notrf!
--
73
Hank WD5JFR

"Michael Black" wrote in message
om...
Jon Noring wrote in message
. ..
In Patrick Turner's latest message replying to the thread I started on
building a high-performance AM tube tuner, he stated his skepticism
that a pure TRF circuit will, in a practical sense, meet the specs I'm
looking for (to meet the specs will require an impractical number of
RF amplifier stages, such as six or more.)

Within the design constraints he selected, I cannot disagree with his
conclusion.

However, one of the constraints he made, and that most make, is that
tuning is to be done by an infinitely variable multigang air
capacitor.

I wonder if the same conclusion would be reached if we approach this
from a single channel perspective? That is, what if we fix the
frequency we want to receive (e.g., 830khz, or 1420khz), and then for
each stage optimize the parallel RLC circuit (or use some other tuned
circuit with the right resonance and bandwidth response -- crystals?)

BUt you need to realize that receiver design is about looking at
the overall picture, not some fine point here or there.

There are tradeoffs when using a superheterodyne, but not using
it has to be the rare exception. The tradeoffs have nothing to
do with issues of selectivity.

You have somehow become fixated on TRF receivers, as if it is
the grail to solve some problem that you haven't really defined
yet.

But once you start building good filters, then what's the point of
not using a superhet, and putting that filter at a fixed frequency,
where selectivity will be constant? You're not going to get
"higher fidelity" by moving the filter to the front end. You haven't
come up with reasons why a superhet is not suitable. You're not
really talking "high performance" as many people would think of it,
you are talking "high fidelity". For that purpose, and until you
come up with reasons of image rejection or front end overload, you
won't even notice that you are using a superhet rather than a TRF.

The issue isn't where the filter lies, the issue is getting your
"high performance filter". Realistically, that can be accomplished
far more easily at a fixed IF than by putting it at the front end.
What you need to be concerned about is proper care in designing
that filter. Design it for wide bandwidth but with good skirt
selectivity, and that's all that matters to your end game.

Michael