The very same series resonant circuit connected
between the line and ground would provide a "band
stop" or notch filter. The above schematic does not
show points A, B, C, and D. If you break the line
between SW1A and L1 and call the left side "A" and
the right side "B" then break the line between C1 and
the Atten pot and call the left side "C" and the right
side "D," you can then connect the mode switch, at
left. Switch "up" is bandpass filtering, switch "down"
gives band stop (notch).
This arrangement works fine, but it limits you to either
the bandpass or band stop functions.  







This next schematic shows two additional
improvements:
1. The bandpass and notch functions are available
full time. However, to do this, they had to be
electrically isolated. Instead of using active isolation (
a unity gain transistor stage) I elected to use
interstage transformers (Micro Circuits or Coil Craft).
Note that the L and C components have been
graphically simplified for ease of explaining this circuit.
In reality they must be just the same as those in the
top schematic.
2. In order to increase the selectivity of both filters,
B/P and notch, I decided to reduce the circuit
impedance by a factor of 8:1, or 6 Ohms.
The results are as expected with an effective
selectivity of +/- 5 KHz, or a total skirt width of 10 KHz.
This is not as sharp as a crystal or mechanical IF
filter, but very effective on the crowded bands...even
very usable on AM broadcast!
Bottom line: Since the bad stuff is attenuated and the
good stuff is passed at unity levels, you get the
benefit of having a circuit with gain, even though this
is a passive device.
Although this is the second version of an on-going R&D project, it still represents the basic design principles.
The circuit is fairly straight forward and really shouldn't need any explaination:
But, here's the basics:
1. Cap in series with your antenna makes ir electrically shorter.
2. Inductor in series with your antenna makes it electrically longer.
3. Cap and coil in series equals zero impedance at resonance ( C x L = a passband filter)
Subsequent improvements and fine tuning upgrades have been added to the circuits below.
This schematic speaks for itself. it's the next phase of
this evolution.This circuit allows two bandpass filters
to be superimposed or slightly offset from one
another. Now you will have adjustable selectivity level
along with adjustable bandwidth. Of course, you'll
have to incorporate all of the component switching
and attenuators as in the top schematic. Then, when
you are done with that, add the notching
feature...viola.  



I don't have the space here to describe how to use a
bandpass and notch filter...it's pretty much instinctive:
peak what you want, notch the undesirable stuff. Note
that if you tune both filters to the same frequency, you
create a very large attenuator...that's a good way to
calibrate the circuit if you are so inclined. You really
have to "play" with the controls for a while until you
get a "feel" for the operation. But you'll wonder how
you ever got along wothout a preselector...once
you've used one!  

The evolution of preselector technology:
The two schematics above make use of lower
"internal" impedances, within the circuit. Here is a
response plot of the filter tuned for the high-end of
the BC band with different source and load
impedances. Note the increase in "Q" between 50, 12,
and 6 Ohms as taped from the interstage
transformers. From a practical standpoint, the 6 or 12
Ohm circuit is about a tight as it's going to get without
going to active stages (and their associated
problems). The next, and probably last phase of this
project will be adding a series resistor in the form of a
potentiometer which will allow "Bandwidth" control
which is represented by the 6 Ohm and 50 Ohm limits
of the circuit.
Preselector Design and Theory
Next  Back  Home
Counter