[TenTec] Explanations of the Different Types of Filtering on Radios

[Rick - DJ0IP / NJ0IP]

I will take a stab at explaining IF Shift.  Then contrast it to passband
tuning.

For my analogy, I will compare the IF filter to a pair of sliding glass
doors, like you might find between a living room and the outdoor patio.  To
open the door, one door slides to the left, the other door slides to the
right.

Now let's assume one foot of opening between the doors is equal to one KHz
bandpass of a filter.
So 2.4 ft. of opening between the doors would equal 2.4 kHz of bandpass for
a our filter.

Normally we open both doors simultaneously by pushing them apart, so we
normally would open each door by 1.2 feet to get a 2.4 ft. opening, centered
on the center of the door.

Our filter is 2.4 kHz centered on the desired frequency we want to receive.

Since both of our doors slide, it is possible to slide them both at the same
time, at the same pace, either left or right of the center of the door,
while maintaining our 2.4 ft. opening.

Now let's say we have two guys standing side by side, shoulder to shoulder,
both wanting to come in through the door at the same time.  With our door
open just 2.4 ft., they will not be able to both get through the door at the
same time.

If we slide both doors 1.2 ft. to the left, the opening is still 2.4 ft.,
and the guy on the left can pass through, but the guy on the right cannot.
If we slide both doors to the right by 1.2 ft., the guy on the right can
pass through, but the guy on the left cannot.

IF Shift enables us to move the center of the bandpass up or down, and thus
can help to let one signal in, while rejecting the other.  Of course this
only works if the two signals are not on the same frequency.  It also
results in a bit of changing of the tone of our desired signal.

With IF Shift, the bandwidth of the filter always remains the same (in the
example above, 2.4 kHz).

NOW LET'S LOOK AT PASSBAND TUNING.

For this example we're going to change the assumption.
Let's assume our sliding doors are open 4 ft. this time.

The same two guys are standing shoulder to shoulder and can both pass
through the door at the same time.
However we only want to let one guy in at a time.
This time we will slide only one of the two doors.
If we slide the door on the right back towards the center a ft. or so, the
guy on the right cannot pass through the door, but the guy on the left still
can.  And vice versa.
That is passband tuning.

The technology:

For IF Shift, all we have to do is vary the frequency of the local
oscillator up or down and we shift the center frequency of the bandpass up
or down.

For Bandpass Tuning, we must have two IF filters (usually of the same
bandwidth),  AND  they must be on DIFFERENT frequencies.  Let's place one on
9 MHz, and the other on 6 MHz.  Our signal will pass through both filters.
To do that we need another local oscillator and another mixer to convert
from one IF to the next. When both filters are centered on our desired
passband frequency and both are 2.4 kHz, then it will work (and sound) just
like a signal passing through just one filter, BUT we will have a higher
ultimate attenuation of signals off the sides of the bandpass.

To operate the BPT, we apply normal IF Shift to just one of the filters.
This slides the pass frequency of the filter being shifted across the pass
frequency of the filter that is not being changed.  This results in a
reduction of the overall bandwidth of the radio, by reducing it from one
side.  It moves just one flank of the filters, and it can be either the
upper or the lower flank of the filter that is moved.  

If we only have IF Shift and shift the bandpass a little, we can reduce the
signal strength of a neighboring signal, but the audio tone of the signal we
want to received will change a bit. However if we use BPT and adjust just
one flank of just one of the filtes, then the neighboring signal is also
reduced in strength, but the tone of our desired signal does not change.
This is the advantage of BPT over IF Shift.  (along with increased ultimate
attenuation)

Obviously BPT is superior to IF Shift, but at a substantial increase in
cost. It requires an additional filter, additional Local Oscillator, and
additional Mixer.  As a side product of deploying BPT, (where the two
filters are on different IF [frequencies], we get about 20dB better ultimate
attenuation than we would get if we were to place two identical filters
(same IF) in series.

I think we should leave it at that.
If we continue and try to explain Hi-Cut and Lo-Cut, we will probably
overwhelm the new guy.

73
Rick, DJ0IP


-----Original Message-----
From: TenTec [mailto:tentec-bounces at contesting.com] On Behalf Of Will Jones
Sent: Friday, August 30, 2013 4:34 PM
To: tentec at contesting.com
Subject: [TenTec] Explanations of the Different Types of Filtering on Radios

There are a few of us who meet together every Wed. morning at a local
restaurant to talk about ham radio and stuff.
Some are new hams. One asked this week a question which I think would be an
interesting study in comments.
He asked, "I have an I.F. shift on my radio and I have read about what it
does but don't really understand "what" it does. Also how does it compare to
the other features I see on radios such as PBT, notch filters, RIT, DSP
noise reduction and noise blankers. Without having a radio with these
features on them it is hard to explain just what your signal will sound like
when these devices are used.
So can anyone give us some ideas how to explain these different types and
why we would want one and not another and why radios don't have them all on
one radio. Bill, N4LB in TN _______________________________________________
TenTec mailing list
TenTec at contesting.com
http://lists.contesting.com/mailman/listinfo/tentec