wrote:
http://www.kongsfjord.no/dl/dl.htm
Dallas Lankford has done some serious research on the cause and cure
of/for
the distortion caussed by ionospheric "hops".
All of his PDFs are informative, but his "Elliptic Low Pass Audio
Filters" series
are must reads. His conclussions in his "Elliptic Low Pass Audio
Filters (Amplified) -
Simplified and improved", 29-MAY-2006, are downright fascinating. I
have been
playing with his design and a couple of steep cut off 3600Hz filters. I
don't
have the tools to duplicate his research. I knew from the "get go" that
my
remote/weak signal source with which I use to test detector, and to a
lesser
degree, antenna/feedline combination suffered from the major weakness
that
there was no multipath effects. All of my experiments were local ground
wave
and I couldn't, and still can't, "messure" the effects of such fading.

I would love to have a Drake R8B, but I was forced to deal with the
receivers I do
have. I used R390, R392, R2000(modified),
R2000(stock),(borrowed)AOR7030,
and a DX398 for some simple tests. I am just out of the ground wave
for several
MW stations and around dawn and dusk I get serious and nasty fading. So
for
the last few weeks I have been comparing stock, ie non-synchronous,
detectors
with synchronous detectors, and the addition of a brick wall 3.5KHz low
pass fitler.

As Mr. Lankford concludes a synch detector is only (and that might
ought to be
"may") be slightly better then a AF LP good filter. It is rather
frustrating to have
spent the last 18 months building an outboard synchronous detector to
find that
a simple LP filter offered so much improvement. Don't get me wrong, a
synch
detector is a usefull addition, but not the end all I had hoped.

I will disagree with his use of a simple bipolar 2W AF amp. The one
thing I have
descerned is that after AF detection, any additional distortion rapidly
degrades
intelligibility. I found MOSFETs, and vacuum tubes, amps allowed me to
understand
signals better then 6dB down from a "good" bipolar AF amp.

"My" third R2000 was siezed by my wife. I had added a MOSFET amp,
redesigned
the treble cut to a tone-tilt control and just completed adding
switchable 3KHz/4KHz
Filters as designed by Mr. Lankford. My wife has been doing some casual
SWL for the
last few days and agrees the filters are very good for nasty band
conditions.

Email is abandoned and dead.
Terry

It is an interesting idea, but nobody builds LCR filters. Rather, you
use the LCR filter as a prototype, then build a leapfrog active filter
from signal flow graphs based on the physical LCR filter.

The problem with elliptic filters with sharp cutoffs is they ring. I
think that would be more annoying than the flutter it is trying to
remove.

On 26 Jul 2006 23:00:58 -0700, wrote:



It is an interesting idea, but nobody builds LCR filters. Rather, you
use the LCR filter as a prototype, then build a leapfrog active filter
from signal flow graphs based on the physical LCR filter.


Really? What kind of filters does the Drake R8 series use?

David wrote:
On 26 Jul 2006 23:00:58 -0700, wrote:



It is an interesting idea, but nobody builds LCR filters. Rather, you
use the LCR filter as a prototype, then build a leapfrog active filter
from signal flow graphs based on the physical LCR filter.


Really? What kind of filters does the Drake R8 series use?

As a demod filter? I would image a low order active filter to clean
things up. Remember, this is the audio band, not RF. I've seen some
write ups on 455khz IFs being done with active filters.

Kiwa sells an active filter for 455Khz
http://www.kiwa.com/kiwa455.html

Note the AR7030 has "tone controls", so certainly it has an active
filter past the demod. The problem with building LCR filters in the
audio band is they are bulky, not to mention often inaccurate. With
active filters, you have more flexibility over component values.

wrote:

As a demod filter? I would image a low order active filter to clean
things up. Remember, this is the audio band, not RF. I've seen some
write ups on 455khz IFs being done with active filters.

Kiwa sells an active filter for 455Khz
http://www.kiwa.com/kiwa455.html

Note the AR7030 has "tone controls", so certainly it has an active
filter past the demod. The problem with building LCR filters in the
audio band is they are bulky, not to mention often inaccurate. With
active filters, you have more flexibility over component values.

Drake uses an LC filter in the IF. They "Get away" with it becuase of
the
lower IF they use. R390s, original not the R390A, and the R392 use
several staged of LC filters and have excellent skirts.

The Kiwa filter you refference is not a "active filter", but a ceramic
filter
with amplification. To me active filter means opamp or norton amp
with feedback to control pass/reject charactoristics. The premium Kiwa
unit is nearly as good as a crystal or mechanical filter and MUCH
easier
to connect. I installed one in a friends R2000 and was impressed by
the quality and how well it worked.

A big advantage of passive LC filters is they are much less "fussy"
then active filters. I like not having to mess with power and proper
bypassing.
And if you are willing to wind your own torroids, it is pretty easy to
get the
L very close to what you want. The C can be built with standard value
caps in parallel.


The Tone-Tilt filter I used in all 3 of our R2000s is active because it
would be
VERY difficult to use LC filters effectively.

Terry

wrote:
wrote:

As a demod filter? I would image a low order active filter to clean
things up. Remember, this is the audio band, not RF. I've seen some
write ups on 455khz IFs being done with active filters.

Kiwa sells an active filter for 455Khz
http://www.kiwa.com/kiwa455.html

Note the AR7030 has "tone controls", so certainly it has an active
filter past the demod. The problem with building LCR filters in the
audio band is they are bulky, not to mention often inaccurate. With
active filters, you have more flexibility over component values.

Drake uses an LC filter in the IF. They "Get away" with it becuase of
the
lower IF they use. R390s, original not the R390A, and the R392 use
several staged of LC filters and have excellent skirts.

Except we are not talking about IF filters .The "fading" filter is at
the end of the chain, i.e. past the demod.


The Kiwa filter you refference is not a "active filter", but a ceramic
filter
with amplification. To me active filter means opamp or norton amp
with feedback to control pass/reject charactoristics. The premium Kiwa
unit is nearly as good as a crystal or mechanical filter and MUCH
easier
to connect. I installed one in a friends R2000 and was impressed by
the quality and how well it worked.

A big advantage of passive LC filters is they are much less "fussy"
then active filters. I like not having to mess with power and proper
bypassing.
And if you are willing to wind your own torroids, it is pretty easy to
get the
L very close to what you want. The C can be built with standard value
caps in parallel.

Except this is at audio frequencies, where the component sizes are much
larger. Again, this is not at IF frequencies.




The Tone-Tilt filter I used in all 3 of our R2000s is active because it
would be
VERY difficult to use LC filters effectively.

Terry

wrote:
snip
Except we are not talking about IF filters .The "fading" filter is at
the end of the chain, i.e. past the demod.
snip
Except this is at audio frequencies, where the component sizes are much
larger. Again, this is not at IF frequencies.

Mr. Lankford's main concept is that by using a narrow enough IF filter,
a narrow filter with with a step attenuation skirt, and by offset
tunning to
only get the carrier and the desired sideband, and with a following
suitable AF LP fitler can do wonders. It is not magic, and doesn't work

with every receiver and under every condition. In the ret of this, and
all
future posts, I will simply call it "ELPAF".

I can say is that it is a usefull technique, and will even help when
used
premium receiver like an AOR7030 or R390, the filter can really reduce
the effects of "fading". With a modest receiver like the R2000 that has

been upgraded with a suitablely narrow IF filter, the results are
impressive.
With a "marginal" receiver like the ATS909/DX398 the results are
nothing
short of amazing.

Terry

On 30 Jul 2006 09:06:34 -0700, wrote:

wrote:
snip
Except we are not talking about IF filters .The "fading" filter is at
the end of the chain, i.e. past the demod.
snip
Except this is at audio frequencies, where the component sizes are much
larger. Again, this is not at IF frequencies.

Mr. Lankford's main concept is that by using a narrow enough IF filter,
a narrow filter with with a step attenuation skirt, and by offset
tunning to
only get the carrier and the desired sideband, and with a following
suitable AF LP fitler can do wonders. It is not magic, and doesn't work

with every receiver and under every condition. In the ret of this, and
all
future posts, I will simply call it "ELPAF".

I can say is that it is a usefull technique, and will even help when
used
premium receiver like an AOR7030 or R390, the filter can really reduce
the effects of "fading". With a modest receiver like the R2000 that has

been upgraded with a suitablely narrow IF filter, the results are
impressive.
With a "marginal" receiver like the ATS909/DX398 the results are
nothing
short of amazing.

Terry

I used a similar technique with my R-390A. I'd tune one side of the
carrier and use the BFO to provide a local one. You have to back way
off on the RF gain as the 390A BFO is sort of weak.

wrote:

wrote:
snip
Except we are not talking about IF filters .The "fading" filter is at
the end of the chain, i.e. past the demod.
snip
Except this is at audio frequencies, where the component sizes are much
larger. Again, this is not at IF frequencies.

Mr. Lankford's main concept is that by using a narrow enough IF filter,
a narrow filter with with a step attenuation skirt, and by offset
tunning to
only get the carrier and the desired sideband, and with a following
suitable AF LP fitler can do wonders. It is not magic, and doesn't work

with every receiver and under every condition. In the ret of this, and
all
future posts, I will simply call it "ELPAF".

I can say is that it is a usefull technique, and will even help when
used
premium receiver like an AOR7030 or R390, the filter can really reduce
the effects of "fading". With a modest receiver like the R2000 that has

been upgraded with a suitablely narrow IF filter, the results are
impressive.
With a "marginal" receiver like the ATS909/DX398 the results are
nothing
short of amazing.

Almost sounds like the 'syncho-phase' detector on the R7/A. Use of a narrower
filter and tilting the passband control to either USB/LSB.

dxAce
Michigan
USA

In article .com,
wrote:

wrote: snip
Except we are not talking about IF filters .The "fading" filter is
at the end of the chain, i.e. past the demod.

snip

Except this is at audio frequencies, where the component sizes are
much larger. Again, this is not at IF frequencies.

Mr. Lankford's main concept is that by using a narrow enough IF
filter, a narrow filter with with a step attenuation skirt, and by
offset tunning to only get the carrier and the desired sideband, and
with a following suitable AF LP fitler can do wonders. It is not
magic, and doesn't work with every receiver and under every
condition. In the ret of this, and all future posts, I will simply
call it "ELPAF".

I can say is that it is a usefull technique, and will even help when
used premium receiver like an AOR7030 or R390, the filter can really
reduce the effects of "fading". With a modest receiver like the R2000
that has

been upgraded with a suitablely narrow IF filter, the results are
impressive. With a "marginal" receiver like the ATS909/DX398 the
results are nothing short of amazing.

Any time you narrow the frequency range with a filter you lower the
noise floor possibly making the faded carrier large enough so the
detector does not greatly distort the audio.

--
Telamon
Ventura, California

wrote:
wrote:
snip
Except we are not talking about IF filters .The "fading" filter is at
the end of the chain, i.e. past the demod.
snip
Except this is at audio frequencies, where the component sizes are much
larger. Again, this is not at IF frequencies.

Mr. Lankford's main concept is that by using a narrow enough IF filter,
a narrow filter with with a step attenuation skirt, and by offset
tunning to
only get the carrier and the desired sideband, and with a following
suitable AF LP fitler can do wonders. It is not magic, and doesn't work

with every receiver and under every condition. In the ret of this, and
all
future posts, I will simply call it "ELPAF".

My point still holds in that nobody serious builds LCR filters for
auido (speaker crossovers exempted). His LCR filter IS in the audio
chain, not the IF.




I can say is that it is a usefull technique, and will even help when
used
premium receiver like an AOR7030 or R390, the filter can really reduce
the effects of "fading". With a modest receiver like the R2000 that has

been upgraded with a suitablely narrow IF filter, the results are
impressive.
With a "marginal" receiver like the ATS909/DX398 the results are
nothing
short of amazing.

Terry