Several weeks ago I asked if capacitors could make that much difference

in intelibility and that such a simple question would move me to
complete,
then redesign a project from over 20 years ago.

I found that on weak and noisy signals capacitors can make a world of
difference. A little scrounging has given me a Pioneer KEHP9500 MOSFET
automitive stereo unit, with very low distortion when operated below
10W
and a set of Minmus 7 speakers. And 10W in 2 speakers is painfully
loud.

I brought the audio straight out from my R2000 and found that by
bypassing
all of the audio chain the audio was much clearer. 2nd best was to use
the
Rec out. The difference here is slight and I wouldn't swear in court
that
I am positive.

I then compared the detector direct signal with different types of
capacitors. And Craig at Kiwa is mostly correct. My only disagreement
is with the Panasonice bypolar electrolytics. I douldn't hear any
diffrence bewteen
them and "good" single ended caps. Perhaps my samples were bad. And all
electrolytics sound worse then mylar and related caps.

I then compared the best direct signal with the improved detector.
Again a hands down victory for better technology. I did a quick
comparison between
2 different improved detectors.
http://www.amwindow.org/tech/htm/lowdisdet.htm
is a simpler single IC unit, and the unit shown at
http://www.amwindow.org/tech/htm/alowdisdet.htm.
While the simple unit is much better then the stock diode, the 2nd,
more complex unit is much better then the simple unit.

I then compared the
http://www.amwindow.org/tech/htm/alowdisdet.htm unit with the AD607
synch detector. The synch detector is the clear winner.

The first comparisons showed the advantage under weak/noise conditions.

The sycnh detector is better under ALL conditions. My wife plays the
flute
and we listened to a folk music broadcast from SA and my wife said is
it was the first time a flute really sounded like a flute on the radio.

The "Vinyl Cafe'" on the CBC (13.655 10EDT) today sounded great!

Many radios would sound much better with some attention to the audio
chain.
I used to think that a "little" distortion didn't matter. After all the

signals get pretty mangled by the ionosphere. And I was wrong. It makes
a
great difference. I have spent the day listening to HF and going back
and
forth just to get a feel for it.

I experimented with better, lower V-forward, and tried biasing the
diode.
While these steps can make very minor improvements, they are hardly
worth
the minimal effort.

Most audio sections are now built from ICs. Most ICs have nasty notch
or
"cross over" distorion. If I was crazy enough to decide to rebuild the
complete audio chain I would start with a better output stage, a
discrete,
MOSFET design. I would then rework the stages to eliminate all the
electrolytics in the signal path. I would then add an AD607 synch
detector
and redesign the BFO.

My R2000s are like new receivers. I can uderstand signals that were
lost
in the noise before. For the time being I am going to live with an
outboard detector BFO combo. Kind of like the Kiwa MAP, or the Sherwood
SE-3 MK III. I
say "kind of" because I doubt that my unit approaches the quality of
either
of those units. I am still working out the details and so far have
~10KHz, 4.8KHz, 3KHz 2.1KHz and 500Hz filters with the last 2 being
crystal fitlers.
A word of advice to anyone deciding to add filters, forget the older
Collins mechanical filters. Mine at least "rings like a bell".

I am still working to intergrate the BFO, "tone tilt" filter switching
and trying to come up with a decent project case. But the effort has
been well worth the effort.

While I would love to buy a really "good" receiver like the AOR7030+,
or
one of the better Drakes, or even an NRD, money being rather tight, I
am
happy to get any improvements I can.


Terry






I finished the AD607 synhcronous detector and have
been testing it and comparing it to the stock diode
detector and to an "improved" diode detector.

Excellent report Terry! We need more info like this instead of the
slack-jawed political and religious bull**** from the mutant gallery!

Frank

I have been doing some more research on "tone tilt" and found a
refference
that states that 75uS is the standard pre/de-emphasis for MW, and by
context HF. I thought 75uS was teh standard for non dolby FM broadcasts

in the US, and ~?50uS was standard in Europe.

Am I really that lost and there is a pre/de-emphasis for AM?

Terry

wrote in message
oups.com...

[snip]

Am I really that lost and there is a pre/de-emphasis for AM?

Terry


AM is usually preemphasized as part of the audio processing. They do it to
more or less compensate for the normal frequency roll-off in the IF section.

Frank Dresser

wrote in message
ups.com...

[snip]

Many radios would sound much better with some attention to the audio
chain.
I used to think that a "little" distortion didn't matter. After all the

signals get pretty mangled by the ionosphere. And I was wrong. It makes
a
great difference. I have spent the day listening to HF and going back
and
forth just to get a feel for it.

The Radio Amateur's handbooks have some interesting articles about
increasing intelligibility in radio transmissions.

The older ones will have articles about AM ham radio transmissions.




While I would love to buy a really "good" receiver like the AOR7030+,
or
one of the better Drakes, or even an NRD, money being rather tight, I
am
happy to get any improvements I can.


Terry


You might want to consider a tube radio. A tube radio will get the best
performance from a simple diode detector. The diode detector works quite
well, except when the voltage across it gets low. As the voltage gets low,
the diode's equivalent resistance goes up rapidly. Working the diode at a
higher average voltage and into a high impedance load minimize these
problems.

In general, the consumer SW sets will sound better than the
communications/ham receivers.

Frank Dresser

Hi Terry,

Polypropolyne capacitors are the best sounding types for audio use, with
polycarbonate following closely. It you take a look at the dissipation
factor vs frequency characteristics, you will find out that the
polypropolyne caps are the flattest in the regard.
As far as mechanical filters, you don't run into ringing with these units,
but if the filters are not terminated properly they will not have a
flat-topped response characteristic, and that can sound like ringing.
What I used to do if I didn't have the data on those old disc-wire filters
was to either hook up a network analyzer or an LCR bridge (not the kind that
is an extra function on those inexpensive multimeters) and measure the input
inductance of the filter.
This would allow me to determine the value of the required I/O resonating
capacitors. In addition to this, buy calculating the inductive reactance of
the filter, I would be able to roughly calculate in I/O inpedance.
Nowadays, with the newer Torsional Mode filters, this is not required as
long as you keep the strays in your PC board design below 30pF. The newer
filters (especially the 526-9695-XXX series) have a 2k in/out impedance so
you can use them where those Murata ceramic filters are used.

Pete

"Frank Dresser" wrote in message
...

wrote in message
ups.com...

[snip]

Many radios would sound much better with some attention to the audio
chain.
I used to think that a "little" distortion didn't matter. After all the

signals get pretty mangled by the ionosphere. And I was wrong. It makes
a
great difference. I have spent the day listening to HF and going back
and
forth just to get a feel for it.

The Radio Amateur's handbooks have some interesting articles about
increasing intelligibility in radio transmissions.

The older ones will have articles about AM ham radio transmissions.




While I would love to buy a really "good" receiver like the AOR7030+,
or
one of the better Drakes, or even an NRD, money being rather tight, I
am
happy to get any improvements I can.


Terry


You might want to consider a tube radio. A tube radio will get the best
performance from a simple diode detector. The diode detector works quite
well, except when the voltage across it gets low. As the voltage gets
low,
the diode's equivalent resistance goes up rapidly. Working the diode at
a
higher average voltage and into a high impedance load minimize these
problems.

In general, the consumer SW sets will sound better than the
communications/ham receivers.

Frank Dresser

skrev i en meddelelse
ups.com...

I then compared the
http://www.amwindow.org/tech/htm/alowdisdet.htm unit with the AD607
synch detector. The synch detector is the clear winner.


Do you have a link to the synch detector schematics?


--
Anders Henriksen - born without teeth
supermule [at] lite [dot] dk

Whosoever is delighted in solitude, is either a wild beast or a god

Anders Henriksen wrote:

Do you have a link to the synch detector schematics?

-----------------------------
Pete KE9OA kindly sent me a diagram for a MW
receiver that he designed and built. The AD607 PDF
gives pin out and other data.
Discrription
http://www.analog.com/en/prodDesc/0,2895,AD607%5F0,00.html
PDF can be found on this page:
http://www.analog.com/en/prodDesc/0,2895,AD607%5F0,00.html

Terry

skrev i en meddelelse
ups.com...

Pete KE9OA kindly sent me a diagram for a MW
receiver that he designed and built. The AD607 PDF
gives pin out and other data.
Discrription
http://www.analog.com/en/prodDesc/0,2895,AD607%5F0,00.html
PDF can be found on this page:
http://www.analog.com/en/prodDesc/0,2895,AD607%5F0,00.html


Thanks for the info. Now we'll see, if Pete is willing to send me the
schematics?


--
Anders Henriksen - born without teeth
supermule [at] lite [dot] dk

Whosoever is delighted in solitude, is either a wild beast or a god

Anders:
Check out thse links:

http://home.worldnet.att.net/%7Ewa1sov/technical/sync_det.html
http://home.worldnet.att.net/~wa1sov/technical/allpass/allpass.html
http://home.worldnet.att.net/%7Ewa1sov/ver2/Software_Rado.htm

The AD607 has a PLL built in and I don't know why this designer
didn't employ it instead of adding a NEnnn PLL. But at least it shows
one way to use the AD607.
I am presently using a ARRL BFO deign from the early 1970's that
works, but has developed some nasty drift issues in the 30+ years
it has mainly set idle. I am debating using 3 Epson programable
for the BFOs. I can either go with them running at 3 times the desired
Fout,
or hihger mutliples. At 13.nnnMHz the phase noise is on teh edge of
being too high,. but for less then 5MHz the phase noise is minimal. I
have a query
to Espon tech support to find out at which freqs they have minimal
phase noise. Supply them with +5 or +3.3V at 10mA and they give very
stable
Freqs. I have one 455KHz crstal, but it is sluggish in every circuit I
have
tried it in. I can get three of the Epson uints for less then a "good"
crystal.
I want one at 455.0000MHz and one at +1.300 and -1.300 KHz.
I am building this as a series of modules each in it's own Altoid Mints
tin.
A friend gave me a BIG Altoid tin that has more then enough space for
everthing.

Terry

Terry