I was going to stay out of this fight,but it has just gotten too silly.
So;"Can twisted wire replace shielded wire?"
The simple answer is yes.
But there is a big BUT.

In a very quite RF locaiton, like the middle of the woods where there
are no electric/phone/CATV lines, no TVs, no motors, no flourescent
lamps a "twisted pair" can work ery well. Slightly lower losses, but my
experiments in such a RF quite locaiton show no practical or detectable
effects for LF through F (say 50KHz through 30MHz).

If you have any RF noise sources, in many cases this will include your
receiver as many modern recievers have some (to a lot) of RFI
potiential,
you will need to reduce the noise that gets into/onto your signal.

Balanced open wire, AKA "ladder line" is better then twisted pair as
far as
line losses are concerend. But open line is more susceptable to ingress
unless
great care is taken to keep it well away from "anything" that will
cause imbalance.
Sadly the "anything" really means anything. At LF through HF the
effects aren't
all that important.

A trick I learned back in the days of tall towers for VHF an UHF TV
antennas was
to put a 1 twist per foot in the twinlead. This often greatly reduced
ghosting from
very bad to not visible.

So back to the orignal question:
"Can twisted wire replace shielded wire?"
Yes, under the right conditions. But in the real world, at least where
I live, Coax,
or Twinax or Triax, is much better because it reduces any additional
noise I could
pick up as my lead in leaves the antenna and reaches my receiver. I
have used
common twisted "bell" wire when I didn't have coax and if the location
was quite, it
worked great.

The real question is not can twisted wire be used, but how well can you
isolate
twisted or parallel(think lamp cord) from your noise generators. It
can be done,
but coax is a lot easier to keep quite.

I have enough RF noise that I can't reduce, there is no way I want to
add any more
RF crud to my 1uV signals.

Terry

On 27 Nov 2005 14:57:43 -0800, wrote:

I was going to stay out of this fight,but it has just gotten too silly.
So;"Can twisted wire replace shielded wire?"
The simple answer is yes.
But there is a big BUT.

In a very quite RF locaiton, like the middle of the woods where there
are no electric/phone/CATV lines, no TVs, no motors, no flourescent
lamps a "twisted pair" can work ery well. Slightly lower losses, but my
experiments in such a RF quite locaiton show no practical or detectable
effects for LF through F (say 50KHz through 30MHz).

If you have any RF noise sources, in many cases this will include your
receiver as many modern recievers have some (to a lot) of RFI
potiential,
you will need to reduce the noise that gets into/onto your signal.

Balanced open wire, AKA "ladder line" is better then twisted pair as
far as
line losses are concerend. But open line is more susceptable to ingress
unless
great care is taken to keep it well away from "anything" that will
cause imbalance.
Sadly the "anything" really means anything. At LF through HF the
effects aren't
all that important.

A trick I learned back in the days of tall towers for VHF an UHF TV
antennas was
to put a 1 twist per foot in the twinlead. This often greatly reduced
ghosting from
very bad to not visible.

So back to the orignal question:
"Can twisted wire replace shielded wire?"
Yes, under the right conditions. But in the real world, at least where
I live, Coax,
or Twinax or Triax, is much better because it reduces any additional
noise I could
pick up as my lead in leaves the antenna and reaches my receiver. I
have used
common twisted "bell" wire when I didn't have coax and if the location
was quite, it
worked great.

The real question is not can twisted wire be used, but how well can you
isolate
twisted or parallel(think lamp cord) from your noise generators. It
can be done,
but coax is a lot easier to keep quite.

I have enough RF noise that I can't reduce, there is no way I want to
add any more
RF crud to my 1uV signals.

Terry

Common mode rejection.

Look it up.

David wrote:
snip
Common mode rejection.

Look it up.
++++++++++++++++++++++++
Just for grins lay a 100' of twisted leads, lamp cable and coax out on
the ground.
Attach 50 Ohm resisitors at the far end of each.
Swtich from one to the other and see which one picks up the most RF.
MW stations will be received very well with both the twisted pair and
lamp cord.
Very close/strong MW stations may be recieved with evenpoor quality
coax.

Next build one of these to go between the feedline and your receiver:
http://www.dxing.info/equipment/coax_leadin_bryant.pdf
Repeat the experiment.

If your results are anything like mine the twisted pair and the lamp
cord will
still have significant reception.

True "common mode" only occurs when there is perfect balance between
the
conductors, with no "realworld" artifacts to effect that balance.

The results I provided are real world, not from some SPICE simulation.

I tried this with both a R2000 and a AOR 7030+. And I got nearly the
exact
same result with the AOR giving slightly better MW rejection with the
coax
and the in line filters.

I preffer to use real world experiments whenever I can.

I was surprised at how little difference the quality of the coax, or
even the
Z mattered. I expected no measurable difference between 50 and 75 Ohm
coax.
But even 93 ohm, low capacitance, coax acted the same. And the type of
shield
made no difference. By type of shield I am reffereing to: single braid,
double braid,
braid + foil and even some real cheap foil and drain only.

I even did a repeat of this test at a ham friends QTH where he was
putting up a
repeater and had about 200' of hardline stretched out on the ground.
Worked pretty
well, but I had to add the chokes in the PDF link I gave.

Now Triax did make a big difference. But only if you only grounded it
at one end,
the receiver end being about 20dB better. And Triax did not benifit
from the common
mode supression. It amde no difference.

Terry

On 27 Nov 2005 18:23:39 -0800, wrote:


David wrote:
snip
Common mode rejection.

Look it up.
++++++++++++++++++++++++
Just for grins lay a 100' of twisted leads, lamp cable and coax out on
the ground.
Attach 50 Ohm resisitors at the far end of each.
Swtich from one to the other and see which one picks up the most RF.
MW stations will be received very well with both the twisted pair and
lamp cord.
Very close/strong MW stations may be recieved with evenpoor quality
coax.

Next build one of these to go between the feedline and your receiver:
http://www.dxing.info/equipment/coax_leadin_bryant.pdf
Repeat the experiment.

If your results are anything like mine the twisted pair and the lamp
cord will
still have significant reception.

True "common mode" only occurs when there is perfect balance between
the
conductors, with no "realworld" artifacts to effect that balance.

The results I provided are real world, not from some SPICE simulation.

I tried this with both a R2000 and a AOR 7030+. And I got nearly the
exact
same result with the AOR giving slightly better MW rejection with the
coax
and the in line filters.

I preffer to use real world experiments whenever I can.

I was surprised at how little difference the quality of the coax, or
even the
Z mattered. I expected no measurable difference between 50 and 75 Ohm
coax.
But even 93 ohm, low capacitance, coax acted the same. And the type of
shield
made no difference. By type of shield I am reffereing to: single braid,
double braid,
braid + foil and even some real cheap foil and drain only.

I even did a repeat of this test at a ham friends QTH where he was
putting up a
repeater and had about 200' of hardline stretched out on the ground.
Worked pretty
well, but I had to add the chokes in the PDF link I gave.

Now Triax did make a big difference. But only if you only grounded it
at one end,
the receiver end being about 20dB better. And Triax did not benifit
from the common
mode supression. It amde no difference.

Terry

Try it with a true balanced input.

David wrote:
Try it with a true balanced input.
++++++++++++++++++++++++Ihave.
I used a home made 1:1 torroid tranformer with minitmal effects.
I even tried it with a 1:1 balun at the far end, with no effect.

The only point I am trying to make is that the twisted pair and
paralled
conductor that I tried were not completly balanced. Perhaps high
quality
CAT6 ethernet cable would be.

We are moving into the realm of "how many angels can dance on the
head of pin". Interesting from a philosphical point of view, but not
much
use in the real world.

I have been trying to find better ways to reduce noise since before
I got my EE. In my experience better reception, IE receiving the really

weak signals, has more to do with reducing the QRM and QRN then
"magic" feedlines. I am not in love with coax, it is just the
easiest/best
solution to keep my feedline from picking up noise, and with common
mode supression, keeping noise from finding it's way back up the shield

and getting into the antenna.

As I have said, if I lived in the woods, with no electrical equipment
to
be sources of RF noise, then the type feedline I used would not matter.

BTDT But in the world that I live in, there are a variety of noise
sources in
my house that I have mitigated as far as possible. One step in
mitigation
was to reduce QRM ingress as much as possible. With one, or more, PCs,
AC to DC power supplies, DSP(rather noisy) and other noise sources
in my shack, coupled with the other household equipment, "smart
furnace"
fridge, wifes PC, TV/VCR/DVD, microwave, alram clock, and every light
switch,
I found that I could not get a balanced RF feed to work.

However the audio that I distribute around our home is balanced for the

reasons you listed. At work we distribute some video via balanced
tiwn-ax
up to 2 miles. We are replacing these with fiber not because balanced
doesn't
work very well, but because lighting strikes keep frying the line
drivers.

In a situaiton where one was using a balanced antenna,like a dipole,
with
no significant local noise sources parallel, twisted or orpen ladder
line
would be great. For extremely high power transmit operations open
ladder line
is the best choice. But there will more leakage from the open line then
from coax.

Terry

On 28 Nov 2005 08:59:00 -0800, wrote:


David wrote:
Try it with a true balanced input.
++++++++++++++++++++++++Ihave.
I used a home made 1:1 torroid tranformer with minitmal effects.
I even tried it with a 1:1 balun at the far end, with no effect.

The only point I am trying to make is that the twisted pair and
paralled
conductor that I tried were not completly balanced. Perhaps high
quality
CAT6 ethernet cable would be.

We are moving into the realm of "how many angels can dance on the
head of pin". Interesting from a philosphical point of view, but not
much
use in the real world.

I have been trying to find better ways to reduce noise since before
I got my EE. In my experience better reception, IE receiving the really

weak signals, has more to do with reducing the QRM and QRN then
"magic" feedlines. I am not in love with coax, it is just the
easiest/best
solution to keep my feedline from picking up noise, and with common
mode supression, keeping noise from finding it's way back up the shield

and getting into the antenna.

As I have said, if I lived in the woods, with no electrical equipment
to
be sources of RF noise, then the type feedline I used would not matter.

BTDT But in the world that I live in, there are a variety of noise
sources in
my house that I have mitigated as far as possible. One step in
mitigation
was to reduce QRM ingress as much as possible. With one, or more, PCs,
AC to DC power supplies, DSP(rather noisy) and other noise sources
in my shack, coupled with the other household equipment, "smart
furnace"
fridge, wifes PC, TV/VCR/DVD, microwave, alram clock, and every light
switch,
I found that I could not get a balanced RF feed to work.

However the audio that I distribute around our home is balanced for the

reasons you listed. At work we distribute some video via balanced
tiwn-ax
up to 2 miles. We are replacing these with fiber not because balanced
doesn't
work very well, but because lighting strikes keep frying the line
drivers.

In a situaiton where one was using a balanced antenna,like a dipole,
with
no significant local noise sources parallel, twisted or orpen ladder
line
would be great. For extremely high power transmit operations open
ladder line
is the best choice. But there will more leakage from the open line then
from coax.

Terry

Up until about 30 years ago, almost every radio station on the air was
feeding audio to their transmitters via unshielded wire phone lines.
No buzz.

Optoisolators as suicide devices.

David wrote:
On 28 Nov 2005 08:59:00 -0800, wrote:


David wrote:

Up until about 30 years ago, almost every radio station on the air was
feeding audio to their transmitters via unshielded wire phone lines.
No buzz.

Optoisolators as suicide devices.
+++++++++++++++++++++++++++++++++++++
Now I get it.
I am trying to recieve signals near or at the noise floor which is set
first by the local QRN. there are as amny different levels as
refference
books, so I will use John Doty's figures which are wildly optimistic in
my
experience. I will post links to my refferences so others may follow
them.

From:http://anarc.org/naswa/badx/antennas/grounding.html

"According to "Reference Data for Radio Engineers"
(Sams, 1975), the wintertime level of natural noise in
my area at 10 MHz should be about 32 dB above the
thermal reference level: this would produce a noise
floor of -104 dBm in this bandwidth with a perfectly
efficient antenna. "

A very handy chart that converts dBm to Volts can be found at:
http://www.minicircuits.com/dg03-110.pdf

Sherwood Engineeringhttp://www.sherweng.com/table.html gives
the R2000 noise floor as -130dB in the SSB Mode.

So I am trying to capture signals in the range of -130dB=0.058uV to
-119dB= 0.251uV.

Your refference to audio being delivered over balanced line is at 0dBm
(nominal) for a voltage of 0.775V. (*600Ohm) or 0.225 at 50 Ohms.

There is more then a slight voltage level diffference.

Can we agree that:
Adding 1uV of noise to a 1uV(-107dBm) signal will significantly degrade
the
SN ratio.

Adding 1uV of noise to a 0dBm 50 or 600 ohm line will result in
minimal,
as in I would like to see it measured, degradation of the overall SN
ratio.

And I bet you won't try to run the output of a moving coil phono
cartridge
over ~70' of balanced audio line.

In a uV world I will stick to single ended, unbalanced,
coaxial/shielded cable
interconnects. If you are lucky enough to deal with signals in the
0.1V range
then twisted or parallel cable will work just fine.

Terry

On 28 Nov 2005 10:47:37 -0800, wrote:


David wrote:
On 28 Nov 2005 08:59:00 -0800, wrote:


And I bet you won't try to run the output of a moving coil phono
cartridge
over ~70' of balanced audio line.

In a uV world I will stick to single ended, unbalanced,
coaxial/shielded cable
interconnects. If you are lucky enough to deal with signals in the
0.1V range
then twisted or parallel cable will work just fine.

Terry

You can't send a MC phono signal 70' on co-ax either.

David wrote:
You can't send a MC phono signal 70' on co-ax either.
++++++++++++++++++++++++++++++++++
You mihgt find the story
at:http://broadcastengineering.com/mag/broadcasting_future_video_utp/
intersing.
It discribes using UTP (unshielded twisted pair) for video.


Mention is made of crosstalk.
For CAT5 cable 53dB of isolation at 4.2MHZ is acheived.
CAT6 gies 63dB at 4.2MHz.
Crostalk increases with frequency.

Another good refference that shows "failure" modes for twisted pair:
http://www.extron.com/technology/archive.asp?id=utptechnology

I have lost my bookmarks to a couple of links for coax cable "leakage"
and crosstalk specificitions. I will have to pull them from a backup
later this
week.

Why do you suppose that 99.99% of all CATV operations use coax for
the coper connections instead of twisted pair? Twisted pair is less
expensive,
so if it was "as good" as coax I suspect they would switch to it.

It has never been my contention that twisted and parallel balanced
pairs
have no place in the world. I just accept that while they do have many
usefull applicaitons, low level RF distribution isn't 'one of them.

By the way we are rolling out some component HDTV in house distribution
with CAT6 UTP. Looks pretty darn good after a 200' run. The CEO wanted
this and it was much cheaper to pull in good CAT6 then 4 coax cables.

Terry

On 29 Nov 2005 06:52:04 -0800, wrote:


David wrote:
You can't send a MC phono signal 70' on co-ax either.
++++++++++++++++++++++++++++++++++
You mihgt find the story
at:http://broadcastengineering.com/mag/broadcasting_future_video_utp/
intersing.
It discribes using UTP (unshielded twisted pair) for video.


Mention is made of crosstalk.
For CAT5 cable 53dB of isolation at 4.2MHZ is acheived.
CAT6 gies 63dB at 4.2MHz.
Crostalk increases with frequency.

Another good refference that shows "failure" modes for twisted pair:
http://www.extron.com/technology/archive.asp?id=utptechnology

I have lost my bookmarks to a couple of links for coax cable "leakage"
and crosstalk specificitions. I will have to pull them from a backup
later this
week.

Why do you suppose that 99.99% of all CATV operations use coax for
the coper connections instead of twisted pair? Twisted pair is less
expensive,
so if it was "as good" as coax I suspect they would switch to it.

It has never been my contention that twisted and parallel balanced
pairs
have no place in the world. I just accept that while they do have many
usefull applicaitons, low level RF distribution isn't 'one of them.

By the way we are rolling out some component HDTV in house distribution
with CAT6 UTP. Looks pretty darn good after a 200' run. The CEO wanted
this and it was much cheaper to pull in good CAT6 then 4 coax cables.

Terry

I think the context is shortwave radio.