Newsgroups: comp.dcom.modems
From: floyd@polarnet.com (Floyd Davidson)
Subject: Re: Longish cable runs?
Date: Mon, 14 Jul 1997 09:49:22 GMT

rcollins@cyberus.ca wrote:

>floyd@polarnet.com (Floyd Davidson) wrote:
>
>>Your connection will not be as good with a 25' extention using
>>flat non-twisted pair cable.  I personally would not recomment
>>anything longer than 6 feet using flat cable.
>
>On the other hand, I have about 50 feet of cheap, flat cable that runs
>along the floor and is a source of endless amusement to the cats.  I
>connect at 33.6 99% of the time, and 46+ with X2.  All in all, my 50
>feet represent a very small part of the telco circuits I'm using, and
>the only rule would appear to be "if it works, use it".

That assumes you can tell if it works (and even then it wouldn't
be a very good rule anyway...).

The connect speed is perhaps a very good indication of the line
condition at the moment the two modems performed the
initialization sequence.  In many cases the performance in the
next few seconds or minutes causes a drastic (or maybe not so
drastic) change in data rate.

Hence your measuring stick is fine, but it measures the wrong
thing rather well.  How well a connection passes data for an
hour is a much more useful measure than connection speed.  Very
few people accomplish the equivalent of a Bit Error Rate Test
(BERT) on their modem connections, so few people really know how
well their line is actually working.

And too, your 50 foot of local cable has a pretty fair potential
that might not be obvious.  It is connected to some several
thousand feet of telco cable, and if your cats (I've got three
myself, so lets not be too hard on the nature of feline
curiosity here) manage to chew it up in a way that allows
exposed wires to make a connection with objects it comes into
contact with, the result is the entire length of your telephone
line works more as an antenna than as the transmission line it
should be.  You might not only pick up noise induced locally in
your home, but also a mile down the road where the splice box
is...

More likely however the cats won't chew it that good, and the
only impairment you suffer is noise induced by having a stretch
of non-twisted pair cable.  If you are quite far from any large
motors or fluorescent lights, it might not be too bad.  But the
fact is noise will be induced every time there is any electric
field nearby that changes.  It might only last long enough to
cause a single packet error on your data, or it might mess up a
great deal of data.  It will likely not be enough to disconnect
the modem, and at any given time you choose to measure
throughput it won't be there.

Hence it is very easy to trick yourself into believing that all
is well and things are working fine.  But a controlled test with
appropriate equipment would demonstrate otherwise.

Plus in using rules of thumb to hand out advice to people trying
to avoid problems it does seem well to suggest ways that are
guaranteed to work rather than methods that work if you get lucky.

  Floyd

Newsgroups: comp.dcom.modems
From: floyd@polarnet.com (Floyd Davidson)
Subject: Re: Longish cable runs?
Date: Mon, 14 Jul 1997 07:59:58 GMT

George Henry  <gachenry@worldnet.att.net> wrote:

>Floyd Davidson wrote:
>>
>> Your connection will not be as good with a 25' extention using
>> flat non-twisted pair cable.  I personally would not recomment
>> anything longer than 6 feet using flat cable.
>
>Hmm.. I just made up 120 silver-satin flat cables w/dbl ended RJ-45
>Plugs. The
>run is from a Nortel PBX to racks of NEC D4 Channel Banks. You shouldn't
>have
>any trouble.

You don't mention how long these cords are, but I would highly
recommend, because there are so many of them, they be replaced
with twisted pair, even if the length is short.  Put it this
way: no telephone company would do what you are doing!  We use
twisted pair even for 6 inch long patch cords, and flat cable
only from a modular jack to an actual individual telephone set.

What you have described is an invitation for crosstalk problems.

>If it'll set your mind at ease just put 4 or 5 twists in it.

Eh?  You can't!  Twisting the cable is NOT the same as twisting
the pairs.

>> It is very difficult for the average modem user, who is not a
>> telecom technician, to determine quickly if a given cable run is
>> going to cause problems.  Just making a few test calls and
>> checking the connect speed will give absolutely no indication
>> that a problem exists.  Later on it might be noticed that every
>> time the furnace starts the modem gets dumped, or everytime the
>> overhead flourescent light is on the throughput for file
>> transfers is down by 25%.
>
>Make sure all connections are tight and clean as a loose connection
>is noisy and represents a swinging high resistive load.

A poor connection upsets longitudinal balance, which causes
noise induction just the same as any other cause of imbalance.

>> A little crosstalk on a voice connection because you have a long
>> extention cord is fine, but for use with a modem it pays well to
>> avoid anything that is a potential problem.
>
>De-modulate RF occurs in the telset and not the cord. X-talk as in
>crossed lines is another matter. This occurs with actual physical
>contact with another working line and is referred to as X-battery
>or a foreign ground. This can be tested by your loac Repair Service
>Bureau.

I don't think you quite understand what crosstalk is.  When the
current in one pair causes a differential current in another
pair, that is crosstalk.  If it is really bad the party on one
line can actually hear and understand a party on the other line.
In less severe cases only high level signals, such as ringing or
power line hum, can be heard.  It has nothing to do with RF, and
it is not necessarily caused by physical contact (though that
can be one cause). And it is never ever refered to as X-battery
or a foreign ground.

The "local Repair Service Bureau" will charge you an arm and a
leg to test it too, as well they should.  If a person is able to
do the work it would probably be cheaper to rewire an average
home using proper engineering practices than it is to have
someone test the existing cabling!

The most common instance of crosstalk in house wiring is running
two phone lines on a two pair cable that does not use twisted
pair.  Ringing, modem tones, dialtone, touch tones, etc. can
commonly be heard on the "other" line.  In many areas older homes
were wired with 4 wire JK cable (red, green, yellow, and black
wires) that was not twisted, and while random pops and crackles
on voice calls were never upsetting enough, few subscribers want
to hear their modem or another conversation after they have put
in a second line...  hence the old cable is now a big problem.
(Not to mention that everyone wants 33.6Kbps connections on that
modem too!)

  Floyd
  AT&T Alascom, Fairbanks Testboard

Newsgroups: comp.dcom.telecom.tech
From: floyd@tanana.polarnet.com (Floyd Davidson)
Subject: Re: Cabling length limits
Date: Sun, 18 Jan 1998 21:00:05 GMT

[emailed and posted]

Aaron Leonard <Aaron@cisco.COM> wrote:
>chip76@ix.netcom.com (Jeff Vinocur) writes:
>|floyd@icefog.polarnet.com (Floyd L. Davidson) wrote:
>|:Sam Hall <slhall1@airmail.net> wrote:
>|:>chip76@ix.netcom.com (Jeff Vinocur) wrote:
>|:>
>|:>>I need to run an extension temporarily about 50 yards from
>|:>>an existing phone jack.  Two ideas come to mind:
>|:
>|:>Just plug it in, fifty feet should not be a problem
>|:
>|:That is true _if_ twisted pair is used.  I wouldn't use
>|:more than a 6 foot length of the common flat satin extention
>|:cable.  A twisted pair extention that is plugged in or
>|:spliced either one, will work fine.
>|
>|If I go to, say Radio Shack, what do I need to look for?  It
>|it labelled "twisted pair", or what?
>
>Oh, anything called Category 3 or Category 5 will be fine.
>I'd be willing to use quad (red/green/black/yellow) for this,
>but you might as well use cat 3 or better.

Good advice.  (One would hope that the sales people would be
of some assistance!)

>I'm curious though: why only 6' of flat satin?

To induce noise with a 6 foot length of flat cable is reasonably
difficult.  It requires that you wrap it around a source of
noise, like a flourescent light or whatever.  Anything longer is
enough to make it easy, and perhaps 15 or 20 feet is long enough
to practically guarantee too much noise induction for a v.34 modem.

Consider that for voice use the twist can be anything less than
about 30", so a six foot cable is 2 times the longest twist.
Perhaps not much harm, but obviously a cable that has more than
4 times the distance in which there should be a twist to
effectively balance induction will suffer from a lack of
balance.

  Floyd




--
Floyd L. Davidson   <floyd@tanana.polarnet.com>   Salcha, Alaska

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Advice on finding line noise?
Date: 26 Sep 1998 11:07:19 GMT

<soundguy@.ix.netcom.com> wrote:
>Chad McHenry > <red_sea@earthlink.net> wrote:
>
>>Are there phone wiring specs I should attempt to meet in the house
>>before I can expect to see v.90?  (Distance of run, gague of wire)  Is
>>there some way to test or "certify" the amount of noise on the internal
>>wiring?  Where would that be documented?
>
>The sure fire "quick & dirty" test method is to go out and buy a
>loooong RJ11 wire & hook your modem up directly to the demarc

There is no such thing as "RJ11 wire".  Twisted pair telephone
wire or better yet CAT3 data grade twisted pair would be the
appropriate choice.

>modems. Most new house wiring is 4 pair 26 gauge twisted, but the drop
>to your house is probably industry standard untwisted flat wire (14-16
>gauge copper coated steel). If the drop itself is twisted, it's
>several decades old and should be replaced(by the phone company). If

Industry standard drop wire is twisted pair.  That "flat" drop wire
has a sheath shaped for strength, and inside is a twisted pair.

>BTW, I have a 100 foot flat drop and can run 52k all day long over a
>50 foot flat wire, so don't worry too much about wire length. Just
>keep it under 6000 feet or you'll have to install a "load coil" to get
>your phone to ring.

Cable loading coils have nothing to do with whether the phone
will ring or not.  They reduce cable loss at voice frequencies
at the expense of amplitude distortion (rolloff at both the high
and low end of the voice spectrum) and several other parameter
changes that are less important (phase shift, envelope delay, and
characteristic impedance of the cable).

>>What else can I do if all inside the house check's out, and what else am
>>I missing.

I believe that the original article mentioned a considerable length of
flat satin type of extension wire.  For a modem that is probably the
worst possible cable.  The best way to check out your house wiring is
to eliminate it for a test.  Either take the modem as close to the
interface point as possible, or purchase a length of twisted pair cable
and disconnect all other cable at the interface while you run a test
connecting the interface to the modem using known good quality cable.

>You might try calling the phone company and complaining about noise on
>the line (after you've checked the house wiring). They may or may not
>be interested in helping you. (after all, it's not like you can just
>switch to another local loop provider...) Many of them will only
>guarantee 9600 connections. A few don't "support"  modem usage.

If the phone company is recalcitrant, call them back every few days.
They _will_ get tired of it quicker than you!

  Floyd




--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Advice on finding line noise?
Date: 27 Sep 1998 01:59:12 GMT

<soundguy@.ix.netcom.com> wrote:
>floyd@tanana.polarnet.com (Floyd Davidson) wrote:
>>I would not recommend any length of flat cable longer than 6-8 feet.
>
>As I stated in my original post, I can get 52k 100% of the time over a
>50 foot  flatwire(not a good one either..Radio Shack). I live 3 miles
>from the CO. I'm on GTE property. My ISP is 30 miles away on US West
>property.
>To be fair, and to test the validity of your statement, I just made
>several connections with 6 foot and 50 foot flatwire. There was NO
>difference in connect speeds whatsoever.

And a leak in your roof wouldn't be there when the sun shines,
either??? :-)

Untwisted cable is a very poor idea for telephone cable.  Not
that it can't work, but it won't work under a large number of
circumstances where twisted pair cable will work.  Every time
your refrigerator's compressor goes on, or even when a light
switch is thrown, that non-twisted pair is likely to cause a
serious hit for a modem connection.  You won't notice that doing
the kind of useless test that you just described.  Likewise, if
it just happens that the person you have advised to use flat
cable runs it in parallel with an extention cord (which wasn't
part of your test) or happens to have flourescent lights (also
not part of your test) or just happens to have an exceedingly
noisy power source due to a machine shop that does arc welding
next door (not part of... ) then his "test" is going to be
totally useless too, but in a different way than yours was.

Simple solution: use twisted pair for all applications where a
telephone line is extended more than a few feet.  (I.e., flat
satin extentions should not be used for more than 6-8 feet.)
VF quality telephone twisted pair is OK, but CAT3 is better and
does not cost much either.  CAT5 gets expensive, but for a
relatively short length (a few dozens of feet) it still is not
overly expensive and certainly should be considered if there is
any chance that a data network will be added in the next decade
or so (which should always be given a reasonable chance!).

>>GTE where I live uses twisted pair (which looks for all the world just
>>like "twinlead").
>
>The old standard drop wire is dull rubber with fully rounded edges.
>The newer multi-pair twisted cable is shiny plastic, has squarer
>edges, and a shallower center groove. Eventually, everyone will have
>it, but it's only been spec for a couple of years.

I've never seen a drop wire that was not twisted pair!  I'm not
saying there are none...  just that I haven't seen any.  (And
I'm counting spiral-4 as "twisted".)  I just moved to a new
house, and it has twisted pair.  The house I was in for the past
20 years had twisted pair too.  So did the one I lived in for 10
years before that...

> There are still
>thousands of miles of open wire in this country that need to be
>replaced first. ( open wire only "twists" once every 500 feet)

Open wire?   *thousands of miles*?  Where?  Who can afford to
maintain the stuff?

(We still have some strung on poles in Alaska, but all the old
linemen are now either retired or in any case too old to
refurbish any of it.  I think the last that was actually in use
was turned down more than 20-25 years ago now.)

However, the thing you want to note very carefully about drop
wires and open wire transmission lines is that the environment
in close proximity to the line is very carefully controlled too.
No power lines, electric motor lines, or metal objects are
allowed.  Likewise each pair is separated by some distance
greater than the two pairs within flat satin cable.  That keeps
the crosstalk down; and is not generally the way that flat satin
extention cords are run in a normal home environment.

  Floyd


--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)

From: floyd@tanana.polarnet.com (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Advice on finding line noise?
Date: 28 Sep 1998 06:10:27 GMT

Chad McHenry  <red_sea@earthlink.net> wrote:
>I read somewhere that not connecting all computer equipment w/in the
>same bldg to the same ground, can be a source of noise in ethernet
>connections.  Would having the electric meter and phone interface box on
>a different ground than my computer equiptment cause an issue?

That could cause problems.  Just be aware that grounding is a
complicated subject...  In your home it is very unlikely that
you have the computer on a different ground than the electrical
system.  If your house and the house next door are on different
distribution transformers (those pole mounted cans), you could
do it by running an extention cord over and plugging into an
outlet in that house...  otherwise it should be a little
difficult.  Everything in your house should end up being tied to
the same ground as the electrical system, by the very fact that
you plug everything into the distribution of that system.

>Also, the test mentioned by Floyd and Patrick talk of testing straight
>to the demarcation point, using (preferably) cat3 or twisted pair
>telephone wire. I've got some 8-con (4 twisted pair) 24 Ga. wire, and
>after stripping the sheathing back about a foot, I still see no "twists"
>or even pairings to the wires...  what gives?

It can be hard to see the twist, because it might be 20-30
inches per twist.  Usually you can see the twist in the wires
through the sheath easier than you can see it looking at the
wires.  Also you can strip back the sheath by taking one wire
and pulling it backwards, ripping the sheath.  On a twisted pair
cable that wire will twist around the cable as you pull it.
(Get a couple inches of wire loose by cutting the sheath with a
sharp blade, then pull one wire in one direction, and all others
and the sheath in the other direction.)

The tighter the twist the easier it is to see, so with telephone
wire meant only for voice frequencies it is harder to see than
on cable meant for data, which has a tighter twist.

>I figure I'll get some cat3, but how would I wire it?  Should I:
>
>A) Use one pair for the red/green wires and one for the black/yellow
>that are normally found in phone wiring, or

Yes!  What you will have is ordered pairs, Blue, Orange, Green,
Brown, and Slate.  The Tip has more white, the Ring has more
color, although it really makes no difference at all if you get
the tip and ring backwards.  The important part is keeping each
circuit on a twisted pair, and not split between two pairs.

If you have any interest in a slightly more more technical
description of how twisted pair cable works, go to
www.dejanews.com and look up an article that I posted to
comp.dcom.telecom.tech on September 20th, in the thread "2wire,
4-wire".

  Floyd


--
Floyd L. Davidson                                floyd@ptialaska.net
Ukpeagvik (Barrow, Alaska)

From: Floyd Davidson <floyd@ptialaska.net>
Newsgroups: comp.dcom.modems
Subject: Re: Phone Line Length
Date: 17 Jul 2000 23:48:52 -0800

Jack Daniels <nobody@nowhere.com> wrote:
>Does the use of non-twisted flat cable for the drop from pole
>to the house have any effect on connect speed? I didn't have a
>choice when they installed it but that is what Ameritech used.

Are you sure that it is actually non-twisted?  It can be very
difficult to tell without actually looking at the cable, as the
non-twisted and the twisted drop cable might look almost
identical.

However, in many instances it makes little difference which is
used.  If the drop to your demarc is a straight shot from a
pole, and not very close to any power lines or to any metal
objects, then there will likely not be any interference for the
non-twisted drop line to pick up.

That is very different than using a non-twisted and very
flexible cable inside a house where it has relatively sharp
bends, is directly in contact with other computer cables and
power cords, not to mention that it might be wrapped around a
flourescent lamp...  ;-(

>Are the better telephone companies using twisted pair for the
>drop now that modems are so popular?

Both are used, but twisted pair is much more common.

--
Floyd L. Davidson                          floyd@barrow.com
Ukpeagvik (Barrow, Alaska)

From: Floyd Davidson <floyd@ptialaska.net>
Newsgroups: comp.dcom.telecom.tech
Subject: Re: twisted pair
Date: 02 Aug 2000 01:20:17 -0800

"Al" <austin_powers_groovybaby@NOSPAMhotmail.com> wrote:
>I have I understood the thread correctly ....?  Plain old voice
>can travel miles without a problem twisted or not (as noise,
>cross talk etc is not really noticable) whereas data cable is
>twisted

Plain old voice can probably go for feet, and maybe even yards,
on untwisted pairs.  But that long cable from your house to the
telephone company is twisted pair, otherwise you would hear so
much 60Hz power line hum that talking on the phone would be
impossible.

>to reduce interference between the conductors on the same
>circuit and genrally screened from adjacent circuits especially
>at high data rates.

Twisted pair transmission lines provide noise immunity compared
to a non-twisted pair (generally in excess of 60 dB of immunity,
which is a _huge_ amount).  "Noise" is any externally induced
signal that appears in the output which was not at the input,
hence in this case that includes crosstalk from other cable
pairs, power line hum, ignition noise from nearby motor vehicles
with high voltage spark ignition systems, and just a myriad of
other noise sources.

However, for short runs it is OK to use non-twisted pair.  The
trick is knowing how short!  Basically as the frequency of the
signal goes up, the amount of twist in the cable needs to be
higher, or the cable needs to be shorter.  For cables with no
twist at all we might consider 6 feet to be short enough for
voice (i.e., modem) use, though up to 12 feet might work just
fine.  Some people actually use 25 foot lengths of untwisted
cable with no problems.  Telephone companies sometimes hang 100
feet or more of untwisted cable from a pole to a house and get
away with it.  But each added foot increases the chances that
noise will be picked up.  (If the line is not near any source of
noise, there will be none...  until next year when someone
installs a noise generator right next to that line which hasn't
been a problem!)

Basically, for voice a few feet of untwisted pair will almost
always work, and a few yards might.  For higher frequency
signals, such as T1 digital telephony signals, a few inches is
usually fine and a few feet might work without being twisted.
For even higher frequencies, such as 10baseT ethernet a few
fractions of an inch are usually OK, and a few inches might even
work.

Various different kinds of cables are used for different signals
and for different environments.  A higher quality cable than is
necessary may be used without harm (other than costing more).
CAT5 cable has low loss and a very high twist, making it
suitable for very high frequency data signals.  It also works
great for plain old voice lines too!  Since it barely costs more
in the quantities used to wire telephone lines in a home, it is
commonly recommended for that purpose even though it represents
a several times overkill.

  Floyd



>The Old Bear <oldbear@arctos.com> wrote:
>>
>> This got me to thinking about the very old telephone
>> installations which were done up through the 1950s.  The
>> installers used to use three-wire twisted cable with no outer
>> jacket and the conductors designated by a colored thread
>> underneath the insulation.  This stuff came in dark brown or
>> "ivory" to match the dark or light colored woodwork in most
>> American homes and offices.  The installer neatly tacked this
>> to baseboards and door frames using large head tacks in
>> matching color.
>>
>> It was not until the late 1950s or early 1960s that the
>> telcos started using the plastic jacket station wire with the
>> familiar green red yellow and black inner wires.  This was
>> installed with the now-standard staple gun which fires a
>> U-shaped cable tack.
>>
>> Now we're back to using twisted pair inside a vinyl jacket in
>> the form of Cat 5 or its earlier UTP cousins.
>>
>> What happened here?  Why did the telcos stop using twisted
>> wire for inside stations?  Was it a manufacturing limitation
>> of the station wire or was it determined that twisted
>> conductors were less important for shorter interior runs?
>>
>> Any thoughts?

--
Floyd L. Davidson                          floyd@barrow.com
Ukpeagvik (Barrow, Alaska)

From: floyd@ptialaska.net (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Maximal length of a serial (V.24/RS232) cable?
Date: 8 Dec 1999 15:05:41 GMT

budgie  <budgie@cantech.nowhere.net.au> wrote:
> floyd@ptialaska.net (Floyd Davidson) wrote:
>
>>>>  4) Twisting signal pairs will reduce the distance.
...
>>It will never improve noise rejection.  Instead it will
>>degrade it.  The two twisted wires are coupled more tightly
>>than two wires in a non-twisted cable (which would also tend
>>to use thicker insulation specifically to separate the
>>conductors farther, as opposed to twisted pairs where closer
>>proximity is made up for by noise immunity due to common mode
>>rejection provided by balanced circuits).
>
>there you go again, comparing apples with polar bears.  Now
>watch my lips and say slowly after me - For a given cable
>construction.  Got it now?

Here's an apple for ya, pard,

It will *never* improve noise rejection.  Instead it _will_
degrade it.

If you don't understand that, it will be a bear until you do.

Could you substantiate your earlier statement that "it WILL
improve rejection of noise somewhat, even though it is still an
unbalanced TL."  What noise rejection improvement will there be?
Please describe the mechanism.

You also stated, "It isn't going to increase the pair cap".  Can
you enlighten us about how two wires can be twisted together and
not increase the capacitance between them?

Of course, "for any given cable construction"...  but it is
reasonable to assume cables that are actually available for use.
For example, to avoid induction between circuits, cables
intended to be unbalanced transmission lines are constructed to
maintain a relatively large physical spacing between any two
conductors.  On the other hand, to accomplish the same results,
construction of a twisted pair cable will provide different
twists for different pairs to reduce induction.  Twisted pair
cable, to counter the physical bulk increase caused by the
twist, will use thinner insulation, which is effective because
common mode rejection is the primary noise reduction mechanism
for balanced circuits and the capacitance between conductors
does not cause the significant increase in noise which would be
seen with a non-balanced circuit.

Of course, attempting to match twisted pair to non-balanced
circuits or non-twisted pair to balanced circuits negates all of
the optimizations in cable construction.  It is not recommended.

  Floyd


--
Floyd L. Davidson                          floyd@barrow.com
Ukpeagvik (Barrow, Alaska)

From: floyd@ptialaska.net (Floyd Davidson)
Newsgroups: comp.dcom.modems
Subject: Re: Maximal length of a serial (V.24/RS232) cable?
Date: 9 Dec 1999 12:37:46 GMT

budgie  <budgie@cantech.nowhere.net.au> wrote:
>floyd@ptialaska.net (Floyd Davidson) wrote:

You seem to have clipped out the first question, which was
posed in relation to using single ended RS-232 circuits on
twisted pair cable:

   "Could you substantiate your earlier statement that "it
    WILL improve rejection of noise somewhat, even though
    it is still an unbalanced TL."  What noise rejection
    improvement will there be?  Please describe the
    mechanism."

Please do describe the mechanism, or admit there is none.

>>You also stated, "It isn't going to increase the pair cap".  Can
>>you enlighten us about how two wires can be twisted together and
>>not increase the capacitance between them?
>
>You first - you explain how twisting two wires will increase the
>coupling between them.  Any transmission line model you use will
>reject your theory.

Simple fact: it happens.

If you twist the wires it takes more wire to go the same
distance, with a higher twist rate requiring more original
conductor length to obtain the same twisted pair length.  That
alone will increase the coupling for any given distance of cable
when comparing two conductor non-twisted pair cable only.  But
we are talking about multi-pair cable, and in that case there is
even more reason.  If the two conductors are not twisted, they
should be constructed such that proximity to all pairs in the
cable becomes roughly equal over some minimum distance.  With
twisted pair cable, each pair is treated the same, which is why
the bundle is swirled and each pair has a different twist.  It
basically means two conductors should not lay next to each other
for the entire distance of the cable.  But with twisted pair
that is exactly what happens for those two conductors.  When
they are used with a balanced circuit that is beneficial, but
with a different single ended circuit on each conductor it is
clearly detrimental.

Hence the coupling between pairs of a twisted pair are higher
than between either of those conductors and any other conductor
in the cable, as originally stated.

If you would like to explain how you can twist two pairs
together and not cause that effect, be my guest.


>>Twisted pair
>>cable, to counter the physical bulk increase caused by the
>>twist, will use thinner insulation.
>
>You may find that "bulk" is not the real driving force behind multiple
>twisted pair cables using thinner dielectric.

But you aren't going to grace us with your concept of what that
might be???  In fact, physical size is in fact the primary benefit
of thinner insulation.  Every other effect can be accomplished in
other ways, but not a reduction in size.

>> which is effective because
>>common mode rejection is the primary noise reduction mechanism
>>for balanced circuits
>
>right
>
>> and the capacitance between conductors
>>does not cause the significant increase in noise which would be
>>seen with a non-balanced circuit.
>
>Exposed to an external noise source (where else?) why would the pair
>cap cause ANY increase in noise?

If you put the receive data signal on the tip and the transmit
data signal on the ring of a twisted pair, each signal is
nothing but noise to the other.  If those signals are placed on
conductors with less coupling, there will be less "noise" for
each signal.  That is true because with the exception of clock
signal lines, which are usually not used with asynchronous
RS-232 anyway, the two data lines are the most active as far as
the rate of signal transitions.

>>Of course, attempting to match twisted pair to non-balanced
>>circuits or non-twisted pair to balanced circuits negates all of
>>the optimizations in cable construction.  It is not recommended.
>
>That optimisation was for the intended purpose indeed, but that
>doesn't preclude cable selection for a non-designed purpose, nor does
>it automatically render that selection/use either invalid, unworkable
>or unwise.

It does if one doesn't have an extremely broad based understanding
of cables.  Lacking that, using cable designed for the specific
purpose will avoid the type of errors your understanding of cable
use and abuse will cause.

>  You must sometime in your career think outside the
>box/envelope.

A lot of my career has been spent tracking down the cause of and
correcting unfortunate mistakes made by people who didn't know
where the box/envelope was, much less have any idea when their
thinking was either inside or outside of it.  Unfortunately,
your perception of cable functionality fits that description,
and tends to leave those kinds of problems for trouble shooters
to correct at some later date when small problems have grown to
significant proportions.

  Floyd




--
Floyd L. Davidson                          floyd@barrow.com
Ukpeagvik (Barrow, Alaska)