For the exchange of information about telephones and telephone circuitry.

99 responses total.

I'm considering having a second telephone line installed, and need to
trace where the 5 interior lines on the block go. Considering that they
are "hot", or could ring, I'd like a way to identify each line without
unwiring it from the block before I'm ready to move it. For example, a
*magnetic compass* might indicate the 50 - 100 ma off-hook current in a
line. Does anyone have any other ideas for such a passive detector, which
is sensitive enough? The block is up in the joists, so even using a compass
would be very awkward: a display on a multimeter would be better.

I've never heard of a 5-wire block, however, phone lines are
pairs of wires, do you mean 5-pr on a block? 
  
IN any event, here is a -48VDC on a hot pair, when the voltmeter is
connected, red/black (most often), the red lead is tip and the black
lead is ring (and the v-meter reads -48). Notto worry too  much
about polarity these days though, almost all stuf is polarity independent.
  
If you are pretty close to downtown the DC voltage on a open circuit
could be as high as -56 VDC, but when there is a load onteh line (when
you lift the receiver to hear a dialtone) the voltage drops to the
5-8 VDC range, with 20-100 mA loop current.
  
Does this help?

TS, I don't know who wired *your* house, but the telephone signal should
be on the red/*green* pair, not red/black. Red is ring and green is tip.
Black and yellow are either used for a second line, or more usually,
to carry 18VAC from a transformer to power the lights in an old style
princess phone.

Phone trivia: Referring to the 2 phone signals as "tip" and "ring" comes
from the fact that old-style operator panels used to function by the
operator physically plugging circuits together by hand. The pysical
connection was done with 1/4" phone plugs/jacks. These are the same
kinds of plugs that you see today on Full-sized stereo headphones. That's
why they are called "phone" plugs even though most people only see
them associated with stereo gear these days. Those plugs have a connection
at the "tip", a small black insulator, and then a long metal sleve
connection that is refered to as the "ring". It's a common mistake to
assume that the wire called "ring" carries the signal that rings the phone.
In accuality, *both* wires carry *all* signals.

If you come across a functioning 2-wire line, and get dial tone, but don't
know whose line it is, dial 0 and ask the operator for a "drop wire ident-
ification."  That's the jargon to obtain the phone number.

Or you can just dial 311 and it will tell you what number it is automatically.

Yes, TS, I should have called them "pairs", not "lines". Boy, these
purists.... And I know about the ca. 50 volts, and green-tip-(+) and
the red-ring-(-), and the 90 V AC ringer zappo. So, I have 5 pairs
on a single drop, which go to various jacks around the house. I want
to find out which is which, without unwiring the block (yet). If there
is a magnetic field probe I could clip to one wire of a pair, I could
go around the house taking phones off-hook, or plugging in a dummy
load, to see which jacks are serviced by that pair. But is there such
a probe? (I think my alternative is to wire all the pairs to terminal
blocks now, open them one at a time to find the corresponding jacks,
and jumper them later for two incoming drops. But this is no plebian...)

Rane, there are various ways of tracing the wires, but every way I can think
of entails dsiconecting the wires from the block first. What signal you
tried to place on a given pair, it would back propagate to *all* the pairs
unless you disconnect them. The only way I can think of tracing a given pair
without disconnecting it from the block, is to give one pair a real hard
tug, and then see which phone moves. :-)

Heh! Greg, if I can detect the *current* in a pair, I can separate them
without disconnecting, since this doesn't flow into the other pairs. How
small a current in a wire could a Hall Effect transistor be used to
detect?

Don't know. I know Hall effect devices can be very sensitive, but I don't
have much experience with them. OTOH, it seems to me that building such a
device would be *alot* more work than simply loosening 2 nuts and removing the
5 pairs from the block. This isn't really a whole lot of work, in fact, it's
pretty trivial, unless of course you are trying to solve this as a mental
exercise and not the real thing.

In the long run, I'd have lot of uses for a flux probe. In this case, there
would be a lot of wiring and unwiring, as each pair is traced, and with the
probe, my daughter could watch the meter in the basement, and let me know
by radio when I find the associated jacks. But Greg does have a point: I
could have run around a lot, and gotten it done, even though the telephones
would be out of service for a while, in the time I've spent discussing how
I can do it simply, here. Of course, though, if we found there were such
a probe, everyone else (reading this) would be able to save lots of time.
Look to the future (or, "never do today what you can put off for tomorrow").

"Never put off till tomorrow anything you can get out of doing entirely."

Ummm, what I said, correctly, is "when the voltmeter is connected,
red/black (most often), the red lead is tip and the blacklead is ring
 (and the v-meter reads -48)."  Most VOLTMETERs use red/black for
polarity identification and since the tip is negative wrt ring, and
tip is above ring on a punchdown block, the VOLTMETER leads, correctly
connected (red/black) will produce a -48 VDC on the VOLTMETER.
  
There is NO red/green pair coming INTO the demarcation point. There is
a red/green/black/yellow two-pair INSIDE WIRING  - AFTER the demarc,
unless 3-pr/6-pr/12-pr/25-pr is run instead, in which case there STILL
is NO red/green wiring until the outlet is installed, and the red/green
goes from the block to the rj-11/rj-14, etc jack.
  
I've punched 300-pr cables and their ain't no red/green anywhere.
  
Further along the trivia line, the reason the polarity was reversed
in the first place, and tip became (-), was because the operators
would take their phone plugs and run the tip along the patch panel
to see which one "popped" intheir headsets - thus indicating the hole
to plug into. Too often, before the polarity as reversed, the  tip <at
+48V> would scrape across ground (!!!) and zap fuses and sprks would
fly, and all sorts of mean, nasty, ugly things (apologies to Arlo)
would result.
 
since the operators were only looking for a "pop" from the panel, the
panel stayed at ground potential (same as "tip") and the ring was
put at +48. And ring was Insulated from the panel. 
  
So, when the operators scraped across all the ring contacts they
would get their requisite "pops,' but if the tip scratched across
the metal panel itself ...... no potential diference, no sparks,
no blown fuses ... !
  
And linemen had to reverse their either their probes (RED/BLACK) when
teh checked for battery, ... or ...  get used to -48v when there
was battery voltage. The latter became the standard for historical
reasons.
 
And until the late 60's early 70's, polarity was critical for your
phone to work. Eventhe early TouchTone (r) phones were polarity
sensitive. 
  
If you couldn't "break the dialtone," the fix was to reverse the
leads in the wall jack. 
  
Oh., rcurl, there is a toner and probe gizmo-combo that traces
wires perfectly. You can pull a hot pair out of a thousand-pr
cable in about 5 seconds. Yes, I have the gizmo-combo, and it
is on loan at the moment.
  
I'l try to get it back, rsn, I need it too.

Rane.  How about getting hold of an old clamp-on AC current probe and 
then putting an O-scope on the coil in the probe.  Then put the probe
on any one wire of a pair and run around the house with a phone and 
plugging it into each of the, otherwise empty, connection boxes while
someone is ringing your line.  As soon as the phone is connected to 
the pair with the probe on it, you will see a 20 Hz sine wave on the
scope.  I suspect a phone with a high ringger current would work the
best.  Many of the electronic phones with "crickett" ringgers draw
very little current.
 
(I also have all the parts to build a flux gate magnometer sensitive
enough to detect the earth's magnetic field.  It doesn't use Hall
effect devices, though I understand some are sensitive enough to do
that.)

BTW.  Is there a No. I can dial that will ring the phone I'm calling
from?  I remember being able to do this about 20 years ago.

Going in reverse: I also used to know the "ringback" number in some other
exchanges, but don't know it here. The idea to detect the ring is to
be considered, if I could find a clamp-on AC probe. Its not quite as
"passive", or quiet, as looking for the off-hook DC current, though.

TS, your information differs from my sources, which say that tip is
positive and on the green wire, and ring is negative and on the red
wire. Maybe the confusion comes from the tip (green) being at *ground*
in the on-hook state (but still positive w.r.t. ring (red)). My sources
are the ARRL Handbook, and Sokolowski's _The Talking Telephone_ (Tab
Books, 1990 - not a particularly good source, but useful). We better
get this straightened out before we carry this discussion further.

 .... and teh voltage reading is NEGATIVE 48 V, because when the 
voltmeter red/positive probe is on tip and teh voltmeter black/negative
probe is on the ring .... it is reading a +negative+ from the 
phone company battery which has the negative pole of the battery connected
to the tip and the positive pole of the phone company battery 
connected to the ring.
  
It's really not a consequential difference at this point since polarity
is functionally irrelevent with today's polarity-insensitive devices.
  
As I already said I have a toner and probe that finds wires, standard
issue stuff for just what is needed.
  

I presume the toner puts a signal on a line, and the probe detects it.
However my five pairs are connected in parallel, so they could not be
distinguished. (Unless I disconnect them, which is what I wanted to delay.)

Re polarities, here is what it says in 1991 ARRL Handbook, p. 28-23:

"The equipment leads are called "tip" and "ring", or just T and R. For the
idle or on-hook state, the tip (positive) lead (green) is grounded. The
CO ringing generator and the -48 V dc supply connect to the ring lead (red)."

Say, were you born with reverse polarity, TS? ;-)

Hmmm, ok, I see the problem here. You are both right. :-)
The tip lead *is*  positive, but they decided to use a positive ground
system, that's the backwards element. Most electronis use a negative
ground system.

There are TWO errors, then, in the AARL Handbook. NEITHER lead is
grounded - phone lines are a floating, balanced pair, equally  isolated
from ground (earth). 
  
If you want to blow up your telephone service, ground one wire
of your pair and try to use that line.  There should be an
available ground connection available on the block. You can 
either unearth the facts with direct experimentation, or pay 
Bell $60/hr to stop by and unearth it for you. 
  
Ummm, all five of your +drops+ are connected in parallel ???? Or,
are there 5 in-house pairs connected to one Bell drop-wire ?
The latter sounds more plausible, actually.
  
And yes, if the second hookup is true, then you do have to
disconnect the in-house wiring at the block to isolate a
specific in-house pair running to a specific jack. No biggie,
standard practice. However if the black/yellow wires are NOT
connected to anything at the block, THEN you use this spare
pair for trace identification, no biggie again, and no dis-
connection necessary either.
  

gregc slipped in with #17. He must edit fasterthanido.

I have taken TS's challenge, and measured the voltages on the drop to my
house: with everything "on hook", one side is *exactly* at ground
potential (and has the green wires), and the other side is at -50 V (and
has the red wires). ARRL is sustained! 

However, off hook, the difference is about 7 V, and both sides are
negative, at ca. -31 and -37 w.r.t ground. This fits with the *source*
impedance being approximately balanced. 

That was a good suggestion, to follow the black/yellow pairs! That is, if
they are connected at the jacks. In fact, it may solve the whole problem.
Now, why didn't I think of that? I was looking at the hanging black/yellow
wires, but it just didn't occur to me. 

(If the green wire really is grounded and is really connected to the tip
and the red wire is at a - potential... That would put the operators panel
, of era gone bye, at some other potential besides ground, were it made of
something conductive.  Even if non-conductive, the exposed portion of the
jack and significant portion of the plug would be something besides ground.
I wonder what Rane would measure, w.r.t.gnd., when the line rings?)

Sorry to slip that one in on you, rcurl, ..... but wrt measuring the
voltages: yes, one of the wires shows NO potential difference to ground.
  
However, that does nOT mean that they are the +same point+, ahem.
 
What it means is that (way back there when I explained WHY the
poliarities were reversed) the tip striking ground will NOT blow
a fuse, there is no potential difference. 
  
If they are the same point, then an ohmic measurement is the validity
check. I think that it takes a megger to measure that though. Phone
pairs are balanced, isolated and floating - not unbalanced to ground.
 
Besides, my test was not to measure teh potential differnce, but
rather to ground whichever wire and try to make a phone call, you
changed the test and got an accurate measurement with a VERY misleading
conclusion therefrom. 
  
With the phone off-hook, the battery drops to about minus 5 to 9 volts, and
your difference of (-37) - (-31) = -6 VDC, which is just fine, tip to ring.
  
Since a phone pair is a loop (thus, "loop current" of 20-100 mA) and
the phone instrument is in series with that loop, you can see that
the source impedance is rather high, proportionately. 
  
It is through padding the source impedance that telco adjusts loop
current to compensate for 500 or 25,000 feet of series copper wire
between you adn the switch.
  
ARRL has not yet been vindicated. Great organization, I was a member
for quite a while, but that particular section .... well, they're human.

~p
  
Oh, it doesn't matter whether the bk/yl is connected in the jack,
teh wires are there! Sure, you might have to open a jack in some
room womewhere and connect something to bk/yl in order to trace,
but, it'll work with no problem (unless either bk or yl have been
used to "replace" a goofy rd or gn wire at some earlier time)
  
Keep usinformed please.

I did a little reading about telephone systems, and learned that the
"tip-ring" thing goes *way* back, but it is mostly the terminology,
not the hardware, that remains. What my source called the "modern"
common-battery exchange uses a 3-wire plug, with the "tip" positive,
the "middle" collar the "ring", and negative, and the plug sheath (the
old ring) is used only for local signaling purposes (off-hook light,
in particular). There are many variations on this, even with 4-wire
jacks. And yes, the jacks are/were insulated from the panel.

All of the telephone system circuit diagrams in all of my sources show
the CO *battery* positive electrically grounded. My measurement on the
drop to my house agrees with this. In what I've posted here, I have
only said that with the phone *on hook*, the tip/green/positive wire
is at ground potential. However the local circuit "floats" on the CO
balanced line impedances. Do we agree on that?

Without question, the " 'tip-ring' thing goes *way* back," let me 
emphasize, way, WAY, W A Y back! Mu guess (guess) is that by the
time of hte third installation, the polarity was reversed. 
  
Thank you, "the jacks are/were insulated from the panel." 
  
Please recognize that there are TWO types of ground - I'll try
a freehand ascii representation, but who knows .......?
                    |  
signal ground       -
                    _
  
That supposed to be the three decreasing width horizontal lines 

                       |
                      __ 
and   earth ground   /// 
  

As to the 3 & 4-wire plugs, perhaps both grounds are included, signal
and earth. However, tip is still negative wrt ring, by 48 VDC, and the
phone pair is floating, isolated, balanced, ABOVE earth ground.
  
Wrt "the CO *battery* positive electrically grounded" that is correct
ONLY in respect to +signal ground+, not +earth+ ground, (see previous
test suggestions not taken) even though the potential difference, measured,
between the two IS  zero volts, dc, the tip does float above ground
in ohmic measurements (no answer yet, nor tests made - probably no megger
available). 
  
You are not goint to blow Bells line by grounding one or the other
of the balanced pair - they are used to that happenning accidently
and nothing untoward will result, trust me, but you won't be able
to complete a call if you earth-ground either signal wire.
  
We can agree that tip (signal ground) "is at ground potential," 
even as it is ohmically    W A Y   above earth ground.  And we
can agree that "the local circuit "floats" on the CO
 balanced line impedances," but we don't agree, yet, that tip is
negative wrt ring, or that proper polarity on the VOLTMETER shows
a -48VDC CO battery.  
  
Ummmmm, you introduced a new acronym, CO, which means Central 
Office (CO) and is interchangable with "the switch"in most situations.
in most situations.

  

Positive.

We had an interesting problem not too long ago.  We got a dial tone and
could call out without problem and the audio quality was as good as ever.
People calling in, however, got two short rings and then got a msg. from
the CO saying that the call could not be completed and to try again.  If
we caught it after the first short ring, we could talk to the other party.
Sometimes it would ring normally, etc.

I connected a phone to the box outside the house and the problem was gone.
I unplugged *everything* plugged into our many phone jacks.  The problem
persisted.  I disconnected all the red wires to the junction in the basement
and had someone ring our line.  I monitored the rings with a phone clipped
to the junction points and connected one red wire at a time till the problem
came back.  After pulling all the jacks on that line, I determined that a
10 ft. length of plenum phone wire had developed an low resistance path
between the red and green wires!  The wires closest to the last jack in 
the line also looked a bit stressed, almost melted.

Why?  Can lightning do this sort of dammage?  How do I replace this length
of wire, burried behind the wall and stapeled to the 2 x 4's?

What is "plenum phone wire"? (Can you drop a cord or run a snake between
the end points? I dropped a phone wire down a cold-air return plenum,
after I drilled into it at each end.)

Plenum wiring comes in all shapes, sizes, and kinds. You can get plenum phone
wire and plenum coax for running ethernet. Plenum wiring is designed to be
run *inside* of heating ducts to take advantage of that natural pathway between
rooms. It is rated for higher temperatures and as such, is ussually made out
of teflon and costs *alot* more than normal wiring of the same type.

Yeh, the specs on that stuff changed a few years ago and the insulation
on it is *&*&^%(#$^&$(*@(#&% to remove.
  Oh, it does its job alright, and it is "safer" than normal stuff,
but grrrrrrRRRRRRR, working with it is horrid.
  
To my mind it is a radicaly unecessary "improvement," dictated by gummint
to featherthe nests of some company that wanted a lock on a market they
couldn;t penetrate without gummint dictates to an otherwise competitive
marketplace.
  

Oh, on that wiring problem, the only fix is to splice and/or route around
that section of the wire  - - - - OR ....... use the  bk/yl pair for that
particular run. That's why they are there in the first place, not for a second
phone, although that's convenient after the fact.  And you probably only
have to use one of the extra wires in a pair with one of hte original
rd/gn pair. 
  
Nice  job of trouble shooting  .

I've got another phone problem.  When we got a second line at my house a
few months ago, we routed it through the black/yellow pair on the existing
wiring so we wouldn't have to rewire the house for it.  Now, when both
lines are in use, people on the two lines in my house can hear eachother
very quietly.  The people on the other ends of the conversations, however,
don't hear anything out of the ordinary.

It's called "crosstalk". It's caused by inductive coupling. Cheap home 4wire
(notice I didn't say "2 pair"? That's because it's not, it's "4 wire") with
red/green/yellow/black will do that to you. They are just simple straight
runs inside with nothing to isolate them the way a twisted pair would.
The short answer is: "That's the breaks". If you don't like it, run new wire.

Try switching pairs: instead of g-r/y-b, try g-y/r-b or g-b/y-r. One
will have the *least* crosstalk.

Hmm, you *might* get lucky that way, but I wouldn't hold my breath.

I wouldn't either. I'm hoping Steve will try it, and let us know ;-).

<sigh>
I'm wishing I had time to try anything on this right now.  I'm thinking
the best thing to do on to have the electrician my parents are about to
have do some other things rewire the phone line as well.  Otherwise maybe
I can get to it in a few months, but I don't think anybody around here
wants to wait that long.

Another option, probably better, is to reverse *one* pair at about the
middle of the run. (If polarity matters, it can be reversed back at either
end.) One could also build a inductive coupler between the two pairs,
with the phase reversed from the present coupling - might be a little
hard to "tune" it for bests crosstalk cancellation, however. 

Good note, gregc, about the 4-wire, not 2-pr, for rd/gn/bk/yl  typical
home wiring. But in a house that is a pretty massive amountof inductive
coupling taking place in what would seem to be some mighty short runs
of wire.
  
gregc's short answer is pretty much right, rewire or live with it. It
might be possible to improve things the way rcurl suggested +if+ there
is a single run throughout the house and not a bunchof "home runs" to
some of the jacks (or +mostly+ a single run). 
  
Polaritiy shouldn't matter too much these days, it's almost an artifact
unless we are trying to use older equipment. I'd leave all the rd/gn
pairs alone, and for a first reversal, switch the bk/yl at the demarc,
somewhere in the basement probably. That may do the trick all by itself.
The crosstalk should +change+ anyway. Then if more is neeeded, find
an approximate midpoint inthe wire run, and flip the bk/yl as it
heads out to the rest of the house.

Any word yet on these ideas?

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss