💾 Archived View for gemini.spam.works › mirrors › textfiles › magazines › PRIVATELINE › privateline-… captured on 2022-06-12 at 14:04:37.
-=-=-=-=-=-=-
Volume 1, Number 2 -- private line -- a journal of inquiry into
the telephone system
Table of Contents
General Information
I. Editorial Page
II. Update and Corrections
III. Telco Payphone Basics, Part II
IV. The Coin First Coin Line
V. The Dial Tone First Coin Line
VI. Tip, Ground and Ring Explained
VII. California Cell Fraud Law
----------------------------------------------------
1. General Info on private line: ISSN No. 1077-3487
A. private line is published six times a year by Tom Farley.
Copyright (c) 1994 It runs 24 to 28 pages. It's done in black and white.
B. Subscriptions: $24 a year for subscriber's in the U.S. $31 to
Canada or Mexico. $44 overseas. Mailed first class or equivalent.
(1) Make checks or money orders payable in US funds to private
line.
(2) Back issues are five dollars apiece.
(3) A sample is four dollars.
(4) The mailing list is not available to anyone but me.
C. Mailing address: 5150 Fair Oaks Blvd. #101-348, Carmichael, CA
95608
D. e-mail address: privateline@delphi.com
E. Phone numbers: (916) 488-4231 Voice (916) 978-0810 FAX
F. Submissions: Go for it! Anything semi-technical is strongly
encouraged. I don't run any personality pieces. I pay with
subscriptions.
G. Ads: Yes, I'm taking electronic related ads. A full page is
$75.00, a half page $37.50 and a quarter $18.75. Subscribers get
free classified ads of 25 words or less.
------------------------------------------------------------------
The front cover illustration is of an line finder rack for a step
by step exchange. The photo is from a 1943 Popular Mechanics
Yearbook. The caption reads, "In an automatic telephone
exchange many sets of selectors are required, and when a call is
made a vacant line must be found automatically. This apparatus
finds one within a few seconds."
----------------------------------------------------
I. EDITORIAL PAGE
Sing Ho For The Life of A Zine; On Explaining the Unexplainable;
Vegas Bound
Welcome to the second issue of private line. I hope you enjoy it.
The first issue was well received and I am encouraged. I am now
sending samples to magazine wholesalers. I may find a nation wide
distributor by the October issue. That would lead to more
readers, more comments and more information. Until private line is
more reader driven, however, you are stuck with me. And that means
fundamentals. I finish the discussion of Telco pay phone
basics in this issue. The mystery of ground start is examined as
well as the different arrangements of tip, ground and ring.
These explanations are my best attempt to make sense out of
seemingly nonsensical ideas. They are starting points for a
conversation to begin. They are not The Last Word. I worry
terribly, however, about my writing. It seems that I have two poor
choices. I can provide a precise answer that is too complicated to
understand or a simple one that is too general to be accurate.
So, something in the middle is presented instead. People have
been very forgiving. They appreciate the effort that it takes to
get a discussion going. I appreciate that consideration.
A local scanning article will be featured in the next issue of
private line. People always want interesting numbers to call. The
problem is that such numbers are often of regional interest only.
I will, therefore, describe some ways that everyone can use to
search for test numbers, voice mail boxes, governmental telephone
system numbers and so on. This article will be done with the help
of an Oakland hacker. It will use numbers from the 415, 510, 707
and 916 area codes as examples. People in the Bay Area will be
able to use the numbers given, but people everywhere will be able
to use the techniques. It will even have some worksheets to help
you systematically explore a prefix and an area code.
For now, though, it's off to Def Con II in Las Vegas. A
gathering of the clan is taking place in the burning hot desert.
It might be a hacker's Woodstock or a recreation of the last
scene in The Stand. I don't know. But I'm going. I can't afford
the trip. But I'm going. My car may not make it. But I'm going.
Next month and next month's finances will have to take care of
themselves. For all the right and wrong reasons, people are now
going to Las Vegas. And so am I. I'll tell you what happens.
Thank you,
Tom Farley
Carmichael, California
II. UPDATES AND CORRECTIONS
This update column will be a regular part of private line.
Material comes from the last year of Telephony.
The local switch
1. I didn't write much about central office switches in the last
issue. I thought others had done a better, more complete job so I
spent time writing about CDO's and remotes. There are, however,
some new CO switches coming on line.
An article in early 1994 stated that NEC was one of only two
vendors with a large, ATM based central office switch that is
ready to be installed. Fujitsu is apparently the other vendor.
They did not state, however, the names of the switches. NYNEX was
reported in a later article to be installing Fujitsu's Fetex-150
broadband switch for a field trial. Broadband does means ATM. Bell
South is also playing with the Fetex-150. They are going
into North Carolina and soon to Atlanta. But Telephony doesn't
state whether the 150 is the switch that was referred to earlier.
If NEC or Fujitsu does deliver a CO then they may offer some sort
of coin line service.
As I understand it, ATM or asynchronous transmission is a way to
handle many kinds of information fairly quickly. Video services,
in particular, benefit from ATM. The No. 5ESS, by comparison, is
a time division switch. It handles most data files and voice
traffic in a faster way than ATM. But it can't handle multi-media
or video as well. Read more about ATM in the June IEEE Spectrum.
Latest upgrade to the No. 5 , by the way, is apparently the
5E9(1), which went to customers in November, 1993. This now
provides the so-called National ISDN-2 capabilities. NYNEX
is now able to offer services such as residential voice dialing
service and its phonesmart caller ID and call trace. Lovely.
As of April 11, 1994, 72% of NYNEX lines were served by digital
switches. Half of the remaining lines will eventually be served by
5ESS's or NT S/DMS SuperNode switches. The company expects its
network to be 100% digital by 1998. 18% of its lines, therefore,
are still served by electronic or analog switches. That's fairly
large considering that NYNEX, the Darth Vader of the baby bells,
is so well financed. You can tell by this that smaller markets
will have a far higher percentage of older equipment.
2. The Remote Switching System
Current practice calls a digital remote switch a module. These
correspond to the CO switch. For example, when you buy a central
office switch you get a module to go along with it if you need a
remote. An example would be the No. 5A Remote Switching Module to
go along with the No. 5ESS. Remote switching modules are also
known as RSMs. Siemens Stromberg Carlson also makes a module for
its central office EWSD switch. This switch and its attendant
remotes have been installed recently in Puerto Rico. An
independent Telco named Alltel has also bought an EWSD switch and
one remote unit to serve rural Eclectic, Alabama. It might be
interesting to call Eclectic sometime to hear the new switch in
town, possibly the only one of its kind in America.
I referred to Northern Telecom's DMS-10 as a remote switch and a
collection of components. Not exactly. The Digital Matrix Switch-
10 is primarily a switch for rural use. Any components that go
with it can be thought of as accessories and not a part of the
switch itself. I mentioned several times that a low volume of
calls makes rural service expensive, along with the higher costs
of building and maintaining the local loop. This low volume
works against upgrading since revenue is low. A way around the
problem is by offering a switch like the DMS-10. It may generate
greater revenue in rural areas by providing services that step by
step offices can not. Things such as call forwarding and call
waiting. Still, are there that many people that need call waiting
in Gabbs, Nevada?
The term Community Dial Office is falling out of favor. CDO's
refer to older equipment rather than an operating method. Remotes
and modules, though, are still dependent on a larger switch. Even
basic terms are being redefined. Pac Bell doesn't refer to central
offices anymore. They are, instead, a dial tone producing end
office.
3. The subscriber loop network
How expensive rural service can be is demonstrated by a US West
(the old Mountain Bell) field trial. 35 miles from Jackson,
Wyoming are 40 customers who live near the town of Bondurant. They
are now being supplied phone service by satellite. Subscriber
lines terminate at two small satellite earth stations which then
connects the customer to US West's switching center in Jackson.
U.S. West wants to see if this is less expensive than installing
fiber or cable out to these homes, many of which have party line
service or no service at all. Now, that's expensive.
4. Coin deposit tones
I doubted last issue that operators listened to tones anymore. I
speculated that the CO probably listens for the tones instead and
sends the amount on a data circuit to the TSPS console. Such
nonsense. An attendee of the last San Francisco 2600 meeting
gently pointed out the obvious fact that a voice channel exists
when you are talking with the operator. Of course. Yes, the amount
of money does totalize on the console but you are talking with the
operator at the same time. If they hear a bogus tone then they'll
do something about it. I don't know what I was thinking of when I
wrote that.
5. The GTE RTSS phone
This phone interfaced with many other pieces of equipment.
Somewhere in Kansas wrote in the Summer issue of 2600 that KG and
KY prefixed machines were discussed in a Scientific American
article with photos a few years ago. I looked in Carl, Uncover,
Inspec and Current Contents for it. Nothing. I then looked on the
shelves. The last index S.A. published was in 1978. Nothing. The
article probably lies, therefore, between 1979 and about 1988.
I'll keep looking. AT&T Technology, however, does have an article
on STU III. This article came out in 1989 in volume four. The page
numbers are 36 to 40. STU III is apparently a crypto product that
AT&T makes which can interface with the GTE RTSS. The magazine
was missing when I went to check it out. And so it goes.
6. Interesting numbers
The ANAC for parts of 415 has been submitted as 760-7760 and 760-
7761. This agrees with the old ANAC list floating about the
Internet. 924--0036 may be a loop disconnect number for 415. In
916, 440-1212 gets you a second dial tone. If you dial additional
numbers you may get a long distance operator who doesn't identify
her company when she comes on the line. 484-0001 is a strange one.
No connection is made. I don't think this is a quiet termination
test number. I usually hear a connection and then silence with
those. This one never makes a connection. Some Pac Bell numbers to
modem into in the 916 are 481-0022 and 484-0022. Possibly 481-
0078. The third issue of private line will be about local
scanning. There will be many, many more numbers.
III. TELCO PAYPHONE BASICS, PART 2
The Subscriber Loop Network
7. We looked at the subscriber loop network briefly in the June
issue. As you may recall, the network is made up of all those
elements which constitute the local loop. This includes
the twisted pairs that run to each phone, the local switch,
overhead cable, amplifiers, multiplexers and so on. In other
words, all the elements of switching and transmission. Let's look
at what I think is the most confusing part of the subscriber loop.
Circuits and the subscriber loop
8. We know that a circuit is a connection with the central
office. It carries a call. A circuit exists through the twisted
pair or in a channel within a wire to the central office. A
circuit can also be a connection between offices, between
equipment or within the equipment itself. These circuits may or
may not carry a conversation. The word circuit is also used to
describe a particular way that the local loop is arranged. I know
this sounds confusing. Let's look at three examples of circuits in
the subscriber loop.
The ringdown circuit
9. For this example we must turn away from pay phones momentarily
to consider a semi-public phone. Some supermarkets in Sacramento
have taxi phones installed near their front entrances. Lifting the
handset rings the dispatcher at Yellow Cab a few miles away. It
keeps ringing until it is answered. This is a ringdown circuit. It
is possible that Yellow Cab ran its own wire years ago from each
market to their headquarters. But not likely. They would then
need to power the line, rent space on utility poles for the wires
and maintain the system. That doesn't make sense. What does
makes sense is having the Telco engineer a solution. This means a
relay or circuit board at the central office for the supermarket.
The twisted pairs providing cab service are routed by the relay to the
headquarters' number. The Telco can probably program a switch to do
the same thing today without any hardware.
10. I've heard that some remote places use ringdown circuits. Like
isolated ranches. Perhaps. That means, however, that an operator
would be signaled whenever someone wanted to make a call. Party line
service would be more likely. Party line service is not the same as
ringdown. There is no dial with ringdown. An emergency phone on the
street might use a ringdown circuit. It may even use a dedicated line that
goes directly to a dispatcher. An elevator phone is another example. It
also rings until it is answered.
The field exchange circuit
11. The field exchange circuit or foreign exchange circuit is
often used by businesses. It provides a local phone number for distant
customers. Let's say I'm a landscape contractor in Davis, California.
Half my work comes from Sacramento which is twenty five miles
away. My Davis number has a 752 prefix. My Sacramento number,
though, starts with a 371. That's an exchange in West Sacramento
which is the closest office to the Davis CO. The 371 a free call for most
Sacramento residents. A local call for long distance. I doubt that Telcos
use these for pay phones. (1)
Dial long line circuits
12. A dial long line circuit or DLL is often used by pay phones.
It enables a coin phone to be placed further from the central office than it
might otherwise be. Most phones are located within three miles or so of
the CO or its connecting point. That's about the distance that pay
phoneproduced signals start to fade. Picking them up beyond that point
is difficult for the central office. It's a matter of resistance. The
resistance of the twisted pair increases with length. At about 2.8 miles
the telephone circuit builds to around 1300 ohms. That's acceptable.
This figure includes the resistance of the phone, the central office
equipment and the twisted pair itself. A coin phone at the six or seven
mile mark might have to signal through as much as 3500 ohms of
resistance. Amperage falls from about 23 miliamperes to 14
milliampsor less. All signals from the payphone become weak. A
dial long line for coin service has special equipment which steps up or
amplifies these weak signals. It then sends them to the switching
equipment at the central office. This is called repeated signaling. (2)
This central office solution may be a cheaper than installing heavier
gauge cable or multiplexing equipment to reach distant customers.
Signaling
13. The telephone system uses many kinds of signals. Direct
current signals, acoustical tones and digital signals are all employed. All
three kinds may be used to complete or conduct a call. This variety
makes signaling hard to understand. The central office controls Telco
pay phones with DC signals. Acoustical tones address a call, signal
the coins deposited and perform a number of network functions. Digital
signals are indispensable for long distance working. Let's look at DC
signals first.
DC signaling in the local loop
14. The simplest form of DC signaling is performed when you take the
handset off the switch hook. It's called the off hook signal or the off
hook condition or more often just off hook. Lifting the handset causes
the switch hook buttons to rise. These cause contacts in the phone to
close the circuit with the telephone line. They are normally open. This
simple act is a signal. It is electrically based. It tells the CO that a phone
has gone off hook and that a dial tone should be returned. Another
example is the operator attached signal. It disables a pay phone's key
pad by changing the polarity of the coin line from a negative charge to a
positive one.
15. A rotary dial also produces DC signals. Some refer to this
process as loop disconnect signaling. A rotary dial disconnects and
reconnects the current in the telephone line as it speeds in a circle. Five
interruptions means the number five. But why use DC signals to begin
with? Why not control a pay phone with tones? Why not digital signals?
DC signals are used for many reasons:
(a) They're simple. Manipulating a coin line's electrical
status seems complicated. But it's easy to do. DC signaling depends on
relays to do the work. These are simple, bulletproof mechanisms that
work reliably for years;
(b) They're quick. Electricity travels near the speed of
light in a circuit without resistance. It's not that fast in the local loop.
But it's quick enough. An electrical signal at 60% of that speed is
traveling at over 100,000 miles per second. Most pay phones lie within
three miles or so of a central office or its connecting point. DC signals,
therefore, act almost instantaneously;
(c) They're cheap. DC signals don't require expensive
equipment. Tone signaling requires finely tuned oscillators to send tones
and complex circuits to decode them;
(d) They're resistant to fraud. This is a side benefit of DC
signaling. It's more difficult to manipulate wires and to generate
different voltages than it is to produce tones. Never-the-less, such
manipulation is possible. The direct current initial rate signal is
simulated by punching a pay phone. Black boxing was an early activity
in which physical control of the line was. (3)Direct current signals are
treated further later on in this issue.
Tones in general
16. DC signals are used unless there is a good reason not to. Or if it is
impractical. Keypads are an example of the former reason. The simple
and sturdy method of rotary dialing was replaced by the complicated
and expensive method of using touch tones. (4) Touch tones are
produced and processed faster than rotary dial pulses. Switching
equipment is tied up for less time. Milliseconds are vital to the telephone
system because of the hundreds of millions of calls a day. They travel
more efficiently over microwave links and they make end to end
signaling easier. (5) So, touch tones are replacing DC signaling for
addressing a call.
17. Tones are also used where DC signals are impractical. DC signals
are not very loud by themselves. They might exist as a click for a
second or perhaps a soft hum. None would make, for example, a good
dial tone. A pleasant, clearly audible signal is needed. The dial tone, the
busy signal and ringback (the central office produced sound that
represents a ringing phone) are examples of network call progress
tones. These are the common everyday tones that signify the current
status of the call. Feedback, in other words, for the calling party.
18. Similarly, an audible coin deposit tone is needed to represent a coin
when a call is in progress. A DC signal might interfere with the call
itself since it affects the electrical status of the line. A digital signal
requires a modem inside the phone. Telcos don't favor that approach. A
deposit tone or a redbox tone is still a good approach even though it
interrupts conversation. Let's look briefly at some other signals.
Multi-frequency or MF tones
19. I covered coin deposit tones in detail last issue. There are also some
specifics about them later in this issue. ACTS and operators control
other parts of coin operation through MF tones. Older offices that don't
receive digital signals for coin control use these. Again, the central
office controls the payphone with DC signals. The central office is
controlled in turn by ACTS or an operator. They use acoustical tones or
digital signals to do this.
20. Tones by themselves don't do very much. A dial tone or a busy
signal is rather passive. Tones that actually control equipment are
different. (6) They are part of a coordinated signaling method or
system. You can guess that such signaling systems predated digital
working. That's why many analog offices such as step by step and
crossbar still use them. MF tones provide automatic number
identification or ANI for long distance calls from some of these offices.
ANI is essential for billing. It accompanies a call. ANI is put into a
digital form at the first properly equipped toll office. Never-the-less,
ANI exists in an acoustical form until that time. Creative use of MF may
disrupt or alter ANI. In addition, telephone companies use MF tones
extensively for internal use. An operator, for example, may address a
call to another operator using these tones. Access to inward operators,
therefore, is another possibility with home grown MF.
21. Most MF tones in current use are founded on an international
agreement called C5. Tones are called codes. Code six stands for
the number six, code seven for the number seven and so on. Numbers
are represented by different frequencies than DTMF. Three special
control tones are used for different functions. MF signaling depends
on special receivers just like DTMF signals. MF, though, works
differently than DTMF. Touch tones are sent at a pace that varies from
person to person. MF tones are often sent in bursts by a machine. 10
tones may be sent in a little more than a second. DTMF signaling is
straightforward. MF, on the other hand, depends on a strict protocol.
The KP or key pulse code is sent first. It tells the decoder that tones will
follow. The ST or start code indicates that all digits have been sent. This
shuts the decoder off. The basic tones are shown on page 29. The
chart on this page shows how the same frequencies are used for pay
phone control once a call is in progress. Actual working of C5 is
beyond the scope of this introduction.(7) If there is enough interest,
however, I could devote an entire issue to multifrequency tones. A
good understanding of MF seems essential to traveling the world by
telephone.
Digital signals
22. Digital signals help the Telco route a call, trace a call and identify a
pay phone's location. Among other things. These signals are not
directly accessible to hackers like MF tones. That's because digital
signals are produced at the switch and not at the pay phone. Access to
the switch itself is needed before any modification can begin.(8) In
addition, digital signals are put on a different channel than the voice path
on which most hacker signaling takes place. Simply blasting modem
tones down the line won't to do any good. It is this inaccessibility that
makes digital signals so frustrating.
Trunks, Circuits and Links
23. A trunk is a communication channel between switching offices or
between equipment at a switching office. It may be a single wire but
only rarely. It is most often a channel within a wire or cable. A trunk is
distinguished from a line which carries traffic between a customer and
an office. Trunks tie offices and equipment together. A subscriber line
and a trunk are both transmission lines. The phrase trunk line is correct
but redundant. It is always thought of as a trunk first. A line is always
thought of as carrying traffic to a local switch. A trunk always passes
traffic
24. A trunk may use different signals than a line. Most DC signals
can't be used in trunks, for example, because you can't vary the
voltage of a particular channel within a trunk. The same current
powers all of the channels within the cable. Think of a cable TV
line. It may carry fifty channels of programming but you can't
vary the voltage on channel 21 and not affect channel 22. The
cable has to carry about 60 volts to power the entire line.
Different kinds of signals, therefore, may be used between
offices than the kinds used between a coin phone its end office.
25. A trunk forms a circuit. But not all circuits are trunks. A
trunk usually carries conversations. A circuit usually doesn't.
For example, a no test trunk is used to tell whether a line is
busy. It's what the operator uses to break into your call when
there is an emergency.(9) It may use a circuit or relay to work
but it has always been considered a trunk. By comparison, the
Automatic Number Announcement Circuit or ANAC is a circuit
between switching equipment at the central office. But that
doesn't make it a trunk. It tells you the phone number you are
calling from. It does not carry, though, any real voice traffic.
I wrote about other circuits later. The field exchange circuit
would appear to be a trunk since it connects two switching
offices. Perhaps. I think it is best described as a hybrid. It has
always been called a circuit but it has all the attributes of a
trunk. You'll find people using the word trunk less and less these
days.
26. A link has several meanings. A data link is fairly self-
descriptive. It can be simple. A private, leased line might carry
company data from a field office to headquarters. It might
be complicated. Most of the telephone network uses data links to
carry control signals and routing information for calls which run
on trunks. On the other hand, a link is also a collection of
circuits. The first push-button long distance operator console
used a complex of four circuits. They were known collectively as a
position link. You'll also hear about A-links, B-links, off links
and so on. They are a collection of circuits. Connections by radio
to a switch are also called links. As in a microwave link.
Common channel signaling
27. A system that utilizes links, data circuits and trunks
together is called common channel signaling. CCS is poorly named.
Signaling and conversations are not placed on a common channel.
Putting the call on one path and the signals that control the call
on another is a part of C6 and C7, the signaling system currently
handles most calls.
28. C5 controls trunks with tones. These tones are different than
MF but the principle is the same: controlling equipment from a
distance with the right signal. C5 carries control codes and
conversations together. This was standard practice until the
digital age. C5 requires a tone decoder for each trunk. An analog
office with 100 trunks needs 100 decoders. They are not cheap. C6
and C7 doesn't need tones to control trunks. Most common channel
signaling uses something like a Signal Transfer Point or STPs
instead. These are routing computers distributed about the
network. STPs direct each call to a toll office. Hundreds and
hundreds of multiplexed calls are individually managed through
these computers.
29. Routing and other features are enabled by the digitally
encoded markers that are put on each call. Among other things,
these headers identify the origin of a call and its destination.
Data bases can be queried automatically while a call is placed.
An operator knows that you are calling from a Telco payphone as
soon as you are connected to them. They may even know that you are
using an airport pay phone. Automated coin toll service or ACTS,
the automated operator you get with a 1+call, is also made
possible by accessing these line information data bases or LIBDs.
(10)
30. MF controlled trunks still exist for a great deal of operator
traffic and perhaps to as many as twenty five per cent of America's
central offices. (11) Many still use single frequency tones like 2600.
Such a tone might gain control of the trunk or seize it. Remember,
though, you are seizing a particular channel in a cable, not the entire
cable. A sweep generator at one end may be one way to test for a MF
trunk from a pay phone.(12) These system 5 trunks have to interface
with system 6 and 7 at some point for long distance calling. Don't think
that remote signaling is impossible because your area has gone digital
in the form of 6 and 7. Yes, your call to Ryde, California may be
split up when sent from your area but both voice and control signals
must reunited on one path when getting to the analog office. As long as
you have a voice path to an old crossbar or step by step you may
be in luck. Here is an example of how convoluted this can be.
31. Most common channel signaling methods give you a local busy
signal if a distant phone is busy. Let's say that you dial Gabbs, Nevada.
CCS races ahead to see if the line is busy before a voice connection is
set up. If it is busy then the data link is brought down and your
CO is told to generate a busy signal for you to hear. No need to
provide a 600 mile path for you to hear a busy signal. The old Bell
System method was called CCIS or common channel interoffice
signaling. It used 2400 baud modems to pass information back and
forth. Specialized modems still send the routing information back and
forth. Let's say, though, that the central office in Gabbs isn't equipped
to handle system 6 or 7. Like much of the rural west. What then?
32. It's my understanding that the nearest properly equipped toll
office would stand as the interface point. A pay phone call from Gabbs
to Sacramento might go something like this: the pay phone would
communicate with the central office using DC signals, the CO might
communicate with the toll office by tones and the toll office would
communicate with the network by digital signals. The STP might send
the voice path from the toll office to Reno and then Sacramento. Or
maybe to Bakersfield and then back to Sacramento. Depends on the
traffic on the net. The STP might be in Fresno. Still, a home
brewed tone should be faithfully reproduced over the network to the
tone sensitive area you are investigating. To do whatever it may.
References
1. Might it be possible for the skillful hacker to use such a
circuit? An older central office that still uses tone signaling for trunks
might provide a stepping stone for the telephone enthusiast. A call
placed here might attract less attention than an 800 number. I invite
comments and speculation.
2. Schillo, Robert F. "A Circuit That Stretches Coin Telephone
Service' "Bell Laboratories Record." 51:4 (April 1973) 123
3. Billsf mentions black boxes in "True Colors" 2600, The Hacker
Quarterly. 10:3 (Autumn 1993) 11. Black boxing seems impossible
today but I am open to hearing about how it could done. Still, what
would be gained if you were successful? A local call? Physical control
of a Telco pay phone is either complicated or impossible. They are
usually in public view and subject to surveillance by the Telco. It seems
that an ordinary subscriber line would be a better choice for reinventing.
I have read, though, of people using pay phone lines to carry their local
calls by wiring in part of a cordless phone. You would need to be fairly
close and willing to be dropped out whenever someone made a call. . .
4. Touch Tones and DTMF stand for the same thing. They are both
dual tone multi-frequency signals. The phrase TOUCH TONES was a
trademark of the Bell System. They did pioneering work on tone
signaling through Bell Laboratories. Do not confuse them with MF
tones. Multi-frequency tones are also dual tones but they are mostly
used for internal Telco use.
5. Fike, John L. and George E Friend. "Understanding Telephone
Electronics." 2d ed. Carmel, SAMS 1990
6. Most tables describe tones in a confusing way. The dial tone,
for example, is a combination of 350 Hz and 440 Hz. Charts state it like
this: 350 + 440. You might think that the resultant tone is 790 Hz. Not
so. Common sense tells us that two low tones put together will not
produce a higher tone. Yet every table I've seen makes it look like an
addition problem. I use the ampersand symbol instead. 350 "&" 440.
Two tones combined. This is not a minor, pedantic point. It goes to the
definition of what a tone is. A single tone is represented by a single sine
wave. Two sine waves put together produce a complex sine
wave. What then is the frequency? The baffling answer is that it
isn't any particular frequency. That's why tables use two tones to
describe MF or DTMF signals. I find electroacoustics difficult. What if
you combine two radio frequencies together? Couldn't you get a
frequency counter to tell you the result? Why can't that be done with
audio tones?
7. Billsf "hitchhikers guide to the phone system" 2600 The Hacker
Quarterly 9:2 (Summer 1992) 10. Everything written by Billsf is
fascinating. This article is about international signaling. It emphasizes
MF tones. see also Billsf "True Colors" 2600 The Hacker Quarterly
10:3 (Autumn 1993) 9. More information on the actual working of
MF signals. NB: All 2600 back issues are for sale. See any copy of
2600 for details. Or, call their office at (516) 751-2600. Fax line (516)
474-2677.
8. In "A Guide to The 5ESS" 2600, The Hacker Quarterly, Crisp
G.RA.S.P details the inner workings of a digital switch and describes
ways to program it. It is a very impressive and advanced article. I
understand little of it. Those with a good command of UNIX will fare
better.
9. This procedure is called a busy line verification or BLV in the
trade. A skillful hacker may drop into conversations as well by using the
right tones. Read more about BLVs in Agent Steal's classic article
"Central Office Operations" in the Winter, 1990 issue of 2600. It's also
available through the Legion of Doom's Technical Journal gopher.
10. The trend is to store more and more information in these data
bases. This can enable a company maintaining the data base to provide
additional services but it can also lead to more fights among the
different Telcos and private carriers over who should get that
information and who should pay for it. A completely digital network
might be operating in our lifetime but you can bet that it won't be
flawlessly implemented because of turf wars. 500 companies provide
long distance service according to the FCC report referenced
below; competition is a zoo. Local competition when implemented
will be like letting open the gates of the zoo. Even with call trace a
hacker should be able to get some breathing room by going through as
many companies as possible when placing a call.
11. "Semiannual Report on Telephone Trends in Telephone Service,"
May, 1994. Industry Analysis Division, Federal Communications
Commission. Available on the Pac Bell gopher and I think Bell South's.
The gophers take out the 34 interesting tables. For them you have to
modem to the FCC itself, which maintains the world's worst
bulletin board at (202) 632-1361. Good luck . . .
12. Such as, perhaps, the one available through the Edlie
Electronics ('Always Something New') catalog for around seventy
dollars? The "pocket size" sweep generator perhaps? Model 125B?
Write for a catalog: 2700 Hempstead Turnpike, Levitown, L.I. NY
11756-1143. I'm sure your Telco will love you for it.
IV THE DIAL TONE FIRST COIN LINE
33. I've made many references to the dial tone first coin line in this two
part series. I think I have explained it enough by comparison and
contrast. Dial tone first is the operating method for at least 90% of the
coin telephones in the United States. One thing I haven't done yet is to
explain some of the terms on the dial tone first table.
34. TSPS stands for Traffic Service Position System. It is a grotesque
phrase the Bell System coined to describe their operator service. Before
1965 most operators worked at manual switchboards. A long distance
board might be called a toll board. The Bell System a push button
console in 1965 that eliminated the cords and jacks and automated some
parts of coin telephone service. It was quite an accomplishment. They
called the new console a traffic service position. That made a little sense
because you could argue that an operator did indeed work at a position.
Years later the Bell System improved the console but not the name. It
was now a system or TSPS. I understand that Northern Telecom or
Northern Electric makes a similar product called TOPS for our Canadian
friends. These operators must then work at a traffic operator position
system? I understand that US West has their own kind of automated
console for their operators. In any case, all of these consoles have
dozens and of buttons and lights to control calls. A display tells them
how much money you should deposit for a certain call and then they can
watch it ring up or totalize on another display.
35. Wink or multi wink is an important part of computer signaling
as well as a method used in the telephone industry. Carefully timed
pauses turn a signal in a channel off and on. You can tell by the table
that coin phones may be first signaled with this method. It works great
for optic fiber trunks since no tones or voltage are required to operate it.
It is sort of like flashing the switch hook except that each wink must be
the same. And I doubt you can access this since it is triggered at the
TSPS position. That may be hundreds of miles from the central office.
V. THE COIN FIRST COIN LINE
An introduction
36. I wrote in the first issue that coin first pay phones was the
standard operating method from the 1920's. Do any remain? I consider
coin first a defunct operating system, as dead as panel switching.
Deploying 911 throughout the country would be hindered by coin first.
There are some interesting details to coin first but I won't describe
many because I think it's obsolete.
37. Coin first phones required a deposit before they would operate,
although not necessarily a dime. I remember flashing the switch hook
after putting in a nickel. That got you a few Pacific Bell numbers. The
grace period was also nice. If you dialed a wrong number you could
quickly hang up and the pay phone returned your dime. This
disappeared in the 916 after dial tone first was introduced. That may
have been related, however, to the installation of newer switches and
not to a special feature of coin first.
38. There were some problems. The worst was that you needed a coin
to call an operator in an emergency. There was no 911 in the early to
mid 1970's. Call boxes existed but there was no centralized emergency
service. The operator called the right agency when you dialed 0 for help.
I remember worrying as a kid about always having change with me.
Otherwise, you might find yourself in real trouble and really alone.
Another problem was that you couldn't tell if a pay phone was out of
order until it took your money. No soothing dial tone to confirm
operation. They were dead as a rock without a dime.
39. Some contend that coin first was more susceptible to fraud
than dial tone first. I'm not so sure. Blue boxing occurred during the era
of coin first. But coin first did not give rise to blue boxing. Instead,
single frequency coin deposit tones, non armored handset cable and
less sophisticated totalizers all contributed to make coin first pay phones
more susceptible than the current models. Coin first operation is not
inherently suspect, even if the implementing hardware at the time was.
Single frequency trunks were not a part of coin
first but instead were accessed by them.
Ground Start
40. Memories aside, however, coin first did contribute something
that's used to this day by every dial tone first Telco pay phone. It's
called ground start. Ground start did two things with coin first. It
signaled that 1) the pay phone was off hook and 2) that a coin had been
deposited. Dial tone first, by comparison, only uses ground start
to signal an off hook. Coin first assumes a coin has been deposited
since the phone won't operate without one. Dial tone first provides a
dial tone to begin with. It needs a related signal called the initial rate test
to indicate that a coin has been put in. Let's look at the mysterious
sounding ground start.
41. We usually think of grounding as a way to keep people and
equipment safe from electrical shock. The issue of grounding for safety,
however, is a different matter than using grounding to get a telephone
connection going. Consider what happens when a normal or a post pay
coin phone goes off hook. Removing the handset causes the switch
hook buttons to rise. This closes the tip and ring contacts in the
phone set. They are normally open. Current flows into the loop from
the central office. The phone starts consuming power like any other
electrical appliance. Voltage drops from 48 volts DC to, say, 10 volts
DC. This current flow is detected by a line relay at the CO. It signals
other equipment to return a dial tone when a strong enough voltage
drop is detected. This is loop start. It's named after the twisted pair that
forms a loop connection with the CO.
42. Ground start works differently. With coin first, a relay in the
phone grounded the ring wire when a coin was deposited. Current then
flowed to the pay phone over the tip wire and into the ground. A dial
tone followed shortly thereafter. A little later the ground was removed.
This might not make sense at first. We think of electricity as flowing in
a loop. We associate circuits with circles. Yet here we have a
connection in the local loop in a straight line. No return wire to the CO.
But this is the way that telegraphs worked for decades. A conversation
can certainly work over one wire. The ground provides the complete
path that defines an electrical circuit. Electricity flows to a good ground
as easily as water flows downhill. The local loop uses two wires to
provides a better sounding call. Not necessarily to provide a complete
electrical circuit. A loop is more efficient as far as conducting electricity
but you can talk on one wire if you can tolerate some noise. Certainly
it is enough to get a connection. But why use this technique for pay
phones?
43. Fike and Friend say that "ground start lines are used on loops
connecting PBX's to the central office, and in other situations (pay
phones) where it is desirable to detect a line that has been selected for
use (seizure of the line) instantaneously from either side of the line."(1)
Unfortunately, they do not say why it is desirable to so seize a line.
44. I think that coin first used ground start for speed. (NOTE: I'M
INCORRECT ON THIS POINT -- SEE THE THIRD ISSUE) It's
about getting a dial tone as quickly as possible. That's why it is still
used. Ground start ties up equipment less than loop start. I wrote in the
first issue that the Bell System chose pre pay operation instead of post
pay because of the time it saved its operators. This decision can be
traced back to 1906.(2) The simpler post pay was discarded in favor of
coin first because an operator had to wait for a customer to coins. With
coin first an initial deposit was already placed by the time an operator
handled the call. Switching equipment can also be held up. The Bell
System still worried about this 60 years later when they decided to go to
dial tone first nationwide. Dial tone first would return them to the kind
of delays that they feared at the turn of the century. Here's a cry of woe
from the Record in 1969:
"Making modifications to existing equipment is not the only problem.
Some additional equipment must also be provided in the central office to
convert to dial tone first operation. For example, holding time of
crossbar registers and subscriber senders can increase up to 60 percent
for each completed coin call with the new service. This is due to the time
taken by customers to deposit coins after the register or sender is
attached and furnishing dial tone. Moreover, some calls -- those without
the correct initial deposit -- will not be completed and will have to be
redialed. Registers and senders must therefore be added to compensate
for the increased holding time as the office is converted. Similarly, coin
calls handled by ESS offices are subject to a 5 to 15 percent increase
in processing time. This increase plus longer equipment holding time
will result in a decrease in call handling capacity and require more coin
control circuits." (3)
45. Boo hoo. It's obvious that holding time was the most important
thing to the Bell System. Ground start would continue to be used with
DTF since it is the fastest way to set up a connection. Why is it faster?
It uses fewer steps. The central office does not have to power the entire
loop immediately to provide a dial tone. Let's say the CO is five miles
from a pay phone. Five miles of tip wire and five miles of ring wire.
Same 48 volts DC under a pressure of perhaps a hundred milliamps.
Pay phone goes off hook. CO supplies power on one wire. Current
runs to ground. Dial tone right behind it. No waiting for the rest of the
loop to power up. But it can't be that much quicker. It does helps with
part of the problem. Not much can be done, though, about someone
fumbling for a coin. Or a telephone company drumming its fingers.
46. Switches like the 5ESS return a dial tone before we can put
the handset to our ear. Ground start, though, was developed in the era
of crossbar, panel and step by step. It might have made a difference
then. COCOTS certainly aren't bothered with a wait for a dial tone.
But these milliseconds and microseconds are of concern to the Telco
since they are the local provider of phone service. Several thousand pay
phones in a large city could add up to the that the Bell article described.
A Telco pay phone now requires a good ground to properly function.
Many signals have been developed which utilize grounding. I explain
these on page 39.
References
1. Fike, John L. and George E. Friend. "Understanding Telephone
Electronics." 2d ed. Carmel, SAMS. 1990 191
2. Fagen, M.D., ed. "A History of Engineering and Science in The
Bell System: The Early Years, 1875 -- 1925." New York: Bell
Telephone Laboratories, 1975. 156
3. Ruppel, A.E. and G. Spiro 'No Dime Needed' "Bell Laboratories
Record" October, 1969 293
VI. TIP, GROUND AND RING EXPLAINED
47. The central office controls Telco pay phones by direct current
signals. I discussed why in the basic signaling article. We now look at
how DC signals are produced, some terminology about them and a short
description of each one.
48. Changing the electrical status of the telephone line produces
DC signals. This is done by manipulating the ends, or leads, of the tip
and ring wires. That, in turn, is done by relays. These simple, remotely
controlled switches are located in the central office and in the pay
phone. A coin phone relay can fit on a circuit board. Central office
relays are much larger. They may be mounted in racks.
49. Relays work by opening, closing or grounding the tip or ring
wire to produce a signal. Opening a circuit breaks the connection.
Closing a wire completes it. Grounding a wire shorts it out. Grounding
one wire, however, doesn't necessarily short out the entire circuit with
the central office. Current and conversations can still flow over the
remaining wire.
50. Depending on the signal needed, tip or ring may be opened,
closed or grounded at either the central office or at the pay phone. There
are nine ways to manipulate tip, ground and ring. Just a few are used
for signaling. But we'll look at all of them for comparison. Here's the
list:
1. Tip open and ring open.
2. Tip open and ring closed.
3. Tip open and ring grounded.
4. Tip closed and ring open.
5. Tip closed and ring closed.
6. Tip closed and ring grounded.
7. Tip grounded and ring open.
8. Tip grounded and ring closed.
9. Tip grounded and ring grounded.
1.) Tip open and ring open. On hook. The circuit is open because the
handset is on the switch hook. This tells the central office that a
particular phone isn't being used.
2.) Tip open and ring closed. -48V DC. Coin first idle. The normal
polarity of the now defunct coin first line.
3.) Tip open and ring ground. A dead line or an open circuit. No
current flows. Not used for coin line signaling. Automatic testing
equipment may remove the coin line from service. (1)
4.) Tip closed and ring open. This common DC signal has many
variations:
(a) The initial rate test signal. -48V DC. An important part
of dial tone first operation. Tells the CO that a coin has been put in.
Depositing a valid coin trips two pay phone relays. One adds a thousand
ohms of resistance to the circuit with the central office. The other
grounds the circuit itself.(2) Thus, a coin deposit is represented
by a grounded circuit with, supposedly, a certain amount of
resistance.(3) The CO, possibly tone, opens the ring lead on its own
end. Detecting the coin ground over the tip wire causes a central office
relay to close the ring side again. The initial rate signal, therefore, is the
action of opening the ring wire to detect the ground. I do not
know why it is necessary to disconnect the ring side and not the tip.
(b) The stuck coin test signal. +48V DC. Positive current is applied
if a coin relay ground persists. That was described above. If successful,
the coin will fall into the coin box, resetting the relay and thus removing
the ground. The line returns to normal. Automatic equipment may take
the line out of service if the ground persists.
(c) The coin return signal. -130V DC. The coin relay directs
coins to the coin return hopper. Why 130 volts? Later crossbar switches
used this voltage. Bell Labs may have used it for coin line signaling
since many central offices could produce it.
(d) The coin collect signal. +130V DC. The coin relay senses
the change from negative to positive current. This directs coins to the
coin box. Why doesn't the stuck coin test signal use the same higher
voltage? They both use positive current. I don't know. This is difficult
to reconcile since the same relay, I think, is being used in both cases.
5.) Tip closed and ring closed. Off hook. Normal operation and dial
tone.
6.) Tip closed and ring grounded. Reverse battery. -48V DC. Prompted
by the called party going off hook. The first issue discussed reverse
battery in detail. This signal may trip a pay phone relay which shorts out
the DTMF key pad.
7.) Tip ground and ring open. A dead line. No path for electricity to
flow.
8.) Tip ground and ring closed. Current flows on the ring side but
the tip side is shorted out. There are a number of variations:
(a) Post pay idle? -48V DC. Normal polarity of the post pay line,
according to Reeve, before a call is connected. I'm not sure anymore.
Few post pay phones should utilize a grounded circuit.
(b) Dial tone first idle. -48V DC. Normal condition of the
line until a valid coin is deposited or a free call is placed.
(c) The operator attached signal. +48 V DC. ACTS or the
operator applies positive voltage to the line. This puts the pay phone into
the toll mode. Coin deposits are then totaled automatically by ACTS or
they show up on the operator's console.
(d) The operator released signal in dial tone first. -48 V DC. ACTS
or the operator removes positive voltage from the line; restores normal
negative voltage after a call. Pay phone goes back to local mode and the
totalizer resets itself to zero.
(e) +48V DC. The key pad inhibit signal. A coin first signal, similar
to the operator attached signal. Disables key pad, perhaps, and resets the
pay phone totalizer.
9.) Tip grounded and ring grounded. Dead line.
References . . . .
(1) Martin, John T. "Chilton's Guide to Telephone Installation
and Repair." Radnor. Chilton Book Company. 1985 140
(2) Detailed in Reeve, Whitman D. "Subscriber Loop Signaling and
Transmission Handbook: Analog." New York: Institute of Electrical and
Electronics Engineers. IEEE Press. 1992 221
(3) Why such a complicated process? Preventing fraud, perhaps?
Adding resistance to the initial rate signal may prevent someone from
merely grounding the circuit to get a dial tone. Yet, there are many
stories of punching pay phones with a pin or nail to simulate the initial
rate test.* NYNEX, in fact, claims millions in damage from
punching.** That's why so many transmitters are now sealed. We may
conclude then that 1) grounding alone works, despite the resistance
that's theoretically required or 2) that the human body itself provides
the needed resistance, when the punch is held.