HOW TO DO AN EXCHANGE


The  following  document was written in about 1980 and is somewhat out
of date.  It refers, in  the  main,  to  Strowger  non-director  Group
Switching Centres.  

Since  it  was  written,  Strowger GSCs have been replaced by System X
units - but many local units  are  still  Strowger  and  it  is  worth
playing with them in the ways indicated.

Also,  there are a number of new services available on 1xx levels such
as Mercury on 131 (and 132!).

None of the information is particularly applicable to  Director  Areas
(STD codes commencing 021, 031, 041, 051, 061, 071, 081, 091).

Nevertheless, TRY all codes ....






SPU 210192




































                HOW TO DO AN EXCHANGE - BY "RON ASTON"

I am sometimes asked what exactly is meant by "doing" an exchange. The
short  answer  is  simply  that  the object of the game is to find out
everything about the exchange that can be  found  out.  Obviously  the
first  step  is to consider the information about the exchange that is
actually   published   -   the    telephone    directory    and    the
local-dialling-code  list.  In  the  case of a moderate-sized GSC, for
example, a quick glance at the  directory  may  establish  that  local
subscribers   have   numbers   in  the  range  2xxxx  and  3xxxx;  the
dialling-code  list  may  show,  in   the   simplest   case,   perhaps
half-a-dozen  small  exchanges (identified below as Red, Orange, etc.)
with 8x codes, two neighbouring GSCs with 9x codes, and  usually  also
some  service  codes  such  as 191, 192, etc. The first thing to do is
rearrange this information into numerical order; this will  produce  a
table something like this:- 

11         191 EQ           5             87           96 
12         192 DQ           6             88           97
13         190 Telegrams    7             89           98
14         19x              81 Red        80 Clock     99 Emergency
151 Faults 100 Assistance   82 Orange     91 Black     90
15x        10x              83 Yellow     92 Brown
16  Disc   2xxxx            84            93
17         3xxxx            85 Blue       94
18         4                86 Violet     95 

The  dialling-code  list  may  also include something like "Green - no
code". 

If in fact  the  exchange  area  is  served  by  a  main-and-satellite
linked-numbering  scheme,  it is best to ensure that all the following
steps are carried out from a telephone on the main exchange - normally
one in the town centre, and usually, (but  not  invariably)  having  a
number beginning with 2. 
       
First,  check that all the advertised codes actually exist and do what
they are said to do. Determine if possible the nature of the equipment
at distant exchanges. In the case of small exchanges on 8x,  determine
whether it is possible to dial "0" on the incoming grading for STD, or
"9"  for  a  route  back  to  the  GSC.  In the case of adjacent GSCs,
determine whether it is possible to dial "1" or "0" incoming and  note
the  result; and determine whether it is possible to dial "9" incoming
for routes to more distant GSCs or back to the  originating  GSC,  and
determine the effect of "1" or "0" on those routes. 

Having done this, the operator-service codes should be checked. It may
be  found,  for  example,  that  15 is teed to 19 so that 151 which is
advertised for "Faults" in fact merely calls the Enquiry  position  at
the  AMC,  in which case it is clearly unnecessary to test spare codes
both on 15x and 19x since these will  be  the  same.  It  will  almost
certainly  be  found  that  99  is  sufficient  to reach the Emergency
position, so that in fact there are no 99x codes. 

The next step is to dial everything which is NOT advertised, ie 11, 12
and so on up to 90. It will probably be found that there is  a  series
of  engineering  type  codes  on  17x  (see below) and an unadvertised
operator-service code on 198. It  may  also  be  found  that  (in  the
hypothetical  example  above)  84  and  87  also  appear  to  be used;
reference to the Directory may show that  subscribers  in  Green  have
numbers  of the form Green 84xxxx, and that subscribers in the village
of White, very close to the GSC, have numbers of the form  87xxxx  but
with  the  same exchange name as the GSC - ie there is in fact a White
exchange, but it forms part  of  a  linked-numbering  scheme  and  its
existence  is  not advertised. All other codes, including single-digit
codes 4 to 7 may be found to give NU tone. 

The codes which may be found on level 1 are as follows (it is stressed
that it is very unlikely indeed that all of these codes would be found
at any one exchange):- 

151  Faults (or Enquiries if 15 is teed to 19)
152  DQ (if 15 is teed to 19)
153  International Telegrams (if 15 is teed to 19)
155  International Service
156  International Service
157  International Service
158  Reverted calls operator (if 15 is teed to 19)
150  Inland Telegrams (if 15 is teed to 19)
16   Dial-a-disc
171 Charge ckeck - NU tone interrupted  at  intervals  which  are  ten
times  as frequent as charging pulses; by dialling a "1" the equipment
may be stepped-on to measure the charge rate for each charge  band  in
turn  -  ie local, a rate, b rate, any applicable international rates;
dialling further "1"s will repeat the sequence, and the equipment  may
at  any  time be stepped over one or more charge rates without testing
them by dialling either more than one "1" or a digit greater than 1.
172  Test Desk - often, ringing tone will change to 1000 c/s test tone
without any challenge by an engineer, since it  seems  to  be  assumed
that  that  is  what  is  wanted; but any such change indicates manual
intervention by an engineer and consequently is an indication that the
test desk is staffed.
173  Usually NU, but may be CPI for the second 10,000 lines of a large
exchange.
174  FRB
175  SALT
176  CPI
177 Routiner-answer - short pause followed by reversal, and  any  tone
or  noise  in  the forward direction, even a shout of Boo!, produces a
short burst of 1000 c/s tone and a de-reversal; no further action  has
any effect).
178  Repair Service Centre, or CPI for third 10,000 lines.
179  TMO - may be a route to the PBX operator at the TMO or a route to
the PBX if there is one.
170   Test Code - results vary from place to place but a common result
is a series of six reversals and de-reversals accompanied by either NU
tone or inverted ringing tone,  followed  by  EET  for  an  indefinite
period. 

191  Enquiries
192  DQ
193  International Telegrams
198  Reverted-calls operator
190  Inland Telegrams
100  Assistance 

Having  done  all  this and found nothing unusual, the next step is to
repeat anything which gave NU tone and dial at least four "0"s at  the
NU,  to  look  for concealed accesses. Accesses have been found in the
past where (a) two "0"s dialled at an NU tone gave access  to  trunks,
and  (b)  any four digits dialled at an EET tone gave dial tone giving
free calls. It is thus necessary to treat EET or  busy  tones  in  the
same  way  as  NU  tones;  and  although  as  yet  no  access has been
discovered hidden behind a ringing tone, any ringing tone  which  does
not  answer in a reasonable time during normal working hours should be
treated in the same way (except, of course,  192  -  DQ  almost  never
answers in anything like a reasonable time). 

This  just  about  exhausts  all the possibilities. However, any tests
involving junctions to other  exchanges  should  be  repeated  several
times, since "accidental" accesses often occur on only a single choice
or  a  small  number of choices; it is necesary to try dialling "1" at
GSCs and "0" at UAXs for example on a weekday  morning  (busy  hour  -
only  at  this  time  are late choices likely to be obtained) and also
late at night or early  Sunday  morning  (quiet  period  -  reasonable
chance of obtaining early choices). 

If  time  permits,  it  is  worth  while  checking the levels used for
subscribers' numbers. Thus if numbers  are  in  the  range  2xxxx  and
3xxxx, dial each of the hundred combinations 211 to 200; assuming that
no tone is received, dial 1; if no tone is received, dial a further 1.
Obviously  if  it  is known that, say, 24680 is a working number it is
unnecessary to test 246, and so on. Repeat for 311 to 300 or as far as
necessary - it may well be found that 36, 37, 38 and 39 give NU, there
being subscribers only on 31xxx  to  35xxx  and  30xxx.  Dial  SLOWLY,
listening for a tone after each digit. 

Any NU tome encountered on a 2xx code would be suspect - accesses have
been  hidden  in the subscribers' numbering range before now and might
well be covered by an NU tone.  There  appears  to  be  no  legitimate
reason  why  all  the  hundred  2xx  codes  should  not  be  used  for
subscribers. Dialling tone obtained after dialling one, two  or  three
digits  may  indicate (1) a fault condition causing open-trunking; (2)
an unadvertised outgoing route,  perhaps  to  a  temporary  or  mobile
exchange on which the subscribers have main-exchange numbering; or (3)
a  fraudulent  STD access. Ringing tone or NU tone after dialling 2xxx
could indicate connection to an eleven-and-over final  selector;  this
fact  should  be noted to keep records complete as possible, but it is
unlikely to be of any use.  No  tone  after  dialling  2xxxx  probably
indicates  a fault, but in such a case go on dialling (more or less at
random, but remembering what has been dialled - 1 2 3 4 5  ...  is  as
good  as  anything  else) until it is obvious that nothing is going to
happen. 

Ringing tone after six digits (or NU tone) on what was believed to  be
a  five-digit exchange may merely indicate that the exchange is bigger
than you though it was, but  could  indicate  DDI  to  a  large  PABX.
Ringing  tone  after more than six digits (and therefore not available
by STD unless a special code is provided which gives access to  second
or  third  selectors),  or  NU  tone,  is  very  suspicious;  the only
legitimate reason would be a subscriber-dialled Freefone facility,  eg
to  the  local Police HQ which is normally arranged deliberately so as
to be unavailable over STD, but only one  such  facility  is  in  fact
known  to  exist; any such result after dialling an abnormal number of
digits may equally well indicate a fraudulaent access  of  some  kind.
Any  ringing tone encountered in the subscribers' numbering range when
fewer digits  than  necessary  to  operate  an  eleven-and-over  final
selector have been dilled is also suspect; it may represent a route to
the  test  desk  or  other  engineering  position  but could also be a
concealed access - as mentioned  above,  no  acceses  concealed  by  a
ringing  tone  have  been discovered, but there has to be a first time
for everything. 

One feature which may be very useful is frequently found at GSCs where
there is a dependent numbering group containing one or more TXE2s.  If
such a group is available for example by codes of the type 93x, it may
be  found  that  the  code 939 returns to the GSC; and cases have been
found where such routes are to subscribers' grading (0  STD),  to  UAX
grading (0x STD), or to an incoming grading on which 1 gives trunks. 

Obviously,  all  the  above will not be done in the course of a flying
visit - a complete investigation of this nature is  only  possible  if
one is living in the exchange area. When it has been done, however, it
is worth while keeping it up to date. For example, if the subscribers'
numbers  go up to 35599 it may be found that 356 and 357, the next two
levels to be used, open-trunk instead of giving NU. In a  few  months,
when  these  levels  have been opened up, it may be found that 358 and
359 open-trunk, and so on. It is thus worth-while keeping  a  note  of
the  highest  number  used and up-dating the information regularly, as
open trunks are very useful on incoming routes. 

Assuming that all that has been done from the  main  exchange,  it  is
worth while repeating some of it from any satellite, and most of it on
incoming  routes  from  (1)  other  GSCs,  (2)  UAXs, and (3) small ND
exchanges, SAXs and TXEs. 

It is unlikely that the main-exchange grading reached by dialling 1, 8
or 9 at the  satellite  will  be  in  any  way  different  from  those
available  at  the GSC itself, but the effect of initial digits 2 to 7
at a satellite may well be different and it is as well to be aware  of
these  differences  even  if  they  are unlikely to be of any use; for
example, there may be local ring-backs, testers, or other  engineering
services;  it is always possible that there may be some kind of access
(probably an STD access, since a trunk access would  be  difficult  to
arrange) on first selectors at a satellite although none has ever been
found. 

On  routes  incoming  from  other  exchanges in the home charge group,
conditions on levels 2 to 9 are usually identical with  those  on  the
subscribers' grading at the GSC. (This incoming grading is referred to
as  the  NDH  grading).  From  ND exchanges in adjacent charge groups,
access is to the so-called NDA grading  on  which  there  is  often  a
separate  grading  of  level-9  second selectors giving access only to
those exchanges which are local to the originating group. 

There may  also  be  further  differences  between  the  NDH  and  NDA
gradings,  especially  at  GSCs  adjacent to director areas. All these
matters should be checked.  Further,  although  level  0  of  incoming
selectors  on  NDH and NDA gradings is normally identical with level 1
at the GSC, this is not necessarily the case; it  is  understood  that
the second-selector gradings are distinct, although usually identical,
so that differences may occur. 

Incoming  routes from UAXs are usually to a grading which is identical
with the local subscribers' grading except for level 0; however, where
combined level 1, 9 and 0 working is NOT used, and there are  separate
STD  junctions on level 0, level 9 at the UAX normally gives access to
a grading at the GSC which does not have 0 going to STD; it should  be
determined  in  such  cases  whether  this  is  a special "UAX grading
without STD" or whether it is simply the appropriate ND grading. 

Finally, when testing from a dependent ND exchange, remember  that  it
is  just  possible  that  there  may  be an access of some kind on the
dependent exchange itself. This is not likely, but remember Deal 66! 