Two 1A ESS COs to be Replaced in 2010; 59 Remain [telecom]
Mark J. Cuccia
related Yahoo Groups and "Listserves", but I hadn't included Telecom
Digest (comp.dcom.telecom). Many participants in Telecom Digest are
also on one or another of these Yahoo Groups and/or the listserves,
but there are still some TD participants or visitors who are not, and
they might still have an interest in this.
~ ~ ~ ~ ~ ~ ~ ~
WECO/Lucent/Alcatel 1AESS switches still exist in the US. There are
around 60 such 1As remaining, basically all within at&t/SBC/Ameritech,
sbc's at&t/BellSouth, and at&t/SBC/Southwestern-Bell territory. From
what I can tell, there are _NO_ more 1As in at&t/SBC/Pacific*Telesis
(Pacific*Bell in CA nor Nevada*Bell), nor Qwest/US-West territory, nor
Cincinnati Bell territory, nor at&t/SBC/SNET (Connecticut).
VeriZon/Bell-Atlantic/NYNEX does not seem to have any 1As remaining
(nor does long-time VZ/BA/NYNEX/NET&T-now-FairPoint in ME/NH/VT), but
VeriZon/Bell-Atlantic in C&P seems to have one in Baltimore MD, and
two in Virginia (Richmond, Norfolk). It doesn't seem like
VeriZon/BA/Bell-of-PA nor NJ-Bell have any more 1As still in service.
There were VERY FEW (no more than about four or five) Northern
Electric NE-1ESS switches manufactured/installed in the mid/late
1960s-era for Bell Canada in Toronto ON and Montreal PQ. These were
replaced LONG ago. These were manufactured/installed back when
Northern Electric and Bell Canada still had a VERY CLOSE working
relationship and licensing arrangement with Western Electric/AT&T/Bell
Labs of the US. Back then, virtually everything developed by the "US"
Bell System was also made available under license (although there
might be some modifications) to Bell Canada and Northern
Electric. That licensing arrangement for new developments ended around
1975/76, some twenty years following the 1956 Consent Decree that the
Bell System entered into with the US DOJ. It SEEMS that it was the
_US_ federal government more than the Canadians who back then wanted
AT&T to withdraw from Canada! Prior licensing arrangements w/r/t
Western Electric and Northern Electric, now known as Northern Telecom
in the post-1975 period (Nortel) would be honored, but there would no
longer be any almost automatic licensing of US Bell and Western
innovations directly to Bell Canada as such.
Some other non-digital SPC (Stored Program Control) switches include
the Northern Telecom SP(x) series, but I think that all of these in
the US and Canada have since been replaced with digital switches. And
the (AGCS) GTE-AE (x)EAX switches in the US and Canada (except for the
5EAX which is really the digital GTD-5) all seem to have been replaced
with digital offices of one kind or another. There are still quite a
number of 5EAX/GTD-5 switches still in service, but these are digital
offices, not "analog" non-digital yet still SPC offices...
In the early 1970s, since Bell Canada and Northern Electric knew that
the day was fast approaching for the separation between Bell/Northern
of Canada and the "US" Bell System, especially as AT&T was selling off
more and more of its holdings of Bell Canada and NECo, Bell Canada and
NECO created "Bell Northern Research", sort of like a Bell Labs for
Canada, their "own" Canadian R&D unit. It was BNR that developed the
SP(x) series of stored program switches heavily used in Canada (and by
many non-Bell telcos in the US and elsewhere in the North American
network), as well as the early truly digital switches, the DMS series,
also heavily used in Canada, by independent telcos in the US (and
elsewhere), and even by Bell telcos in the US both prior to
divestiture but also more-so after 1984 divestiture.
I also seem to think that there are no longer any more WECo 2(x)ESS or
3ESS analog-non-digital-yet-still-SPC offices still in service. The
4ESS and 5ESS are digital switches though -- the 4ESS is mainly for
toll and tandem functions, and these are slowly being replaced with
more recent model digital (tandem/toll) switches in the AT&T Long
Lines network and some BOC/ILEC networks which inherited them
post-divestiture. The 5ESS digital is quite versatile, handling local,
tandem, toll, operator (OSPS) services, or combinations thereof. And
Lucent/Alcatel keeps coming out with new features and models. But the
old 1AESS (which enhanced/replaced the 1ESS of the
mid-1960s/early-1970s era) is still around, although there aren't many
left. However, "back in the days", the 1/1AESS was in _ALL_ Bell
territories, even SNET in Connecticut and in Cincinnati Bell
territory, both as replacements for SXS, Panel, Crossbar, and as brand
new wirecenters or "expansions" of existing central offices, but to
think that it's now down to around 60 such offices still in service,
and only in a few of the old Bell telco territories!
But of at&t/SBC/Ameritech (Michigan-Bell and Illinois-Bell, but
nothing left in Ohio, Indiana, Wisconsin), sbc's at&t/BellSouth (both
Southern Bell and South Central Bell), and at&t/SBC/SW-Bell (in the
St.Louis MO area and scattered about Texas), there are still several,
although these are SLOWLY being replaced by digital switches and
packet switches.
In October and November 2009, at&t's technical notices website (I
don't visit this each and every day... that's why I haven't posted
this until now) had notices about the elimination of a Dearborn MI and
a Livonia MI 1AESS.
The main page for the at&t technical notices is:
http://www.att.com/gen/public-affairs?pid137
On Thursday 22-October-2009, at&t (ILEC) issued ATT20091022L.1 which
can be downloaded from:
http://www.att.com/public_affairs/regulatory_documents/ATT20091022L.1_Web.doc
Sometime during 4Q/2010 (exact date not shown), LIVNMIMNCG0 Livonia MI
"Main" 1AESS is to be replaced with a new Nortel Packet Remote switch,
LIVNMIMNRP0. The Packet Remote will be hosted by WAYNMIMN20T/DS1 Wayne
MI "Main" Nortel DMS-200 tandem, which doesn't appear to have any
c.o.codes. (Wayne MI DS0 does have local c.o.codes though).
LIVNMIMNCG0 Livonia MI "Main" 1AESS has SS7 Point Code 250-050-051.
LIVNMIMNRP0 Livonia MI Nortel Remote Packet Switch will use the same
SS7 Point Code 250-050-045 as its Wayne MI Nortel-DMS-200 host/tandem
WAYNMIMN20T/DS1.
The "default" c.o.codes involved in the switch replacement are:
313-937 for the Detroit-Zone-05 MI ratecenter;
734-261,421,422,425,427,458,513,522,524 for the Livonia MI ratecenter.
On Monday 02-November-2009, at&t (ILEC) issued ATT20091102L.1 which can
be downloaded from:
http://www.att.com/public_affairs/regulatory_documents/ATT20091102L.1_Web.doc
Sometime during 3Q/2010 (exact date not shown), DRBRMIDBCG0 Dearborn MI
"Main" 1AESS is to be eliminated, its 313-NXX c.o.codes, lines, customers
all migrated over to the co-located DRBRMIDBDS0 WECO/Lucent/Alcatel 5ESS.
SS7 Point Codes involved:
DRBRMIDBCG0 1AESS: 250-050-048
DRBRMIDBDS0 5ESS: 250-050-043
The "default" c.o.codes on the 1AESS include:
313-277,561,562,563,565,724, all on the Detroit-Zone-06 MI ratecenter.
The "default" c.o.codes on the 5ESS include:
313-274,278,359,730,791,792, all on the Detroit-Zone-06 MI ratecenter.
NOTE that in BOTH switch replacements, I mention "default" c.o.codes.
Remember that with portability, there might be customers with OTHER
c.o.codes who ported-in to the old 1As being replaced, or there might
be customers who have these indicated c.o.codes who have now
ported-out of the old 1As and are now already on some digital or
packet switch.
Additionally, I have compiled the following listing from NUMEROUS
different sources, of what seems to be all of the other (some 59)
1AESS offices remaining in the US as of early 2010 -- the three in
VZ/C&P, and the several in at&t/MI-Bell, at&t/IL-Bell,
at&t/Southern-Bell, at&t/South-Central-Bell,
at&t/Southwestern-Bell. The list below does NOT include the two
at&t/MI-Bell 1As mentioned above that are scheduled to be replaced
sometime later during 2010 (Livonia MI, Dearborn MI).
I hope I don't have any typos here. I tried to get ALL of the (BOC)
NPA-NXX c.o.codes on these 1As. I did NOT include "paging" prefixes of
other paging providers. There might still "appear" to be a BOC 1AESS
in service in some resources, but the NPA-NXX c.o.codes associated are
_NOT_ those of the BOC, however, and those are not included here, as
their inclusion in some resources is probably an anomaly.
I have included the c.o.switch "building names" as well. If a building
name is _NOT_ shown, then it is assumed to be known as "Main". Some
building-ID-codes in the 7th/8th positions of the CLLI are 'MA' or
'MN' and these seem to always be known as "Main". 'MT' could mean
"Main/Toll" by some BOCs. Some BOCs use a two-alpha abbreviation
reflecting the ratecenter or locality name again for the "building"
code, such as Oak Park 'OP' in Illinois listed below, so I consider
that to be "Oak Park IL -- Main", and don't give any ADDITIONAL
reference to "Oak Park" nor "Main" for the building name.
VeriZon/Bell-Atlantic/C&P:
--------------------------
BLTMMDEDCG0 Baltimore MD
"Edmondson Avenue"
410-233,362,566,624,945,947
RCMDVAHLCG0 Richmond VA
"Hull Street"
804-230,231,232,233,291,319
NRFLVAGSCG0 Norfolk-Zone-02 VA
"Granby Street"
757-480,531,583,587,588
at&t/SBC/Ameritech/Michigan-Bell:
---------------------------------
BRHMMIMNCG0 Birmingham MI
248-258,433,540,642,644,645,646,647,901,988
PNTCMIWSCG0 Pontiac MI
"West"
248-681,682,683,706.738
WYNDMIMNCG0 Wyandotte MI
734-246,281,282,283,284,285,324
LNNGMISOCG0 Lansing MI
"South"
517-272,393,394,882,887
GDRPMIBL770 Grand Rapids MI
"BEll"
(the co-located GDRPMIBLDS1 Nortel-DMS-100 has some 616-23x codes, which
were known as 'BEll-x' in the 2L-5N days)
616-770
at&t/SBC/Ameritech/Illinois-Bell:
---------------------------------
CHCGILAUCG0 Chicago IL
"AUStin"
773-261,287,378,379,473,626,854,921
OKPKILOPCG1 Oak Park IL
708-209,366,383,386,445,488,524,660,763,771,848
at&t/BellSouth/Southern-Bell:
-----------------------------
ATLNGAAD69F Atlanta GA
"Adamsville"
404-472,505,691,696,699
ATLNGACD28F Atlanta GA
"Columbia Drive"
404-282,284,286,289,534
ATLNGAGR24F Atlanta GA
"Gresham"
404-212,241,243,244,328,381
ATLNGAHR79E Atlanta GA
"Hollywood Road"
404-792,794,799
ATLNGAWE75F Atlanta GA
"West End"
404-752,753,755,756,758
CRTNGAMA83C Carrollton GA
770-214,830,832,834,836,838
678-796
FRBNGAEB96A Fairburn GA
Atlanta-South ratecenter
"East Broad"
770-306,774,892,964,969
LGRNGAMA88C LaGrange GA
706-242,812,837,845,882,883,884,885 LaGrange GA ratecenter
334-982 Oakland (Chambers) AL ratecenter
SVNHGADE35C Savannah GA
"Derenne Avenue"
912-303,351,352,353,354,355,356,691,692
WYCRGAMA28C/02T Waycross GA
912-283,284,285,287,338
AGSTGAFL79C Augusta GA
"Fleming"
706-560,771,772,790,792,793,796,798
WRRBGAMA92C Warner-Robins GA
478-322,328,329,542,918,922,923,929,975
JCBHFLMA24E Jacksonville Beach FL
904-241,242,246,247,249,270
JCVLFLRV38E Jacksonville FL
"Riverside"
904-381,384,387,388,389,981
FTLDFLSU74E Fort Lauderdale FL
"Sunrise"
954-572,578,741,742,746,747,748,749
HLWDFLHA45E Hollywood FL
"Hallandale"
954-454,455,456,457,458,516
MIAMFLBA85E Miami FL
"Bayside"
305-250,285,854,856,857,858,859,860
786-314
MIAMFLME32E Miami FL
"Metro"
305-324,325,326,545,547,548,549,550,560,585
WPBHFLRB84E West Palm Beach FL
"Riviera Beach"
561-494,840,841,842,844,845,848,863,881,882
at&t/BellSouth/South-Central-Bell:
----------------------------------
NSVLTNINCG0 Nashville TN
"Inglewood"
615-226,227,228,258,262,650
BRHMALEN78E Birmingham AL
"Ensley"
205-206,780,781,783,785,786,787,788
BRHMALEW95E Birmingham AL
"Eastwood"
205-951,956,957
BRHMALTA84E Birmingham AL
"Tarrant"
205-808,841,849
BSMRALMA42E Bessemer AL
205-424,425,426,428,481
SHPTLAHDCG0 Shreveport LA
"Highland"
318-219,861,862,864,865,866,868,869
SHPTLAQBCG0 Shreveport LA
"Queensboro"
318-526,621,631,632,635,636,638
LFYTLAMACG1/04T Lafayette LA
337-231,232,233,234,235,236,237,261,264,265,266,267,268,269,262,572,920
LFYTLAVMCG0 Lafayette LA
"Vermillion"
337-216,273,294,295,406,981,983,984,988,989,991,993
at&t/SBC/Southwestern-Bell:
---------------------------
STLSMO04CG0 St.Louis MO
"FOrest"
314-361,367,454,758,823,826,829,848,855,879
STLSMO05CG0 St.Louis MO
"JEfferson"
314-286,289,321,371,531,533,534,535,652,658
STLSMO08CG0 St.Louis MO
"PRospect"
314-268,577,664,762,771,772,773,776,865
STLSMO23CG0 Ladue MO ratecenter
"Overland"
314-253,423,426,427,428,429,538,733,890
STLSMO24CG0 Ladue MO ratecenter
"Riverview"
314-388,867,868,869
STLSMO40CG0 Ladue MO ratecenter
"Florissant"
314-830,831,837,838,839,921,972
STLSMO43CG0 Ladue MO ratecenter
"Hazelwood"
314-551,731,895
ELPSTXNECG0 El Paso TX
"Northeast"
915-744,745,751,755,757,759
ODSSTXLICG0 Odessa TX
"Lincoln"
432-331,332,333,334,335,337 Odessa TX ratecenter
432-580,532 Odessa 'EACS' TX ratecenter
DLLSTXGPCG0 Dallas TX
"Grand Prairie" TX ratecenter
972-237,262,264,266,282,504,642 Grand Prairie TX ratecenter
972-260,263,269,558,901 Grand Prairie 'EMS' TX ratecenter
DLLSTXHACG0 Dallas TX
"HAmilton"
214-421,426,428,565,928
DLLSTXWHCG0 Dallas TX
"WHitehall"
214-779,941,942,943,944,946,947,948
FTWOTXATCG0 Ft.Worth TX
"ATlas" TX (ratecenter and 2L-5N name)
817-284,580,590,595 Atlas TX ratecenter
817-589 Atlas 'EMS' TX ratecenter
817-280,282,285 Euless TX ratecenter
817-268 Euless 'EMS' TX ratecenter
FTWOTXGLCG0 Ft.Worth TX
"GLendale" TX (ratecenter and 2L-5N name)
817-446,451,457,492,496 Glendale TX ratecenter
817-429,654,930 Glendale 'EMS' TX ratecenter
FTWOTXJECG0 Ft.Worth TX
"JEfferson"
817-413,531,534,535,536
FTWOTXWACG0 Ft.Worth TX
"WAlnut"
817-207,920,921,922,923,924,926,927
HSTNTXADCG0 Houston (Suburban) TX
"Aldine"
281-219,442,449,590,985,986,987
HSTNTXGLCG0 Houston TX
"GLendale"
713-330,450,451,453,455,637
HSTNTXIDCG0 Houston TX
"IDlewood"
713-413,433,434 Houston TX ratecenter
713-340,436 Houston Suburban TX ratecenter
HSTNTXWLCG0 Houston (Suburban) TX
"West Ellington"
281-464,481,484,921,929
BUMTTXTECG0 Beaumont TX
"TErminal"
409-212,654,757,784,785,813,827,832,833,835,838,839,841,868,880
~ ~ ~ ~ ~ ~ ~ ~
Mark J. Cuccia
Sam Spade
~
>
> WECO/Lucent/Alcatel 1AESS switches still exist in the US. There are
> around 60 such 1As remaining, basically all within
> at&t/SBC/Ameritech, sbc's at&t/BellSouth, and
> at&t/SBC/Southwestern-Bell territory. From what I can tell, there
> are _NO_ more 1As in at&t/SBC/Pacific*Telesis (Pacific*Bell in CA
> nor Nevada*Bell), nor Qwest/US-West territory, nor Cincinnati Bell
> territory, nor at&t/SBC/SNET (Connecticut).
Do you (or anyone) know why Pacific Bell, et al, decided to replace
perfectly good 1AESS platforms with either DMS-100s or 5ESSes? I can
understand changing to digital for further replacement of remaining
5XBAR and SXS, but some of those 1AESS platforms had been in service for
as little as 15 years.
What does a digital end office do that a 1AESS won't (wouldn't) do?
markj...@yahoo.com
> Mark J. Cuccia wrote:
>> WECO/Lucent/Alcatel 1AESS switches still exist in the US. There are
>> around 60 such 1As remaining, basically all within
>> at&t/SBC/Ameritech, sbc's at&t/BellSouth, and
>> at&t/SBC/Southwestern-Bell territory. From what I can tell, there
>> are _NO_ more 1As in at&t/SBC/Pacific*Telesis (Pacific*Bell in CA
>> nor Nevada*Bell), nor Qwest/US-West territory, nor Cincinnati Bell
>> territory, nor at&t/SBC/SNET (Connecticut).
> Do you (or anyone) know why Pacific Bell, et al, decided to replace
> perfectly good 1AESS platforms with either DMS-100s or 5ESSes?
No specific reason, unless it was a tax write-off? Nortel (or
AT&T/Lucent) "sold" them on the DMS (or 5E)?
> I can understand changing to digital for further replacement of
> remaining 5XBAR and SXS, but some of those 1AESS platforms had been
> in service for as little as 15 years.
Well, when was the cutover? If it was past 2000, then the 1A was
originally installed after 1985. I don't know if WECO/AT&T was still
making 1As for _NEW_ installations or complete cutovers replacements
of electromechanical switches (SXS, Panel, XB) after the mid-1980s.
In 1987 in New Orleans, the last two #5XBs were cutover to "ESS".
Broadmoor (NWORLABM---) became a 5ESS. Michoud (NWORLAMU---) became a
DMS-100.
> What does a digital end office do that a 1AESS won't (wouldn't) do?
1AESSes and similar non-digital, yet still electronic/SPC offices
apparently can NOT do ISDN, nor other more modern/enhanced packet-type
functions. Of course, for the average residential or small business
customer of the general public, this really isn't much of an issue.
But for larger business customers, if there aren't any 5Es or DMSes or
other digital switches nearby to get FX/FCO from, then the ILEC will
need to replace the 1A with a digital. Or else that business customer
will port away to a CLEC willing to provide service off of their
digital or packet switched local network!
Also, Lucent/Alcatel might not be able to provide continued
assistance/etc. for 1As (and 2(x)ESS, 3ESS) anymore, although I don't
know for certain. While I can still surf-the-web/etc. with Win-95 and
such. there are more and more webpages that won't function properly
with older OS' and/or software associated wtih older OS'. The older
software might not work anymore neither, or if I have an older OS, I
can't integrate newer software. I don't like it, but that's how things
are these days. Similarly, I have had to buy brand new cellphones
instead of having older ones simply repaired, or at times even buying
brand new cellphones to simply use my wireless provider. And then
there was the FORCED converson from NTSC/analog to digital last year
(don't get me started on that!), even though you can have
conveters/etc.
So, I guess the same things apply to 1As vs. digital/packet c.o.
switches as well.....
mjc
Steven
support. I wanted something and was in a 1A; don't remember what it
was, but after a long protracted dispute; I wanted them to port my
number to the 5E in the same office, the finally agreed to move me with
a new phone number to the 5E with no charge. What was funny is that the
1A was replaced less then a year later. Switches don't have the life
time (25 years or more) and can be replaced it 5 years. I like that
because I'm a CO Installer and it keeps me very busy.
--
The only good spammer is a dead one!! Have you hunted one down today?
(c) 2010 I Kill Spammers, Inc., A Rot in Hell. Co.
Sam Spade
> Well, when was the cutover? If it was past 2000, then the 1A was
> originally installed after 1985. I don't know if WECO/AT&T was still
> making 1As for _NEW_ installations or complete cutovers replacements
> of electro-mechanical switches (SXS, Panel, XB) after the mid-1980s.
> In 1987 in New Orleans, the last two #5XBs were cutover to "ESS".
> Broadmoor (NWORLABM---) became a 5ESS. Michoud (NWORLAMU---) became
> a DMS-100.
The two I am familiar with: one was an early 1ESS, installed in 1970
or so. Upgraded to 1AESS in the late 1970s. Changed out to a 5ESS
about 1986. The other was installed in 1975 (I recall the year on
that one with certainty). It was upgraded to a 1AESS perhaps 4 years
later, and changed out to a 5ESS in the early 1990s.
> 1AESSes and similar non-digital, yet still electronic/SPC offices
> apparently can NOT do ISDN, nor other more modern/enhanced
> packet-type functions. Of course, for the average residential or
> small business customer of the general public, this really isn't
> much of an issue. But for larger business customers, if there
> aren't any 5Es or DMSes or other digital switches nearby to get
> FX/FCO from, then the ILEC will need to replace the 1A with a
> digital. Or else that business customer will port away to a CLEC
> willing to provide service off of their digital or packet switched
> local network!
My city (San Clemente, CA) cut from 5XBAR to DMS-100 in 1985. The
next town over was the one that cut from 1AESS to 5ESS about 1986. In
the early 1990s when ISDN was being promoted I decided to go for it.
My local DMS-100 could not handle ISDN. Apparently that is a major
upgrade. So, Pacific Bell provisioned my two ISDN lines to the Irving
DMS, some 27 circuit miles away. The 5ESS close by was ISDN capable,
but Pacific Bell's policy was to place you on an ISDN platform of the
same switch type as your local switch. I guess there were sufficient
differences between the 5ESS and the DMS 100 that resulted in that
policy.
I also understand that all the 5ESS could do ISDN unlike the DMS-100s.
That was an expensive arrangement to provide me two 27 effective FX
circuits at no change. I have up on ISDN after two years because the
adapter I bought for $700 was lousy and did not perform to specs. I
did get the data speed but most other features did not work. I
finally sued the adapter vendor in small claims court and got my money
back. ISDN, at least for me, was indeed Is Still Doing Nothing. I
understand it's still around, but why escapes me, with it's painfully
slow speed compared to DSL. (and DSL is not nearly as good as my
local cable broadband).
What else do most people want other than broadband. I wonder whether
DSL could have been made to work on 1AESSes with adjunct hardware?
Garrett Wollman
Sam Spade <s...@coldmail.com> wrote:
> ... ISDN, at least for me, was indeed "Is Still Doing Nothing". I
> understand it's still around, but why escapes me, with it's
> painfully slow speed compared to DSL.
It is still very important for broadcasters, because it provides
better reliability and performance guarantees than IP-based technology
-- important if you're doing a live broadcast from a remote location
or a home studio. Many radio stations also use ISDN as a backup
studio-transmitter link, although T1 circuits are more popular now
(because they have enough spare bandwidth to remotely operate all
sorts of IP-based equipment now used at transmitter sites).
A few years ago, I was with a friend at a live broadcast of "A Prairie
Home Companion" from the Koussevitzky Music Shed at Tanglewood in
Lenox, Mass. We asked the show's engineer how they were getting the
audio to the network, and he said that they had three ISDN circuits to
Minnesota Public Radio HQ in St. Paul (where the show normally
originates), and for emergency backup, three ISDN circuits to NPR
Satellite Services in Virginia. Each one of those circuits had both
"B" channels in use, since the broadcast feed is stereo. Two of each
set of circuits were used for the audio feed and one was used for
confidence monitoring.
-GAWollman
--
Garrett A. Wollman | What intellectual phenomenon can be older, or more oft
wol...@bimajority.org| repeated, than the story of a large research program
Opinions not shared by| that impaled itself upon a false central assumption
my employers. | accepted by all practitioners? - S.J. Gould, 1993
T
says...
It really comes down to maintenance expense. Plus the fact that I think
Alcatel isn't making many parts for 1AESS these days.
Sam Spade
>>What does a digital end office do that a 1AESS won't (wouldn't) do?
>
> It really comes down to maintenance expense. Plus the fact that I
> think Alcatel isn't making many parts for 1AESS these days.
Understood, but it's sort of the chicken-egg syndrome. Had several
hundred 1AESSes remained in service, parts wouldn't be an issue.
***** Moderator's Note *****
I think they retire 1A ESS offices because only the oldest techs can
understand Type 3 E&M.
Bill "Protected member of the class" Horne
Moderator
John Levine
>> understand it's still around, but why escapes me, with it's
>> painfully slow speed compared to DSL.
ISDN is still circuit switched 64K channels, known as B for Bearer
channels. It's a substitute for POTS, not for broadband.
Since it's digital end to end, the voice quality is better than
analog, and as has been noted you can bond multiple B channels
together to get better fidelity. ISDN lines also have a D channel
used for call setup and (in theory although rarely in practice)
moderate speed data, so ISDN provides a full set of spiffy calling
features via the D channel.
My impression is that in the US the main use of ISDN is on PBX trunks,
where a T1 is typically configured as one D channel and 23 B channels.
The implementation of North American ISDN was seriously botched.
Elsewhere in the world, it's plug and play so in Japan they have ISDN
pay phones into which you can plug your ISDN laptop modem. The
version in the US requires non-trivial configuration of each end of a
circuit, so for most purposes it's more hassle than it's worth. Oh,
and the Bells grossly overpriced it, too.
R's,
John
***** Moderator's Note *****
I think the U.S. ILECs were reluctant to push ISDN because they had an
acute shortage of 8-bit trunks to serve ISDN calls. Even after the
implementation of SS7, most interoffice trunks were served via
ordinary DPO and DPT cards that still used robbed-bit signalling.
There were also serious provisioning issues: when I was doing SS7
Engineering, I worked at home for an extended time following an
accident, and at one point (this was waaaay before cable was
available, and ADSL wouldn't stretch to my house yet) I ordered ISDN
service in hopes of getting somewhat higher data rates. I had to sign
up with an ISP that provided ISDN service, and since they were in
Boston I had to resign myself to paying by the minute on every data
call.
I bought a Motorola BitSurfer ISDN "T" adapter, and hooked everything
up at the appointed hour. As John says, the voice quality was great,
and I got an extra phone number. Although it was always a crapshoot as
to whether or not I'd get a 64kbps or a 56kbps conneciton on a data
call, I found the added speed to be a great advantage. Also, since the
ISP had a much better backbone than my previous provider, I noticed
dramatically reduced latency during data calls.
But -
The phone wouldn't ring on incoming calls. I tried a different T
adapter, putting a voltmeter across the RJ-11 jacks, and using a phone
with a 0.1 REN. Nothing worked.
After several days of frustration, N.E.T. changed me back to POTS,
refunded the ISDN fees, and admitted that the provisioning system
wasn't capable of correctly setting the IDSN bit mask needed to add
ringing capability to an ISDN line. The CO foreman told me point-blank
that none of his techs knew how to provision the service by hand, and
said I was chaising my tail trying to use ISDN at all.
I took the hint: ISDN is, at least in the U.S., a monument to the
elephantine inertia and shortsightedness of the former monopoly.
Bill Horne
Moderator
Thad Floryan
>>> ... ISDN, at least for me, was indeed "Is Still Doing Nothing". I
>>> understand it's still around, but why escapes me, with it's
>>> painfully slow speed compared to DSL.
>
> ISDN is still circuit switched 64K channels, known as B for Bearer
> channels. It's a substitute for POTS, not for broadband.
>
> Since it's digital end to end, the voice quality is better than
> analog, and as has been noted you can bond multiple B channels
> together to get better fidelity. ISDN lines also have a D channel
> used for call setup and (in theory although rarely in practice)
> moderate speed data, so ISDN provides a full set of spiffy calling
> features via the D channel.
>
> My impression is that in the US the main use of ISDN is on PBX trunks,
> where a T1 is typically configured as one D channel and 23 B channels.
Correct, and in that usage it's known as a Primary Rate Interface aka PRI.
I ordered a PRI for one client who wanted the Nortel BCM system (a PBX),
and also PRIs for clients who wanted the asterisk VoIP system. Amazingly
(to me) all PRIs were actually fiber lines to the MPOE where they were
converted to <something> (2-pair) over Cat5 that connected to the PCI
card(s) in the asterisk system(s).
***** Moderator's Note *****
Please tell us what a "MPOE" is, and what the "<something> (2-pair)"
turned out to be. I'd also like to know what the PCI cards for the
Asterisk system cost, and what hardware they're running on. TIA.
Bill Horne
Moderator
Thad Floryan
> On 2/10/2010 9:09 PM, John Levine wrote:
>> My impression is that in the US the main use of ISDN is on PBX trunks,
>> where a T1 is typically configured as one D channel and 23 B channels.
>> [...]
>
> Correct, and in that usage it's known as a Primary Rate Interface aka PRI.
>
> I ordered a PRI for one client who wanted the Nortel BCM system (a PBX),
> and also PRIs for clients who wanted the asterisk VoIP system. Amazingly
> (to me) all PRIs were actually fiber lines to the MPOE where they were
> converted to <something> (2-pair) over Cat5 that connected to the PCI
> card(s) in the asterisk system(s).
>
> ***** Moderator's Note *****
>
> Please tell us what a "MPOE" is, and what the "<something> (2-pair)"
> turned out to be.
Not sure if it's California/PacBell terminology. but:
MPOE = Minimum (or Main) Point Of Entry, aka DEMARC. Locally it's a/the
building's wiring closet to which all outside "wire" (incl. fiber) would
terminate and LEC responsibility ended, and from which CPE wiring would
be run to floor and/or individual company wiring closets.
The "<something> (2-pair)" is just that: one analog pair for transmit and
one pair for receive (which is which depends on the point-of-view of the
endpoint devices, with a CSU/DSU CPE at one end; this is a T1) which typically
would be run using 2 (of the 4) pairs in Cat3 or Cat5 cabling for convenience.
I've never heard it termed anything other than "the T1 line" or "the CSU/DSU
wire". :-)
> I'd also like to know what the PCI cards for the Asterisk system cost, and
> what hardware they're running on. TIA.
System hardware is really quite modest. A Dell Optiplex (~ 1000-2000 MHz
single-core) is more than adequate for most purposes (small company, less than
100 phone instruments, etc.) and running a several-release-old "stable" Fedora
though CentOS would also be suitable. The PCI cards are/were generally in the
range of US$500; recommended cards can be seen (and purchased) here:
<http://store.digium.com/telephony_card_selector.php>
Note the system "should" have two NICs (Ethernet ports): one for the normal
data Ethernet traffic and one for the VoIP traffic. Believe me, contrary to
all the industry hype, you really want separate networks for a business so
the shipping guy downloading his p0rn or playing a game doesn't bring down a
high-stakes conference call. A home or (really) small business setup can work
fine over a single network. Security concerns would be another reason to have
separate networks (data and VoIP); the cost of (most) networking gear is so
relatively low nowadays (vs. the n * $1000 multiples of only a few years ago)
that I'd recommend 2 networks "just to be safe".
I don't have any of the (old) purchase orders listing exactly what cards I've
used over the years. The big concern is how one is going to power all the
phone instruments. Cisco 7960 and Polycom IP4000 (conference room)
instruments seemed to be the overwhelming and near ubiquitous favorites. The
Polycoms required an external power supply, the 7960s would work either with
an external power supply or via PoE (Power over Ethernet). Be aware that
most, if not all, PoE switches (even from Cisco) probably will not be able to
supply power to more than 20-30 phones (which is really insane for a 48-port
switch) -- the ((Cisco) 7960) phones are real power hogs -- but newer phones
"might" be more "green" (energy efficient). I've been "out" of this business
for several years now and haven't actively monitored ongoing progress.
Robert Bonomi
wrote
> ***** Moderator's Note *****
>
> Please tell us what a "MPOE" is,
<M>ain <P>oint <O>f <E>ntry.
> ... and what the "<something> (2-pair)" turned out to be.
'bastardized' DS1. Works fine for relatively short distances.
DS1 cable spec calls for shielded TP, but you can get away with UTP for
limited distances.
> I'd also like to know what the PCI cards for the Asterisk system
> cost,
Dialogic or Brooktrout, basic functionality cards, circa $250-300 on the
secondary market. (You can pay a bunch more, an get more 'smarts' on the
card, which means less CPU required.)
Digium (the folks behind asterix) -- circa $800 list (new, obviously).
John Levine
> PBX), and also PRIs for clients who wanted the asterisk VoIP
> system. Amazingly (to me) all PRIs were actually fiber lines to the
> MPOE where they were converted to <something> (2-pair) over Cat5
> that connected to the PCI card(s) in the asterisk system(s).
I don't think anyone's still installing traditional two-pair T1. My
T1 was two-pair when they installed it in 1995, but a couple of years
ago they switched to HDLC, so there was one pair coming to the house,
and a little line powered box that turned it back into the traditional
two pairs I plugged into my CSU/DSU.
There's tons of used T1 equipment available, on ebay and specialized
places. I got a WANIC T1 card on ebay for under $100, probably
because nobody else knew what it was.
R's,
John
***** Moderator's Note *****
Please tell us how "HDLC" is used in this context, and which
physical-layer protocol is used: the highest-speed two wire data line
I'm familiar with is IDSL, but that tops out around 144 kbps.
Bill Horne
Moderator
P.S. Why is it that everyone _else_ has the best toys? ;-)
Garrett Wollman
Thad Floryan <th...@thadlabs.com> wrote:
>Cisco 7960 and Polycom IP4000 (conference room) instruments seemed to
>be the overwhelming and near ubiquitous favorites.
MIT did a big rollout starting two years ago (which I think is still
ongoing... something about maintenance renewal on the 5ESS looking to
cost $BIGNUM) and chose to go all Polycom. The desk phones --
four-button 550s and six-button 650s -- seem to be pretty decent, and
run better software than the now-obsolete IP4000 conference phone.
(My lab is getting IP6000s now but we still have mostly 4000s.) A
sore point has been compatibility with wireless headsets; it took a
lot of complaining before the central phone people would make
available a Polycom software load that properly supported current
models of headset, and users' existing "handset-lifter" headsets were
mechanically incompatible with the Polycom handsets. (We had used
AT&T/Lucent/Avaya 7506 ISDN desk sets before the switch.)
For a cheap, simple, configurable desk phone, I preferred the Linksys
(ex-Sipura) models, but unfortunately, Linksys won't support end-users
who have them, and "business" features require use of their
proprietary PSTN gateway. I still have one on my desk, but I think
it's the last. The Polycoms do have slightly better voice quality,
but don't support cool stuff like end-to-end encryption.
-GAWollman
N.B.: I don't speak for MIT.
--
Garrett A. Wollman | What intellectual phenomenon can be older, or more oft
wol...@bimajority.org| repeated, than the story of a large research program
Opinions not shared by| that impaled itself upon a false central assumption
my employers. | accepted by all practitioners? - S.J. Gould, 1993
***** Moderator's Note *****
Please tell us more about the phone(s) and the choices each offers.
I'm especially interested in the encryption tool(s) which each
vendor and/or system has available.
I'd also like - very much - to know what factors (other than money)
drove the cutover: it's very rare for a large organization to make so
fundamental a change, so please share what you know about the
decision-making process.
Bill Horne, who remembers when the Dormphone SxS was in the basement of
Walker Memorial and had "pheeping" flags on it.
David Clayton
>>> ... ISDN, at least for me, was indeed "Is Still Doing Nothing". I
>>> understand it's still around, but why escapes me, with it's painfully
>>> slow speed compared to DSL.
>
> ISDN is still circuit switched 64K channels, known as B for Bearer
> channels. It's a substitute for POTS, not for broadband.
>
> Since it's digital end to end, the voice quality is better than analog,
> and as has been noted you can bond multiple B channels together to get
> better fidelity. ISDN lines also have a D channel used for call setup and
> (in theory although rarely in practice) moderate speed data, so ISDN
> provides a full set of spiffy calling features via the D channel.
The ISDN D channel is used in Australia for at least one service by
the major telco here, and that is to cheaply transport small amounts
of EFTPOS data from retailers to banks via an X25 network that
connects to the D channel in the exchange (the Telstra "Argent"
product).
It saves the cost of an individual data link, but nowadays is being
outmoded by secure Internet based EFTPOS links.
--
Regards, David.
David Clayton
Melbourne, Victoria, Australia.
Knowledge is a measure of how many answers you have, intelligence is a
measure of how many questions you have.
John Levine
>T1 was two-pair when they installed it in 1995, but a couple of years
>ago they switched to HDLC, so there was one pair coming to the house,
Sorry, too many TLAs today. That's HDSL, a flavor of DSL that is
symmetric in both directions and is intended to replace T1s. It runs
on one pair, can go 12Kft between repeaters, and is less sensitive
to funky cable.
I just droped by the local telco who tells me their low end business
fiber product is 5mb down/512kb up for $69/mo which is not that much
more than the so-so DSL I have now.
R's,
John
David Lesher
> I don't think anyone's still installing traditional two-pair T1. My
> T1 was two-pair when they installed it in 1995, but a couple of
> years ago they switched to HDLC, so there was one pair coming to the
> house, and a little line powered box that turned it back into the
> traditional two pairs I plugged into my CSU/DSU.
>***** Moderator's Note *****
> Please tell us how "HDLC" is used in this context, and which
> physical-layer protocol is used: the highest-speed two wire data
> line I'm familiar with is IDSL, but that tops out around 144 kbps.
The LEC {be it I or C} sells you a DS1 circuit at your location. They
deliver it over 2 pairs from their SmartJack [tm, I'm sure] on your
wall.
How they get it to you is THEIR call. It could be circa 1961
[yes.. JFK, mini-skirts, etc] T-1, two pairs, repeater in the middle
of each 6Kf segment [amazingly, right where the "paint can" loading
coils were already, hmmm.] of cable, etc.
OR they could use other means. Maybe there's fiber in your basement
already. [If you're in a skyscraper....] Or a SLC-FO next door. Or
they use single pair HDSL, and a different color module in the
SmartJack.
I was told by an installer that is was an Engineering option. T-1
needs two pairs, stomps over ADSL sharing the same bundle, and such;
but it can used repeaters to get a LONG way, up to {ISTM} 150 miles
before the jitter gets out of hand. [Such must use alternte power
arrangemnets..]
The HDSL was good to ~18Kft, but did not allow repeaters. But I was
just digging into the ADC catalog, and there they are; I assume they
are a new offering. I know someone at ~40Kft who must have two
repeaters, but I've seen only the one.
--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433
Garrett Wollman
Bill Horne appended the following to my text:
> Please tell us more about the phone(s) and the choices each offers.
There's really not much to say.
> I'm especially interested in the encryption tool(s) which each
> vendor and/or system has available.
The only ones I know about that have crypto built-in are the Linksys
ones, and they can have private keys installed as a part of
provisioning. Since I never had access to their provisioning tools, I
was never able to set them up. They used a proprietary (or at least
undocumented) key-exchange protocol transported in the SIP HELLO
message. They also used fairly short keys and something they called a
"mini-certificate", not standard PKIX or OpenPGP.
> I'd also like - very much - to know what factors (other than money)
> drove the cutover: it's very rare for a large organization to make so
> fundamental a change, so please share what you know about the
> decision-making process.
I was/am not privy to information of that sort. Different division of
the Institute, at a much higher pay grade. The only thing that I was
ever told in a public forum was that the cost of maintaining two
parallel infrastructures, one for voice and one for data, was deemed
unsupportable.
-GAWollman
Bill Horne
>
>> I ordered a PRI for one client who wanted the Nortel BCM system (a
>> PBX), and also PRIs for clients who wanted the asterisk VoIP
>> system. Amazingly (to me) all PRIs were actually fiber lines to the
>> MPOE where they were converted to <something> (2-pair) over Cat5
>> that connected to the PCI card(s) in the asterisk system(s).
>
> I don't think anyone's still installing traditional two-pair T1. My
> T1 was two-pair when they installed it in 1995, but a couple of years
> and a little line powered box that turned it back into the traditional
> two pairs I plugged into my CSU/DSU.
>
> There's tons of used T1 equipment available, on ebay and specialized
> places. I got a WANIC T1 card on ebay for under $100, probably
> because nobody else knew what it was.
Do you prefer to use a CSU/DSU because that's what you had on hand, or
it's what your router is equiped for, or are there advantages, such as
V.35 access for test equipment, that make it a better setup?
Bill Horne
--
If music be the food of love
then laughter is its queen
and likewise if behind is in front
then dirt in truth is clean
- Keith Reid
John Levine
> John Levine wrote:
>> There's tons of used T1 equipment available, on ebay and specialized
>> places. I got a WANIC T1 card on ebay for under $100, probably
>> because nobody else knew what it was.
>
> Do you prefer to use a CSU/DSU because that's what you had on hand,
> or it's what your router is equiped for, or are there advantages,
> such as V.35 access for test equipment, that make it a better setup?
a) It's what my cards were set up for.
b) Easier to test. The last CSU/DSU I used had an Ethernet jack and I
could telnet into it from behind my firewall.
c) Sacrificial relatively cheap equipment in case of lightning strike
***** Moderator's Note *****
I hadn't thought about the lightning protection aspect: I'll keep that
in mind if I'm ever in the market for HDSL or other similar wire-based
services.
What, by the way, does an HDSL line cost in round numbers? Does your
ISP use it to deliver your phone connections as well?
Bill Horne
Moderator
John Levine
> ISP use it to deliver your phone connections as well?
It cost about $200/mo for the T1, and $300/mo for the Internet
service. That was a great price in 1995 but it's ridiculous now. It
was just data, the phone was (and is) on a separate pair. The T1
price is distance independent which in my case was bad since the
distance was only three blocks.
Telco is now touting their business fiber service, where I can get 5Mb
down/512Kb up for $69/mo with a three year commitment and a rather
vague pricing for IP addresses. Or I can pay $129 for 10Mb/768Kb, or
$128 for the bundle of 10Mb/768Kb, a phone line with 1c local, 5c toll
calling, and free installation.
This is biz service. They say eventually they'll have resi fiber, but
not for a while. Maybe I'll wait, the DSL is only $40 and is
moderately sucky, about 3M/300K.
R's,
John
***** Moderator's Note *****
My DSL is about 600k/100k, and I think that qualifies as _really_
sucky: too slow even for Vonage.
My brother, however, has FiOS, and he gets about 5,800k/1,280k: about
the same as cable around here, but far from what fiber _can_
deliver. He says that Verizon is throttling usage to sell higher-tier
services, but that the tier he's at now beats the separate internet &
phone prices.
Whatever happened to the electricity that was going to be too cheap to
measure?
Bill Horne
Moderator
Michael Grigoni
<snip>
> Telco is now touting their business fiber service, where I can get
> 5Mb down/512Kb up for $69/mo with a three year commitment and a
> rather vague pricing for IP addresses. Or I can pay $129 for
> 10Mb/768Kb, or $128 for the bundle of 10Mb/768Kb, a phone line with
> 1c local, 5c toll calling, and free installation.
>
> This is biz service. They say eventually they'll have resi fiber,
> but not for a while. Maybe I'll wait, the DSL is only $40 and is
> moderately sucky, about 3M/300K.
>
> ***** Moderator's Note *****
>
> My DSL is about 600k/100k, and I think that qualifies as _really_
> sucky: too slow even for Vonage.
I have had 256k symmetric DMT for a number of years at about $39.00/mo
with a voice pair (Qwest, separate ISP) and have never experienced any
reduction in the allocated bandwidth. Perhaps I am lucky in how they
provisioned it at my QTH?
> Whatever happened to the electricity that was going to be too cheap
> to measure?
Environmentalists. Lack of national focus and urgency. Energy should
have the same priority as health care and war-making and is a lynch
pin of "national security". How about a discussion for a little while
here regarding a proposal to roll-out a significant number of fission
reactors for a "25-year plan" during which time a real effort to
mature fusion would be undertaken?
Michael
***** Moderator's Note *****
I opened the door, so I'm allowing your question. I'm also closing the
door: a debate about nuclear power isn't related to telecom. Sorry.
Bill Horne
Moderator
T
says...
> It cost about $200/mo for the T1, and $300/mo for the Internet
> service. That was a great price in 1995 but it's ridiculous now. It
> was just data, the phone was (and is) on a separate pair. The T1
> price is distance independent which in my case was bad since the
> distance was only three blocks.
>
> Telco is now touting their business fiber service, where I can get 5Mb
> down/512Kb up for $69/mo with a three year commitment and a rather
> vague pricing for IP addresses. Or I can pay $129 for 10Mb/768Kb, or
> $128 for the bundle of 10Mb/768Kb, a phone line with 1c local, 5c toll
> calling, and free installation.
>
>
Yesterday I did some work for a client who is orginally from France. She
was telling me that for 30 euros ($40) you can get 100mbps down/50mbps
up net service along with 300 channels of video service, and telephone
service that covers not only France but 100 other countries.
This is for residential service of course but still, if you could ever
get 100mpbs service here in the U.S. I'd imagine the price would be
astronomical. My 20/5 service is $53.99 a month.
And Vonage just hiked their bogus fees by $1.72 a month. Nice of them.
Dan Lanciani
>
> My DSL is about 600k/100k, and I think that qualifies as _really_
> sucky: too slow even for Vonage.
Until recently I had 768k/128k. Eventually I was able to upgrade to a
whopping 1M/384k but the service still does not seem suitable for
VOIP. The funny thing is that the problem appears to be mainly in the
downstream direction. The various web sites that offer tests for VOIP
suitability seem to concentrate on the upstream path. I'd really like
to understand better what is going on...
Dan Lanciani
d...@danlan.com
T
says...
Who is providing your DSL service? I've often suspected the likes of
Verizon and other companies of deliberately blocking VoIP traffic if
it isn't destined for THEIR servers.
***** Moderator's Note *****
In the case of Comcast, port blocking has already been proven. I wrote
an article some time back, asking readers to contact their congressmen
and support the "net neutrality" doctrine.
Bill Horne
Moderator
David Clayton
On assumes that your VoIP connection directly connects into the last
piece of network equipment that connects to the ADSL modem and that it
has QoS working correctly?
Dan Lanciani
|In article <2010021709...@ss10.danlan.com>, d...@danlan.com
|says...
|> Until recently I had 768k/128k. Eventually I was able to upgrade to a
|> whopping 1M/384k but the service still does not seem suitable for
|> VOIP. The funny thing is that the problem appears to be mainly in the
|> downstream direction. The various web sites that offer tests for VOIP
|> suitability seem to concentrate on the upstream path. I'd really like
|> to understand better what is going on...
|Who is providing your DSL service?
Verizon MA
|I've often suspected the likes of
|Verizon and other companies of deliberately blocking VoIP traffic if
|it isn't destined for THEIR servers.
It's probably not that simple. I see similar problems regardless of
whether I go directly or through an encrypted tunnel. The latter
hides the details from the local ISP and it's a rather non-standard
setup (IPSEC inside L2TP) so I doubt they are looking at, e.g., the
packet sizes.
The problem is not total failure, and in fact sometimes it works
acceptably. But a lot of the time I get long dropouts and the router
providing the voice port shows lots of silence fill. I've gone so far
as to write programs to simulate packet streams similar in size and
timing to RTP streams, but at least when I run them I'm not seeing
either drops or out-of-sequence packets. Of course, the end points
are not the same as those of any of the VOIP services I've played
with so I suppose the problem could be far-end rather than near.
The simplest explanation is that the DSL downstream is just way
oversubscribed and happens to perform poorly often enough to make
for a bad VOIP experience. If this is typical, though, I wonder
how things like MagicJack work acceptably. Maybe I'm just too
sensitive...
Dan Lanciani
d...@danlan.com