[telecom] Re: Are any of the #5 Crossbar offices still in service?
Curtis R Anderson
> I was talking with an old friend from my time at Ma Bell,
> and the subject of Number 5 Crossbar came up.
>
> Are any of the 5XB offices still in service? If so, where?
Other than the Museum of Communications in Seattle and other similar
places, I wouldn't think so, at least on the PSTN in the US and Canada.
My possible reasons include:
1. The originating registers can't hold all the digits to support
dialaround to all those long distance carriers.
2. Finding paper tape for the centralized automated message accounting
punches wich record toll calls can be problematic. Finding and
maintaining equipment to read the paper tape is more problematic.
3. Originating a call to another switch would require in-band MF pulsing
the originating telephone number to the terminating switch or toll
tandem, so caller ID can be maintained.
4. Users want custom calling features like call waiting.
5. Users want a clean connection so when they do dial up a modem
connection, they can get speeds approaching 56kbps.
Where I live, the central office decommissioned their last #5XBs in the
early '90s. Half the customers at that time were on a #1AESS, and I
believe we were all cut over to a #5ESS at that time, which we still use
today.
--
Curtis R. Anderson, Co-creator of "Gleepy the Hen", still
"In Heaven there is no beer / That's why we drink it here ..."
http://www.gleepy.net/ mailto:gle...@intelligencia.com
mailto:gle...@gleepy.net (and others) Yahoo!: gleepythehen
**************************
Moderator's note:
That's a shame. I always loved "number 5": there were two 5XB offices
in Boston's Back Bay CO when I started at NET&T in 1972: they
installed 1A soon afterwards, but it was to replace the panel offices
that were in service when I joined.
There were some computerized replacements for the paper punch system,
and IIRC they were used toward the end of the crossbar machine's
life.
BTW, the connections through 5XB were clean enough for 56Kbs DDS
traffic: if V.whatever modems couldn't use them, that's not the
fault of the switch.
<soapbox>
If uses want custom calling features, it's because they never stopped
to think that they were paying good money just to allow someone else
to impose on them. When did we all get the notion that someone else
was entitled to put us in charge of managing their time? I don't care
if you want to talk to me when I'm not home: that's your problem. I
don't care if you want to talk to me while I'm talking to someone
else: that's life.
Caller ID is just a way to avoid bill collectors. If I feel like
answering the phone, I do it. If I don't, I don't. I don't care who's
on the other end of the line: _I_ make a choice about how to spend my
time, and if that choice proves wrong, I hang up.
And when did people forget that a busy signal _IS_ a message? It means
"I'm busy. Call back later!"
</soapbox>
Sigh. I still have a four-column Touch Tone pad that I got out of the
Crossbar offices after they were retired. I used to use it for ham
radio repeater control, until all the radios came with it built-in.
I feel old. Sic transit technology.
Bill Horne
Diamond Dave
<telecom...@comcast.net> wrote:
>Are any of the 5XB offices still in service? If so, where?
The last working #5XB that I am aware of is at the Museum of
Communications in Seattle, WA.
A similar working #3XB (smaller compact version of the #5XB) is at The
Telephone Museum in Ellsworth, ME.
As for the public network, last time I heard of a #5XB was around 1992
or so. The last step by step switch in the US was changed to a digital
switch in 1999. The last step by step switch in Canada was changed to
a digital switch in 2001.
There are a handful (approximately 50 to 70) Western Electric #1AESS
switches in the public network. Three in Verizon territory (one each
in Baltimore, MD; Richmond, VA and Norfolk, VA), none in Qwest, and
the bulk of the rest are in old BellSouth areas. I'm not sure about
SBC/at&t territory, but there may be a few there.
The reason why the old step and crossbar switches are gone is
primarily for Equal Access requirements. I think the FCC mandated that
Equal Access had to be available nationwide by the year 2000. Hence
the last handful of step switches (and XY cousins) were taken out by
1999.
Crossbar switches had rigid dial registers and routings. They tended
to be the first switches to go in the mid to late 1980s. They also
could not handle IDDD (International Distance Direct Dial) and
couldn't handle the extra digits for Equal Access codes. The dial
resgisters had a hard limit of 10 digits. Some select XB witches were
upgraded as test cases, but Western Electric found it was too
difficult and costly.
Independent analog switches (AE, SC, North Electric) came next (same
issue as crossbar - rigid dial registers and routings - and couldn't
handle SS7).
#2 and #3ESS switches couldn't be upgraded either (would have cost too
much to redo the programming - cheaper to replace with a #5ESS
remote).
Step by Steps came later (they normally didn't handle toll by
themselves, usually routed & billed via tandem)
The #1AESS analog switches are last (since they could upgraded to
handle Equal Access and SS7 signalling with an additional stand alone
processor).
The first #4 crossbar toll tandem was installed in 1943, the last #4A
crossbar toll tandem was installed in 1976. The first #4ESS toll
tandem was installed in 1976 as well. The #4ESS replaced the #4A
crossbar by the late 1980s.
The last #4ESS tandem was installed in 1999. These days, #5ESS and
DMS-250 switches as well as packet switches are now being used as toll
tandems. As of 2007, there are still over 140 #4ESS tandems stil in
the AT&T long lines network. MCI (now Verizon Business) and Sprint LD
use primarily DMS tandems, with some older DEX tandems in MCI (unless
they've been replaced with DMS).
Also, the TDM (time domain) switch (5ESS, DMS-100, etc.) is
approaching obsolesence. Packet switches are now being installed in
some places, and we all know the dreaded four letters - VoIP. Packet
switching is much more efficient, and can handle both voice and data
on the same network (voice is treated like data).
So to quote a famous radio host, "... and now you know the Rest of the
Story."
**************** Moderator's Note ****************
Packet switching is not new: certainly ATM, which is the best
packet switching technology IMHO, is not. What _IS_ new is the
public's willingness to accept gaps in calls, lat
en
cy,
and other interuptions, and abyssmal voice quality - all the
result of a generation which has grown up with cellular trans-
ceivers glued to their ears.
Ma Bell did a lot of things wrnog, but they got the technical
part right: any call on a Bell switch was guaranteed at least
4,000 Hz bandwidth (I exclude Mode 2 SLC, sorry). In addition,
there were good standards for echo return loss, echo, etc.,
which made the U.S./Canada network the envy of the world.
This is the only place on the globe where people pick up a
telephone and dial a number without bothering to listen
for a dial tone. That's going to change, because VoIP is
not a "Bell System" grade service, and it can't be made into one,
no matter how hard they try.
Bill Horne
Sam Spade
> From: Moderator <telecom...@comcast.net>
> Subject: Are there any #5 Crossbar offices still in service?
> Message-ID: <200708...@billhorne.homelinux.org>
> Errors-To: telecom...@comcast.net
> Approved: mode...@telecom-digest.org
> Newsgroups: comp.dcom.telecom
> Organization: TELECOM digest
> Sender: telecom...@comcast.net
> Precedence: list
> Date: Mon 13 Aug 2007 02:43:00 -0400 (EDT)
>
> I was talking with an old friend from my time at Ma Bell,
> and the subject of Number 5 Crossbar came up.
>
> Are any of the 5XB offices still in service? If so, where?
>
> Bill Horne
As I understand it, they are long gone from any U.S. service territory.
They cannot handle SS7, which is a show stopper for starters.
Some of them were probably crated up and sent to other parts of the
world where there is no money for electronic switching.
Sam Spade
>
> This is the only place on the globe where people pick up a
> telephone and dial a number without bothering to listen
> for a dial tone. That's going to change, because VoIP is
> not a "Bell System" grade service, and it can't be made into one,
> no matter how hard they try.
>
> Bill Horne
>
>
I have had Vonage since its inception. There were issues the first
year, such as echo, etc. Now, it is every bit as good as any "Bell
System" line.
AT$T has gotten so greedy, at least in California, that the Los Angeles
Times wrote an editorial recommending that everyone who could switch to
wireless or VOIP dump AT&T because of their predatory pricing.
Here is what AT&T in California now charges for everything I get with
Vonage for $24.99 per month:
Line 10.69
Message Center 8.95 (activation fee $19.95)
Call forwarding 5.00 (activation fee $7.50)
Call waiting 5.00 (activation fee $7.50)
Caller ID 9.00 (activation fee $7.50)
Three-way 5.00 (activation fee $7.50)
Unlimited nation-wide calling $30.00
Unlimited calling to Canada 5.99
Total $70.63 per month.
Plus, with Vonage, I have much more sophisticated options with voice
mail and call forwarding.
Also, I can have virtual numbers wherever I please. I live in Southern
California and my primary number is in Washington, DC. (I have a
virtual number local to my California location).
The DC number is like having a multi-thousand dollare AT$T foreign
exchange line.
Bill Horne
Sam,
I know another businessman who's had good luck with Vonage: if it works,
don't fix it. My personal opinion (I'm "stepping out of the chair" here)
is that the monthly cost of your Cable or DSL service must be added to
the $30/month you pay Vonage in order to get the real cost: you're
probably paying from $60 to $90/month for the combined total, but
I hope you continue to have good luck with the Vonage service.
Of course, Vonage isn't the only player: I use the Comcast phone
service, but when the $100/month "Triple Play" promotion is over, I'm
going back to copper. The service has very noticeable latency and
dropouts, and it lacks features that I like to have, such as the ability
to turn off call waiting while on a call, and being able to erase
messages before I've listened to the entire recording.
My objection to VoIP as a replacement for Virtual Circuit switching is
theoretical as well as practical: there's no specification in the
Internet Protocol for minimum transit time, which means that there's no
way to prevent latency and other degradations caused by packet transit
times without compression and queuing delays at the sending station.
Even then, there's no QOS quarantee: it's just not in the protocol.
Bill
--
Remove QRM from my address to email me directly!
T
gle...@gleepy.net says...
> Bill Horne wrote:
>
> > I was talking with an old friend from my time at Ma Bell,
> > and the subject of Number 5 Crossbar came up.
> >
> > Are any of the 5XB offices still in service? If so, where?
>
> Other than the Museum of Communications in Seattle and other similar
> places, I wouldn't think so, at least on the PSTN in the US and Canada.
> My possible reasons include:
>
> 1. The originating registers can't hold all the digits to support
> dialaround to all those long distance carriers.
>
> 2. Finding paper tape for the centralized automated message accounting
> punches wich record toll calls can be problematic. Finding and
> maintaining equipment to read the paper tape is more problematic.
>
> 3. Originating a call to another switch would require in-band MF pulsing
> the originating telephone number to the terminating switch or toll
> tandem, so caller ID can be maintained.
>
> 4. Users want custom calling features like call waiting.
>
> 5. Users want a clean connection so when they do dial up a modem
> connection, they can get speeds approaching 56kbps.
>
> Where I live, the central office decommissioned their last #5XBs in the
> early '90s. Half the customers at that time were on a #1AESS, and I
> believe we were all cut over to a #5ESS at that time, which we still use
> today.
>
Interesting history! I'm reading the "Switching Technologies" book right
now and I seem to recall that Bell experimented with common control for
things like SxS and even better CC for #1 and #5 Crossbars.
In the end it was all for naught though since the #1 ESS and more
importantly #4 ESS spelled the end of crossbar.
Garrett Wollman
Bill Horne <telecom...@comcastQRM.net> wrote:
>My objection to VoIP as a replacement for Virtual Circuit switching is
>theoretical as well as practical: there's no specification in the
>Internet Protocol for minimum transit time, which means that there's no
>way to prevent latency and other degradations caused by packet transit
>times without compression and queuing delays at the sending station.
>Even then, there's no QOS quarantee: it's just not in the protocol.
It doesn't *need* to be in the protocol; that's the beauty of it. A
properly-engineered network (perhaps not the easiest thing to come by,
particularly from a cableco, but they do exist) and a small playout
buffer at each end (which you need anyway for echo cancellation) are
all you really need.
Now I wouldn't switch my home phone service to VOIP, but that is as
much because (as a network engineer myself) I need to be able to reach
people even when the Internet connection is broken. But my office
phone is VOIP and I've been quite happy with it. (Our campus is
finally getting ready to give the heave-ho to the 5ESS when it next
comes up for renewal and replace it with not one but two independent
VOIP infrastructures. There's something like 12,000 lines to
convert, and I'm thinking of starting a pool on when the last AT&T
7506 will leave service. Is there anyone else on Earth still using
those things?)
At some point, things will get reliable enough (or I'll have enough
redundant connectivity) that I'll be able to switch to VOIP at home.
I wonder whether that will happen before or after Verizon itself
switches to VOIP?
-GAWollman
--
Garrett A. Wollman | The real tragedy of human existence is not that we are
wol...@csail.mit.edu| nasty by nature, but that a cruel structural asymmetry
Opinions not those | grants to rare events of meanness such power to shape
of MIT or CSAIL. | our history. - S.J. Gould, Ten Thousand Acts of Kindness
Bill Horne
> In article <B-6dnUylqfAsKl7b...@comcast.com>,
> Bill Horne <telecom...@comcastQRM.net> wrote:
>
>> My objection to VoIP as a replacement for Virtual Circuit switching is
>> theoretical as well as practical: there's no specification in the
>> Internet Protocol for minimum transit time, which means that there's no
>> way to prevent latency and other degradations caused by packet transit
>> times without compression and queuing delays at the sending station.
>> Even then, there's no QOS quarantee: it's just not in the protocol.
>
> It doesn't *need* to be in the protocol; that's the beauty of it. A
> properly-engineered network (perhaps not the easiest thing to come by,
> particularly from a cableco, but they do exist) and a small playout
> buffer at each end (which you need anyway for echo cancellation) are
> all you really need.
I'm going to disagree with you on this point. A properly engineered
network has to have provision for prioritization of traffic during
mass-call events, which the Internet can't provide. Even in "normal"
periods of traffic, there's no viable way to distinguish VoIP from data
from video, etc.: at the heart of the Internet's design is the belief
that traffic can be queued _at the originating end_ until bandwidth is
available, and the net was never meant for real-time communications.
After all, nobody cares if their email arrives at 09:00:00 or 09:00:30,
but such delays are anathema to VoIP.
>
> Now I wouldn't switch my home phone service to VOIP, but that is as
> much because (as a network engineer myself) I need to be able to reach
> people even when the Internet connection is broken.
That's the point I hope I'm making: the Internet connection is _already_
broken as far as VoIP is concerned - there's no way to guarantee you a
reliable circuit, even during an emergency.
> But my office
> phone is VOIP and I've been quite happy with it. (Our campus is
> finally getting ready to give the heave-ho to the 5ESS when it next
> comes up for renewal and replace it with not one but two independent
> VOIP infrastructures. There's something like 12,000 lines to
> convert, and I'm thinking of starting a pool on when the last AT&T
> 7506 will leave service. Is there anyone else on Earth still using
> those things?)
I'm not sure an office phone system, or even a campus-wide network, is a
valid model for using VoIP on the Internet. I agree that VoIP can be
made to work in a controlled environment where all the factors are
accounted for and planned into the system, but that's like saying that
solar power will replace fossil fuels: it doesn't allow for pushback
from the embedded interests whose ox is going to be gored, nor for the
(often hidden) costs of updating phones and other infrastructure to
connect to the new system, and especially not for the net's inability to
assure bandwidth to any given customer at any given instant.
> At some point, things will get reliable enough (or I'll have enough
> redundant connectivity) that I'll be able to switch to VOIP at home.
> I wonder whether that will happen before or after Verizon itself
> switches to VOIP?
I think you're making my argument for me here: at some point, there
_might_ be reliability in VoIP, but it'll need changes to IP to make it
happen, and I don't see the IETF as being behind such a paradigm switch.
As it stands now, the Internet is effectively an "Aloha" network, where
everybody sends and then sends again if the first time didn't work.
There's no capability in the net, at any point, to deliver virtual
circuits - there are just a lot of under-utilized pipes that are "OK"
for now, but which will fill up as more uses are found for them.
In any case, Verizon won't be switching to VoIP anytime soon: they have
plenty of ATM switches to do the heavy lifting if they want to go with
an all-packet network, and ATM was designed with real-time traffic in mind.
Bill
--
Remove QRM from my address to email me directly!
Remember, if your post doesn't have "[telecom]" (no quotes) in the
subject line, I'll never see it.
Sam Spade
> Interesting history! I'm reading the "Switching Technologies" book right
> now and I seem to recall that Bell experimented with common control for
> things like SxS and even better CC for #1 and #5 Crossbars.
I can only speak to the history of Pacific Telephone. In the Los
Angeles area they had a whole bunch of both urban and suburban steppers.
Bell Labs/Western Electric could never get common control to work for
steppers. So, they installed a crossbar switch in each C.O. to handle
the common control requirements for regional toll and nationwide direct
dialing. Those crossbars were also growth end-office machines.
>
> In the end it was all for naught though since the #1 ESS and more
> importantly #4 ESS spelled the end of crossbar.
>
General Telephone of California wouldn't invest in crossbar. So,
Automatic Electric developa piece of poorly working equipment called
"The Director," which handled toll for all their steppers. Those
directors had a high failure rate in processing toll calls, which made
it appear the steppers were failing. It wasn't the steppers, though,
because call completion on non-toll calls was very good.
Geoffrey Welsh
> the monthly cost of your Cable or DSL service must be added to the
> $30/month you pay Vonage in order to get the real cost
... unless you were going to pay that for an internet connection anyway (and
will continue to pay for it if you switch back to POTS.)
> My objection to VoIP as a replacement for Virtual Circuit switching is
> theoretical as well as practical: there's no specification in the
> Internet Protocol for minimum transit time, which means that there's
> no way to prevent latency and other degradations caused by packet
> transit times without compression and queuing delays at the sending
> station. Even then, there's no QOS quarantee: it's just not in the
> protocol.
QoS is a bit of a myth: it takes a certain minimum amount of time to get a
packet (or any other signal; see Relativity) from London to Tokyo no matter
what features the protocol implements. What QoS does is slow down some
traffic more than others where significant congestion is encountered; as
other have pointed out, the solution to this (and the only way to keep all of
your customers happy, QoS or not) is to engineer a network where congestion
is not a normal event. It is, on the other hand, valid to observe that
packet and TDM networks degrade differently: while you may experience latency
on a VoIP call, you might in stead hear a fast busy (i.e., be unable to place
the call at all) on an SS7 network. It is left as an exercise for the reader
to determine which is preferrable in their environment.
--
Geoffrey Welsh <Geoffrey [dot] Welsh [at] bigfoot [dot] com>
Just when you think something is behind you, that's when it's in the
perfect position to bite you in the ass.
.
T
says...
And my two favorite Vonage features are Do Not Disturb, and the freebie
calls to Italy, UK, Spain, etc. I managed to track down a particular
vintner in Spain and get a case of wine for relatively short money
because of that.
And the SO has gotten info on a suit found at an estate sale by calling
the UK.
T
dig...@comcastQRM.net says...
It's not an apples-apples thing regarding ISP charges. For example, I
get MUCH more use out of my cable net service than most people. My
Vonage service runs off it as well as my Vudu service and even Joost.
T
wol...@bimajority.org says...
> In article <B-6dnUylqfAsKl7b...@comcast.com>,
> Bill Horne <telecom...@comcastQRM.net> wrote:
>
> >My objection to VoIP as a replacement for Virtual Circuit switching is
> >theoretical as well as practical: there's no specification in the
> >Internet Protocol for minimum transit time, which means that there's no
> >way to prevent latency and other degradations caused by packet transit
> >times without compression and queuing delays at the sending station.
> >Even then, there's no QOS quarantee: it's just not in the protocol.
>
> It doesn't *need* to be in the protocol; that's the beauty of it. A
> properly-engineered network (perhaps not the easiest thing to come by,
> particularly from a cableco, but they do exist) and a small playout
> buffer at each end (which you need anyway for echo cancellation) are
> all you really need.
>
> Now I wouldn't switch my home phone service to VOIP, but that is as
> much because (as a network engineer myself) I need to be able to reach
> people even when the Internet connection is broken. But my office
> phone is VOIP and I've been quite happy with it. (Our campus is
> finally getting ready to give the heave-ho to the 5ESS when it next
> comes up for renewal and replace it with not one but two independent
> VOIP infrastructures. There's something like 12,000 lines to
> convert, and I'm thinking of starting a pool on when the last AT&T
> 7506 will leave service. Is there anyone else on Earth still using
> those things?)
A 7506? I'm familiar with the 7406D because we still use those, along
with 8410D and 6408D sets.
The nice thing is that our office is wired so that we could switch the
phone wiring terminations around and use VoIP throughout without the
traffic riding on our main network.
We've got a Prologix G3iV11 and an older Definity G3iV6 at another
location.
Sam Spade
>
> Sam,
>
> I know another businessman who's had good luck with Vonage: if it works,
> don't fix it. My personal opinion (I'm "stepping out of the chair" here)
> is that the monthly cost of your Cable or DSL service must be added to
> the $30/month you pay Vonage in order to get the real cost: you're
> probably paying from $60 to $90/month for the combined total, but
> I hope you continue to have good luck with the Vonage service.
I need broadband for the Internet, both business and personal. Thus, as
an accountant, I view the pipe is already there "for the taking" for
VOIP. The accounting case becomes overwhelming when I take my Vonage
adapter with me and plug it into a "free" Ethernet port at a hotel, or such.
>
> Of course, Vonage isn't the only player: I use the Comcast phone
> service, but when the $100/month "Triple Play" promotion is over, I'm
> going back to copper. The service has very noticeable latency and
> dropouts, and it lacks features that I like to have, such as the ability
> to turn off call waiting while on a call, and being able to erase
> messages before I've listened to the entire recording.
Call waiting is even better with Vonage. I am, in effect, the
"switchmeister" because I can go on-line and enter the feature and
remove the feature at will. The wireline LECs should have done this
years ago, but they didn't. Instead, they are making calling features a
big reveue honeypot for their greedy benefit and it is going to bite
them. Not so many people I know are dropping wireline for VOIP, but a
lot of them are dropping wireline for wireless, at least for residential
use.
>
> My objection to VoIP as a replacement for Virtual Circuit switching is
> theoretical as well as practical: there's no specification in the
> Internet Protocol for minimum transit time, which means that there's no
> way to prevent latency and other degradations caused by packet transit
> times without compression and queuing delays at the sending station.
> Even then, there's no QOS quarantee: it's just not in the protocol.
>
The end user just doesn't care about such issues unless it affects them.
I have a dedicated Vonage fax port. I recently bought a new fax
machine that does 32,800 and that works just fine with Vonage. But,
even the fax machine is dying out. Secured Acrobat files are a whole
lot better; the limiting factor being who you can send them to.
The question of latency and other issues in the Internet will ultimately
rise or fall on reasonably priced, really fast bandwidth being available
like it is in other more developed countries. (that stings, doesn't it
;-) Verizon is technically going there but their pricing makes it quite
unattractive.
Bill Horne
[snip]
>
> The question of latency and other issues in the Internet will ultimately
> rise or fall on reasonably priced, really fast bandwidth being available
> like it is in other more developed countries. (that stings, doesn't it
> ;-) Verizon is technically going there but their pricing makes it quite
> unattractive.
>
Sam,
I guess we'll have to agree to disagree on this one: I don't think that
building a large pipe is appropriate when most of the time it has only a
modest amount of flow: as I said, the Internet is designed around the
idea of time-shifting traffic to get around bottlenecks.
At the moment, we're enjoying the surplus of bandwidth that resulted
from the "dot bomb" speculation of a few years back. That bandwidth
will be absorbed as other players come in and more people connect to
the net, and then it'll be back to "business as usual", which is,
in the case of the Internet, having to wait for your traffic to move
at whatever pace the net has available at that time.
Let's remember that data links were _ananlog_ lines when the Internet
was designed, and the whole idea of a packet-switched network was to
obtain reliable data transfers using multiple unreliable links. It
matters not that the links are now more reliable: IP just doesn't
have the capability of supporting a virtual circuit, and that's what's
required to have reliable telephone service.
YMMV.
Bill Horne
--
Remove QRM from my address to email me directly!
Remember, if your post doesn't have "[telecom]" (no quotes) in the
T
says...
> Bill Horne wrote:
>
> >
> > Sam,
> >
> > I know another businessman who's had good luck with Vonage: if it works,
> > don't fix it. My personal opinion (I'm "stepping out of the chair" here)
> > is that the monthly cost of your Cable or DSL service must be added to
> > the $30/month you pay Vonage in order to get the real cost: you're
> > probably paying from $60 to $90/month for the combined total, but
> > I hope you continue to have good luck with the Vonage service.
>
> I need broadband for the Internet, both business and personal. Thus, as
> an accountant, I view the pipe is already there "for the taking" for
> VOIP. The accounting case becomes overwhelming when I take my Vonage
> adapter with me and plug it into a "free" Ethernet port at a hotel, or such.
I was rather shocked to learn that ever since true competition came to
play in RI, Verizon has lost 55% of their customers. And from what I'm
hearing, not many people are signing up for their FIOS service.
RI is one of the only places where this could actually happen since the
rules for network sharing are pretty one sided in favor of competition.
jsw
>steppers
Actually, Ma Bell did have a small-scale successful deployment of
common control for SxS offices, known as the Director system.
The 'Directorized' SxS systems were not that common, and of course
they did not scale nearly as well as the 'real' common control
switches, but they did exist and they did work quite well.
712-366 was an example of this, the Manawa office in Council Bluffs,
Iowa, which was cut to a DMS-10 in 1984 or so.
The 'director' equipment (IIRC) sat between the linefinder and
first incoming selectors and recorded the dial pulses or DTMF
tones (yes, 712-366 had Touch Tone on step) and then either drove
the selectors for an intra-office call or handled the trunk
selection and outpulsing for an inter-office call.
This scheme reduced the number of selectors necessary for a full
seven-digit numbering plan without the need for the messy
digit-absorbing schemes.
Good day JSW
Al Gillis
> traffic but there is an operating #5 Crossbar at the Museum of
> Communications in Seattle, Washington. There is also an operating #4
> Crossbar, a Strowger stepper and, get this, an operating PANEL switch in
> the
> Museum. All these CO switches operate and you can call from one telephone
> to another while the thing does it's work!
>
> The Museum also has numerous Teletype Corporation machines, gadgets from
> the
> network radio and television Control Centers, microwave radio and carrier
> equipment, switchboards, testboards, cable and other Outside Plant items,
> etc.
>
> It's just a fantistic place and things are maintained by Telephone
> Pioneers
> with some support from Qwest.
>
> If you're in Seattle it's worth a Tuesday morning to go see! Here's a
> URL:
> http://www.museumofcommunications.org/
T
j...@ivgate.omahug.org says...
Bell experimented with lots of things from WW II on. Here's a gas tube
controlled line/link network for a #1 Crossbar:
Roy
>T wrote:
>
>> Interesting history! I'm reading the "Switching Technologies" book right
>> now and I seem to recall that Bell experimented with common control for
>> things like SxS and even better CC for #1 and #5 Crossbars.
>
>I can only speak to the history of Pacific Telephone. In the Los
>Angeles area they had a whole bunch of both urban and suburban steppers.
> Bell Labs/Western Electric could never get common control to work for
>steppers. So, they installed a crossbar switch in each C.O. to handle
>the common control requirements for regional toll and nationwide direct
>dialing. Those crossbars were also growth end-office machines.
I was a Switchman in one of those offices in the late 60's.
Wilmington CA
[snip]
Sam Spade
In that era Pasadena had a giant stepper serving almost all of the area,
with one small XBAR unit doing the toll and seving only one office code.
I believe the XBAR was installed in time for the cut to nationwide
direct dialing (1959 or so?)
Al Gillis
"Roy" <rfo...@nospam.gmail.com> wrote in message
news:2007082603...@massis.lcs.mit.edu...
So Roy... Spin us a yarn or two about the good or the bad days, the really
interesting things or the really dumb things that happened during your time
with steppers or crossbar...
hanc...@bbs.cpcn.com
> T wrote:
> > Interesting history! I'm reading the "Switching Technologies" book right
> > now and I seem to recall that Bell experimented with common control for
> > things like SxS and even better CC for #1 and #5 Crossbars.
>
> I can only speak to the history of Pacific Telephone. In the Los
> Angeles area they had a whole bunch of both urban and suburban steppers.
> Bell Labs/Western Electric could never get common control to work for
> steppers. So, they installed a crossbar switch in each C.O. to handle
> the common control requirements for regional toll and nationwide direct
> dialing. Those crossbars were also growth end-office machines.
In essence SxS remained pretty much the same until the end. However,
the Bell System did implement a number of electronic modern
improvements to SxS to improve its performance and efficiency and
reduce cost.
Full "senderzation" for SxS was complex, but a partial mode was
practical. For plain vanilla calls, the digits were outpulsed as
normal to the steppers. But for toll calls and others , they were
routed to the tandem switch and possibly translated which made for
better trunk usage. Inter-exchange and toll trunks were costly and
much Bell System effort was intended to make good use of trunks--
enough to meet service needs of peak demand, but not so much as to be
wasteful. Also of course ease of maintenance, debugging for trouble,
and network control.
The Bell System history provides many details on this "add-ons" given
to SxS over the years. This included things like Touch Tone, ANI, and
some AMA. SxS could and did also support Centrex.
Remember when DDD first came out there was no ANI, they used ONI where
an operator keyed in the calling number. This continued in use well
into the 1970s until ANI became universal. The modern #4 ESS toll
switch provided for ONI.
Remember that the type of equipment in a given SxS exchange varied
greatly. Some exchanges had many add-ons, but others couldn't even
support DDD even into the1970s. (When the Bell System went to cheaper
dialed direct rates, customers in exchanges that didn't have DDD still
go the cheaper rates).
We forget that electronics were still extremely expensive even in the
1970s, so relay technology remained cost-effective in an existing
application. In certain small exchanges or PBXs, SxS remained the
cheapest way to go since common control, be it crossbar or ESS
required some mandatory overhead which was wasteful if not many lines
were served. It took a while to make small switches cost effective
with electronics. (The same applied to computers in general, right
into the 1980s classic punched card tabulating machines (1940s design)
were most cost effective than electronic computers in certain limited
applications).
BTW, crossbar could and did handle overseas calls.
Crossbar could also provide Call Waiting and was successfully tried
out, but it was too expensive to implement in production service.
> > In the end it was all for naught though since the #1 ESS and more
> > importantly #4 ESS spelled the end of crossbar.
New generations of electronics significantly reduced the cost and
footprint size of ESS making it desirable to replace electro-
mechanical equipment faster than originally planned. For instance,
many places needed more switching capacity and going to ESS was
cheaper than building a new CO. Technology exploded between 1970 and
1980. ESS also offered special service features that brought in more
revenue. Somewhere in that time frame digital switches and trunking
as well as SS7 came out as well.
>
> General Telephone of California wouldn't invest in crossbar. So,
> Automatic Electric developa piece of poorly working equipment called
> "The Director," which handled toll for all their steppers. Those
> directors had a high failure rate in processing toll calls, which made
> it appear the steppers were failing. It wasn't the steppers, though,
> because call completion on non-toll calls was very good.
Didn't AE have their own version of crossbar?
********* Moderator's Note *************
As a young, poor student studying under the G.I. Bill at Santa Barbara
City College in 1978, it took me about 3 calls to realize that the GTE
payphones on campus had a serious security flaw: the phone worked
normally with or without a coin in the slot, except that without a
coin deposit, the dial didn't function.
The local step exchange had been converted to Touch-Tone(tm) ...
Bill Horne
Temporary Moderator
T
@bbs.cpcn.com says...
> New generations of electronics significantly reduced the cost and
> footprint size of ESS making it desirable to replace electro-
> mechanical equipment faster than originally planned. For instance,
> many places needed more switching capacity and going to ESS was
> cheaper than building a new CO. Technology exploded between 1970 and
> 1980. ESS also offered special service features that brought in more
> revenue. Somewhere in that time frame digital switches and trunking
> as well as SS7 came out as well.
No doubt, in the book Switching Technology 1925-1975 they show the
general improvements for #1 ESS.
Universal Trunk Frames went from 13 feet down to 4.4 feet. The Trunk
Link Network went from 26 feet down to 6 feet 6 inches.
Fred Goldstein
>...
>I can only speak to the history of Pacific Telephone. In the Los
>Angeles area they had a whole bunch of both urban and suburban steppers.
> Bell Labs/Western Electric could never get common control to work for
>steppers. So, they installed a crossbar switch in each C.O. to handle
>the common control requirements for regional toll and nationwide direct
>dialing. Those crossbars were also growth end-office machines.
Steppers and crossbars (#5) had compatibility issues. They
apparently could not have direct trunks between each
other. Steppers, of course (unlike crossbars), passed dial pulses
over the trunks as they were dialed. So in order to get between
them, a tandem was needed. Hence a lot of crossbar tandems (XBT)
were in use in some areas.
In general, Ma Bell avoided mixing the types when possible. So in
crossbar-heavy areas, there were no steppers, period. AFAIK a lot of
older urban areas, like New York, New Jersey and Chicago were like
that. Los Angeles though was heavily step.
Boston was crossbar but its suburbs were steppers. So there were
lots of XBTs. My pre-divestiture New England Telephone CO listing
shows Class 4 offices (XBTs) in, among other places, Dorchester, Fall
River, Boston-Franklin, Boston-Harrison, and Malden. Those are all
places that no longer have any tandems. So not every stepper CO got
an XBT, but the biggest ones did (some of the tandem sites, like
Framingham, had steppers for Class 5) and the other local switches
subtended them.
> > In the end it was all for naught though since the #1 ESS and more
> > importantly #4 ESS spelled the end of crossbar.
During the 1980s, as the network was reconfigured post-divestiture,
tandems were consolidated. The XBTs all went away, mainly replaced
by 4Es and DMS-200s (including dual-use 100/200s). And as the
steppers retired, the replacement switches (mostly 5E and DMS-100
host-remote clusters) all had compatible trunks, so direct trunks
between hosts became the norm. By 1995, LATA 128 (E. Mass.) was down
to five tandems or so (Cambridge, Framingham, Worcester, Lawrence,
Brockton). Traffic growth during the dial-up era led to the addition
of a tandem sector in Newton (to relieve Cambridge) and an additional
switch in Cambridge.
Nowadays, many rural LATAs only have one Bell tandem. Maine, for
instance, has a Portland tandem which reaches all the way north into
The County (Aroostook) and "down east" towards the maritimes,
hundreds of miles away. But GTE, in contrast, never bought in to
tandem consolidation. So Verizon-Illinois, with fewer than 1M lines
(a lot of square miles though, mostly cornfield), has more tandems
today than Verizon-Massachusetts, with several times as many lines.
hanc...@bbs.cpcn.com
> Steppers and crossbars (#5) had compatibility issues. They
> apparently could not have direct trunks between each
> other. Steppers, of course (unlike crossbars), passed dial pulses
> over the trunks as they were dialed. So in order to get between
> them, a tandem was needed. Hence a lot of crossbar tandems (XBT)
> were in use in some areas.
>
> In general, Ma Bell avoided mixing the types when possible. So in
> crossbar-heavy areas, there were no steppers, period. AFAIK a lot of
> older urban areas, like New York, New Jersey and Chicago were like
> that. Los Angeles though was heavily step.
There were special "instant" trunks set up so that a SxS could dial
directly into another type of gear. In later years, SxS switches got
electronic front end buffers that could hold digits if need be. The
Bell History explains this better than I can.
The Phila area had a variety of switching in service at the same time:
Manual with call indicator lights for inward dialed calls*
Manual, no call indicator, operator needed on inward calls*
SxS
#1 Crossbar
#5 Crossbar
Panel
*For dialable manual exchanges, the dialed code was shown, ie FLanders
2-5000 (converted roughly 1962)
For non-dialable manual exchanges, it was shown as Willow Grove 1234
(converted roughly 1959)
A busy town might stay manual longer than a quiet one since it was
easier to add operator positions to an existing exchange; a dial
conversion required everything. A quiet town required operator
presence 24/7 even if traffic was light.
But I would to know, for a given city, how the Bell System looked at
it for upgrades at the end of WW II. Which exchanges would get
crossbar, which would stay manual, etc.? Who would get DDD in what
order? Why were the decisions made as they were? Unfortunately, the
Bell Labs history is weak on operating company decisions.
Years ago I tried asking Bell for historical information but they said
they had no records that far back, I'd have to go to AT&T archives.
(Maybe they actually did but didn't want to open them up.)
Sam Spade
> There were special "instant" trunks set up so that a SxS could dial
> directly into another type of gear. In later years, SxS switches got
> electronic front end buffers that could hold digits if need be. The
> Bell History explains this better than I can.
>
In the 1970s I was served by a GTE stepper in the LA area that was on
the boundary of Pacific Tel territory. It was local calling between the
nearby GTE and Pac Tel offices so they didn't go through a tandem. The
GT stepper had MF trunks connecting directly to those Pac Tel offices
and the Pac Tel XBAR and 1ESS machines would dial pulse on direct
connections to the GTE steppers.
I suspect this all cost a lot of money to keep GTE for many years from
having to bite the big bullet. ;-)
Steven Lichter
before that is was dial pauses. GTE added MF translators to their
directors in about 1973, th also used a system by IBM which did
converting, I don't remember much about those since the only one I had
to deal with was in Edgement near March AFB, I know it was a big pain in
the neck and sat there for years after we cut it out, it was owned by
Bank Of America and they never told us what to do with it, finallly we
took it out a scraped it. I was on a COE crew that went around to all
of the offices in the Southern area of LA an converted them
--
The Only Good Spammer is a Dead one!! Have you hunted one down today?
(c) 2007 I Kill Spammers, Inc. A Rot In Hell Co.
hanc...@bbs.cpcn.com
Note that the use of a tandem office had nothing to do with whether a
call was toll or not, tandems were used to simplify and improve the
usage of inter-office trunks. If calling volume was heavy enough,
there'd be direct trunks. Often offices were linked by a series of
trunks, some direct, some through a tandem, some following a relay
route, etc. The method of inter-office connections depended on the
switch within the offices and traffic volume, as well as future plans.
As mentioned, back around 1970 adding some electronic front ends to
extend the life of a SxS office was clearly cheaper than building an
entirely new office. Remember, the number of raw lines served by SxS
_peaked_ in 1974; that is, they were adding lines and capacity to many
SxS offices, despite replacing others with crossbar or ESS. (The
percentage of all phones served by SxS was going down.)
Sam Spade
tch within the offices and traffic volume, as well as future plans.
>
> As mentioned, back around 1970 adding some electronic front ends to
> extend the life of a SxS office was clearly cheaper than building an
> entirely new office. Remember, the number of raw lines served by SxS
> _peaked_ in 1974; that is, they were adding lines and capacity to many
> SxS offices, despite replacing others with crossbar or ESS. (The
> percentage of all phones served by SxS was going down.)
>
>
GTE of California didn't stop installing steppers until the mid 1980s.
By that time they were choking Pacific Telephone's regional networks in
California.
Steven Lichter
offices that were added to while an EAX office was installed. GTE also
had 3 of its own toll centers which handled most of the inter and intra
lata. I spend my whole time with GTE in COE installing equipment and
since retiring I do some part time contract work.
T
says...
Not surprised that Bell never got common control to work on steppers,
they invested their energy into Panel, #1XB as well as #4XB and #5XB.