FCC now planning "all-IP" phone transition [Telecom]
John Mayson
coupon program for converter boxes, was a migration nightmare, wait
until it's time for the phone system to dump its legacy
circuit-switched system and move to an all-IP communications network.
That day could be coming sooner than you think; the Federal
Communications Commission has just requested comment on its planning
for the transition.
Or here: http://z.mayson.us/dlufy
John
--
John Mayson <jo...@mayson.us>
Austin, Texas, USA
***** Moderator's Note *****
The FCC isn't planning any transition to IP telephony. What it _is_
planning is to show the Congress that it should be careful whose
bureaucratic ox it tries to gore.
Congress dictated progress. The FCC is sabotaging the idea, publicly
and crudely, in order to teach capitol hill just who is in charge of
the phone network. As soon as the "comments" have been received, the
FCC will report to Congress that it is "making progress" and then
shelve the whole thing.
Bill Horne
Moderator
hanc...@bbs.cpcn.com
> If you thought that the digital TV transition, with its billion-dollar
> coupon program for converter boxes, was a migration nightmare, wait
> until it's time for the phone system to dump its legacy
> circuit-switched system and move to an all-IP communications network.
> That day could be coming sooner than you think; the Federal
> Communications Commission has just requested comment on its planning
> for the transition.
What would be costs of such a transition? How much conversion would
be required in the local loop plant, especially in older areas? How
much in central offices such as in the cable vault to convert analog
loop lines to IP digital?
FIOS requires every subscriber to provide their own power and have a
battery backup. Right there is an additional cost to the subscriber.
While I'd say the number of traditional Western Electric sets (ie
2500, 2554) out there aren't too many, there is a huge number of more
modern analog sets that are not compatible with IP.
Unlike broadcast television, where most subscribers were already using
cable TV delivery, not over the air, far more people will require
converters. Further, conversion will be a nuisance since commercial
power would be needed.
Who is gonna pay for all that?
After that is done, what will the benefits be and who will be the
beneficiaries.
I can't help but suspect, given tech history, ordinary subscribers
will pay more and get less. Specialized subscribers who will benefit
will lobby hard for this.
Dave Haber
> If you thought that the digital TV transition, with its billion-dollar
> coupon program for converter boxes, was a migration nightmare, wait
> until it's time for the phone system to dump its legacy
> circuit-switched system and move to an all-IP communications network.
> That day could be coming sooner than you think; the Federal
> Communications Commission has just requested comment on its planning
> for the transition.
>
> More here: http://arstechnica.com/tech-policy/news/2009/12/fcc-plans-for-death-of-circuit-switched-phone-networks.ars?utm_source=rss&utm_medium=rss&utm_campaign=rss
>
> Or here: http://z.mayson.us/dlufy
First they came for the NTSC televisions, and I did not speak out -
because I was not an antique television collector.
They can have my analog phones when they take them from my cold, dead
hands.
--
Dave Haber
Massapequa Telephone, part of the global C*NET System
C*NET 1-798-7619
Paul Hoffman
traditional "switched technology" telephone networking, in favor of
packet switched (internet-style) networking. [AT&T claimed that]
maintaining two parallel networks that accomplish essentially the same
thing was wasteful and uneconomical, and this has caused quite a bit
of posts, especially between collectors and users of 'old telephone
technology' such as me.
At the time I read about AT&T's request I did not realize the petition
was for the technology in and between central offices, and not a bid
to remove all copper twisted pair out to consumer's premises. However
with the ever-increasing amount of fiber either to premises or to
centralized communication near the end users, my [concerns] remain the
same.
Right now I have home [phone] service [from] my CATV provider, and
they give me an RJ11 "dial tone source" jack, that I merely connected
into my existing house IW wiring after lifting the IWs from my
landline protector at the side of my house. So if the outside plant
copper were to go away, or even the CO switch located a block from my
house were to change, big deal. I'm already not using them (at least,
not for my local loop.)
Also, as far as I can tell, POTS stuff is pretty much copper ONLY from
the last CO to the houses, and in many cases not all that way either
as fiber is pushed closer and closer to the customer premises, or
directly into them in some cases. So if switched technology were to
be phased out, AT&T would beef up their internet backbones, surplus
dozens of backbone ESS switches and probably hundreds of local CO
switches, and start to recover enough copper strung throughout the
country to probably defray most of the transition costs.
And as it is now for me, over CATV Coax on the same feed that gives me
my home internet network and cable tv transmissions, things work fine,
right up to and including running it to my 555xbrd as it's one and
only "trunk". So changing from 'switched technology' to 'packet based
technology' does NOT mean the end of grandfathered analog (switched)
telecommunications gear.
Those who have actually built their own outside plant may have another
story (unless they get IP access way down the road where their "feed"
interfaces with existing POTS cabling.)
The real losses:
1. Transmission quality - MAYBE ...although quite frankly my home line is
working as well (if not better) over CATV coax as it did
w/traditional T&R copper back to the CO.
2. Service during power outages and/or other disasters. Even if the
future continues to have battery plants backed up by diesal
generators in central offices, it won't be of much use to those who
are wired up like me. BUT, even if they are backed up well in the
central offices, how about the fiber feeds to local distribution
units in the field? How much battery backup do those
end-of-the-fiber boxes have with them? 6 hours? 12? Two years
ago a transformer blew during a 110-115F heatwave, and it took the
local power company 5 days to replace it and restore power to my
house. If I had still been on T&R copper from the central office
my home phones would have worked because (at least some of them) do
not depend on local power. As it was, since I was off CATV coax
and had a very small UPS to back it up, my home phone was dead
about 3 hours into the commercial power outage. Fortunately it was
not a wide area power grid disruption, so my cell phone still
worked...
In a major earthquake, we'll be back [to] depending on ham radio
operators, as I strongly doubt all those cell towers will remain
operational...
Paul H.
hanc...@bbs.cpcn.com
too.comcast.net> wrote:
> Recently AT&T asked the FCC to set a date to transition completely
> off traditional "switched technology" telephone networking, in favor
> of packet switched (internet-style) networking. [AT&T claimed that]
> maintaining two parallel networks that accomplish essentially the
> same thing was wasteful and uneconomical, and this has caused quite
> a bit of posts, especially between collectors and users of 'old
> telephone technology' such as me.
>
> At the time I read about AT&T's request I did not realize the
> petition was for the technology in and between central offices, and
> not a bid to remove all copper twisted pair out to consumer's
> premises. �However with the ever-increasing amount of fiber either
> to premises or to centralized communication near the end users, my
> [concerns] remain the same.
Where does Verizon fit in all of this? Isn't today's at&t a
relatively small company?
There are several issues in play here, some transparent to the home
subscriber, some possibly of import.
First, today's telephone network is almost all digital even though
most telephone sets are analog. Shortly after your analog signal
leaves your telephone set the telephone company converts it to
digital. This may be done at a line concentrator on a pole, at a
converter box within or near your home (such as with VOIP), or at the
central office.
The question then becomes what kind of digital format will be used to
transmit the signal through the network. Again, that probably will be
transparent to the user. The phone company can choose to make that
signal very high fidelity if it wants, though it will most likely keep
it at the 4 KHz it has been for years for land line and whatever
miserly bandwidth it gives up for cell phones.
That is, if the phone company were to use a new kind packet switching,
it could make it high fidelity or crap, but it could do so now with
its existing transmission.
Second, we must remember the real difference between fibre and copper
(and coax) is capacity. If they replace your copper line with coax or
fibre there is no reason your home telephone set _has_ to change.
As far as I know, other than Touch Tone and fancy features, a
telephone set today is functionally exactly the same as a 1938
telephone set (300 set with F handset). A 1938 set works fine today
and a modern set would work fine on the 1938 network (except for Touch
Tone and some sets offer pulse as an switchable option).
To me, a big change _may_ come where the subscriber's individual
telephone set will do the analog-digital conversion and so emit
digital signals. It likely will use a new carrier-signal and ringing
current instead of the 48V DC and 90V AC 20 Hz used now. But at that
point almost all telephone sets will be obsolete. Subscribers may
have to buy adapters just as rabit ear TV owners had to do. (I
suppose some business sets are digital now.)
As electronics have gotten so cheap, conversion to a uniform standard
isn't as important as it once was. For instance, Amtrak's Northeast
Corridor is powered by ancient 25 Hz current. Years ago there was
consideration to convert this to modern 60 Hz current but it isn't as
big an issue as in the past, certainly not one to justify the cost of
conversion. They're upgrading the power supply but keeping it at 25
Hz.
I'm not sure the at&t claim (it's lower case now, right?) of parallel
networks being a pain is correct. It's digital signals using a
protocol. When we make a telephone call, we do not get dedicated use
of a given digital line between two points. Rather, our digital
stream is multiplexed with lots of other digital streams over a high-
bandwidth high capacity trunk. Obviously the terminal equipment has
to be able to separate down individual calls, but that's what
computers do.
> Also, as far as I can tell, POTS stuff is pretty much copper ONLY
> from the last CO to the houses, and in many cases not all that way
> either as fiber is pushed closer and closer to the customer
> premises, or directly into them in some cases. �So if switched
> technology were to be phased out, AT&T would beef up their internet
> backbones, surplus dozens of backbone ESS switches and probably
> hundreds of local CO switches, and start to recover enough copper
> strung throughout the country to probably defray most of the
> transition costs.
I'm not sure the above is accurate. Reclaiming copper is labor
intensive, and copper can multiplex now. Going to fibre is irrelevent
to the switching technology.
I'm not sure there is a separate _physical_ "internet backbone" from
the switched network and the private line network. To make good use
of economies of scale, they simply electronically divide up various
high-volume trunks as needed for various services. That is, it's a
_logical_ division. I strongly doubt it's something like the public
network runs along old US 1 while the internet backbone runs along
I-95.
If i'm not getting this correctly perhaps someone could correct me in
layman's terms.
> The real losses:
>
> 1. Transmission quality - MAYBE ...although quite frankly my home
> line is � �working as well (if not better) over CATV coax as it did
> � �w/traditional T&R copper back to the CO.
Since you're using a "new" technology of CATV, it damned better be a
superior transmission and service quality than what you had before.
(Several cable users have complained to me about service qualtiy,
though. However, they like the lower price.)
> 2. Service during power outages and/or other disasters. �Even if the
> � �future continues to have battery plants backed up by diesal �
> �generators in central offices, it won't be of much use to those who
> � �are wired up like me. �BUT, even if they are backed up well in
> the � �central offices, how about the fiber feeds to local
> distribution � �units in the field? �How much battery backup do
> those � �end-of-the-fiber boxes have with them? �6 hours? �12?
> �. . .
"Competition" has been the big mantra driving public policy before and
after Divesture. But competition works in multiple ways and not
necessarily in the public interest. For instance, in the regulated
world, telephone and power utilities generally had very strong
networks and recovery labor forces since the rate base had no choice
but to pay for it and the PUCs encouraged it.
But with competition, consumers have fled from high cost to low cost,
not worrying about recovery during troubles. In the advent of power
competition, many electric companies have become too lean and needed
much longer time to repair storm damage, plus had less redundancy and
would fail more easily. (Retire power people told me they were very
glad to get out since the modern world is skimpy and ugly).
That has already happened to the landline telephone business. Support
is harder to get; repairs take longer. I'm afraid those expensive
C.O. diesel generators will be scrapped because the baby bells will
figure, "Hey, the other guys are stealing our customers by lower
price, so we have to become lean, too". I strongly doubt the cable
companies have diesel generators and huge batteries in their terminal
rooms.
[public replies, please]
Robert Bonomi
<hanc...@bbs.cpcn.com> wrote:
> On Jan 4, 12:31�am, "Paul Hoffman" <prhkgh.remove-t...@and-this-
> too.comcast.net> wrote:
>
>> Recently AT&T asked the FCC to set a date to transition completely
>> off traditional "switched technology" telephone networking, in favor
>> of packet switched (internet-style) networking. [AT&T claimed that]
>> maintaining two parallel networks that accomplish essentially the
>> same thing was wasteful and uneconomical, and this has caused quite
>> a bit of posts, especially between collectors and users of 'old
>> telephone technology' such as me.
>>
>> At the time I read about AT&T's request I did not realize the
>> petition was for the technology in and between central offices, and
>> not a bid to remove all copper twisted pair out to consumer's
>> premises. �However with the ever-increasing amount of fiber either
>> to premises or to centralized communication near the end users, my
>> [concerns] remain the same.
>
> Where does Verizon fit in all of this? Isn't today's at&t a
> relatively small company?
AT&T:
Current assets $265 *Billion*
2008 gross revenues $124 *Billion*.
(operating expenses $100 Billion)
2008 gross profit (revenues less expenses) $24 Billion
Doesn't sound like a 'small' company to me.
Others may have a different opinion.
> There are several issues in play here, some transparent to the home
> subscriber, some possibly of import.
>
> networks being a pain is correct. It's digital signals using a
> protocol. When we make a telephone call, we do not get dedicated use
> of a given digital line between two points.
False to fact.
One _does_ get dedicated use of a given digital 'circuit' between the two
points. This is what the 'connection set-up' when a call is placed establishes
It may well be time-division-multiplexed (e.g. a DS-0 in a DS-1) on a common
physical connection, but the full capacity of that circuit *IS* dedicated
for your use, regardless of how much, or how little, use you actually make
of it during the call. That connection is _yours_, for your exclusive use,
until the 'connection tear-down' at the end of the call.
This is why the telco backbone _is_ referenced as a 'circuit-switched'
architecture, not a 'packet-switched' one.
David Clayton
.......
> This is why the telco backbone _is_ referenced as a 'circuit-switched'
> architecture, not a 'packet-switched' one.
In Australia, the major telco has been using VoIP as a backbone for
business voice traffic for quite a while now. No one has noticed much
difference.
http://www.computerworld.com.au/article/9981/telstra_flicks_switch_1_5_billio
n_ip_network/
--
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.
hanc...@bbs.cpcn.com
>> Where does Verizon fit in all of this? ��Isn't today's at&t a
>> relatively small company?
>
> AT&T:
> �� �� �� Current assets $265 *Billion*
> �� �� �� 2008 gross revenues $124 *Billion*. ��
> �� �� �� ��(operating expenses $100 Billion)
> �� �� �� 2008 gross profit (revenues less expenses) $24 Billion
>
The question was _relative_ to other carriers, such as Verizon. And
isn't it properly spelled as "at&t"?
> False to fact.
Nope. It's true to life.
> One _does_ get dedicated use of a given digital 'circuit' between
> the two points. This is what the 'connection set-up' when a call is
> placed establishes It may well be time-division-multiplexed (e.g. ��a
> DS-0 in a DS-1) on a common physical connection, but the full
> capacity of that circuit *IS* dedicated for your use, regardless of
> how much, or how little, use you actually make of it during the
> call. That connection is _yours_, for your exclusive use, until the
> 'connection tear-down' at the end of the call.
>
> This is why the telco backbone _is_ referenced as a
> 'circuit-switched' architecture, not a 'packet-switched' one.
Actually, certain multiplexing techniques, such as on an underseas
cable, do NOT dedicate the full capacity of an assigned channel.
You snipped some key explanations in my earlier post.
The "full capacity of a 'circuit'" these days is enormous, far more
than one needs for a voice telephone conversation. That's why we have
'multiplexing' which is a way for multiple conversations to share an
individual circuit.
There are many different ways to multiplex multiple telephone
conversations or data streams on an individual circuit. The
particular technique chosen does not normally matter to the individual
caller. What matters is the assigned bandwidth--that determines the
quality of the connection.
When an audio sound is digitized for transmission, it could be a high
or low quality. (Will it sound like a telephone voice or CD voice?)
Certain telephone users need higher bandwidth and pay more to get it.
For instance, a radio show transmitted over phones needs a higher
quality sound so a better connection is provided. Television needs a
much higher bandwidth for the video signal and better sound, so an
even bigger connection is provided. (Not as much TV and radio use the
phone network these days; this is just to show an example to
illustrate the point.)
"Packet switching" is just another multiplex method; a way to stuff
multiple conversations or data streams through a single big pipe.
Once again, when an audio sound is digitized for transmission, it
could be a high or low quality. They can make packet switching very
high quality or very low quality as they deem fit.
Another poster describes good quality packet switching in Australia.
What will matter is how the packets are defined, just as the digital
sampling rate varies.
***** Moderator's Note *****
This is a complicated subject: let's try to shed more light than heat,
OK?
Bill Horne
Moderator
Neal McLain
> Recently AT&T asked the FCC to set a date to transition
> completely off traditional "switched technology" telephone
> networking, in favor of packet switched (internet-style)
> networking. [AT&T claimed that] maintaining two parallel
> networks that accomplish essentially the same thing was
> wasteful and uneconomical, and this has caused quite a bit of
> posts, especially between collectors and users of 'old
> telephone technology' such as me.
>
> At the time I read about AT&T's request I did not realize the
> petition was for the technology in and between central
> offices, and not a bid to remove all copper twisted pair out
> to consumer's premises. However with the ever-increasing
> amount of fiber either to premises or to centralized
> communication near the end users, my [concerns] remain the
> same.
hanc...@bbs.cpcn.com wrote:
> There are several issues in play here, some transparent to the
> home subscriber, some possibly of import.
> First, today's telephone network is almost all digital even
> though most telephone sets are analog. Shortly after your
> analog signal leaves your telephone set the telephone company
> converts it to digital. This may be done at a line
> concentrator on a pole, at a converter box within or near your
> home (such as with VOIP), or at the central office.
The A-D conversion can be done by digital loop carrier, but I've never
seen a DLC small enough to be mounted on a pole. Do the such things exist?
As for doing it in a "converter box within or near your home," the old
question remains: how do you power it? With VOIP, the customer
understands that he/she must provide the power, including battery backup
power if desired. But if the telco provides the conversion, who
provides the power?
> Second, we must remember the real difference between fibre and
> copper (and coax) is capacity. If they replace your copper
> line with coax or fibre there is no reason your home telephone
> set_has to change.
Except that somebody -- either the subscriber or the telco -- still
has to provide the operating power for the phone, possibly including
backup batteries. How do you propose to power it?
> As far as I know, other than Touch Tone and fancy features, a
> telephone set today is functionally exactly the same as a 1938
> telephone set (300 set with F handset).
True, including the requirement that the copper pair that connects it
to the network also provides operating power.
> A 1938 set works fine today and a modern set would work fine
> on the 1938 network (except for Touch Tone and some sets offer
> pulse as an switchable option).
They even require the same kind of power to function.
> To me, a big change _may_ come where the subscriber's
> individual telephone set will do the analog-digital conversion
> and so emit digital signals. It likely will use a new
> carrier-signal and ringing current instead of the 48V DC and
> 90V AC 20 Hz used now. But at that point almost all telephone
> sets will be obsolete. Subscribers may have to buy adapters
> just as rabit ear TV owners had to do.
What kind of adapter? A D-A converter that includes a power supply
and a ring generator?
> (I suppose some business sets are digital now.)
They are. They operate on local power and may or may not have backup
batteries. Numerous small business systems are digital, but if you
dial 9 to get an outside line, you get an analog subscriber line.
It's usually ground start, as opposed to loop start, but it's still an
analog loop. DC voltage and ringing voltage are superimposed on the
loop for signaling purposes, but the customer's phone system itself is
powered locally.
Hoffman continued:
> Also, as far as I can tell, POTS stuff is pretty much copper
> ONLY from the last CO to the houses...
Or the last DLC.
> and in many cases not all that way either as fiber is pushed
> closer and closer to the customer premises, or directly into
> them in some cases. So if switched technology were to be
> phased out, AT&T would beef up their internet backbones,
> surplus dozens of backbone ESS switches and probably hundreds
> of local CO switches, and start to recover enough copper
> strung throughout the country to probably defray most of the
> transition costs.
AT&T could recover pole-mounted cables relatively easily (so could
scavengers). Recovering buried cables, particularly in urban
neighborhoods, would be far more expensive. I suspect the cost of
recovering it would exceed its value, even at today's copper prices.
> I strongly doubt the cable companies have diesel generators
> and huge batteries in their terminal rooms.
Diesel generators, yes; huge batteries, no. Most of the headend
equipment that cable companies use is powered from 115 VAC, not -48 VDC,
so DC is not needed for normal operations. Batteries are needed only in
UPSs that keep essential equipment running until the generator fires up.
Neal McLain
John Levine
> things exist?
Sure. A decade ago the local telco had pole mounted boxes that broke
out a T1 into 24 POTS circuits. They were line powered, so the POTS
voltages were somewhat lower than normal. Their engineer told me that
he wanted to avoid equipment outside the CO that needed its own power.
I haven't checked to see what they do now, but since they offer DSL to
nearly everyone, it must be something else.
R's,
John
Neal McLain
> The A-D conversion can be done by digital loop carrier, but I've
> never seen a DLC small enough to be mounted on a pole. Do such
> things exist?
John Levine wrote:
> Sure. A decade ago the local telco had pole mounted boxes that broke
> out a T1 into 24 POTS circuits. They were line powered, so the POTS
> voltages were somewhat lower than normal. Their engineer told me that
> he wanted to avoid equipment outside the CO that needed its own power.
>
> I haven't checked to see what they do now, but since they offer DSL to
> nearly everyone, it must be something else.
Thanks for the info. I'm surprised that it was line powered, because
every DLC I've seen was pad-mounted and had an electric meter. But I
can certainly understand why they wanted to avoid equipment that
required its own power. Cable TV networks have numerous power
supplies, and maintaining them has always been a headache.
Neal McLain
hanc...@bbs.cpcn.com
> As for doing it in a "converter box within or near your home," the old
> question remains: how do you power it? �With VOIP, the customer
> understands that he/she must provide the power, including battery backup
> power if desired. �But if the telco provides the conversion, who
> provides the power?
Good question.
For decades consumers just assumed the phone company would power their
home telephones. Business telephone systems* usually needed
commercial power, although some basic call ability continued with CO
power if a direct line could be established.
But in the last decade cordless phones became very popular and
consumers have to plug them in and depend on commercial power for them
to work. Consumers were advised to keep at least one traditional
wired phone for power failures. Likewise, consumers realize that cell
phones have to be recharged with commercial power.
So, the idea of using commercial power to run a telephone isn't a
strange idea to consumers anymore; they're used to it.
Consumers still will need to be educated that a backup battery is
absolutely required for modern telephones. I believe Verizon's FIOS
installation includes such a battery backup.
There is of course a slight cost in electricity consumers must pay for
these phones; but I don't think it's something consumers think about.
But all the rechargeable devices and modern electronic always-on
devices do eat power that adds up.
> Except that somebody -- either the subscriber or the telco -- still
> has to provide the operating power for the phone, possibly including
> backup batteries. �How do you propose to power it?
I would expect new phones to come with power cords just as answering
machines and cordless phones do now. FIOS devices may have a
dedicated power source. Historically, business key systems plugged
into an 120V outlet that was mounted for that purpose in the utility
closet. (Older Princess and Trimline phones used an AC incandescent
dial lamp and those required a plug and transformer; newer models used
LEDs powered by the phone line. Interesting how we're going
backward.)
> �> A 1938 set works fine today and a modern set would work fine
> �> on the 1938 network (except for Touch Tone and some sets offer
> �> pulse as an switchable option).
>
> They even require the same kind of power to function.
True. If someone had a cordless phone and went back in time to 1938,
they would plug it into a nearby 120V wall outlet (that most homes
had) and would be in business. I don't think cordless phones use much
current.
> �> To me, a big change _may_ come where the subscriber's
> �> individual telephone set will do the analog-digital conversion
> �> and so emit digital signals. �It likely will use a new
> �> carrier-signal and ringing current instead of the 48V DC and
> �> 90V AC 20 Hz used now. �But at that point almost all telephone
> �> sets will be obsolete. �Subscribers may have to buy adapters
> �> just as rabit ear TV owners had to do.
>
> What kind of adapter? �A D-A converter that includes a power supply
> and a ring generator?
Probably.
> �> I strongly doubt the cable companies have diesel generators
> �> and huge batteries in their terminal rooms.
>
> Diesel generators, yes; huge batteries, no. �Most of the headend
> equipment that cable companies use is powered from 115 VAC, not -48 VDC,
> so DC is not needed for normal operations. �Batteries are needed only in
> UPSs that keep essential equipment running until the generator fires up.
Cable companies actually have diesel (or other fueled) generators?
Those units are big and heavy, plus there is the fuel storage, and
maintenance.
As an aside, these units need frequent testing. I've seen a lot of
mission critical places go dark when they were testing their
generators--the control circuitry failed to make a proper transition.
*Old PBXs sometimes had a hand crank to run the ringing signal in the
event of a power failure having used one, they're not that easy to
turn and get tiring pretty quickly. An alternative was just
connecting extensions to outside lines directly as done for night
connections.
Gary
news:4B45FB69...@annsgarden.com...
>
> Thanks for the info. I'm surprised that it was line powered, because
> every DLC I've seen was pad-mounted and had an electric meter. But I
> can certainly understand why they wanted to avoid equipment that
> required its own power. Cable TV networks have numerous power
> supplies, and maintaining them has always been a headache.
That's one of the reasons Verizon is pushing FiOS. It is a PON, which
means "passive optical network." Passive is the key word here; PON's
have no powered equipment between the endpoints. All power is at the
"CO" or premise. Couple that with the fact that fiber doesn't
rust(*), and you can see they are creating a very reliable, low
maintenance cost infrastructure.
Cable networks use a hybrid fiber coax system (HFC). HFC does require
powered equipment in the field. There are some powered amplifiers,
but the biggest powered pieces are the "nodes," which convert between
fiber and coax. Thus, HFC infrastructure is inherently costlier to
maintain than a PON.
However, HFC infrastructure has been built out over a much wider area
of the country than FiOS, so it has the advantage of a much larger
installed base to amortize costs across.
-Gary
(*) When I had FiOS installed last spring. the connection between my
drop and the neighborhood fiber was made in an underground box that
was flooded. After screwing the connectors together, the tech simply
dropped the connection into the water. I asked if he was concerned,
but he wasn't. The connection is water tight (at least to a few
feet), and with no metal in the cable, water is no problem.
Robert Bonomi
<hanc...@bbs.cpcn.com> wrote:
>On Jan 5, 11:01��pm, Robert Bonomi <bon...@mail.r-bonomi.com> wrote:
>>> Where does Verizon fit in all of this? ��Isn't today's at&t a
>>> relatively small company?
>>
>> AT&T:
>> Current assets $265 *Billion*
>> ��2008 gross revenues $124 *Billion*. ��
>> (operating expenses $100 Billion)
>> 2008 gross profit (revenues less expenses) $24 Billion
>>
>> Doesn't sound like a 'small' company to me.
>
> The question was _relative_ to other carriers, such as Verizon.
Well, AT&T revenues are roughly -30% smaller than Verizon, Inc. and assets
are $-63 billion smaller. and that's after Verizon acquired GTE and AlTel.
> And isn't it properly spelled as "at&t"?
>
>> False to fact.
>
>Nope. It's true to life.
>
>> One _does_ get dedicated use of a given digital 'circuit' between
>> the two points. This is what the 'connection set-up' when a call is
>> placed establishes It may well be time-division-multiplexed (e.g. ��a
>> DS-0 in a DS-1) on a common physical connection, but the full
>> capacity of that circuit *IS* dedicated for your use, regardless of
>> how much, or how little, use you actually make of it during the
>> call. That connection is _yours_, for your exclusive use, until the
>> 'connection tear-down' at the end of the call.
>>
>> This is why the telco backbone _is_ referenced as a
>> 'circuit-switched' architecture, not a 'packet-switched' one.
>
> Actually, certain multiplexing techniques, such as on an underseas
> cable, do NOT dedicate the full capacity of an assigned channel.
FALSE TO FACT.
Extra compression is applied, such that the bandwidth demand for a
'voice channel' is smaller, *BUT* that (reduced) channel bandwidth
_is_ reserved for the exclusive use of that voice channel.
This is an absolute necessity given the design requirement that
_every_ 'established' call be able to actually _use_ their 'voice
channel' at all times while the call is 'connected'.
> The "full capacity of a 'circuit'" these days is enormous, far more
> than one needs for a voice telephone conversation. That's why we
> have 'multiplexing' which is a way for multiple conversations to
> share an individual circuit.
To misquote Clinton, "that depends on what you mean by 'circuit'".
With the possible exception of the proverbial "last mile" every
physical circuit contains a large number of 'logical' circuits.
*EVERY* such 'logical circuit' has certain resources assigned for it's
_exclusive_ use.
Even on an OC-192, each (logical) DS-0 is guaranteed it's particular
stream of bits.
> There are many different ways to multiplex multiple telephone
> conversations or data streams on an individual circuit.
There are a surprisingly _small_ number of ways to do multiplexing.
Time-division
Frequency-division
and 'statistical'
about cover it.
Time-division and frequency division are both 'synchronous'
techniques, except for a small (fixed!) amount of additional latency,
the recovered signal is an exact duplicate of the original.
Statistical multiplexing (of which packet-switching is a special-case
subset) is an asynchronous methodology. The bits will come back out
in the same order, but that's the extent of the 'guarantee' On any
given 'channel' one data block may have had much more (or less)
latency than the previous block. There's no way to know, nor to
predict in advance, when this happens, nor 'how bad it will be' when
it does happen. The advantage of 'stat muxing' is that you _can_
'over-subscribe' the capacity of the composite circuit -- e.g. put 16
9600 baud terminals on a single 19,200 baud trunk. As long as they
are _lightly_ used, it works fine. BUT if they're all beating up on
the connection, they're each only going to see about 1200 baud
performance. And it will be 'bursty', a bunch of data will come in
fast, then there will be a long pause until the next burst of data.
> The particular technique chosen does not normally matter to the
> individual caller.
It -can- matter a great deal if one is trying to do something other
than 'talk' on the line. Modems, Fax machines, TDD devices,
voice-recognition systems (e.g., IVR) even Touch-Tone(r) control of
remote devices -- these can all be adversely affected by the use of an
'asynchronous' multiplexing technique.
> What matters is the assigned bandwidth--that determines the quality
> of the connection.
'Audio quality' of the connection is irrelevant to the _guaranteed_
availability_, END-TO-END, of the bandwidth (and other resources)
required to handle the particular connection.,
To have a _guaranteed_ end-to-end bandwidth, 'circuit switched'
technology is required. You have to build a 'virtual circuit' (a
locked-down routing decision at every decision point) from end to end,
with a committed circuit information rate on every link, and through
every switching device.
If one does -not- do that, one cannot guarantee that the resources
required by the connection (phone call) will be available at all times
during the call.
Every device, and every path segment, does have a fixed upper bound on
the amount of traffic it can handle. Any attempt (regardless of
whether it is packet-switched, or circuit-switched, technology) to
force 'n+1' things through a resource that can only handle 'n' is
doomed to failure.
The _critical_ difference between packet-switching, and
circuit-switching, in *that* regard is that a circuit-switched network
can tell you _at_ _call_ _initiation_ (aka 'call set-up') that the
required resources are not presently available (e.g. a 'fast
busy'). Also -- assuming the set-up phase completes successfully -- a
circuit-switched network _guarantees_ the availability of that level
of resources for your use until call completion (aka circuit tear-down
time).
A packet-switched network has no 'set up' and 'tear down' phases, and,
therefore cannot tell you whether or not any particular _level_ of
resources is available at the time you start sending. Nor can it
guarantee that the necessary level of resources will be available for
the entire duration of the 'call'.
> "Packet switching" is just another multiplex method; a way to stuff
> multiple conversations or data streams through a single big pipe.
FALSE TO FACT.
"Packet switching" means that the 'routing' for each/every data block
is determined _as_ _that_ _data_ _block_ _is_ _processed_. There is
no guarantee that successive blocks between the same source and
destination will take the same path at any point. And, as a result,
there is no guarantee that the blocks will arrive in the same order as
sent. There is also no guarantee that there will be 'room' for those
blocks on any given physical carrier segment when those blocks arrive.
In contrast, 'circuit switching' figures out all the routing decisions
at the time of call set-up; it _is_ guaranteed that all packets will
traverse the same path, and will arrive in the same order as sent (if
they arrive at all, that is). It is also guaranteed that there _is_
space available on each physical link between the source and
destination for every packet that is sent as part of the call.
These are the _critical_ differences between 'circuit switched' and
'packet switched' technology -
'Circuit swished' technology:
* 'Sets up' a connection (on demand) and allocates resources to it, _before_
it can be used.
* Releases those allocated resources _only_ on call tear-down.
* _All_ the routing decisions are made at call set-up time. *guaranteeing*
a (fixed) path is available, and that all blocks of data between the two
endpoints will take that _specific_ path.
* Required bandwidth is _reserved_ for that 'virtual circuit', =until. the
virtual circuit is 'torn down', at call completion.
+ These last two points, incidentally, guarantee a constant transit time,
a fixed amount of latency, fixed 'jitter', as well as _stable_ values
for a number of other 'important' communications parameters.
* "Lost' data blocks are immediately detectable -- arrival of any block
without the arrival of the preceeding bloc is proof that the preceeding
block was lost.
'Packet-switched' technology:
* Has no concept of a 'connection' within the network. Endpoints may
agree among themselves, that certain blocks of data are to be
treated as part of a common stream, but it is _only_ the end-points
that are aware of that relationship between those data blocks.
* Never 'reserves' any resources for the specific use of any particular
connection. Thus, cannot 'guarantee' availability for anyone, at any
given time.
* Therefore, never needs to 'release' any resources.
* _All_ routing decisions are made _at_the_time_ =each= data block arrives
at each routing point. There is no guarantee that subsequent data blocks
will be routed the same way at _any_ routing decision point.
* There is *NO* guarantee that any path between the endpoints _is_ available
at the time any data block is sent.
* There is no guarantee that the path that was available for the prior
data block is _still_ available for the current data block.
* There is no way for any intermediate point to *autonomously* notify
_either_ end-point that a required link in the path is 'full', has
errors, or has completely stopped working.
* There is no guarantee that the data blocks will arrive at the destination
in the same sequence that they left the origin.
* Detecting data blocs that have actually gotten "lost" in route, as
opposed to merely 'delayed' is virtually impossible.`
Some_ of the characteristics of a circuit-switched technology CAN be
built, and relatively easily, on top of a packet-switched one. Some
of the more critical ones -- end-to-end _guaranteed_ bandwidth for the
lifetime of the 'connection', timely detection of 'lost' data, and
autonomous notification to the endpoints -- are *very* difficult to
add.
> Once again, when an audio sound is digitized for transmission, it
> could be a high or low quality. They can make packet switching very
> high quality or very low quality as they deem fit.
Again, quality of the 'audio signal' is irrelevant, and IMMATERIAL, to
the quality of the communications _network_. Network 'quality' is
based on primary considerations of the reliability, and consistency of
behavior of the infrastructure, not the quality of '[what] passes
through it'.
One can have a 'high quality' network, carrying a very low quality
audio signal, and one can try to push 'high quality audio' across a
low quality network.
wes...@aol.com
hanc...@bbs.cpcn.com writes
> Likewise, consumers realize that cell phones have to be recharged
> with commercial power.
TV stations are warning with the sub-Zero cold expected to take your
cell phone car charger with you if you go out.
> As an aside, these units need frequent testing. I've seen a lot of
> mission critical places go dark when they were testing their
> generators--the control circuitry failed to make a proper
> transition.
Telcos used to run them, usually every Wednesday at 8 a.m. amd
actually transfer the power and run on auziliary power for an hour so
to make suwre the generator is working and so is the transer. I
wonder if they still do.
Wes Leatherock
wes...@aol.com
wlea...@yahoo.com
jmei...@aracnet.com
> Where does Verizon fit in all of this? Isn't today's at&t a
> relatively small company?
Verizon is trying to sell off their land-line operations. Up here in the
Northwest they're planning on getting Frontier to take it all over.
T
I was charged with designing the power redundancy for the Central
Voter Registration System here in RI. I specified an APC Symmetra with
a minimum of 15 minute backup time (Reality gave us 45 minutes),
backed up by a 125kW natural gas fired generator that spooled up
within 10 seconds. It had normal exercise routines that it would do,
and once every quarter we'd do a full scale test. That involved going
to the transfer panel, holding down a button for a set interval and
you'd see everything transition to the generator.
Rob Warnock
+---------------
| <hanc...@bbs.cpcn.com> wrote:
| > Actually, certain multiplexing techniques, such as on an underseas
| > cable, do NOT dedicate the full capacity of an assigned channel.
|
| FALSE TO FACT.
|
| Extra compression is applied, such that the bandwidth demand for a
| 'voice channel' is smaller, *BUT* that (reduced) channel bandwidth
| _is_ reserved for the exclusive use of that voice channel.
|
| This is an absolute necessity given the design requirement that
| _every_ 'established' call be able to actually _use_ their 'voice
| channel' at all times while the call is 'connected'.
(*sigh*) How quickly they forget... :-(
http://en.wikipedia.org/wiki/Time-assignment_speech_interpolation
...time-assignment speech interpolation (TASI) is an analog technique
used on certain long transmission links to increase voice-transmission
capacity.
TASI works by switching additional users onto any channel temporarily
idled because an original user has stopped speaking. When the original
user resumes speaking, that user will, in turn, be switched to any
channel that happens to be idle. The speech detector function is called
voice activity detection.
...
TASI was invented by Bell Labs in the 1960s to increase the capacity
of Transatlantic telephone cables. It was one of their first applications
requiring electronic switching of voice circuits.
Later Digital Circuit Multiplication Equipment included TASI as a
feature, not as distinct hardware.
...
Other sources note that the TASI plan was actually first published
in 1959 [Bullington K, Fraser JM, "Engineering Aspects of TASI",
BSTJ, 38:353-364 (1950)], not "in the 1960s", but that's a minor detail.
In any case, stealing bandwidth from putatively "circuit-switched"
voice calls is now more than a half-century old, and is still used in
applications where the available bit-rate is constrained (e.g. VSATs).
-Rob
-----
Rob Warnock <rp...@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607
Garrett Wollman
Robert Bonomi <bon...@host122.r-bonomi.com> wrote:
> 'Packet-switched' technology:
>
> * Has no concept of a 'connection' within the network. Endpoints
> may agree among themselves, that certain blocks of data are to be
> treated as part of a common stream, but it is _only_ the
> end-points that are aware of that relationship between those data
> blocks.
Nonsense. The original packet-switched networks were all
connection-oriented.
> * Never 'reserves' any resources for the specific use of any
> particular connection. Thus, cannot 'guarantee' availability for
> anyone, at any given time.
Again, nonsense. Numerous mechanisms exist, and have been deployed,
for resource reservation on packet networks.
> * _All_ routing decisions are made _at_the_time_ =each= data block
> arrives at each routing point. There is no guarantee that
> subsequent data blocks will be routed the same way at _any_
> routing decision point.
False.
> * There is *NO* guarantee that any path between the endpoints _is_
> available at the time any data block is sent.
You seem to be confusing the service you can buy from your
consumer-grade ISP for $39.95 a month with the definition of what a
packet-switched network is. It isn't so.
> * There is no way for any intermediate point to *autonomously*
> notify _either_ end-point that a required link in the path is
> 'full', has errors, or has completely stopped working.
Complete malarkey. Even your consumer-grade ISP can and does do that.
> * There is no guarantee that the data blocks will arrive at the
> destination in the same sequence that they left the origin.
Some network technologies provide such guarantees; others don't. Take
off your telco blinders, please.
-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
Steven
crashed, stay tuned for another crash. At least the Northwest is in
pretty good shape, but will Verizon leave enough money in the company
to survive?
--
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.
David
news:hi6hk9$12ic$1...@grapevine.csail.mit.edu...
> In article
> <xuudncghgclS2dvW...@posted.nuvoxcommunications>,
> Robert Bonomi <bon...@host122.r-bonomi.com> wrote:
>
>> 'Packet-switched' technology:
>>
>> * Has no concept of a 'connection' within the network. Endpoints
>> may agree among themselves, that certain blocks of data are to be
>> treated as part of a common stream, but it is _only_ the end-points
>> that are aware of that relationship between those data blocks.
>
> Nonsense. The original packet-switched networks were all
> connection-oriented.
I suggest Robert Bonomi also understand the difference between a
datagram and and a virtual circuit as applied to packet switched
networks.
David
har...@hallikainen.com
in San Luis Obispo, CA in the 1970s. This was the mainland end of an
undersea coaxial cable to Hawaii. The cable had vacuum tube amplifiers
at the bottom of the ocean. TASI does seem to be a sort of analog
packet switching and may suffer from the "pipe is full" problems of
other packet switched or statistical multiplexed systems. I wonder how
much it increased the capacity of the cable and how much capacity they
had to hold in reserve to keep from cutting people off. I imagine a
max increase is 100%.
Harold
Eric Tappert
Harold,
TASI simply switches "circuits" when the call is idle in one direction
or the other. There is a set of circuits that are available in each
direction and the TASI terminal assigns each call to a circuit
depending on whether voice is detected. The absence of voice makes
the receiving terminal insert noise into the circuit to the customer
so there is no need to transmit any signal. In this fashion, more
than twice the number of circuits can be put on a multiplexed circuit.
In the old analog days, a group was not 12 circuits on undersea cable,
but 16 circuits (3 kHz bandwidth instead of 4 kHz) a few groups would
be assembled into a set for TASI. The circuit gain was well above
100%. Digital TASI uses T1 carrier and typically has a circuit gain
well above 100%. In the digital version, if a circuit needs
connectivity (i.e. speech is present) and there is no available time
slot, the link regresses to 7 bit encoding instead of 8 bit encoding
to allow the conversation to continue.
The controls for a TASI system are expensive, thus their use was
restricted to expensive circuits, e.g. undersea cable. Today there is
a real surplus of undersea fiber capacity, so these kinds of systems
are not as attractive as in previous times.
VOIP often uses the same kind of active circuit switching (albeit in
the packet domain...) and doesn't send packets when there is no voice.
BTW, voice detectors have an "attack time", so often the first part of
a syllable is clipped, reducing the quality of the connection. Modems
on TASI circuits resulted in a dedicated circuit, since modems had no
quiet time.
ET
--- news://freenews.netfront.net/ - complaints: ne...@netfront.net ---
hanc...@bbs.cpcn.com
> I suggest Robert Bonomi also understand the difference between a
> datagram and and a virtual circuit as applied to packet switched
> networks.
So, would it be accurate to say that a packet switched network could
have high quality voice transmissions (as good or better than existing
landlines) if they were properly engineered to do so?
Garrett Wollman
<hanc...@bbs.cpcn.com> wrote:
> So, would it be accurate to say that a packet switched network could
> have high quality voice transmissions (as good or better than
> existing landlines) if they were properly engineered to do so?
Absolutely -- most telco voice transmissions today operate over a
packet-switched network, albeit with very small packets. With the
current state-of-the-art in high-speed network technologies (all
packet-based), the serialization delays that motivated the design of
so-called "cell-switched" networks like ATM -- the backbone of most
traditional telco networks -- are much smaller. An appropriate
queueing discipline combined with some sort of resource reservation in
the switches can provide sufficient quality-of-service for
high-definition video circuits, never mind the comparatively miniscule
bandwidth required by voice telephony.
Consider, for the moment, the case of someone using VoIP to make a
local call over naked DSL. Their VoIP terminal (i.e., phone) will
send RTP/UDP/IP packets over the home Ethernet LAN to the DSL modem.
The DSL modem then encapsulates these packets in ATM cells and sends
them to a DSLAM in a telephone central office. Those cells are then
forwarded by the DSLAM over permanent virtual circuits on a carrier
ATM network (possibly even the ILEC's network) to the ISP's POP --
probably located in or next to another CO in a nearby major city --
where a router (probably made by Cisco) accepts the ATM cells and
reassembles them into IP packets, delivering them to the VoIP gateway
(possibly in the same Cisco box) which extracts the digitized
audio,[1] possibly transcodes it, and probably converts it back into
ATM cells again (this time labeled "voice" rather than "data") for
carriage on the ILEC network to the terminating CO switch. The only
reason for the ISP to use ATM between the DSLAM and their POP is the
fact that ATM is readily available in nearly every ILEC CO in their
service area, whereas dark fiber isn't.
-GAWollman
[1] The RTP protocol allows it to accurately reconstruct a continuous
audio stream from the UDP/IP packets it receives.
David Clayton
...........
> VOIP often uses the same kind of active circuit switching (albeit in the
> packet domain...) and doesn't send packets when there is no voice.
Yes and no, there is usually a configurable option on VoIP end devices to
keep sending packets on silence or not to save on bandwidth (Voice
Activity Detection):
http://searchunifiedcommunications.techtarget.com/sDefinition/0,,sid186_gci342466,00.html
har...@hallikainen.com
> but 16 circuits (3 kHz bandwidth instead of 4 kHz) a few groups would
> be assembled into a set for TASI. �The circuit gain was well above
How is the circuit gain well over 100%? If it's a two way conversation
with each person talking 50% of the time, it seems that you'd get a
gain of 100% by dropping in other circuits during the idle times in
each direction. There are, of course, other pauses, and people don't
start talking right when the other person talks (though my wife says I
start before she's finished, but that's another story). So, the
circuit gain COULD go a bit above 100%. I was assuming they would
leave some reserve circuits to handle the times when more than 100% of
the people were talking in one direction, but your mention of dropping
the encoding level in digital TASI makes sense. They could, I suppose,
drop bandwidth per channel on analog TASI, but that seems complex. So,
for analog TASI, did they leave some excess circuits to handle the
peaks?
This sort of multiplexing is interesting. The data stream is bursty
(even in analog, since there are pauses with no speech). Without
multiplexing, you're stuck with allocating a full circuit to the data.
With multiplexing, though, the bursts of one stream can fill in the
valleys of another stream, decreasing the required bandwidth. With
real time streaming, though, there are times where the peaks coincide
and you run out of bandwidth. Moving from real time to near real time
by adding buffers, you can shift the peaks such that they do not
coincide, then restore the timing at the receive end. I think this
works very well for stuff like satellite television where a bunch of
channels can be put in one datastream. A scene change in the video
results in a spike in the data rate since the entire screen has to be
redrawn, instead of just the changes. But, this spike can be time
aligned with a static scene where no data is to be sent. With a large
enough buffer, you can get real high usage of the available bandwidth.
Any excess can, of course, be used for non - realtime data, such as
program guide info, subscriber authorizations, etc. With voice,
however, latency is really a problem since the speakers become
confused. I often get confused on a digital cellphone and start
talking over the person I'm talking to (especially my wife, but,
again, that's another story).
Moving away from TASI a bit, but still staying with statistical
multiplexing, I've been giving more thought to network neutrality and
traffic shaping. If indeed ISPs are suffering from network congestion,
prioritizing of traffic may make sense. Email probably does not have
to be delivered as quickly as interactive video. I believe IP has bits
allocated to packet priority. ISPs could charge different rates based
on the packet priority. They could also go to time of use metering to
try to even out the load. Economics could provide a method of
allocating limited resources.
Anyway, thanks for the TASI discussion. It's been a long time since I
saw that system at the AT&T underground building in San Luis Obispo.
They had a pamphlet entitled "San Luis Obispo, Communications Center
of the World" showing the various Pacific undersea cables.
Harold
Rob Warnock
+---------------
| BTW, voice detectors have an "attack time", so often the first part
| of a syllable is clipped, reducing the quality of the connection.
Though note that digital voice detectors are capable of storing a
FIFO's worth of samples from *before* the "on" decision is made and
can include those in the outgoing stream ... at the cost of delaying
the *entire* "on" period by that much [that is, by the attack time of
the detector]. While increasing end-to-end delay is never desirable,
increasing it by just enough to restore the clipped portion is
probably worth it.
John Levine
>| BTW, voice detectors have an "attack time", so often the first part
>| of a syllable is clipped, reducing the quality of the connection.
>
>FIFO's worth of samples from *before* the "on" decision is made and
>can include those in the outgoing stream ... at the cost of delaying
>the *entire* "on" period by that much [that is, by the attack time of
>the detector].
Why not just speed up the first second or two of the "on" period until
it gets back to real time? That should be well within the capability
of modern DSPs. Don't raise the pitch, just throw away every Nth
sample, analogous to the way they used to use rotating tape heads.
R's,
John
Eric Tappert
Harold,
"Better than 100% gain" is all in the statistics.... Actually very
few people talk 50% of the time and on average it is less than 50%.
If I recall correctly it is about 40%. Of course there are cases when
both parties speak at the same time (little useful information,
perhaps, but they get bandwidth anyway...) and there are other cases
with virtually no active speech (two lovestruck teenagers, for
example. Just an occasional sigh...). Modems demand 100% all the
time, so they screwed things up by requiring a dedicated channel.
If I recall correctly, digital TASI used 5 T-1 lines running ADPCM for
a total of 240 channels per set and they connected to 480 trunks.
That allowed for enough margin that the bit robbing was seldem used,
but most of time, even with 480 active trunks, there were plenty of
spare channels. Not pushing things above 100% was really a quality of
service issue, as the gain was already significant and the incremental
improvement was small relative to the quality of the circuits.
Analog TASI was similarly configured, but I think a set was only 5, 16
channel groups (I could be wrong, it was a long time ago and my
experience was with the digital stuff...). In any event, the larger
the set the better the statistics work. I really don't know how many
of these circuits are still in use as undersea fiber is really cheap
these days...
Before anyone cares to argue that a group was only 12 channels, that
was not true for undersea analog cables. To increase capacity they
used 3 kHz channels and SSB modulation to fill a standard group
(60-108 MHz) with 16 circuits instead of 12, 4 kHz channels.
Hope this helps.
Geoffrey Welsh
> Telcos used to run them, usually every Wednesday at 8 a.m. amd
> actually transfer the power and run on auziliary power for an hour so
> to make suwre the generator is working and so is the transer.
One of my favorite stories, though quite possibly fictional, is about an
organization that tested their backup power systems regularly but failed to
notice that the fuel pump was actually connected to mains power so, while
every test was successful, during a real power outage the generator
spluttered and died...
danny burstein
Which reminds me of my favorite story, which, similarly, may or may
not be true. I've never been able to verify it one way or another, and
the folk involved have pretty much all met up with Father Time.
Back in 1965 we had the Big Northeast Blackout (thank you, Canada)
[a], which included NYC.
While emergency power was pretty rare back then, hospitals generally
had something on line.
Sure enough, Bellevue Hospital in Manhattan, NYC, had some pretty
large (by 1965 standards) backup generators, and they soon came up and
gave the hospital functional power.
Except that... well...
.. as you can well imagine, the generators back then were physically
very big, and heavy, and were sitting on the concrete pads in the
basement. Make that sub, sub, basement.
Bellevue (and Manhattan in general) isn't very high above sea level,
and the generators were pretty far down.
Turned out that the sump pumps weren't hooked up to the backup
power...
(About fifteen years later I did, in fact, get to see the generators
in questions. And sure enough they were pretty deep, and there was a
small pool of water around them from the constant drainage.).
The internet publication "Risks Digest" is filled with examples of
this sort. It's well worth reviewing for anyone considering how to
design things to keep working....
[a] the blackout started at the Sir Adam Beck substation
in Canada, near Niagara Falls, and cascaded from there.
--
_____________________________________________________
Knowledge may be power, but communications is the key
dan...@panix.com
[to foil spammers, my address has been double rot-13 encoded]
***** Moderator's Note *****
At the time, I lived in New Hampshire, in an old house on a
back-country road that suffered frequent power failures: when the
lights started to flicker, my brother and I both ran for the cellar to
get the lanterns and my father lit the kerosene heater that was
built into the kitchen stove, but after a few minutes, the lights were
back to normal and the juice was still on. It seems New Hampshire had
its own generators, and could keep going even when the grid went down.
My father was so impressed with the fact that our power stayed on that
he called the President of New Hampshire power to congratulate him.
I remember David Brinkley on TV, commenting on how the studio was
being lit by a lantern that looked like it should be hanging from a
tree next to a stream while someone scaled fish. They were using a
"miniature" battery-powered camera which had been developed for use on
the floor of political conventions: it was about the size of a
suitcase and had a shoulder-mount and brace to support it.
Some years later, in the Eighties IIRC, there was another outage that
affected mostly New York, and I saw the "Today Show" announcers
sitting crowded together in front of a single camera in the "Emergency
Operations Center": it looked cheap and amateurish. Brinkley, however,
had looked like a pro doing his best in bad circumstances: it's funny
how a Coleman lantern can (literally) set the stage for authenticity.
Bill Horne
Moderator
David Clayton
> One of my favorite stories, though quite possibly fictional, is about an
> organization that tested their backup power systems regularly but failed
> to notice that the fuel pump was actually connected to mains power so,
> while every test was successful, during a real power outage the generator
> spluttered and died...
Or this place where there was one large UPS powering a number of
important servers, but only one server was directly connected to the UPS
for control purposes and all the other servers got their shutdown
instructions via LAN software from this control server.
Various tests of unplugging the UPS power brought all the servers down as
anticipated, and all involved patted each other on the back because of the
completion of (yet) another successful project......... until....
The main power did one day go off and only the directly connected server
shut down properly - because the LAN equipment that all the other servers
relied on wasn't powered by a UPS!
This story has been modified to protect the incomp... I mean innocent.
T
dcs...@myrealbox.com says...
>
> On Thu, 14 Jan 2010 19:31:48 -0500, Geoffrey Welsh wrote:
>
> > One of my favorite stories, though quite possibly fictional, is about an
> > organization that tested their backup power systems regularly but failed
> > to notice that the fuel pump was actually connected to mains power so,
> > while every test was successful, during a real power outage the generator
> > spluttered and died...
>
> Or this place where there was one large UPS powering a number of
> important servers, but only one server was directly connected to the UPS
> for control purposes and all the other servers got their shutdown
> instructions via LAN software from this control server.
>
> Various tests of unplugging the UPS power brought all the servers down as
> anticipated, and all involved patted each other on the back because of the
> completion of (yet) another successful project......... until....
>
> The main power did one day go off and only the directly connected server
> shut down properly - because the LAN equipment that all the other servers
> relied on wasn't powered by a UPS!
>
> This story has been modified to protect the incomp... I mean innocent.
LOL! When we did ours we attached the power whips coming out of the
Symmetra to the server racks and shelves, even extended a whip out to
the telecome gear.
David Lesher
> One of my favorite stories, though quite possibly fictional, is
> about an organization that tested their backup power systems
> regularly but failed to notice that the fuel pump was actually
> connected to mains power so, while every test was successful,
> during >a real power outage the generator spluttered and died...
An indoor generator will have a "day tank" -- one with enough fuel for
a few hours or more. The rest of the fuel will be outside with a
transfer pump.
At an installation I inspected, that pump was not an emergency load,
and so it would have run for no longer than that.
Neither were certain other vital systems; something never uncovered in
3+ years of monthly drills.
After staging a full "pull the utility feed and see what happens.."
test; these items were soon resolved.
--
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