ISDN, DSL, load coils and my screwy Telco...
Gary
I've been using ISDN for a few years now, and my Telco has been touting the
arrival of DSL (SNET now part of SBC). Well great... I call up, give my POTS
line, and am told I'm too far for DSL. On inquiry, it turns out there are
several load coils on my POTS line between my home and the CO. The "cable"
length from the CO is about 13,000 feet but there are a few load coils which
apparently are factored in as several thousand feet a piece. SNET doesn't go
beyond 17,500 feet for their current DSL implementation.
Hmm... OK I think, when I had ISDN installed years ago, the telco left me a
map showing an engineers diagram of load coils which had to be removed for
my line to operate, and remove them from my line they did (I saw this
happening at the time). Upon thinking about this, I returned a call and
informed them that my ISDN line didn't suffer any load coils.
The response was if I wanted to drop ISDN and get POTS on my ISDN line with
DSL it could be worked out.
Problem is I have a contract which requires a penalty on termination with
less than 6 months notice.
The telco folks at the DSL (supposedly the "advanced networking product"
group) seem oblivious to this situation.
Question, if the telco requires my ISDN line be converted to POTS and offers
a free installation (for a 1 year commitment) should I assume this is a
defense against any charge to me for termination of my ISDN contract?
Secondly, if I tell them go ahead, am I terminating it or are they? I mean,
I'm telling them yes I want DSL, but they're telling me I can't have DSL and
ISDN.
Third, would this be better to resolve later, as I'm somewhat concerned if I
make it any tougher for them they'll refuse to install DSL at all.
This is an example of DSL killing ISDN in a way I never fully expected to
report:-o
Regards
Steve Elisberg
Something is screwy, for sure. If you are served by a loop that is less
than 18,000', there should be no load coils. The most common loading
scheme used by the "Bell System", was H88. That is, an 88 millihenry
choke is placed at 6000' intervals. The first is placed at 3000' from
the central office. Thus, the "Load Points" would be at 3kf, 9kf, 15kf,
21kf, 27kf and so on. The practice also calls for strict adherence to
the spacing. If for any reason the load point has to be located at a
closer interval, then a build-out network is placed to adjust the
impedance to make the section appear to be 6000'. Build-outs need to be
removed as well as load coils when a tech conditions a line for digital
services. For voice grade service, the customer can be no closer than
3000' and no more than 12000' from the last load, bridge taps included.
The resistance of the loop is controlled by the engineers by using
standard gauges of 19, 22, 24 and 26. The makeup is determined by the
distance to the serving terminal. If you are close enough to the central
office, it is likely you have some or all 26ga. From experience, 26ga
places a disproportionate penalty on digital services. I have found that
the difference between 22 and 24 is not as bad as the change from 24 to
26. That is one reason that your loop may not qualify for DSL even
though the distance to the CO is relatively short.
The fact is that the telephone companies all have a mostly voice grade
network set up that they're trying to convert to digital. Engineering
practices of placing telephone cable were governed by this fact. The
phone companies thought that twisted pairs were dead meat. Along comes
rapid innovations in DSL. Telco execs get caught with their pants down.
If we assume that your loop is 13,000, then what are loads doing in your
POTS service? It would help to see the cable maps. If you have a copy of
the ISDN work-order, it may have the cable makeup listed showing each
cable section length and gauge. Sometimes, a cable that served a
location that required loading gets transfered to a location closer to
the CO, but the loading is not removed due to an oversight or just plain
screw-up. The latter is usually the case. Another stupid engineering
move is to leave two loads in the loop. That is, 3kf and 9kf. This would
allow you to serve terminals from 11kf to about 21kf. A cheap way to
serve 18 to 21kf loops saving the expense of one load point. If you are
less than 18kf, you are screwed unless the Telco writes up a job to get
the loads out. I'm just throwing out some stuff here. Some of this may
apply to your situation and help you and others with a similar
situation. Without some more specifics, we're all just guessing.
jkatkin...@bellsouth.net
>
> Something is screwy, for sure. If you are served by a loop that is less
> than 18,000', there should be no load coils. The most common loading
> scheme used by the "Bell System", was H88. That is, an 88 millihenry
> choke is placed at 6000' intervals. The first is placed at 3000' from
> the central office. Thus, the "Load Points" would be at 3kf, 9kf, 15kf,
> 21kf, 27kf and so on. The practice also calls for strict adherence to
> the spacing. If for any reason the load point has to be located at a
> closer interval, then a build-out network is placed to adjust the
> impedance to make the section appear to be 6000'. Build-outs need to be
> removed as well as load coils when a tech conditions a line for digital
> services. For voice grade service, the customer can be no closer than
> 3000' and no more than 12000' from the last load, bridge taps included.
> The resistance of the loop is controlled by the engineers by using
> standard gauges of 19, 22, 24 and 26. The makeup is determined by the
> distance to the serving terminal. If you are close enough to the central
> office, it is likely you have some or all 26ga. From experience, 26ga
> places a disproportionate penalty on digital services. I have found that
> the difference between 22 and 24 is not as bad as the change from 24 to
> 26. That is one reason that your loop may not qualify for DSL even
> though the distance to the CO is relatively short.
>
A lot of your information is correct; however, he said his LEC is
SNET. SNET is not/was not a Bell RBOC. Therefore their load coil
rules may be different. (I don't know.) There are other schemes than
H88 for placing load coils, such as "D44" - 44 mH at shorter loop
lengths. One size was not used for all. It was just the most common
usage, since AT&T had most of the customers in the US prior to 1984.
John
Remove "nospam" to reply.
Steve Elisberg
I have never run into D44 in nearly 30 years with "Bell", but I have had
a run-in or two with D66. Once a water logged 300 pair D66 load pot was
replace by an H88 based on presumption. An example of how things can go
wrong. Another point is that a lot of the cable out there doesn't match
what's on the map. I'm just glad that the trend is to get the loads out.
jkatkin...@bellsouth.net
>
> I have never run into D44 in nearly 30 years with "Bell", but I have had
> a run-in or two with D66. Once a water logged 300 pair D66 load pot was
> replace by an H88 based on presumption. An example of how things can go
> wrong. Another point is that a lot of the cable out there doesn't match
> what's on the map. I'm just glad that the trend is to get the loads out.
>
> jkatkin...@bellsouth.net wrote:
> >
Pardon me if I misquoted the loading shorthand - there are several.
It is my understanding that the REA, which makes low interest loans to
rural telcos, as well as to rural electic co-ops, specifies a load
coil standard different from H88. My meaning was there is nothing
sacred about H88, and SNET may not use H88 loading.
John
- 27 years with the big mama, and still learning.
Remove "nospam" to reply.
David Lesher
>A lot of your information is correct; however, he said his LEC is
>SNET. SNET is not/was not a Bell RBOC.
Err... SNET was not majority owned by Ma, but that and Cincy Bell
AFAIK, followed all Bell practices.... Did you think they
did things the GTE way???
--
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
bill
services, DDS, ISDN, xDSL, etc, are the sometime hidden load coils
that everyone forgot about. Not to mention bridge taps between loads.
Case in point. During a 56K DDS circuit install in NYC a few months
ago the existance of a load coil point was detrmined during a Freq
run. The cable was no more than 4 kilofeet long between the CO MDF and
the local terminal. Apparently the sub cable was a re-use from a
retired trunk cable. What we found was apparently the first loading
point out from the CO. In the part on NYC where this circuit went in,
lowere Manhattan, sub cables were never loaded because of the short
lengths of feeder cable.
Interesting note. The DDS installer mentioned that a guy in his gang
was trying to get in a T1 fir the past coule of days on pairs assigned
to the same cable and count. Poor guy never knew what really hit him.
A load coil......