[comp.dcom.telecom] Switching Office Open House

rees@pisa.ifs.umich.edu (Jim Rees) (12/05/90)

I just went to an open house at the Ann Arbor office of Michigan Bell.
If you ever get a chance to do this, I highly recommend it!  This
office has a 1ESS serving five exchanges, a 1A serving four, and a
5ESS serving eight.  The 1A is four times the size of the 5 and serves
half as many lines.  The 1 is already considered obsolete (it's 17
years old, same as my Buick) and is slated to be replaced by the 5 in
six months.

I won't go over the mundane details.  The things that surprised me
were the small size of the battery backup, the large size of the cable
vault (one city block long, filled with 300 pair cables), the extreme
small size of the dry air supply for the pressurised cables, and how
much space is devoted to distribution bays.  I expected the switches
to be small compared to all the cables and wires, and they certainly
were!  The 5ESS has a capacity of 160,000 subscriber lines (not all on
local loops, most come in on SLC-96) and would almost fit in my living
room.  Well, maybe my basement.

I was disappointed that the people we got to talk to didn't know much
about the equipment, they were mostly business office types.  But I
had a nice long talk with an outside plant guy who was full of stories
about cable breaks and how to splice a 1200 pair cable.

I knew that offices were being consolidated but was a little surprised
to learn that this one office covers almost the entire county except
for an island of GTE.  They seem to be tearing out individual 5xbar
switches and replacing them with muxes to the 5ESS.  There is still
one SxS in S.E.  Michigan, in the New Boston office.  In spite of the
concentration of switching equipment in Ann Arbor, there is no one on
duty here at night.

As a computer guy, one thing intrigued me.  Can anyone tell me about
the "1ESS memory card" that is just a piece of aluminum the size of a
sheet of notebook paper?  It doesn't seem to have any electrical
contacts, but you can see little squares on it that might be
individual bits of magnetic memory.

After nearly ten years of reading TELECOM Digest I finally got to see
what some of this stuff looks like.

rcj1@ihlpf.att.com (Raymond C Jender) (12/08/90)

In article <15250@accuvax.nwu.edu>, rees@pisa.ifs.umich.edu (Jim Rees)
writes:

> As a computer guy, one thing intrigued me.  Can anyone tell me about
> the "1ESS memory card" that is just a piece of aluminum the size of a
> sheet of notebook paper?  It doesn't seem to have any electrical
> contacts, but you can see little squares on it that might be
> individual bits of magnetic memory.

> After nearly ten years of reading TELECOM Digest I finally got to see
> what some of this stuff looks like.


The 1E Memory Card is 6 5/8 x 11 1/4 inches.  Each card stores 64
words of 44 Bits. A 45th bit in each word is not used for data
storage.  When a memory card is placed in the module, there are 64
solenoid loops associated with it, one for each row of 45 bar magnets.
A pulse in a solenoid loop interrogates simultaneously the
corresponding row of 45 magnets on the card.

Clear thing up a little?

kabra437@pallas.athenanet.com (Ken Abrams) (12/09/90)

In article <15250@accuvax.nwu.edu> rees@citi.umich.edu (Jim Rees)
writes:

>As a computer guy, one thing intrigued me.  Can anyone tell me about
>the "1ESS memory card" that is just a piece of aluminum the size of a
>sheet of notebook paper?  It doesn't seem to have any electrical
>contacts, but you can see little squares on it that might be
>individual bits of magnetic memory.

Those are indeed very small magnets on the aluminum cards.  This is
something like EPROM on a huge scale.  It is really quite a feat of
precision mechanical engineering.  To change the memory, the cards are
put on a tray and "written" with a 44 head card writer mechanism.  The
whole process is a mechanical nightmare but it has proven to be very
reliable over the years.  The little magnets slide in against a mylar
strip with very fine "read wires".  On the other side of the mylar
strip is a device that produces a weak magnetic pulse across a row of
44 bits.  The ones where the adjacent magnet on the card is "off" will
send the pluse through and produce a "1".  If the little magnet is
"on", then the pulse is blocked and a "0" is the result.  The "words"
come out 44 bits at a time.


Ken Abrams         uunet!pallas!kabra437
Illinois Bell      kabra437@athenanet.com
Springfield        (voice) 217-753-7965

shri@ncst.ernet.in (H.Shrikumar) (12/12/90)

In article <15250@accuvax.nwu.edu> rees@citi.umich.edu (Jim Rees)
writes:

>I just went to an open house at the Ann Arbor office of Michigan Bell.

>were!  The 5ESS has a capacity of 160,000 subscriber lines (not all on
>local loops, most come in on SLC-96) and would almost fit in my living
>room.  Well, maybe my basement.

  Does some one have figures in re the economics of the SLC-96 ? I don't
remember having seen any hard numbers in TELECOM Digest.

  I mean, how much does a SLC-96 cost (equipment + installation +
maintenance) to the telco? And more to the point, how does this cost
compare with (equipment+installation+maintenance) cost of 96
local-loop pairs ?

  Or perhaps, at what point does this break even ... a community of 96
subscribers X (??) miles away are better served by SLC-96 than 96 pair
cable.

  Another question, I recently read about an SLC-120 in an Indian
telecom magazine. They were referring to the US. Any SLC-120s in
service ?


shrikumar ( shri@ncst.in )

williamsk@ncar.ucar.edu (Kevin W. Williams) (12/22/90)

In article <15390@accuvax.nwu.edu>, shri@ncst.ernet.in (H.Shrikumar)
writes:

>   Does some one have figures in re the economics of the SLC-96 ? I don't
> remember having seen any hard numbers in TELECOM Digest.

>   I mean, how much does a SLC-96 cost (equipment + installation +
> maintenance) to the telco? And more to the point, how does this cost
> compare with (equipment+installation+maintenance) cost of 96
> local-loop pairs ?

 You won't see numbers, because there are no numbers. If you are
serving a rural area, a mode 2 SLC-96 running in 2:1 concentration
becomes economical at a fairly close range (2-miles or so).

If you are serving a major metropolitan area, and are weighing the
difference between digging up multiple millions of dollars worth of
real estate to lay in new facility, or reusing a few old analog pairs
to do T1 over, the crossover point can be 50 or 60 feet. I have heard
of SLC-96s being used in the upper floors of buildings just to save on
premises wiring costs.

>   Another question, I recently read about an SLC-120 in an Indian
> telecom magazine. They were referring to the US. Any SLC-120s in
> service ?

I find it difficult to believe that there are SLC-120s in the U.S. The
SLC-120 is based on the European 30+2 scheme (4*30=120, just like
4*24=96).


Kevin Wayne Williams   UUCP : ...!ames!ncar!noao!asuvax!gtephx!williamsk