Ring-No-Answer

[Robert Dinse <nanook@eskimo.com>]

     The ring-no-answer condition on our dial-up lines on the evening of
4/30/98 and morning of 5/1/98 were caused by a power system problem in a
piece of US West equipment called a Discuss. 

     The Discuss is a piece of equipment that contains a number of T1
banks, each of which multiplexes 24 lines over two pairs of wires and is
used when there aren't enough copper pairs to serve each line with an
individual pair.  Nearly all of our lines are served out of this piece of
equipment. 

     It has dual 48 volt power strings and duplicated "rectifiers", which
are actually sophisticated battery chargers that include current limiting,
voltage regulation, and alarm functions.  Each rectifier charges a string
of batteries and the Discuss can draw power from either string so that, in
theory, as long as one is functioning the Discuss will function. 

     Further, there are alarm circuits that are supposed to automatically
alert telephone company personell when a single rectifier fails so they
can replace it before both fail.  The alarm isn't functioning.  This
allowed one rectifier to fail and go unnoticed until the other failed, and
that went unnoticed until the batteries discharged to the point where the
ring generators wouldn't function anymore. 

     When that happened, there was no longer any ringing voltage to cause
the modems to answer the calls or to ring the phones on the support lines. 

     The person that works on the Discuss was able to get one rectifier
operational by power cycling it, but it is still in "failed" condition.
The other is completely non-operational.  The string voltage is right at
the total discharge state so any further deterioration and we will be out
of service again.

     The batteries discharged very unevenly, there are four 12 volt
batteries in each string.  Some are at 12 volts, some are below 9 volts. 
9 volts is normally considered completely discharged for a 12 volt
lead-acid battery.  The fact that they discharged so unevenly really
indicates they aren't well. 

     So this thing is just limping along.  US West power people are now
out rounding up replacement rectifiers and batteries for this unit.  They
are expecting to return around 9:30am to replace the defective batteries
and rectifiers.  Later in the day an alarm person from US West is supposed
to come out and determine why the alarm circuits aren't working to prevent
future failures from resulting in service deterioration and outages. 

     I've got to say this has been really frustrating.  I first heard
there was a problem around 7pm.  I spent about a half hour looking at our
equipment and trying to determine the scope of the problem.  When it
became apparent it was not our equipment and it didn't appear to be
central office wide but in fact appeared to be only circuits on the
Discuss I called this into repair. 

     I called them every 45-60 minutes and escalated it several times
before I finally got to talk to someone who sounded like they had even the
slightest clue.  By midnight I was actually talking to a supervisor in
Seattle that knew what a Discuss was.  By 2am, they were finally able to
identify the unit.  They didn't get anybody here until almost 5am.

     The person they did finally send, Aiken, impressed me as being one of
the more competent people I have had the pleasure of dealing with at US
West, she seemed to know the Discuss pretty well, but she doesn't do
power, so once power was identified as the problem, it was referred to a
power crew, and an alarm crew to fix the alarms. 

     Hopefully batteries and rectifiers will be replaced and the alarm
circuit fixes later today.  If the power has been going south for a long
time this might haved caused some of the intermittant problems we
experienced earlier.