Ray,
>hot dipped galvanized is far more practical from an
estimator's perspective.<
The hot dipped galvanized variant is used in those locations
downtown where street vendors are licensed to operate
their culinary businesses through the use of sidewalk stands.
It has been found to resist the saline residue buildup that
occurs in the underground due to vendors who recklessly
dispose of their perishable liquid waste matter at the end of
the business day.
Large deposits of sodium x-fates are found to collect in
soluble form at certain temperatures during the summer
months. Worse than petrol drippings, these solutions seep
into anything with even the slightest osmotic qualities. They
are the result of hot dog vendors who do a daily run off of
their liquid remains onto sidewalks and street curbs, for lack
of any more feasible means of removal. Not pretty, not
appetizing. Not good for copper circuit splice points.
=====
In 1976, when the four-tone method of nonlinear distortion
was formalized for Bell-approved transmission impairment
measuring sets (TIMS), I was working with the Labs
evaluating HP vs Halcyon, along with some folks in Pactel's
San Diego offices doing the same thing.
During these trials I was called upon to examine a thorny
problem by one of the special services bureaus. After
diagnosing the problem as Harmonic Distortion, I referred it
to the appropriate repair center. Eventually I reported it to
Bell Labs, as well, when it finally dawned on me that the
harmonic distortion was actually manifesting itself on an
all-metallic circuit.
No repeaters, no switch ports, no carrier channel units. Just
wire. I didn't know what could have possibly been causing
it. And I was in much less of a position to recommend a
remedy. So I called the Labs.
This was a line that extended from the Broad Street C.O.,
which serves the NY Stock Exchange, to a Chinatown bank
branch about a mile uptown. It was served out of 32
Avenue of the Americas, the ATT Long Lines HQ building,
where a NY Tel end office also resided.
The circuit had been a frequent repeat offender. Typical
clearances on the trouble reports read as follows: no trouble
found, cleared away while testing, swinging open,
intermittent noise, customer error, customer smoking dope,
etc. It was that kind of frustrating situation.
The circuit could satisfactorily pass data at 2400 bps, but
not at 4800 or above. [This was at a time when 9600 b/s
was really smokin', and used for extraordinary requirements,
only. At the time, modems were priced at about a dollar per
bit. Today, they run about a dollar fifty per kilobit.]
Each time a craftsman (tester) got up on the line to test
it, it seemed to miraculously come clear, i.e., the trouble
vanished. This was due to the 106 vdc [breakdown-causing]
voltage which was applied at the time of test. This
phenomenon has otherwise been referred to, from time to
time, as "zapping" the line. It was the epitome of
ballistic testing until low voltage test sets were finally
mandated for special services test situations.
I was expertly familiar with other threshold syndromes,
whereby data lines could pass traffic up to a certain bit rate,
and then no more. Especially when it occurred on the older
T carrier terminals whose expanders and compressors were
non-hardened (read: unstable semiconductors causing
nonlinear distortion). But I had never seen it before on
circuits which were purely copper-based metallic, end to
end.
The message circuit noise was nil, and there were no
bridged taps. No need to check for phase jitter, or
dropouts, these were not indicated. There was only this
very, very bad signal-to-noise ratio being caused by
nonlinear distortion on an all-metallic circuit. And it went
away, not too surprisingly when you think about it, for a
while, when it was zapped. It later returned, like
clockwork.
The mere fact that it had no active components on it caused
the Labs folks to throw it back at me, suggesting that
perhaps it was I who was smoking dope. Hmph! At
that point I turned to using a Bradley Telecom phasor
domain line analyzer, and sure enough there was this large
three-leaf clover smack in the middle of the tube registering
19 db separation.
With a fundamental to third order harmonic product clearly
visible, the Bradley unit showed six lobes, actually two sets
of three lobes folded back onto themselves, which had a
resultant effect of appearing like a three-leaf clover. The test
clearly indicated a dominant presence of a third order
harmonic. Suddenly, with a Polaroid of this pattern in my
hand and shipped to Holmdel, I had the attention of the
Labs.
The diagnosis? Semiconductor p-n buildup on a good & plenty
splice point caused by salination and oxidation of same in a
hostile street environment. Resolution? None. Actually, we
were going to administer sealing current, but we wound up
changing the pair instead, because the customer was making
higher management level calls by that point, and couldn't
waste any more time. We wound up making the offending
pair (a.k.a., the salty dog) available for burglar alarm circuits
which had constant voltage on them, anyway.
We never received another report from that branch office
again concerning the circuit in question.
Frank_C.
ps - I was only kidding about the Hot DoG merchants. |
