Re: Electrical Wire Safety

[John Denker <jsd@AV8N.COM>]

OK ... I'm a dope.  I retract almost everything I said about
earth-return.

Here is my current understanding of the distribution system
around here.  The wires are observed to have the following
layout on the poles

                 G

               A   B


                 R
                 W
                 K


The wire I'm calling G can reasonably be interpreted as
ground since it is tied to a stake in the ground at every
pole I've observed.  And there is no attempt to insulate
G from the pole itself, unlike A and B which are observed
to be supported by ceramic insulators.  None of the HV
wires (A, G, B) carries insulation along its length.

Ironically, the 120V wires (R, W, K) do carry insulation along
their length.

The step-down transformers have their primaries hooked up
either A-to-G or B-to-G.  Unsurprisingly the two possibilities
appear to used more-or-less alternately.  In all cases the
secondaries are hooked to (R, W, K).

The obvious hypothesis is that (A, G, B) is a two-phase three-wire
distribution system, and that G serves both as a return wire
and as lightning protection.

Note that IR effects in the G wire can be (and presumably are)
greatly reduced by balancing the load, i.e. roughly speaking
balancing the number of A-to-G transformers agains the number
of B-to-G transformers.  Otherwise (without such balancing) the
IR voltage on distal parts of the G wire would be such that
grounding it at each pole would create some earth-return current
whether you wanted it or not.  Even as it is, there will be
*some* IR effects in the G wire, depending on how many A-to-G
customers are running their electric dryers while B-to-G
customers are not.

============

This leaves me unable to answer the question that M.E asked:

> If, in fact, it is not safer, what happened?
....
> Was the grounded system developed for some other reason?

My previous answer -- saving wire -- must be retracted.  Sorry!

In particular, the specific hypothesis that must be considered,
as an alternative to the (A, G, B) system described above, is as
follows:

Using the same number of wires, hook up *all* of the transformers
with the primary going from A to B.  This would require more (but
smaller-diameter) turns on the primary in each transformer ...
which almost certainly would have been easy to arrange.  Keep the
G wire for lightning protection, but don't make it part of the
main circuit.  The key improvement is to have both A and B be
transformer-isolated from ground.

It seems clear that transformer isolation would save lives,
for example the previously-mentioned pool maintenance guy.
Disaster would require a *double* fault, involving an
improper contact to A *and* an improper contact to B.

Also this would eliminate whatever earth-return currents are
created by the present system.

I am at a complete loss to explain why they would have passed
up the opportunity to transformer-isolate A and B.