Re: [Phys-l] home wiring catastrophe

[John Denker <jsd@av8n.com>]

On 05/18/2007 10:12 PM, fred bucheit wrote:
>    Why does the USA use a grounded system in the first place? 

That's a good question.

I don't know the answer.

Are there any non-USA distribution systems that do it differently?
I don't know of any.

I conjecture that history may be part of the answer, i.e. there
were good reasons "back then" ... whereas things might be done
differently if we were starting over today.  It's been said the
US is a first-world country with a third-world power grid.

I further conjecture that lightning protection might be part of
the story.

> If we used 
> a non-grounded system we would not have any of these problems, AND we 
> would not need GFI's. Thousands of lives could be saved.  The entire 
> system would act as if it was on an isolation transformer. 

It's not quite that simple.  We would trade one set of problems
for another.  I'm not saying the new problems couldn't be dealt
with, but it would take some cleverness and some money to deal
with them.

a) You'd need to do something about the lightning issue.

Note that the existing grid offers distinctly less-than-perfect
lightning protection ... but still, you don't want to make things
worse.

b) Not every transformer is a good /isolation/ transformer.
There are things like inter-winding capacitance that commonly
produce nasty voltages and currents at the terminals on
the secondary side.  To say the same thing another way,
achieving adequate isolation requires more than just un-
grounding the center tap of the existing transformers.

You can buy triple-shielded isolation transformers, but they
don't grow on trees.

c) There would need to be a fairly elaborate protocol for
detecting faults.  This is a situation where fault "A" by
itself is harmless, and fault "B" by itself is harmless,
but the two together are deadly.  Given that faults /will/
occur at some rate, it is imperative to find the first
fault while it is still harmless, before the second fault
comes along.

I think something like a GFCI would address this issue, but
I'm not sure I understand all the angles.

  This is a common trait of multi-layer safety systems:  The
  system doesn't work unless the threshold for taking corrective
  action is much lower than the threshold of danger.  You
  mustn't wait until the last layer has failed before taking
  action.

d) Et cetera.  I'm not claiming to have done a complete analysis.
FWIW at one time I earned my living solving grounding and shielding
problems for people ... so I know the difference between a complete
analysis and a half-baked one.

I also know it is relatively easy to defend a particular instrument
(or a particular roomful of instruments) against a specific threat,
and much harder to defend an entire community against all possible
threats.

-------

If you want another mystery, take a look at the circuit diagram of
a typical UL-listed "isolation" product
http://www.carebase.com/pdf-specifications/ILc-240-VA-120V-AC-Isolation-Transformer-Power-Line-Conditioner-cb.pdf
and you will see that on the supposedly "isolated" side, they tie
one of the power wires to ground!  I do not understand that.  I
don't know what problem they're trying to solve with that.  Maybe
they're worried about some sort of up-stream lightning strike or
surge;  it sure doesn't look like they are maximizing protection
against electrocution.  Indeed they have defeated any GFCI that
may have been installed upstream.

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

You can use this to motivate EE students to pay attention in your
physics class.  There are a lot of electrical engineers out there
who suffer because they don't know enough physics.  Even the ones
that do know some physics tend to compartmentalize it, rather than
applying it to engineering situations.  Grounding and shielding is
one area where the "textbook" engineering rules often break down,
and it pays to re-analyze things from first principles.  Wires
have parasitic resistance and inductance;  transformers have
parasitic capacitance, etc. etc. etc.  Even the most basic notion
of how the Maxwell equations work is a big help.

I learned a lot about this from Ralph Morrison.  He's very smart
and very experienced in this area.