[Prev][Next][Index][Thread]

mid-point grounding

I managed to download (all) the archive files, and was
particularly interested in the center-point grounding
thread.  If this hasn't already been beaten into the
ground(no pun intended), I thought I might throw in my
2 cents worth.
Disclaimer:  I am not a licensed electrician.  The
following views are just opinion, OK?  OK.
Under ideal circumstances, grounding the end-run
returns to the transformer case doesn't really make a
wit of difference, in terms of performance, EXCEPT WHEN
THINGS START GOING WRONG.  Let me illustrate by
example:
Nathan Nobucks, owner of Nathan's NoTell Motel out on
the far edge of town, decides to take up an offer from
Larry Lowball (from Lowball Neon) to put up some
boarder tube across the front of the NoTell Motel. 
Nathan wants it in pink, wants it cheap, wants it
yesterday, and really doesn't care how the job is done,
as long as it works.  They negotiate a price, Larry
squeezes a deposit out of Nathan with a promise of
balance to be paid at the completion of install, and
the deed is done.
As it turns out, 50 feet of 15mm blue, pumped neon,
will just stretch across the front of the motel.  To
keep costs down, Larry splits the run exactly in the
middle, 25 feet per side, and quick-mounts a 15,000/30
with the a couple of sheetrock screws to the underside
of the motel eves, out of reach.  A couple of short GTO
runs from the xformer supply the high voltage, and
Larry ties the far ends of each tube run to the metal
rain gutter running the length of the eves (knowing the
voltage of this end-to-end run should be at ground
potential, thus saving the price of 50 some odd feet of
GTO).  To complete the job, Larry strings a length of
14/2 romex from the nearest porch light fixture to the
two 120 volt terminals of the xformer.  Larry finishes
up as darkness falls.  Snapping on the light switch
brings a fuzzy pink neon glow to the front of the
NoTell Motel.  Nathan is suitably impressed, pays half
the balance due in cash, and promises to pay the
remaining by the end of the month.
And all is right with the world, until one evening...
After a long day on the road, Marvin Myopic pulls his
MamouthMotorHome camper into a middle parking space at
the NoTell Motel.  Misjudging the end of the parking
space, he overshoots, neatly wedging the camper's
extended upper berth under the rain gutter of the
motel, cleaning out an 8 foot section of lit pink
tubing in a tinkling shower of broken glass.  With a
hand firmly gripping the door handle, Marvin steps out
of the camper (which is now firmly connected to the
rain gutter of the motel, which in turn is connected to
one end of the 15/30 via 25 feet of still intact glass)
in his saltwater soaked shoes (wet from pulling his
rowboat out of the water at a distant ocean launch
ramp) onto the rain-wet asphalt, to assess damages,
thus completing a circuit from ground (literally) to
one end of the xformer via the above described path. 
As Larry only used 2 wire romex to power this install,
the case of the wood-mounted, 'mid-point grounded'
xformer rises to approx. 7500 volts.  The other end of
the xformer, tied to the remaining short run of broken
glass, rises to 15KV.  After a brief moment, the now
7500 volt xformer core internally arcs to the primary
winding, indirectly completing the circuit to ground
(and starting a small fire inside the xformer).  Marvin
starts dancing the Funky Chicken in time to a
flickering pink light from the unbroken section of
tubing, impressing his concerned wife Edith, who hasn't
seen him step this lively in years.
So what have we learned from all this?
Point: grounding the case of a 'mid-point grounded'
xformer (1) protects the xformer from internal arcing
damage, should one HV terminal become grounded, and (2)
keeps the other un-grounded terminal from rising above
1/2 of the secondary voltage.  I use an actual wire to
ground, rather than relying on a path thru the can
sheet metal.
Point: the end-to-end tubing connection is NOT a ground
- it is a current-carrying conductor.  If tied to the
mid-point of the xformer, we may call it a 'neutral'
conductor, otherwise, it's just another connection,
which may be at a relative low voltage to ground UNLESS
SOMETHING GOES WRONG.  This is why we use GTO for this,
inside conduit (or whatever) for protection.
Point: there should be only one connection from ground
to the case of the xformer.  Any other connection(s)
should be thought of as 'neutrals', and should
electrically float.  The lug on the xformer is the
'mecca point', where everything comes together.
Any comments? (flames?)
p.s. this installation actually exists, in some (to
remain nameless) town in southern Utah, except they
didn't use the rain gutter for the return.  They used
14 gauge THHN nailed under the eves.
p.p.s.  yes, I know, I'm probably warped.  Happy new
year.
Telford Dorr