On 5/11/13 4:51 AM, K1TTT wrote:
You can kind of get away with pouring a bit of household 60hz into the
ground, but don't count on it to work for a primary on a power line
either!
Consider this case... the Navy takes the exact opposite approach in
submarine power distribution, nothing is 'grounded' to the hull of a
submarine. One purpose of this is to prevent stray currents from
magnetizing the hull, but a big reason is for safety. As long as no current
path is available through the hull back to a power source you could
theoretically touch a piece of equipment that has a power line short to the
case and not get shocked (assuming of course the case wasn't properly
connected to the line neutral, which would be a 2nd failure). It takes a
fair amount of effort to maintain this isolation, but since the submarine is
essentially a close electrical system it is worth it.
Spacecraft use a similar scheme. the power is isolated from the
chassis. Some spacecraft (Cassini, I think) use a balanced power scheme
where the 28V power is nominally +14 and -14 relative to the chassis
(even though it's isolated, there are things that aren't... coax shields
from coax/waveguide transitions are going to be at chassis potential).
The power supply is still isolated though (DC/DC converters).
Others have the 28V return (- lead) at close to chassis potential, but
still isolated. Spirit and Opportunity have a 6 Amp fuse between
negative power and chassis. If +28V shorts to the chassis for some
reason, the fuse blows.
There was a weird problem where there was ashort that caused the fuse to
blow in ground testing. The short came when a pyro is fired (since the
pyro is literally "on fire", it's a pretty good conductor) and there was
a temporary short from the 28V bus to the chassis. That's not normally a
problem, but sometimes, the pyro leaves crud that is conductive between
the terminals and the shell.
http://www.nasa.gov/offices/oce/llis/1412.html
The same approach could be used for land based power distribution, but with
all the exposed equipment it is impractical except for some cases where it
has some local advantages. So the opposite method is used, connect
everything to the local earth so that at least if there is a fault somewhere
else there won't be a potential difference between earth and the power
system ground. A common misconception is that the ground rod at the house
entrance protects you from faults in the house... while there may be some
odd cases where it helps a bit for that, it is really the green wire, which
is only connected to the earth through the ground at the entrance panel,
that saves you most of the time... it's function is to provide an alternate
path to the neutral in the panel to blow the fuse or breaker,
And a high resistance ground (through flexible metallic conduit feeding
a pie case n a restaurant) is what caused the disastrous MGM Grand fire.
Aggravated by the use of the space above the suspended ceiling as a
plenum for HVAC, and the lack of provision for evacuation, etc.
http://www.nfpa.org/assets/files/Press%20Room/LasVegasMGMGrand.pdf see
page 46 of the 83 pages for the cause of the fire.
" The flexible metal conduit was not effectively grounded, but it was in
contact with grounded surfaces."
BTW, this is one reason why the codes were updated with new rules for
bonding. Bonding is the preferred term, rather than grounding, by the
way, just because of the confusion about "connecting to earth/soil" vs
"bonding to a common point"
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|