Bud-- wrote:
volts500 wrote:
To make the same earthing sufficient for transistor safety means both
meeting and exceeding post 1990 National Electrical Code. Therefore an
earthing wire must be even shorter than required by code - 'less than
10 feet'.
The NEC actually does require, in NEC 250.30(A)(4), that the grounding
electrode to be "as near as practicable" and "preferably in the same
area" as the grounding electrode conductor.
The NEC can't dictate where a utility will bring in it's service. In
my house, for example, the electric and phone comes in on the right
side of the house, the water comes in the front, and the cable comes in
on the left side of the house, all due to the physical layout of the
utilities.
The IEEE guide on surges and surge protection:
http://www.mikeholt.com/files/PDF/Li...ion_May051.pdf
Thanks for the link! I was pleasantly surprised that there is a IEEE
publication that's free.
starting on guide page 31 shows the problem of having power H-G bond
point and CATV protector block connected with a long wire - potentially
many thousands of volts between the power wiring and CATV lead.
I read that article twice to make sure that I was reading it right.
While it is an excellent article, and the overall point is well taken,
correct me if I'm wrong, but there seems to be a discepancy in the
document. The calculation on the bottom of p. 30 uses 30 feet as the
distance between points A and B (the distance between the cable ground
block and the connection to the electric system ground.)
That being the case, it seems to me that both figures 7 and 8 are in
violation of 820.40(4) : "...the grounding conductor shall be as short
as practicable, not to exceed 20 feet." Of course, the NEC remedy is to
drive a ground rod and bond it to the electric service.
When I get calls to repair and correct wiring from lightning damage, if
the telephone and/or cable demarc is not directly next to the electric
service, I drive a ground rod at each demarc location and bond the
ground rods to the electric service with buried a #4 bare copper wire.
I then install a new (or previously non-existent) surge arrestor at the
meter. I then connect a TrippLite power/phone/cable surge protector,
as the article calls, "multiport protector," at the equipment.
The fix
in the guide is to use a plug-in suppressor that has both power and CATV
wires go through it - the voltage on power and signal wires is clamped
to the common ground at the surge suppressor. Another fix would be to
route the CATV wire from the entrance ground block to the power service
and install a 2nd ground block with short earthing connection to the
power grounding electrode conductor coming out of the service. Then
distribute the CATV from that point. Same with phone, dish, ....
That's a good idea, if one is able to do it. In my area the cable
distribution is locked up and rendered inaccessible by locks similar to
what the gas companies use to secure a valve.
IMHO a
single point ground is more important than the resistance to earth.
I would like to see the NEC require a Ufer ground, though. IMO, a
ground rod (even two, 6 to 10 feet apart) is about the worst grounding
electrode permited by NEC. A Ufer ground (concrete encased electrode)
is easily installed during new construction, and there is no real
excuse, other than laziness, NOT to install one.
Totally agree. Apparently 250.50 starting in the 2005 NEC is intended to
require Ufer grounds in new construction where there is a concrete
footing or foundation. IMHO the requirement is not at all clear and is
more apparent in the exception than the rule. I don't think it was noted
in code change material I read.
--
bud--