Roger,
You make some interesting comments. Please scroll down and see my inserted
remarks. Thank you.
73,
Dale Svetanoff, WA9ENA
Sr EMC Engineer
E-N-A Systems, LLC
Specializing in shielding applications, system grounding, and lightning
protection
> [Original Message]
> From: Roger (K8RI) <k8ri@rogerhalstead.com>
> To: <rfi@contesting.com>
> Date: 9/18/2012 12:39:27
> Subject: Re: [RFI] ISOBAR
>
> On 9/18/2012 10:46 AM, Dale Svetanoff wrote:
> > Pete,
> >
> > Two questions: 1) That 40 feet of bare #2 wire between the entrance
panel
> > and power line ground - I assume it is run outside, and if so, how many
> > ground rods are on it? 2) Does the wire provide all of the grounding
for
> > the entrance panel, or do you have one or more ground rods tied to the
> > panel?
>
> This brings up a number of thoughts and questions.
> Around here, code requires the ground from the panel AND meter are
> relatively short. I don't remember the required minimum distance but
> IIRC it's less than 8 or 12 feet and if additional rods are required
> they must be within that distance. Even with our soil which is wet most
> of the year 3 ground rods are required. I had to put in 5/8 X 8,
> however the required entrance rods are skinny 3/8 X 8.
>
> >
> > While I understand and agree with your comment about non-ideal
> > installations, I think the point has to be made that trying to
accomplish
> > as good of an installation as possible should be a prime goal of every
> > radio installation design. The two most important factors, in ANY case,
> > are: A) Keep the lightning current on the outside of your
> > shack/house/vehicle;
>
> Induced voltage may be higher inside the house than out if a strike is
> close and on the side opposite the entrance panel. But hopefully all
> runs are the same length and voltages rise together.
- - - - - - - - - - - - - - - - - -
Don't bet on lengths being equal, but as long as all of the grounds are
solidly bonded, they will rise and fall (in voltage) together (with phase
differentials being present, of course, on the longer runs). - Dale
- - - - - - - - - - - - - - - - - -
>
> B) Design for as low an impedance as possible between
> > the entrance panel and earth ground and between tower (or antenna
support)
> > and ground. Naturally, net Z will be the total of Xsub-L + wire
resistance
> > + bonding resistance + earth coupling resistance. Net Z should be
figured
> > for frequencies in the range of about 100 kHz to 10 MHz, the main
spectral
> > distribution of lightning energy. (Yes, it extends upwards to nearly
100
> > MHz, but at greatly reduced energy levels.)
>
> Here they "specify" the grounding and it's nowhere near what I'd use.
> It has to be a #6 green insulated wire and it must be continuous which
> means scraping insulation, yet they use mechanical connections which are
> usually loose enough to just lift off the rods after a year or two.
> OTOH you can add additional wires of larger size and there is an
> additional ground out at the power pole.
- - - - - - - - - - - - - - - - -
In my many years in the RF shielding business (shielded rooms and
enclosures), I had one argument after another with electricians who failed
to understand the concept of impedance. They figure 60 Hz is close enough
to DC that it does not make a difference. I can agree on that, but they
have no concept of what a grounding lead (or wire) looks like to lightning
current at 10 MHz. The usual response was "you can't do anything with
lightning. It just goes wherever it wants." - Dale
- - - - - - - - - - - - - - - - -
>
> But when looking at total Impedance and rise times, with the proper
> conditions you can have far more voltage induced into the wiring in the
> house than comes into the panel under some conditions.
>
> The goal is not only grounding, but wiring in such a fashion that all
> equipment connections will rise at close to the same rate, to the same
> potentials. The outlets in a room should be on the same circuit, or
> parallel circuits where the wiring follows the same path with the goal
> of all connections rise and fall together. IE if you have an Internet
> connection to the rigs or to the computer that connects to the rigs and
> a phone line they all are the same length and follow the same path after
> entering the home through a grounded panel.
>
> With older homes wiring is often "daisy chained" from room to room.
> Often where the wiring runs for a particular circuit is a mystery.
> Using my house as an example, I discovered that the outlets on the S
> side of my den where the computers connect, roughly follows a path from
> the panel, up through the wall and over the ceiling to the living room.
> Then down the wall to an outlet, back up and over about 12 feed where it
> descends to another outlet. Some where "up there" is a junction box
> where it feeds two outlets on the West wall and also goes to the back
> bedroom on the West side of the house. From there it goes into the
> bathroom to a switch box with 3 switches and feeds the overhead fan and
> lights. How it gets to the outlets in the den I don't know but the den
> is on the N side of the bathroom. Every thing else in the den is fed
> with relatively short, direct runs from the panel including the phone
> and cable/internet. IE it's about the worst cable routing I could get if
> I did it on purpose. With all those connections, plugging in a small
> canister type vacuum cleaner will dump all the computers so there has to
> be a poor connection in there "some where".
>
- - - - - - - - - - - - - - - - - - - -
My house is full of daisy-chained wiring (run as Romex) by the
electricians. My shack and workshop are in the basement (walk-out level)
and I have used conduit (real EMT) for nearly all wiring down there.
However, I also daisy-chained outlets in some cases - it's almost
inevitable. I also have other conditions which, when I was installing
them, I was not certain what the response would be under actual lightning
strike conditions. Well, thanks to Thor and his friends, I have had 2
chances to find out.
Basically, I have a full perimeter ground ring around the outside of my
house foundation walls (see MIL-HDBK-419A), with 13 size 3/8" by 8' ground
rods either cad welded or clamped to the more than 200 feet of #2 AWG bare
wire. I have 3 towers, each installed with a ring ground and 3 ground rods
around the concrete base. Those ring grounds are tied to the perimeter
ground. I have 2 aluminum entrance panels: a small one on the south end of
the house and a large one on the north end, 54 feet apart. They are both
bonded to the perimeter ground. The perimeter ground is bonded to the
electric power ground right adjacent to the power entrance, on the south
side of the house. My shack is in the NE corner of the basement and is
bonded to the north entrance panel. ALL cable I/Os pass thru protectors in
one of the two entrance boxes. (See the 3-part series by Roger Block in
QST, back in 2002, I believe.)
This works. Current can certainly divide in the loop that goes around each
tower foundation and the house foundation. I agree that there is almost
certainly uneven current distribution, and thus, unequal voltage drop
across a given set of end points. However, experience has shown that
however uneven the voltages may be, it does not appear that any of them
have been sufficient to arc over any in-house wiring or devices, nor to
damage any equipment in the house. Yes, I have a whole-house protector on
the incoming 200 amp mains, and smaller MOV-based protectors at equipment
point-of -use outlets. - Dale
- - - - - - - - - - - - - - - - - - - - - - - -
> I have an electrician coming this week to remedy the mystery path into 3
> circuits that will be a short as possible. Were I 10 years younger I'd
> do the job myself.
>
> Often we don't have a choice such as here the coax cables (TV and ham),
> and rotator cables enter through a common, grounded bulkhead at the back
> of the house while the cable and telephone come in with the power at the
> front of the house. Now the entrance panel and the bulkhead are
> connected with a common #2 ground wire. Unfortunately the house ground
> and bulkhead are almost 30 feet apart. The ground tie does parallel the
> feeds to the den so the rise time better known as Dv/Dt(voltage/time)
> can lead to substantial differences for strikes with rapid rise times.
>
> I agree that just because what we have is less than ideal we still
> should strive to put in the best system possible within our constraints.
>
> 73
>
> Roger (K8RI)
>
> >
> > The latter is essential during the strike because of the huge currents
> > during a strike. Since all of the station equipment is tied (bonded) to
> > the one ground system (or should be), the lower the net Z to ground
means
> > the less voltage developed differentially between grounded connections
and
> > power feeds, antenna inputs, and so forth. (This is in addition to the
> > common mode induced currents from a strike.) Keeping the lightning
> > current on the outside of the building is key to minimizing what I call
the
> > "Kingsford Effect" (conversion of a building into a pile of charcoal)
and
> > is usually handled well by a good entrance panel, well-grounded, and
> > equipped with protective devices for each and every I/O line used in the
> > station. I can personally vouch for the effectiveness of that approach.
> >
> > 73,
> > Dale Svetanoff, WA9ENA
> > Sr EMC Engineer
> > E-N-A Systems, LLC
> > Specializing in shielding applications, system grounding, and lightning
> > protection
> >
> >
> >> [Original Message]
> >> From: Pete Smith N4ZR <n4zr@contesting.com>
> >> To: <rfi@contesting.com>
> >> Date: 9/18/2012 6:36:02
> >> Subject: Re: [RFI] ISOBAR
> >>
> >> I use a piece of 3/4 inch copper pipe running along the rear of my desk
> >> as a ground bus. All of my radios, computers etc. are fed off a single
> >> branch, and each is grounded to the pipe by a short, direct copper
> >> wire. The copper pipe is connected to my aluminum entry panel, and
from
> >> there #2 solid copper goes to my power-line input ground.
> >> Unfortunately, that wire has to be around 40 feet long, so I know it's
> >> not ideal. Not all of us can have purpose-built shacks with perimeter
> >> grounds and copper strap galore.
> >>
> >> 73, Pete N4ZR
> >> The World Contest Station Database, at www.conteststations.com
> >> The Reverse Beacon Network at http://reversebeacon.net, blog at
> > reversebeacon.blogspot.com,
> >> spots at telnet.reversebeacon.net, port 7000 and
> >> arcluster.reversebeacon.net, port 7000
> >>
> >> On 9/17/2012 10:47 PM, Jim Brown wrote:
> >>> On 9/17/2012 7:14 PM, Pete Smith N4ZR wrote:
> >>>> I guess what I was asking was whether the snubbing at the power entry
> >>>> would allow MOVs to work for the branch circuit protection a
> >>>> cost-effective share of the time. I have experienced problems with
> >>>> Ethernet EMP pickup, but so far my cascaded MOVs have done the job,
> >>>> even when I took a direct hit on my tower.
> >>>
> >>> Not if there are the multiple ground point issues with stuff plugged
> >>> into the MOV box.
> >>>
> >>> 73, Jim
> >>> _______________________________________________
> >>> RFI mailing list
> >>> RFI@contesting.com
> >>> http://lists.contesting.com/mailman/listinfo/rfi
> >>>
> >>
> >> _______________________________________________
> >> RFI mailing list
> >> RFI@contesting.com
> >> http://lists.contesting.com/mailman/listinfo/rfi
> >
> >
> > _______________________________________________
> > RFI mailing list
> > RFI@contesting.com
> > http://lists.contesting.com/mailman/listinfo/rfi
> >
>
>
> _______________________________________________
> RFI mailing list
> RFI@contesting.com
> http://lists.contesting.com/mailman/listinfo/rfi
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|