Grounding a metal roof
 

I am having a metal roof put on the house and shop. The installer said it
isn't required to be grounded. The city inspector said the same thing. Is
that really true? I'm thinking of running a #8 from the roof to my common
ground. Does that make sense?

Joe Loucka AG4QC



----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Grounding a metal roof
 

On Sun, 21 May 2006 23:48:59 -0500, "AG4QC"
wrote:

I am having a metal roof put on the house and shop. The installer said it
isn't required to be grounded. The city inspector said the same thing. Is
that really true? I'm thinking of running a #8 from the roof to my common
ground. Does that make sense?

Hi Joe,

It is arguable that the entire roof would even be evenly conducting,
as installed. As conductivity is not a primary concern, the installer
is not motivated to insure that a #8 wire attached to one panel would
be felt by any other with any degree of confidence. Further, even if
all panels exhibited continuity today, this is not to say they would
tomorrow unless some care was taken to insure tight interconnections.

It is not unheard of here in this group to find Hams taking that care
and bonding all panels. The sense of it is to reduce the chance of
developing spurs due to RF currents meeting corroded joints (although,
no one has ever reported this as a problem, only anticipated it).

73's
Richard Clark, KB7QHC

Grounding a metal roof
 

"Richard Clark" wrote in message
...
On Sun, 21 May 2006 23:48:59 -0500, "AG4QC"
wrote:

I am having a metal roof put on the house and shop. The installer said it
isn't required to be grounded. The city inspector said the same thing. Is
that really true? I'm thinking of running a #8 from the roof to my common
ground. Does that make sense?

Hi Joe,

It is arguable that the entire roof would even be evenly conducting,
as installed. As conductivity is not a primary concern, the installer
is not motivated to insure that a #8 wire attached to one panel would
be felt by any other with any degree of confidence. Further, even if
all panels exhibited continuity today, this is not to say they would
tomorrow unless some care was taken to insure tight interconnections.

It is not unheard of here in this group to find Hams taking that care
and bonding all panels. The sense of it is to reduce the chance of
developing spurs due to RF currents meeting corroded joints (although,
no one has ever reported this as a problem, only anticipated it).

73's
Richard Clark, KB7QHC

Joe,
Another concern (though I have no experience with roofs, just autos) is the
intermitent contact (as opposed to a rectifying junction, as Richard points
out) that is possible. If there are joints that can make and break, say in
the wind, you may have noise generated. Chain-link fences would be
susceptable to this as well. Wether or not it produces noise in your
receiver, I would think, is rather unlikely, unless you were somewhere near
a broadcast station or another ham for reasons described next.
Interestingly enough, if the transmit antenna is reasonably close to the
roof, it would manifest itself as noise on your *transmitted* signal also,
but probably low enough such that those listeming to you would hear it.
I had some experience with this phenomenon with things like trunk lids and
motorcycle seat springs causing noise in the 150 MHz car phones of the past.
The transmitter illuminated the noisy joint and the czarrier was literally
modulated by this. The near-by noisy joint can be considered to be a
parasitic element in the transmit antenna system and therefore produce
low-level, but annoying amounts of noise modulation. It appears as
sidebands on the transmitter which duplicate the noise of the joint. This
in turn was wide-band exnogh to extend over to the receive frequency.
Because it was full duplex and both Tx and Rx were on the same antenna, it
was a noise problem in the receiver. Also why I believe it is a bad idea
to use a chain link fence as a counterpoise.

73, Steve, K9DCI

Grounding a metal roof
 

"Steve N." wrote in message
...

"Richard Clark" wrote in message
...
On Sun, 21 May 2006 23:48:59 -0500, "AG4QC"
wrote:

I am having a metal roof put on the house and shop. The installer said
it
isn't required to be grounded. The city inspector said the same thing.
Is
that really true? I'm thinking of running a #8 from the roof to my
common
ground. Does that make sense?

Hi Joe,

It is arguable that the entire roof would even be evenly conducting,
as installed. As conductivity is not a primary concern, the installer
is not motivated to insure that a #8 wire attached to one panel would
be felt by any other with any degree of confidence. Further, even if
all panels exhibited continuity today, this is not to say they would
tomorrow unless some care was taken to insure tight interconnections.

It is not unheard of here in this group to find Hams taking that care
and bonding all panels. The sense of it is to reduce the chance of
developing spurs due to RF currents meeting corroded joints (although,
no one has ever reported this as a problem, only anticipated it).

73's
Richard Clark, KB7QHC

Joe,
Another concern (though I have no experience with roofs, just autos) is
the
intermitent contact (as opposed to a rectifying junction, as Richard
points
out) that is possible. If there are joints that can make and break, say
in
the wind, you may have noise generated. Chain-link fences would be
susceptable to this as well. Wether or not it produces noise in your
receiver, I would think, is rather unlikely, unless you were somewhere
near
a broadcast station or another ham for reasons described next.
Interestingly enough, if the transmit antenna is reasonably close to
the
roof, it would manifest itself as noise on your *transmitted* signal also,
but probably low enough such that those listeming to you would hear it.
I had some experience with this phenomenon with things like trunk lids and
motorcycle seat springs causing noise in the 150 MHz car phones of the
past.
The transmitter illuminated the noisy joint and the czarrier was literally
modulated by this. The near-by noisy joint can be considered to be a
parasitic element in the transmit antenna system and therefore produce
low-level, but annoying amounts of noise modulation. It appears as
sidebands on the transmitter which duplicate the noise of the joint. This
in turn was wide-band exnogh to extend over to the receive frequency.
Because it was full duplex and both Tx and Rx were on the same antenna, it
was a noise problem in the receiver. Also why I believe it is a bad idea
to use a chain link fence as a counterpoise.

73, Steve, K9DCI


Perhaps a more realistic concern would be lightning protection. A friend
down in Fl had just installed a metal roof on his shop just 3 days prior to
it being struck by lightning. Seems the worst damage was a softball size
hole burned in the metal. First installing lightning rods and associated
grounds would provide a good begining to bonding of a metal roof.

Grounding a metal roof
 

"Steve N." wrote in message
...


snip

Another concern (though I have no experience with roofs, just autos) is
the
intermitent contact (as opposed to a rectifying junction, as Richard
points
out) that is possible. If there are joints that can make and break, say
in
the wind, you may have noise generated. Chain-link fences would be
susceptable to this as well. Wether or not it produces noise in your
receiver, I would think, is rather unlikely, unless you were somewhere
near
a broadcast station or another ham for reasons described next.

It's mostly a problem if you are running two or more rigs at the same
time. Your 2M packet rig could interfere with your 2M ragchew.

The Navy, for whom I work, bonds (or isolates) everything topside for
exactly the reason you stated. The intermittent contacts are noise
generators. The tools for detecting some troublesome items are a rubber
mallet and a pair of handhelds. One guy in radio transmits a carrier,
while listening on a nearby freq. If he hears anything, he grabs his
handheld and barks, "What was that you just tapped on?" Primitive but
functional. Every spark, no matter how small, is an RF generator. Chains
are awful. We hate chains.

Grounding a metal roof
 

In article J9adg.27540$QP4.14620@fed1read12,
"Sal M. Onella" wrote:

One guy in radio transmits a carrier,
while listening on a nearby freq. If he hears anything, he grabs his
handheld and barks, "What was that you just tapped on?"

Real Radiomen would use a Spectrum Analyser to look for this type of
stuff........Ears are so linear......

Grounding a metal roof
 

"Sal M. Onella" wrote in message
news:J9adg.27540$QP4.14620@fed1read12...

"Steve N." wrote in message
...


snip

Another concern (though I have no experience with roofs, just autos) is
the
intermitent contact (as opposed to a rectifying junction, as Richard
points
out) that is possible. If there are joints that can make and break, say
in
the wind, you may have noise generated. Chain-link fences would be
susceptable to this as well. Wether or not it produces noise in your
receiver, I would think, is rather unlikely, unless you were somewhere
near
a broadcast station or another ham for reasons described next.

It's mostly a problem if you are running two or more rigs at the same
time. Your 2M packet rig could interfere with your 2M ragchew.

Yes, the "offending" transmitter needs to be in close proximity so that

1- the offending junction(s) are sufficiently illuminated by it to produce
strong enough sidebands to be a problem in the offended receiver.
2- The offended receiver has to be close enough to be offended by the levels
involved.

The Navy, for whom I work, bonds (or isolates) everything ...
Every spark, no matter how small, is an RF generator.

It isn't necessarily a spark, in fact probably infrequently sparks unless
you're dealing with higher power. It goes like this.

Both sides of the offending junction are picking up RF. They are
conductors, so there are RF currents flowing. When there are RF currents
flowing you have radiation. As the junction opens and closes, the RF
currents change. When the junction is making contact, current can flow -
when disconnected, current can not flow (pretty much; let's not get into the
capacitance thing). The changes in current with a noisy contact will mimic
the "noise" of that contact. The radiation will therefore mimic that noise
also. That noisy radiation will sum with the original, desired radiation.
This, in effect, modulates the radiated signal after it leaves the
transmitter. If it is sufficiently high level and the receiver able to pick
it up, you have a problem..

Chains
are awful. We hate chains.

Perhaps you know, but during the Apollo space shots, there were NASA ship at
sea which had really high power HF transmitters. They would interfere with
the radars. It turned out that the deck chains were the offenders. Whether
or not there were sparks, it shows just how wide band the noise can be.

Gee... What's the resonant frequency for a deck chain link?

73, Steve, K9DCI (USN, but never on a real ship untill after I got out)

Grounding a metal roof
 

"Steve N." wrote in message
...


Snip

The radiation will therefore mimic that noise
also. That noisy radiation will sum with the original, desired radiation.
This, in effect, modulates the radiated signal after it leaves the
transmitter. If it is sufficiently high level and the receiver able to
pick
it up, you have a problem..

Correct

Chains
are awful. We hate chains.


Gee... What's the resonant frequency for a deck chain link?


UHF or SHF, but it doesn't much matter. The training aids for the EMI
Awareness course I used to teach at San Diego Naval Station (command =
FTSCPAC, if you know that outfit) included a funky little plywood ship,
within which were two milliwatt-level HF "transmitters" and a receiver.

I/we demonstrated with two links of chain, connected back-to back to make a
continuous current path, that we could generate noise by making and breaking
the connection where the links touched. It was invariably a convincing demo
for the students. The effect only extended out a few inches from the deck
of our plywood model, but scale it up, and you are smack-dab in the real
world.

73,
John, KD6VKW
NARTE-certified EMC Engineer (retired)
USN: 1962 -1982, CTMC, EWCS (Ret.)

Grounding a metal roof
 

"Bart Bailey" wrote in message
...
In Message-ID:lkQdg.27874$QP4.7530@fed1read12 posted on Fri, 26 May
2006 21:12:01 -0700, Sal M. Onella wrote:

The training aids for the EMI
Awareness course I used to teach at San Diego Naval Station (command =
FTSCPAC, if you know that outfit) included a funky little plywood ship,
within which were two milliwatt-level HF "transmitters" and a receiver.

Was that what was called the model pond, has a chocolate chip looking
tower arcing out over it, very unusual sight from the road?
I had a friend Charles Roy WS6F (now SK) that used to work there.

No, but I know what you're talking about. (The demo I did for class was a
desktop affair -- kept in a suitcase between classes) You're probably
referring to the model range(s) at Naval Ocean Systems Center (NOSC) on
Point Loma. NOSC was consolidated, I think, into Space and Naval Warfare
Systems Center, San Diego.

For those not familiar with this process, they used brass ship models, both
existing and proposed, built to 1/48 scale. They were on a lead (Pb) "sea"
at least 100 feet across. At the center of the "sea" was a turntable, on
which the model could present any aspect to the tower you refer to. The
tower supported a log periodic antenna which could be raised along a curved
arc from the horizon up to the zenith, overhead the model. Picture about
halfway down the page at

http://www.spawar.navy.mil/sti/publi...d/1940/photos/

The model ship was fitted with scale antennas, complete with coax that ran
down through the turntable and into the control room. For any given antenna
and location on the model, the engineers could get a three-dimension Antenna
Radiation Pattern at scale for a good idea of how it would work in the real
world. ... and I understand it was quite reliable. 96 MHz scale was 2 MHz
real-world, and so on. I think the LPA did all the transmitting, but I'm
not positive.

I got a guided tour of the range once in the 1980's when I was in their
building for an unrelated conference. (It was more interesting than parts
of the conference, but that's another story.) I wonder, do they still model
antennas this way?

A few of the brass models were regularly on display, visible from the road
about ten feet behind the fence, for many years. It has been five years
since I did any work on Point Loma and I don't know if any models are still
in public view.

73,
John

Grounding a metal roof
 

Bart Bailey wrote:
In Message-ID:F5aeg.27941$QP4.12044@fed1read12 posted on Sat, 27 May
2006 21:58:13 -0700, Sal M. Onella wrote:

You're probably
referring to the model range(s) at Naval Ocean Systems Center (NOSC) on
Point Loma. NOSC was consolidated, I think, into Space and Naval Warfare
Systems Center, San Diego.

I don't know what it's called now, was NOSC, then NRAD, and could be
consolidated with SPAWAR/Lockheed these days.
...and yes it was for doing analyses of maritime antennas and their
mutual effects with the ship's superstructure.

I visited and was given a tour of that site some years ago, when it was
NRaD. They were also doing computer modeling with NEC-4 of the same
ships and structures they were measuring. One of the engineers told me
that when the results from the computer model disagreed with
measurements from the physical model, they'd learned to believe the
computer model.

This was a statement of just how difficult it is to make really good
measurements, even when you have the best equipment. Accurate computer
modeling of complex structures is no trivial task, either, but it was
something they were very good at.

Roy Lewallen, W7EL