Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Lightning Damage

35 views
Skip to first unread message

njrusty

unread,
Apr 28, 2006, 7:35:40 AM4/28/06
to
Well, it is almost summertime again here in PA and that time of the
year when we get thunderstorms. We had our first round here last
week and of course my classic LX200 is now dead due to these storms.

I've been having problems, third year in a row, when it appears if
there are any nearby lightning strikes, I blow the IC's for the slew
functions on the scopes main motherboard (N/S/E/W). I've sent the
board to Tim at http://www.telescopeservice.com twice now BTW: Great
Service!!) and Tim even provided my some extra IC's as replacements.

My question, has anyone else heard of such a problem with the
LX200's? I have my LX200 power supply attached to a line
conditioner and a seperate power surge protector. Still, the damn
thing seems to blow once a year. My observatory has alot of systems
running non-stop (computer, weather station..) which never have any
problems.

I would definintely appreciate some feedback on this.

Thanks,
Russ

Matthew Ota

unread,
Apr 28, 2006, 9:04:08 AM4/28/06
to
How about installing a lightning rod somewhere in the vicinity of your
observatory?
That should prevent any lightning damage.

Matthew Ota

Marty

unread,
Apr 28, 2006, 9:23:34 AM4/28/06
to
I don't know if there's ANYTHING that will really protect against
lightning damage except unplugging everything. I had everything in my
house fried some years back from the TV to the doorbell.
Marty

user

unread,
Apr 28, 2006, 9:44:51 AM4/28/06
to

Uh, no, it won't help in the least. That's not how lightning rods work.

It sounds like the OP is having problems with the nearby strikes inducing power spikes
into the scope cables. If that's the case, then all the surge protectors in the
world won't help, because the voltage isn't coming through the power lines, it's
coming from the scope cabling. Essentially, the cables are acting as unwanted
antennas.

The first thing I'd try is installing a ground rod - which means pounding a
6 ft rod into the dirt, which is, shall we say, not fun. ;-) Then run some
heavy braided cable from the rod to the tripod/mount. Additionally, it wouldn't
hurt to run cable to the OTA as well, if it's metal, though that has
obvious difficulties and you'll need to remove it when observing. ;-)

The second thing to try is, if you have long cables running to the scope,
try shortening them up as much as possible. Lightning strikes generate
primarily low-frequency RF, which means that shorter cables are less susceptible
to acting as antennas, since the shorter the antenna, the higher the frequency
it resonates at.

- Rich


- Rich


Chris L Peterson

unread,
Apr 28, 2006, 10:24:02 AM4/28/06
to

There are three routes for lightning damage. The first is power surges
on the main line. That can be fairly well handled with line surge
suppressors, although a close enough strike will still get past those.
But if the damage is occurring when the scope is off, your problem
probably isn't the line (the LX200 has a physical off switch; when it's
off there are limited current paths available).

The next possible source of damage is via induced currents in cables,
either power cables or signal cables. A close strike produces powerful
EM fields. A few years back I lost most of the equipment in my
observatory to a close (10 meter) strike. It was obvious that induced
currents in the cables were the problem. Killed was the LX200 serial
interface chip, the SBIG parallel interface chip, an ST4 head, an
Ethernet switch, an SBIG parallel to Ethernet converter, and an Ethernet
port server. This stuff was all powered down at the time, and unplugged
from the line. A computer and monitor were running; the computer
survived, the monitor didn't.

Finally, and related to the previous, you can have damage by currents
induced directly in the circuitry (even without cables). This last is
unlikely unless your observatory takes a nearly direct hit.

My solution was cable control. I made all the runs as short as possible.
In most cases I cut the cables to the minimum length, and if that wasn't
possible I coiled them. All cables have ferrite beads. I terminated all
the instruments right at the pier, converted to Ethernet, and run only a
single CAT5 cable from the scope. In the three years since then, despite
numerous major thunderstorms and near strikes (not as close as the
first, though) I've had no problems other than a couple of blown
Ethernet switches. If you aren't in a position to consolidate all your
signals to Ethernet, I'd recommend that you detach all your interconnect
cables when bad weather threatens.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com

Paul Schlyter

unread,
Apr 28, 2006, 11:13:56 AM4/28/06
to
In article <5473-445...@storefull-3338.bay.webtv.net>,

Even unplugging everything won't help against a direct hit by lighting.
Fortunately, direct hits are rare -- power surges propagating along power
lines and telephone lines are much more common.

Unplugging everything can be inconvenient. Installing power surge
suppressors offers some protection, although less effective than
unplugging everything, it will work in less severe cases.

--
----------------------------------------------------------------
Paul Schlyter, Grev Turegatan 40, SE-114 38 Stockholm, SWEDEN
e-mail: pausch at stockholm dot bostream dot se
WWW: http://stjarnhimlen.se/

Paul Schlyter

unread,
Apr 28, 2006, 11:13:57 AM4/28/06
to
In article <de8452hfkigdu8vs2...@4ax.com>,

Chris L Peterson <c...@alumni.caltech.edu> wrote:
>On 28 Apr 2006 04:35:40 -0700, "njrusty" <njr...@yahoo.com> wrote:
>
>>Well, it is almost summertime again here in PA and that time of the
>>year when we get thunderstorms. We had our first round here last
>>week and of course my classic LX200 is now dead due to these storms.
>>
>>I've been having problems, third year in a row, when it appears if
>>there are any nearby lightning strikes, I blow the IC's for the slew
>>functions on the scopes main motherboard (N/S/E/W). I've sent the
>>board to Tim at http://www.telescopeservice.com twice now BTW: Great
>>Service!!) and Tim even provided my some extra IC's as replacements.
>>
>>My question, has anyone else heard of such a problem with the
>>LX200's? I have my LX200 power supply attached to a line
>>conditioner and a seperate power surge protector. Still, the damn
>>thing seems to blow once a year. My observatory has alot of systems
>>running non-stop (computer, weather station..) which never have any
>>problems.
>
>There are three routes for lightning damage. The first is power surges
>on the main line.

... and on the phone line! In my country house, I've had damage due
to power surges twice -- both times the surge came through the phone
line (which runs through the air), not through the power line (which runs
fairly deep in the ground). The first time a small power supply, and
a telephone, got damaged. The second time, a VCR recorder and a
satellite receiver got damaged (the incoming phone line ran parallell
with the antenna line for awhile -- the damage was due to a spark jumping
from the phone line to the nearby antenna line).

So check your phone line too -- sparks can jump from the phone line
to nearby cables of some other kind, depending on how the cables run
in your house.

Chris L Peterson

unread,
Apr 28, 2006, 11:21:02 AM4/28/06
to
On Fri, 28 Apr 2006 15:13:57 GMT, pau...@saaf.se (Paul Schlyter) wrote:

>So check your phone line too -- sparks can jump from the phone line
>to nearby cables of some other kind, depending on how the cables run
>in your house.

Yeah, definitely a point of entry for houses. Most observatories don't
have phone lines, but those that do need to be careful. Still, phone
lines are often much safer than power lines because they are more likely
to run underground. In my rural location, power lines are on poles but
phone lines are underground (for miles). Power line surges are common,
but I don't know of many cases of damage from phone lines (even directly
into modems).

canopus56

unread,
Apr 28, 2006, 11:28:25 AM4/28/06
to
Chris L Peterson wrote:
> On 28 Apr 2006 04:35:40 -0700, "njrusty" <njr...@yahoo.com> wrote:
<snip>
<snip> Finally, and related to the previous, you can have

> damage by currents
> induced directly in the circuitry (even without cables). This last is
> unlikely unless your observatory takes a nearly direct hit.
> My solution was cable control. <snip>

Why not use a lightning rod, either on or near the observatory? -
Canopus56

Chris L Peterson

unread,
Apr 28, 2006, 11:45:54 AM4/28/06
to
On 28 Apr 2006 08:28:25 -0700, "canopus56" <cano...@yahoo.com> wrote:

>Why not use a lightning rod, either on or near the observatory? -
>Canopus56

Because it won't help. A lightning rod provides a current path to
ground, which protects the structure itself, but it doesn't protect
equipment from the EM field. In my case, it was a tree about 10m from
the observatory that acted as a lightning rod- the observatory itself
wasn't hit. A lightning rod might have brought the strike even closer,
producing a stronger field and more damage. If I wanted to protect the
observatory, I'd need to put up several surrounding rods, and they'd
need to be 30m high to provide enough attraction that lightning hits be
kept at least 50m from the structure. That's just not practical. Cable
management is simple and has proven effective.

Dan Mckenna

unread,
Apr 28, 2006, 12:12:30 PM4/28/06
to
Paul Schlyter wrote:
> In article <5473-445...@storefull-3338.bay.webtv.net>,
> Marty <mov...@webtv.net> wrote:
>
>>I don't know if there's ANYTHING that will really protect against
>>lightning damage except unplugging everything. I had everything in my
>>house fried some years back from the TV to the doorbell.
>> Marty
>
>
> Even unplugging everything won't help against a direct hit by lighting.
> Fortunately, direct hits are rare -- power surges propagating along power
> lines and telephone lines are much more common.
>
> Unplugging everything can be inconvenient. Installing power surge
> suppressors offers some protection, although less effective than
> unplugging everything, it will work in less severe cases.
>

I have nightmares about this every summer.


Meet Mr. P.

Most damage is from the EMP, Electromagnetic pulse, like one gets from
atomic weapons. When lightning strikes, the large electric field near the
event collapses and during that phase creates a potential difference
wave that moves at the local speed of light. The field can be 10s of
thousand volts per meter so a short conductor can have 1000 volts across
it for the time it takes to run across. Lightning when it enters the
ground does not go right in but fans out horizontally. lot of pictures
of this as the after effects of strikes on golf courses. So one legged
cows would survive more than the four legged due to the horizontal
ground currents.

The Key is an Iso Potential environment, or , its all relative.

Like most have said, remove the antennas like power cables and guide
paddles etc. Better yet due what the folks at the sloan sky survey do
make and EMP box to fit over the electronics. The Sloan CCD camera is
removed every morning and lowered in to metal box. The Idea is to have a
Faraday cage around your assets. If your observatory has a metal roof or
dome, ground it and create a star ground where the distance to the
"ground" has equal propagation time to your equipment. These are called
iso-potential lines and the idea is to have all the stuff go up and down
together to reduce the potential difference.

As for me, I am attempting to replace all copper data lines with fiber
optic cable and install power switches that disconnect the building
power and ground it at both ends of the run.

Note: UPS devices do not protect your equipment from EMP.

If you have a metal roof (grounded) then usually a lightning rod will
not help as its primary function is to direct the attachment of the main
current flow. In fact our lightning expert said the rods on the obs roof
are a danger from impalement to roof workers and should be removed.

Read Mark Twains account of the lightning rod salesman

d.


Skylook123

unread,
Apr 28, 2006, 12:14:46 PM4/28/06
to
I think the purpose of a lightning rod is to actually be an anti-lightning
rod; to provide a path to dissipate the potential difference between ground
and sky before it builds up sufficiently to catastrophically jump the gap.
If lightening hits a lightning rod, it would pretty much fry/melt/vaporize
whatever is in path or connected to it.

Jim
"Chris L Peterson" <c...@alumni.caltech.edu> wrote in message
news:ood4521l4kip3tnhb...@4ax.com...

Chris L Peterson

unread,
Apr 28, 2006, 12:21:44 PM4/28/06
to
On Fri, 28 Apr 2006 09:14:46 -0700, "Skylook123" <Skylo...@aol.com>
wrote:

>I think the purpose of a lightning rod is to actually be an anti-lightning
>rod; to provide a path to dissipate the potential difference between ground
>and sky before it builds up sufficiently to catastrophically jump the gap.
>If lightening hits a lightning rod, it would pretty much fry/melt/vaporize
>whatever is in path or connected to it.

A lightning rod doesn't do that. It can somewhat reduce the strength of
the strike by letting it begin a little earlier, but it can't discharge
the clouds to the ground and prevent a strike.

Lightning rods take direct hits all the time, and they don't melt or
vaporize, nor does the conductor from the rod to ground (unless it has
developed resistive connections or was undersized to begin with). They
work very well in preventing currents from flowing through structures
(where they cause fires or get into wiring and damage other things). But
the lightning still strikes, and sensitive electronics located within a
few meters of the strike is likely to be damaged by the EM field itself.

Davoud

unread,
Apr 28, 2006, 12:57:50 PM4/28/06
to
Dan Mckenna wrote:

> Most damage is from the EMP, Electromagnetic pulse, like one gets from

> atomic weapons...

Huh!? This happens to you from time to time, the explosion of a nuclear
bomb!?

Davoud

--
usenet *at* davidillig dawt com

AustinMN

unread,
Apr 28, 2006, 1:38:33 PM4/28/06
to
Skylook123 wrote:
> I think the purpose of a lightning rod is to actually be an anti-lightning
> rod; to provide a path to dissipate the potential difference between ground
> and sky before it builds up sufficiently to catastrophically jump the gap.
> If lightening hits a lightning rod, it would pretty much fry/melt/vaporize
> whatever is in path or connected to it.

When lightening strikes the earth (or any structure or living thing on
the earth), it has usually travelled at least a mile to arrive. It
needed sufficient potential to jump that distance. Compared to that
5000+ feet (~1700 meters), the 30 foot (10 meter) elevation of the
lightening rod doesn't make much difference in the energy. A mile-high
lightening rod probably would drain static and reduce the intensity of
lightening strikes. But it would come at the cost of being struck much
more often.

The lightening rod is only intended to deflect a direct strike away
from existing nearby (or below) structures.

As to vaporizing the rod or ground wire, if the lightening lasted long
enough, yes it would. But the energy in the lightening is dissipated
along it's entire length (including it's travels down from the cloud
and after entering earth), not just in the lightening rod and grounding
system. Relatively speaking, the lightening rod only absorbs a small
amount of the lightening's power.

As others have posted, it would do nothing to alleviate the EMP that
probably caused the OP's problem.

Austin

Dan McKenna

unread,
Apr 28, 2006, 2:23:00 PM4/28/06
to
I

Davoud wrote:

I knew a few people who were at A & H tests in the 50's and they spun quite
a yarn.

d

Chris L Peterson

unread,
Apr 28, 2006, 1:43:48 PM4/28/06
to
On Fri, 28 Apr 2006 12:57:50 -0400, Davoud <st...@sky.net> wrote:

>Huh!? This happens to you from time to time, the explosion of a nuclear
>bomb!?

Yeah, and the worst thing about it is the way it keeps blowing out my
serial ports!

Paul Schlyter

unread,
Apr 28, 2006, 2:43:42 PM4/28/06
to
In article <280420061257507656%st...@sky.net>, Davoud <st...@sky.com> wrote:

> Dan Mckenna wrote:
>
>> Most damage is from the EMP, Electromagnetic pulse, like one gets from
>> atomic weapons...
>
> Huh!? This happens to you from time to time, the explosion of a nuclear
> bomb!?
>
> Davoud

Well, in his country I suppose most people carry weapons of one kind
or another. Maybe his neighbour decided to get his own nuclear bomb?
It's his right to defend himself, isn't it? Maybe he detonates a bomb
once every few months or so, when he gets attacked by a burglar or
something....

:-)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))

Dan McKenna

unread,
Apr 28, 2006, 3:44:04 PM4/28/06
to

Paul Schlyter wrote:

That big talk for a country that dropped ABBA on the world !
:-))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))

We will not fall behind on the smile gap, just "look at the big board"
d

starburst

unread,
Apr 28, 2006, 3:17:58 PM4/28/06
to

>>Huh!? This happens to you from time to time, the explosion of a nuclear
>>bomb!?
>>
>>Davoud
>>
>>--
>>usenet *at* davidillig dawt com
>
>
> I knew a few people who were at A & H tests in the 50's and they spun quite
> a yarn.
>

My father also saw a half-dozen or so above-ground shots in Nevada, some
from as close as two miles. They hunkered down in a trench facing away
from the direction of the blast, eyes closed and arm covering the eyes.

He says the flash was so intense you could see the bones in your arm.

Dig it.
Chris

starburst

unread,
Apr 28, 2006, 3:26:39 PM4/28/06
to

>>
>>:-)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
>>--
>>----------------------------------------------------------------
>>Paul Schlyter, Grev Turegatan 40, SE-114 38 Stockholm, SWEDEN
>>e-mail: pausch at stockholm dot bostream dot se
>>WWW: http://stjarnhimlen.se/
>
>
> That big talk for a country that dropped ABBA on the world !
> :-))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
>
> We will not fall behind on the smile gap, just "look at the big board"
> d
>

"Mr. President, I uh... I don't know quite how to put this... but... do
you realize what *serious* breach of security that would be? Why, why
he'll see everything! He'll see the big board!"

Love that movie.

Dan McKenna

unread,
Apr 28, 2006, 4:26:29 PM4/28/06
to

starburst wrote:

Thats the one !

d.

none

unread,
Apr 28, 2006, 4:15:03 PM4/28/06
to
"AustinMN" <tacoo...@hotmail.com> wrote in news:1146245913.707007.312810
@i40g2000cwc.googlegroups.com:

> lightening

You make good points, but your lack of knowledge of how to spell it
correctly drains your credibility.

LIGHTNING.

Skylook123

unread,
Apr 28, 2006, 6:38:30 PM4/28/06
to
Some bad information I had from helping install lightning rods on barns as a
young punk kid back in the late '50s. That was the salesman's pitch; he
claimed the purpose was to prevent strikes.

Never too late to learn, I guess.

Jim

"Chris L Peterson" <c...@alumni.caltech.edu> wrote in message

news:4vf4521q5sknmimuc...@4ax.com...

Davoud

unread,
Apr 28, 2006, 7:24:07 PM4/28/06
to
AustinMN wrote:
> > lightening

"none" replied:

> You make good points, but your lack of knowledge of how to spell it
> correctly drains your credibility.
>
> LIGHTNING.

Many people with cogent ideas ruin their credibility (with some of us)
by their lack of ability to articulate those ideas in written English.
Rightly or wrongly, people like you and me are likely to give short
shrift to such a person's writing even if, as you said, it contains
good points.

Fuhgeddaboudit. You and I are obsolete. Welcome to the new America. No
math, no reading, no spelling, no logic or critical thinking. It makes
sense, in a way -- how much book learning does a person need in order
to watch TV?

My plan to make it a federal crime to expose a child under the age of
21 to an operating television set doesn't appear to be snowballing into
a national movement.

AM

unread,
Apr 29, 2006, 6:35:10 AM4/29/06
to
Paul Schlyter wrote:
> In article <280420061257507656%st...@sky.net>, Davoud <st...@sky.com> wrote:
>
>
>>Dan Mckenna wrote:
>>
>>
>>>Most damage is from the EMP, Electromagnetic pulse, like one gets from
>>>atomic weapons...
>>
>>Huh!? This happens to you from time to time, the explosion of a nuclear
>>bomb!?
>>
>>Davoud
>
>
> Well, in his country I suppose most people carry weapons of one kind
> or another. Maybe his neighbour decided to get his own nuclear bomb?
> It's his right to defend himself, isn't it? Maybe he detonates a bomb
> once every few months or so, when he gets attacked by a burglar or
> something....
>
> :-)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))


Hey, this is America, owning firearms is a constitutional
right ! And dont worry, I am doing my part for all of ya
not buying/owning them. Just got a folding stock Mini 14
yesterday, and a dozen 30 round Ruger mags :)

--
AM

http://sctuser.home.comcast.net

CentOS 4.3 KDE 3.3

Paul Schlyter

unread,
Apr 29, 2006, 6:43:33 AM4/29/06
to
In article <44527084...@as.arizona.edu>,
Dan McKenna <mck...@as.arizona.edu> wrote:

> Paul Schlyter wrote:
>
>> In article <280420061257507656%st...@sky.net>, Davoud <st...@sky.com> wrote:
>>
>>> Dan Mckenna wrote:
>>>
>>>> Most damage is from the EMP, Electromagnetic pulse, like one gets from
>>>> atomic weapons...
>>>
>>> Huh!? This happens to you from time to time, the explosion of a nuclear
>>> bomb!?
>>>
>>> Davoud
>>
>> Well, in his country I suppose most people carry weapons of one kind
>> or another. Maybe his neighbour decided to get his own nuclear bomb?
>> It's his right to defend himself, isn't it? Maybe he detonates a bomb
>> once every few months or so, when he gets attacked by a burglar or
>> something....
>>
>> :-)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
>> --
>

> That big talk for a country that dropped ABBA on the world !
> :-))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))

FYI: I'm just a single individual, not a whole country..... :-)

Btw, one interesting thing with ABBA is that they became much more popular
in a number of countries abroad than they became in Sweden. I myself never
listen to them very much.


> We will not fall behind on the smile gap, just "look at the big board"
> d

DT

unread,
Apr 29, 2006, 7:57:14 AM4/29/06
to
Davoud <st...@sky.net> wrote

> You and I are obsolete.

Don't worry, there are lots of us, just waiting for the right time to
revolt, and then we'll teach these young....
Oh, hello death, who, me? Oh s**t....

Denis
--
DT
change nospam: n o s p a m
v a l l e ys

Marty

unread,
Apr 29, 2006, 10:49:07 AM4/29/06
to
AM was saying

>Hey, this is America, owning firearms is
> a constitutional right ! And dont worry, I
> am doing my part for all of ya not
> buying/owning them.

Ditto here! During the winter months, I've reloaded about 700 rounds of
200gr .45 ACP lead semiwadcutters for my 1911 Kimber! I can punch paper
at the range even on CLOUDY days!
Marty
(the cloudy thing makes this a LITTLE on topic...)

Stephen Paul

unread,
Apr 29, 2006, 12:21:53 PM4/29/06
to

Completely off topic...
/Since the subject came up in friendly conversation./

While I respect and applaud the law allowing us to bear arms for the
purpose of protecting ourselves against organized enemies of the state,
I find the underlying need and purpose of armament unbearably depressing
in a largely civilized society. I guess that means that I'm offended by
people actually bearing arms for entertainment purposes. It seems callus
given the severity of the purpose for which the law grants us the
privilege to participate in our own protection.

That said, no one should be concerned about my taking offense, it's just
how I "feel" about it. I think weapons management and usage should be
part of civilian education in the event that there's a need to take up
arms against the state gone awry, or against organized enemies of the
state otherwise. I have no problem owning a weapon for such purposes,
but in a civilized society, part of that ownership means being held
criminally liable for the safe storage and use of the weapon. I'm in
favor of privately owned civilian armories, stocked with appropriate
weapons, and accessible to private citizens for training, practice, and
purposeful usage should there be a call to arms.

Of course, a lot of this should depend on where you live. Therefore,
while Federal law grants us the privilege, local communities should be
free to have their own bylaws governing weapons access. In my community
there is no reason to believe that there's an impending invasion of, or
by the state that would be cause for the distribution and storage of
weapons in the home. Furthermore, local laws governing the discharge of
a weapon within city limits, and within a safe distance from other
homes, pretty much excludes there use around the home.

Now, someone will surely make the claim that a weapon in an armory two
miles distant isn't going to help you when your life is in jeopardy
during a home invasion...

But that's exactly the point in such privileges being handled at the
local community level. Where I live, there is no reason to believe that
a home invasion is an impending risk. Therefore there is no reason to
have a weapon in the home where the risk of criminal liability is higher
due to theft, or otherwise improper handling. Something which is
placated by a weapon rather stored in a private armory (such as the
local Rod and Gun club facility).

Just my nickel on gun control, the NRA, and the limits of personal
privilege vs. public safety (including the safety of our law enforcement
officials, without whom we'd all necessarily be armed to the teeth).

I may be wrong, but like race and death, being so is one of the few
absolutes in life.

Be careful,
Stephen

Bert Hyman

unread,
Apr 29, 2006, 12:46:39 PM4/29/06
to
In news:9YmdnWc2Z5YBD87Z...@comcast.com Stephen Paul
<smarsh...@gmail.com> wrote:

> I find the underlying need and purpose of armament unbearably
> depressing in a largely civilized society.

Reality is often troubling. That doesn't make it any less real.

--
Bert Hyman St. Paul, MN be...@iphouse.com

Noah Little

unread,
Apr 29, 2006, 12:50:02 PM4/29/06
to
Stephen Paul wrote:
> Where I live, there is no reason to believe that
> a home invasion is an impending risk. Therefore there is no reason to
> have a weapon in the home where the risk of criminal liability is higher
> due to theft, or otherwise improper handling. Something which is
> placated by a weapon rather stored in a private armory (such as the
> local Rod and Gun club facility).

In order to be intellectually honest, though, we need to ask ourselves
if we would be willing to post a sign on our premises declaring "There
are no firearms in this household."

More to the point, if you, with your (presumably) gun-free home, are
getting a "free ride" from whatever protection is afforded by the
implicit possibility that your place is armed, then there are some good
reasons you really shouldn't do anything that would inhibit others'
right to have the actual protection.

Chris L Peterson

unread,
Apr 29, 2006, 1:17:46 PM4/29/06
to
On Sat, 29 Apr 2006 12:21:53 -0400, Stephen Paul
<smarsh...@gmail.com> wrote:

>While I respect and applaud the law allowing us to bear arms for the
>purpose of protecting ourselves against organized enemies of the state,
>I find the underlying need and purpose of armament unbearably depressing
>in a largely civilized society. I guess that means that I'm offended by
>people actually bearing arms for entertainment purposes. It seems callus
>given the severity of the purpose for which the law grants us the
>privilege to participate in our own protection.

The law grants us no such privilege. Rather, it acknowledges a right.
Those are _very_ different things. I am able to have a gun not because
some law allows it, but rather because the Constitution largely
prohibits any law from preventing it (a right which is unfortunately
being chipped away at- like so many other rights).

I find nothing offensive in using firearms for entertainment. I think it
will be a great day when that will be the only useful purpose for them.
Weapons from days past- spears, swords, bows- are now used almost
exclusively for entertainment, and few are offended that these tools,
once used to deadly effect against people, are still used for other
purposes.

Of course, given our history, I can only assume that guns will become
obsolete as weapons when they are replaced by something "better",
phasers or the like. Like you, I find that a bit depressing.

Rich

unread,
Apr 29, 2006, 2:17:50 PM4/29/06
to
On Sat, 29 Apr 2006 12:21:53 -0400, Stephen Paul
<smarsh...@gmail.com> wrote:

>Marty wrote:
>> AM was saying
>>> Hey, this is America, owning firearms is
>>> a constitutional right ! And dont worry, I
>>> am doing my part for all of ya not
>>> buying/owning them.
>>
>> Ditto here! During the winter months, I've reloaded about 700 rounds of
>> 200gr .45 ACP lead semiwadcutters for my 1911 Kimber! I can punch paper
>> at the range even on CLOUDY days!
>> Marty
>> (the cloudy thing makes this a LITTLE on topic...)
>>
>
>Completely off topic...
>/Since the subject came up in friendly conversation./
>
>While I respect and applaud the law allowing us to bear arms for the
>purpose of protecting ourselves against organized enemies of the state,
>I find the underlying need and purpose of armament unbearably depressing
>in a largely civilized society. I guess that means that I'm offended by
>people actually bearing arms for entertainment purposes. It seems callus
>given the severity of the purpose for which the law grants us the
>privilege to participate in our own protection.
>
>That said, no one should be concerned about my taking offense, it's just
>how I "feel" about it.

Speaking as someone who has shot targets (no people, no hunting)
since I was about 4, I think that it's a culturally-induced reaction
to dislike the idea of firearms.
They've done studies and found that around 50% of the people who
express a distaste for firearms do not hate them (at all, or nearly as
much) once they've actually used them. I've seen this for myself.
I'm sure you live in a urban area, on the East or West coast so it's
not likely you'd be able to walk into a gun range (the underclass in
those places are too busy shooting each other) but if you get a chance
to go to one of the ranges if you visit Arizona, or Nevada, etc, or if
you know any friends who have access to one where you live, give it a
"shot." It could change your attitude or at least make you a bit
more personally aware of what a firearm is actually like.

Stephen Paul

unread,
Apr 29, 2006, 5:12:36 PM4/29/06
to

Well, actually I've fired quite a number of firearms, including an AR-13
and .357 and .44 Magnums, 9mm, the obligatory .22 rifle, and a pistol of
same. My brother was a collector for quite some time, and I distinctly
did not grow up in an urban area nor do I live in a particularly
suburban area, at least for the time being. As more and more housing
goes up locally, that is changing.

I think my distaste for weapons can be traced back to my stepfather,
Fredick T. ("Moose") Heyliger, who had a habit of saying, "I ducked
enough bullets in the war". You can read about him in Stephen Ambrose'
Band of Brothers (or watch the HBO special). Not a large amount of
coverage on him, but he was given Command of Easy Company when Dick
Winters was re-assigned, and he was sent to the hospital by friendly
fire upon return from patrol.

Anyway, like I said, it's more of a gut thing ("feel") than an
intellectual one. I have no qualms about folks making a decision to
house firearms, I just choose not too for all the reasons that one
might expect. I would prefer the private armory for the sake of not
having to worry (as much) about a Columbine incident at my local school.

By the way, my brother, who taught gun safety courses on behalf of the
local police, was not immune to an accidental firing of one of his
handguns. Put a hole through the ceiling and lodged the bullet in a
rafter in the attic. :-)

Best,
Stephen

w_tom

unread,
Apr 29, 2006, 10:55:26 PM4/29/06
to
A tree is struck by lightning. Nearby, a cow dies. Did EM fields
kill the cow? No. For same reason that a nearby strike becomes a
direct strike to electronics. Step back to observe a larger picture.
Circuit is a connection from clouds to earthborn charges maybe five
miles diagonally. What is the shortest electrical path? Not five miles
to those charges. Instead, three miles down to a tree and four miles
through earth to earthborne charges.

Meanwhile, another shorter path is up that cow's hind legs and down
its fore legs. Cow is killed by a direct strike - even though
lightning connected to a nearby tree.

Same occurs inside factilities with electronics AND why a single
point earth ground is so essential for protection. Every incoming wire
to a facility must first connect to ground before entering the
building. Not just connect. Connect direct, independent, and short -
ie less than 10 feet. Even how the earthing connection is created is
critical.

This figure demonstrates two structures - both with single point
grounds. Notice how every incoming wire first connects to earth -
either by direct connection or via a surge protector. Even underground
wires can carry destructive transients. Futhermore, to make earthing
more effective, both single point grounds are interconnected:
http://www.erico.com/public/library/fep/technotes/tncr002.pdf

Lightning strikes wires atop a distant telephone pole. Is that a
direct strike to building electronics? If lightning finds a path to
earth via household electronics, then a nearby strike IS a direct
strike. This is, for example, why telephone appliances are damaged by
lightning strikes to AC electric.

Meanwhile protectors adjacent to electronics can even contribute to
transistor damage. Many assume these protectors somehow stop, block,
or absorb surges. Will that little device stop what three miles of
sky could not? Protectors are connecting devices - only as effective
as the connection to earth. Effective protectors are distant from
electronics AND make a short ('less than 10 foot') connection to single
point earth ground. No earth ground means nothing to connect to.
Protectors connect lightning to earth (just like lightning rods) so
that lightning does not overwhelm proetction already inside
electronics. No earth ground means protector is not effective.

You want effective electronics protection? Protection is standard in
telephone switching exchanges, commercial broadcasters, emergency
response centers, and long understood (by example) among ham radio
operators. Direct strikes with no damage. For example "Why a Short
Lightning Ground" in the newsgroup rec.radio.amateur.antenna on 2 Mar
2005 - especially posts from Richard Harrison and Jack Painter at:
http://tinyurl.com/ao36t
Bill Otten demonstrates how enhanced grounding of his amateur radio
station provides lightning protection:
http://www.knology.net/~res0958z/
Orange County Emergency Response system has earthing corrected due to
previous damage:
http://www.psihq.com/AllCopper.htm

You have elecctronics damage? Start with the most critical component
in every protection sysetm - earthing. Electronics damage is due to
direct strikes - often via paths not obvious. Electronics already
contain protection that is effective at the appliance. Internal
protection that assumes major transients are earthed before getting to
a structure. No earth ground means no effective protection. You have
lightning damage? Not from electromagnetic fields. Find that path
lightning used to obtain earthed charges. That path can even be
incoming on one building earth ground, destructively through
electronics, then outgoing through another building earth ground. Many
then would assume this was from EMP fields. Appreciate why earthing
the foundation of AND the most critical part of a transistor protection
'system'. Lightning rods and surge protectors - both are only as
effective as their earth ground.

Chris L Peterson

unread,
Apr 29, 2006, 11:07:28 PM4/29/06
to
On 29 Apr 2006 19:55:26 -0700, "w_tom" <w_t...@usa.net> wrote:

> A tree is struck by lightning. Nearby, a cow dies. Did EM fields
>kill the cow? No. For same reason that a nearby strike becomes a

>direct strike to electronics...

All the same, electronics are at great risk from the EM pulse alone.
When all the equipment in my observatory was blown out, nothing was even
connected to the line. Most of the parts that were damaged were sitting
at one end or the other of interconnect cables between isolated
instruments. No amount of grounding would have helped. Very close
strikes induce currents in wiring. You need to manage the lengths of
cables, and/or provide some extraordinary shielding to reduce the
likelihood of this kind of damage from close strikes.

I originally had the ground circuit for the observatory power connected
to a local ground rod. This resulted in a lot of damaged network
switches (the network wiring runs underground around my property). I
ended up removing the ground from the rod and just letting the
observatory get its ground from the wiring in the conduit. Now I don't
have blown switches anymore. When each end of the power line was
separately grounded, large differences in potential from ground strikes
showed up as differences on signal wires. With everyone sharing a common
ground, this problem has vanished.

w_tom

unread,
Apr 29, 2006, 11:16:53 PM4/29/06
to
njrusty wrote:
> Well, it is almost summertime again here in PA and that time of the
> year when we get thunderstorms. We had our first round here last
> week and of course my classic LX200 is now dead due to these storms.
>
> I've been having problems, third year in a row, when it appears if
> there are any nearby lightning strikes, I blow the IC's for the slew
> functions on the scopes main motherboard (N/S/E/W). ...

Why damage occurs AND how damage is routinely terminated is
described in a earlier post below. John Ferrell is asking exact same
question - even though electronics is different and incoming wires have
different names. More good answers to this post are in that discussion
in rec.radio.amateur.antenna starting 27 Apr 2006 entitled 'Lightning
suppressor?' Note important points such as low impedance ground -
which involves much more than low resistance.

AustinMN

unread,
Apr 30, 2006, 12:12:14 AM4/30/06
to

Gramarian:
http://redwing.hutman.net/%7Emreed/warriorshtm/grammarian.htm

::Quote:: Grammarian usually has little to contribute to a discussion
and possesses few effective weapons. To compensate, he will point out
minor errors in spelling and grammar. Because of Grammarian's obvious
weakness most Warriors ignore him. ::End Quote::

Austin

Paul Schlyter

unread,
Apr 30, 2006, 5:43:06 AM4/30/06
to
In article <ov9852lhgft57a29u...@4ax.com>,

Chris L Peterson <c...@alumni.caltech.edu> wrote:

> On 29 Apr 2006 19:55:26 -0700, "w_tom" <w_t...@usa.net> wrote:
>
>> A tree is struck by lightning. Nearby, a cow dies. Did EM fields
>> kill the cow? No. For same reason that a nearby strike becomes a
>> direct strike to electronics...
>
> All the same, electronics are at great risk from the EM pulse alone.

The cow didn't die from an EM pulse. The cow died because she's got
two pairs of feet a few meters away from one another. Those pairs of
feet will be subjected to a brief voltage difference on the ground
when the lightning strikes. That voltage difference will produce a
current in the cow, which kills the cow.

Creatures who walk on two feet, like humans, are less susceptible to
this (the fact that we wear shoes or boots, which helps serving as an
electric insulator, of course also helps). However, if you're caught
outside during a thunderstorm, far away from a house or a car, the
advice "don't stand under a tree" is well-known. Standing outside in
some field, with tall trees a hundred meters or so away which are
more likely to catch a direct strike, is then preferable. But you
should then also stand with your feet close together, not wide apart.
If you have company, don't touch your fellows - if a lighing strikes
nearby, you may both die, like the cow did.

Likewise, if the cow somehow learned to stand on two feet, she'd be more
likely to survive a lngtning strike nearby.

Brian Tung

unread,
Apr 30, 2006, 12:37:00 PM4/30/06
to
Paul Schlyter wrote:
> The cow didn't die from an EM pulse. The cow died because she's got
> two pairs of feet a few meters away from one another.

Not really anything important, but *damn*--that is one big freakin' cow.

--
Brian Tung <br...@isi.edu>
The Astronomy Corner at http://astro.isi.edu/
Unofficial C5+ Home Page at http://astro.isi.edu/c5plus/
The PleiadAtlas Home Page at http://astro.isi.edu/pleiadatlas/
My Own Personal FAQ (SAA) at http://astro.isi.edu/reference/faq.html

w_tom

unread,
Apr 30, 2006, 3:06:37 PM4/30/06
to
"The cow died because she's got two pairs of feet a few meters away
from one another" was exactly the point. Chris L Peterson demonstrated
why improper earthing can cause damage. He probably has multiple earth
ground - items he assumed were non-conductive. Therefore lightning
came out of the ground - just like the cow - went through electronics
and then back into earth. Table, wall paint, linoleum tile, and wooden
walls are all conductors. Some better than others. It all explains
why a single point ground must be the best earth ground for that
structure.

If we built buildings to protect transistors, then earthing would be
installed before footings were poured. Best and least expensive
transistor protection must be planned for when the architect draws
prints. We still don't build for transistor protection. Therefore
too many earthing solutions are kludge after-thoughts.

Rather than fix the problem, Chris L Peterson only continued to make
the problem bad. He must enhance earthing to well exceed National
Electrical Code (NEC) requirements. One utility provided figures of
bad, good, and ugly solutions:
http://www.cinergy.com/surge/ttip08.htm

Does an EM field contain sufficient energy to damage electronics?
>From junior high school science, experimental evidence (damaged
electronics) is not sufficient. We also need numbers from theory (the
hypothesis). EM fields numbers just do not exist; do not demonstrate
damage from lightning generated fields.

But to demonstrate how trivial those fields are, we use a classic
long wire antenna. A nearby strike can create hundreds (maybe even
thousands) of volts on that wire. Then we connect that wire through an
NE-2 neon glow lamp to earth ground. That neon bulb glows (if room is
dark enough) and remains undamaged when nearby lightning occurs.
Meanwhile that thousands of volts is now only tens of volts. It may be
large voltages, but the current generated by such fields is so trivial
as to barely glow the NE-2. Current does transistor damage. As soon
as a connection to earth exists, that thousands of volts drops to zero
- because current generated from nearby fields is that tiny.

Protection inside appliances makes this trivial current transient
irrelevant since even the neon bulb can 'load' that thousand volts to
near zero. Earthing for direct strikes also makes transient currents
irrelevant.

Figure demonstrates how a fax machine could be damaged because
telephone line and AC electric are earthed at separate locations - too
far apart - at:
http://www.epri-peac.com/tutorials/sol01tut.html
(Yes, text does not discuss this too far apart earthing problem. But
that fax machine can be damaged by such improper earthing.)

Concepts are well documented in IEEE Transactions on Electromagnetic
Compatibility on 4 Nov 1998 by Montandon and Rubinstein entitled "Some
Observations on the Protection of Buildings Against the Induced Effects
of Lightning". They describe additional requirement so that induced
transients -from a direct strike - are made irrelevant.

Notice Montandon and Rubinstein do not just connect things to ground.
How earthing is routed is critical. Notice that Montandon, et al's
figure 9 demonstrates how improper earthing can create interior
electronics damage.

All incoming wires must connect to a single point ground - typically
an earth ground rod only required post-1990. Every earthing connection
(from every wire inside the telephone cable, from all three AC electric
wires, etc) must connect 'less than 10 feet' to that earthing. Even
sharp wire bends will compromise protection. If earthing is not
sufficient, then a single point earth ground must be expanded -
enhanced - so that equipotential is enhanced beneath a structure.

Demonstrated how from one radio station:
http://scott-inc.com/html/ufer.htm

So what is that telescope attached to so as to be stable? Chances
are that telescope is another earth ground thereby violating a
building's single point earthing requirement. Telescope and outside
earth ground rod being same as a cow's fore and hind legs. Concrete
being an excellent conductor when discussing lightning.

If earthing is not sufficient, then a halo ground may be necessary.
Montanon's figure 9 demonstrates same problem also detailed by
Polyphaser in:
http://www.polyphaser.com/ppc_TD1026.aspx

Of course Ufer grounding has long been an ideal solution.

No way around concepts so well proven almost 100 years ago by
telephone systems. Nearby fields do not cause the damage. Bad
earthing means a nearby lightning strike is actually a direct strike to
electronics inside the building - just like the dead cow. Protection
already inside electronics further makes such field induced threats
trivial. Only a direct strike can overwhelm that internal appliance
protection.

Meanwhile, if fields from a nearby strike were so destructive, then
every car and portable radio - antenna connected to a most easily
damaged RF transistor - would always be damaged. Nearby strikes don't
damage those radios because protection inside electronics makes trivial
fields irrelevant.

The original poster is again reminded what is a most critical feature
in transient protection systems: earth ground. Not some protector
that avoids mentioning earth. Earth ground is the most critical
component in every protection 'system' - so that a dead cow example is
made irrelevant. Any solution (lightning rod or surge protector) that
avoids essential earthing is not effective protection.

Chris L Peterson

unread,
Apr 30, 2006, 6:35:47 PM4/30/06
to
On 30 Apr 2006 12:06:37 -0700, "w_tom" <w_t...@usa.net> wrote:

> "The cow died because she's got two pairs of feet a few meters away
>from one another" was exactly the point. Chris L Peterson demonstrated
>why improper earthing can cause damage. He probably has multiple earth
>ground - items he assumed were non-conductive. Therefore lightning
>came out of the ground - just like the cow - went through electronics
>and then back into earth. Table, wall paint, linoleum tile, and wooden
>walls are all conductors. Some better than others. It all explains
>why a single point ground must be the best earth ground for that
>structure.

Not even close. The damage was clearly caused by EM. It was only present
in the interface chips at the end of cables.

I've seen what paint and other materials look like after lightning
passes through them. There was no indication at all that anything like
this happened. The structure itself wasn't hit, and the way it's built
up on posts doesn't provide much opportunity for ground currents to pass
through. No amount of grounding in the world could have helped... the
instruments that were damaged weren't even plugged in!

If you don't believe that the EM pulse from lightning can damage
electronics, you simply don't know much about the subject.

Dan Mckenna

unread,
Apr 30, 2006, 7:19:50 PM4/30/06
to

I learned my lesson,

I had a printer, unplugged from the power with power cable removed and
with the parallel port connector unplugged from the computer but still
attached to the printer, fry when a lightning strike occurred about 1/2
mile away. EMP for sure.

An employee was leaning against a railing and was zapped by a nearby
strike to a tree. Ground currents.

Saw the dead bulls in Texas (McDonald Obs) and it was ground currents.

two..two two phenomena in one.

I ground my dome, use fiber optics whenever I can and have a building
three phase industrial surge protector as well as disconnect as much
copper as practical.


I do believe in spooks


Oh NO here comes summer in Arizona !
Please rain early ( after the mirror is pulled for recoat)

Hope the mountain does not burn down as well

d.


w_tom

unread,
Apr 30, 2006, 9:48:47 PM4/30/06
to
Lightning struck a lightning rod. That lightning rod was connected
to earth via an exterior wire that was less than 4 feet from a computer
inside. Computer never even hiccupped. If a direct strike created a
destructive EM field, then that computer should have, at minimum,
crashed. According to Chris's theory, that computer should have been
destroyed. But it kept working just fine.

But again, if one knows that a lightning induced field is so
destructive, then numbers for that electromagnetic field or numbers for
currents induced on wires can be provided. Numbers essential to know.
Again, without both experimental and theoretical evidence, well,
explain that lighting strike only four feet from a computer? Again,
numbers from that 100' longwire antenna demonstrate that current is
trivial - on the order of milliamps. A six foot peripheral cable
acting as an inferior antenna will be more destructive than a 100' long
wire antenna? Those numbers even exist in IEEE papers. A 'neon lamp
on long wire antenna' example demonstrates the problem. All those
portable radios, those dimmer switches and smoke detectors connected to
much longer (AC electric) wires - all those undamaged electronics again
demonstrates why EM fields just don't explain printer damage.

You stated you know what "paint and other materials look like after
lightning passes through them". Does that mean lightning does that
damage every time? Well again, we consult science studies. Alan
Taylor of the Forestry Service demonstrated that well over 90% of all
trees directly struck by lightning leave no appreciable indication.
Why then would every conductive material indicate damage when carrying
destructive transients? Trees directly struck did not. But again,
that paint damage was only speculation because damage was seen only at
one strike location.

Printer interface chip was not the only item carrying that current.
Remember, first a transient conducts through everything in a circuit.
Only after, is something in that path damaged. Where are all those
other damaged parts - or maybe these transient currents can pass
through many items without indicated damage? IOW it was not just table
top or baseboard heater that carries a transient without damage. That
transient also ran through other items inside the printer - without
damage.

What was that peripheral cable touching when lightning struck? The
desktop? Something behind that table? Fine. Lightning may even pass
through wire insulation. Another direct connection that lightning may
find: from relay coil to relay wiper. How can this be if the wiper and
coil inside a relay as electrically isolated? Do the numbers. Things
you might consider non-conductive can become conductive. Little
surprises are why so many instead assume - yes, only speculate - that
EM fields created by nearby lightning caused damage.

Again, what some assume to be insulators, instead, become
conductors. Why did Orange County stop damage by carefully earthing
and re-earthing everything? Because earthing eliminates nearby strike
damage, as even Polyphaser demonstrates in their app note:
> Lightning strikes somewhere across the street close to the below
> grade West cable vault. ...
This nearby strike is a direct strike to electronics inside a
building.

So what is that building's floor? What is touching the floor?
These can be electrically conductive when discussing damage from
lightning - and why inferior earthing would cause selective damage
inside a building - as even discussed in the IEEE paper from Montandon
and Rubinstein.

Dead cow example demonstrates why damage to interior electronics
occurs. Removing a ground to reduce damage again suggests damage is
due to bad earthing. And finally, where are numbers for that EM field?
Protection that is standard inside electronics - to even make static
electricity non-destructive - and building wide protection makes such
fields irrelevant. Too many reasons added to everything Chris has
posted keep saying EM fields did not create that damage.

Lightning traveled less than four feet away from a computer. That
computer did not even hiccup. Why? Maximum fields only feet from a
computer did no damage - because EM fields don't create that damage.
EM fields created damage without even supporting facts and numbers is
only speculation. Meanwhile, everything posted suggests damaged
occurred for same reasons that lightning killed that cow. Apparently
that building is in desperate need of a minimally acceptable halo
ground - and probably other bonding correction. It is routine to
suffer direct and nearby strikes and not suffier damage - if the
building has been properly constructed. Damage from nearby lightning
only suggests human failure. Such damage is routinely averted where
the earthing system was properly constructed and connected to.
Lightning protection was well understood even long before WWII and
transistors.

Chris L Peterson wrote:
> Not even close. The damage was clearly caused by EM. It was only present
> in the interface chips at the end of cables.
>
> I've seen what paint and other materials look like after lightning
> passes through them. There was no indication at all that anything like
> this happened. The structure itself wasn't hit, and the way it's built
> up on posts doesn't provide much opportunity for ground currents to pass
> through. No amount of grounding in the world could have helped... the
> instruments that were damaged weren't even plugged in!
>
> If you don't believe that the EM pulse from lightning can damage
> electronics,

Let's not forget who is citing IEEE papers, citing numbers, citing
concpets from junior high school science, and even demonstrating nearby
strikes without even a computer crash. One is using science. Another
is using speculation including this wild speculation: "you simply don't
know much about the subject." I guess that proves Chris knows better?
I was simply wasting time reading IEEE papers, building effective
protection, and routinely not suffering damage from nearby lightning..

Chris L Peterson

unread,
Apr 30, 2006, 11:57:14 PM4/30/06
to
On 30 Apr 2006 18:48:47 -0700, "w_tom" <w_t...@usa.net> wrote:

> Let's not forget who is citing IEEE papers, citing numbers, citing
>concpets from junior high school science, and even demonstrating nearby
>strikes without even a computer crash. One is using science. Another
>is using speculation including this wild speculation: "you simply don't
>know much about the subject." I guess that proves Chris knows better?
>I was simply wasting time reading IEEE papers, building effective
>protection, and routinely not suffering damage from nearby lightning..

Good for you. I wish you continued luck, because based on your beliefs,
luck is all that has saved you.

Pierre Vandevennne

unread,
May 1, 2006, 12:45:40 AM5/1/06
to
Chris L Peterson <c...@alumni.caltech.edu> wrote in
news:bm1b52hi1g6455t6v...@4ax.com:

> On 30 Apr 2006 18:48:47 -0700, "w_tom" <w_t...@usa.net> wrote:
>
>> Let's not forget who is citing IEEE papers, citing numbers, citing
>>concpets from junior high school science, and even demonstrating nearby
>>strikes without even a computer crash. One is using science. Another
>>is using speculation including this wild speculation: "you simply don't
>>know much about the subject." I guess that proves Chris knows better?
>>I was simply wasting time reading IEEE papers, building effective
>>protection, and routinely not suffering damage from nearby lightning..
>
> Good for you. I wish you continued luck, because based on your beliefs,
> luck is all that has saved you.

About 10 years ago, when I was still running my company from home,
lightning stuck an oak tree in the garden, about 15 meters away from the
house. I had something 7 personal computers running, and a couple of
"server class" machines protected by grounded UPSes. 3 of the PC rebooted
with no damage. The last 4 could have glitched but didn't reboot. A couple
of mechanical multisync monitors went into "safety" but suffered no damage.
The only piece of hardware that died that day was a 1GB IBM hard drive in
one of the protected servers. The backup had completed 30 minutes before
that.

Since that day, I firmly believe luck and good backups are all I need.

njrusty

unread,
May 1, 2006, 8:11:22 AM5/1/06
to
I wish to thank everyone for their insight and opinions regarding the
blowing of IC's in my LX200. I agree with some of the earlier comments
that it was most likely EMP which damaged the chips AGAIN, which has
destroyed my ability of moving the scope in N/S/E/W.

I believe I have a major problem with the scope setup in that every
year I blow those chips. My observatory has computer equipment,
weather station and remote cameras running 24/7 and still its the LX200
(powered off) that gets the damage. Strange that the EMP does not
effect anything else.

Russ

Chris L Peterson

unread,
May 1, 2006, 11:47:55 AM5/1/06
to

Hi Russ-

Just to summarize, these are the primary ways that lightning can cause
you problems:

1) Direct strike. This is very bad, of course, but fortunately rare with
a small structure. If lightning hits your observatory, you will get
large currents flowing through materials that you would ordinarily
consider non-conductive. That means you get a lot of heat, fires can
start, and the current may end up arcing between interior points. It can
also easily end up in wiring, and will not be stopped by things like
open switches. Ordinary power line protection devices and surge
suppressors will not help. There is really only one way to protect
yourself from a direct strike, and that is to employ lightning rods on
your observatory. Lightning rods will very effectively ensure that you
don't end up with the energy of the strike being dissipated in your
structure and its contents.

2) EMP (and atmospheric transients). This is the result of the rapidly
changing magnetic field that occurs when the lightning channel breaks
down. At that point, you have a current flow of 100 kA/uS or more. The
resulting magnetic field can induce large currents in nearby conductors.
The return stroke itself can directly produce an electric field on the
order of tens of kV within 100m of the strike. The effect these fields
will have depends on your electronics. Interconnect cables are the
biggest problem, since they act as antennas. The most common electronic
failure mode will be the destruction of input protection diodes on
interface chips. While this does destroy the chips, it also provides a
current return path that usually prevents other chips in the device from
being damaged. Also important are details of PC board layout- internal
routes may act as antennas sometimes. This explains why some devices
regularly experience failures and others are more robust. Many modern
devices are sufficiently small that no dangerous voltages are induced
internally. Metal enclosures (especially when grounded) are fairly
effective at shielding the contents from the E field; protection from
the B field is more difficult. The effects of the EMP are mitigated by
reducing cable lengths and employing common mode filtering. Really solid
management of EMP requires complex electrical and magnetic shielding,
and is generally outside of what is practical in an amateur observatory.

3) Power surge. A direct strike on a power or phone line, or current
induced on those lines by the EMP can cause voltage transients capable
of damaging electronics. Such surges dissipate quickly, and are more
likely to cause non-fatal damage (a computer reboot, not damaged
components). Surges are low energy, and won't make it past power
switches. Good quality surge suppressors and UPS devices are generally
effective at managing power line surges.

4) Ground potential rise. This is the result of having different ground
potentials between points with ohmic corrections. It is the most common
cause of damage in cell telephone towers. It can be completely
eliminated by using an isolated signal path- fiber optics or radio. If
ohmic signals run into the structure, GPR damage is mitigated by using a
single point ground. That is, you only ground one end or the other of
the run. If both ends of the run are grounded independently to earth,
the GPR shows up on the signal wires, and is likely higher than the
breakdown voltage of the attached equipment.

Unfortunately, dealing with all these problems involves conflicting
strategies. For instance, you would like your structure well grounded
locally to provide current return paths outside your electronics, but
that can aggravate GPR. There are no generic lightning protection
solutions- each case has to be handled individually. What's the best
solution for a small observatory? I'd suggest:

1. A steel roof, which affords both electrical and magnetic shielding.
2. Isolated data runs (no phone or network wiring, only fiber or radio).
3. A good local ground utilizing a buried copper rod.
4. Good surge suppressors on the incoming power line.
5. Good cable management between instruments.
6. Instruments in metal enclosures when possible.
7. No lightning rods- a direct hit is unlikely, and a lightning rod will
bring what would otherwise by a near strike directly to the structure,
increasing EMP problems.
8. In a severe storm, isolate your instruments not only from the power
lines, but also from each other. Do this at both ends- completely remove
interconnect cables.

w_tom

unread,
May 1, 2006, 11:51:38 PM5/1/06
to
If those UPSes were plug-in, then they did not have effective
earthing. Repeated references to 'less than 10 feet', no sharp bends,
and impedance demonstrate why plug-in protectors don't provide
effective protection.. Plug-in UPSes don't even claim to protect from
surges that typically cause damage. So that you remain unaware, they
even avoid all discussion about earthing and don't even list numbers
for each type of transient in numerical specifications.

Notice brand names on effective protectors such as Square D, Siemens,
Polyphaser, Intermatic, Cutler Hammer, Levitoin, and GE. Effective
protectors also have that dedicated wire to earth ground for technical
reasons cited previously.

Those computers on UPSes should not have crashed or stop running.
Loss of electricity - the purpose of a UPS - should have left computers
running without interruption. Backup is also useful and always
recommended since failures occur for various reasons. But effective
and properly earthed transient protector typically costs about $1 per
protected appliance. Properly earthed 'whole house' protectors are not
only vastly more effective; such devices typically costs tens of times
less money.

Again, if the UPS is plug-in type, then the ground is only safety
ground - not earth ground. Its own specs don't even claim protection
for typically destructive transients. This for engineering reasons
only briefly described in previous posts - especially the all so
important earth ground.

w_tom

unread,
May 2, 2006, 12:08:45 AM5/2/06
to
Please explain why you would believe EMP did damage when even numbers
demonstrate such damage is not possible. Cited were professional
citaions, numbers, and examples how such damage occurs. I am totally
baffled why you think EMP did the damage when Chris only cited
speculation?

If EMP did damage, then you have damage everywhere. Instead, you had
selective damage which is consistent with what was previously posted
(including the dead cow example). Why then do you still beleive EMP
damage exists?

Please explain why you believe one who could not even explain basic
electrical concepts; who also did not have damage to everything -
including car radio - and still declared it EMP damage? Why would
you ignore one who even built and installed effective protection - and
stopped future damage? I am at a complete loss why you ignored posts
that provided layers of supporting facts, professional citations, and
even well proven solutions. Why did you ignore all that? This is a
truly honest question. I am baffled why you would agree with one who
cannot even explain why EMP does not even blow out an NE-2 glow lamp.
Believe me. I am totally baffled. Why do you still believe EMP can
create such damage?

Pierre Vandevennne

unread,
May 2, 2006, 1:44:33 AM5/2/06
to
"w_tom" <w_t...@usa.net> wrote in
news:1146541898.0...@i40g2000cwc.googlegroups.com:

> Again, if the UPS is plug-in type, then the ground is only safety
> ground - not earth ground. Its own specs don't even claim protection

This is getting a bit too complex for me, I am afraid :-). Basically the
whole house's circuit was grounded by copper earth loop in its foundations
(this was a brick house), there was a place where its resistance could be
tested (must be less than 100 Ohms by regulation around here) and there was
a bunch of differential breakers for each sub-circuit.

Chris L Peterson

unread,
May 2, 2006, 2:17:45 AM5/2/06
to
On 1 May 2006 21:08:45 -0700, "w_tom" <w_t...@usa.net> wrote:

> Please explain why you would believe EMP did damage when even numbers
>demonstrate such damage is not possible. Cited were professional
>citaions, numbers, and examples how such damage occurs. I am totally
>baffled why you think EMP did the damage when Chris only cited
>speculation?

Some interesting reading:

http://ewh.ieee.org/soc/pes/lpdl/archive/coax_cable.pdf
http://www.lightningsafety.com/nlsi_lhm/lpts.html
http://www.lightningtech.com/f_sets/laboratory.html
http://www.ursi.org/Proceedings/ProcGA05/pdf/E04.6(0885).pdf
http://www.phoenixcon.com/info/articles/PGT_McCurdyREPRINT_FINAL.pdf
http://64.70.157.146/archive/keitharmstrong/systems_installations5.html#5.2

All of these documents (and thousands more if you want to Google
"lightning lemp" discuss the nature of indirect lightning damage via
high electric fields and/or magnetic fields. Consistent in all is the
idea that lightning rods and building grounding do not provide adequate
protection to electronic equipment in the building because of indirect
effects. Given the relative ease of protecting power lines, it appears
that the primary source of damage to electronics from lightning is by
currents induced by the electric or magnetic fields, including the
fields from cloud-to-cloud lightning several kilometers distant.


> If EMP did damage, then you have damage everywhere. Instead, you had
>selective damage which is consistent with what was previously posted
>(including the dead cow example). Why then do you still beleive EMP
>damage exists?

Electronic failures are complex. There is no reason to expect that if a
given piece of equipment is damaged by induced currents, all equipment
must be. Most damage occurs on the chips directly attached to long
interconnect wiring, and the energies are generally low enough that
damage does not propagate into the circuit past the interface chips.
Input protection diodes tend to fail as shorts, destroying the device
but also saving downstream devices. Small circuits and well shielded
circuits are less prone to damage. Furthermore, different interface
chips provide different degrees of ESD protection. For instance, it is
well known that the serial interface line driver/receiver used in the
classic LX200 is prone to transient induced damage.

The bottom line is that you can play all the grounding tricks you want
on the power mains and building structure, but any electronic equipment
inside that building is still very likely to be damaged by a nearby
lightning strike. There is a reason that most southwestern (U.S.)
observatories shut down during the summer monsoon season, disconnect
their instruments (not just from power, but from each other), and store
the most delicate devices in Faraday cages.

w_tom

unread,
May 2, 2006, 7:28:51 PM5/2/06
to
Most of what was posted is found in basic elementary school science
concepts of electricity and from what Franklin lightning rods also do.
Earth ground is the protection. That is simple. Forget all myths that
promote protectors as protection. The earth ground - wire inside the
foundation - is protection.

Now, how many feet from a UPS to where wire enters the foundation?
Note repeated references to 'less than 10 feet'. Why less than 10
feet? Because wire has impedance - a concept taught in to first year
EE students. But you need not know what impedance is. You know
impedance is too high when wire length is too long, has sharp bends,
has splices, is inside a metallic conduit, etc. If that UPS does not
have a dedicated connection to where wire enters foundation, then UPS
only has a human safety ground; is not sufficient earthed for
transistor safety. Nothing complex about it. Protector is nothing
more than a connection to foundation earthing. A connection that must
be 'less than 10 feet'. What is complex about 10 feet? It is that
dumb simple when we forget myths promoted by plug-in power strip and
UPS protectors.

If you are trying to reconcile effective protection techniques with
myths from plug-in protectors, then you will be confused. Protection -
as even taught in elementary school science about lightning rods - is
earthing.

Building wide UPS systems with that dedicated earthing provides
effective transient protection. A plug-in protector cannot. That UPS
power cord is 6 feet. Is the wall outlet connected less than 4 feet to
where the wire enters foundation?

100 ohms resistance does not report sufficient earthing. High
resistance can report defective earthing. At 100 ohms for a 'wire loop
inside the foundation' (called an Ufer ground) - earthing may have
failed or was not properly installed. 100 ohms resistance is not even
considered sufficient for human safety - let alone for transistor
safety.

Even if that Ufer ground is sufficient, a plug-in UPS is not properly
earthed. If for no other reason - how long is that connection from UPS
protector to where wire enters foundation? It must be 'less than 10
feet' - which is why 'whole house' protectors from responsible
manufactures such as GE, Polyphaser, Siemens, Leviton, Cutler Hammer,
Intermatic, and GE are effective. What makes them effective? Quality
of and connection to that Ufer ground. Ufer ground is the protection.
A protector is nothing more than a connecting device to protection.
It's not complex. Earthing is also why Ben Franklin's lightning rods
work. Note repeated references to a simple concept - 'less than 10
feet'.

w_tom

unread,
May 2, 2006, 9:13:25 PM5/2/06
to
Chris L Peterson - did you even read those papers (and company
promotion brochures)before you posted them? Let's review your very
first citation:
http://ewh.ieee.org/soc/pes/lpdl/archive/coax_cable.pdf
First document does not "discuss the nature of indirect lightning
damage via high electric fields and/or magnetic fields." It discusses
damage due to electricity flowing in earth - another example of why the
cow (in my previous post) was killed.

There is no EMP field damage in your first paper. Do you assume the
word 'induced' means electromagnetic fields transfered to a wire?
Induced means electric currents flow inside earth and adjacent to a
buried wire. Therefore electric currents also flow inside that wire.
This 'induced current' is also why protectors earthed via a wall
receptacle safety ground prong, can create desctrutive transients
elsewhere in the builing.. There is nothing about airborne
electromagnetic fields in your first citation. There are transients
created by lightning (electricity) flowing through the earth - what you
described as GPR - why the cow was killed - and why earthborne
currents can also cause damage inside a building.

Notice why currents got inside the test buildings in that paper.
Wire entered those building by violating - and again - 'single point
earth ground' requirements. That 'single point earthing' violation was
intentional to create and measure what could have been desctructive
currents. If a building ,such as an observatory, is not properly
earthed, then electric currents from earth can damage electronics -
just like the dead cow. There are no airborne electromagnetic waves -
as you claim - in that paper. Did you read the paper? Induced
transient was due to electricity flowing through the earth. Damage
inside the building exactly as I described when building is not
properly earthed.

How do we routinely eliminate damage? We do what your observatory
apparently does not have - single point earth ground. The entire
building cannot be a faraday change when even the earthing is
defective. Therefore - as demonstrated by this your first citation -
this is also how electronics got damage - by electricity flowing
through earth.

Who defectively built an observatory without an Ufer ground? How did
they also fix same problems inside Orange County emergency response
buildings where FL lightning was always problematic? Single point
earth ground - ie halo grounds and other bonding. GPR problems
eliminated using techniques well proven before WWII.

Chris - your damage is classic of a building that has multiple earth
ground points - just like the dead cow example. Damage as even
demonstrated how by your own first paper. EMP is not cited in your
paper; nor damaged electronics. Damage because a building has
defective earthing - ie multiple earth grounds. Damage eliminated, for
example, by installing what even dairy barns sometimes install to
restore milk production - a halo ground.

From your very first citation:
> It is shown that the induced current in the cable shield can
> reach values of about 100 A for close stroke locations
> (within 100 m).
Which is exactly why the cow near a lightning struck tree was
therefore directly struck by lightning. How could cow have been saved?
Halo ground buried beneath and around cow.

From your own citation:
> This can be explained by considering that the horizontal electric
> field along the buried cable that excites the line can be approximately
> related to the magnetic field through the surface impedance.
That described electricity flowing through the earth. Why did you
confuse earthborne electricity with EMP? Problem and solution is
exactly what Polyphaser described in their application notes at:
http://www.polyphaser.com/ppc_TD1026.aspx


>> Lightning strikes somewhere across the street close to
>> the below grade West cable vault. ...

Electronics damage has been studies and learnedy by me for decades. I
am not suddenly seeking posts for my beliefs. I even repaired
electronics by tracing how damage occurs. Electromagnetic fields don't
damage any properly designed equipment. If damage occured as you have
claimed, then portable radios - even those inside cars - are routinely
damaged by such fields. You are only speculating. Now you are
desperately seeking citations for those speculations. One after
another, your citations don't even claim what you have speculated. In
fact many again cite the single point earth ground as essential.

Your other citations debunked in other posts. Electronic damage is
quite simple once the basics are first learned - including proetction
already inside all electronics as even defined by IEC and other
standards.

You, of course, know of the 2000 and 20,000 volt protection already
insde interface chips? Those who need proof should visit manfacturer
datasheets from interface IC manufacturers such as Maxim Semiconductor.
Yes that is 20K volts standard protection. Check out the data sheets
yourself to again understand why EM field damage does not exist.

Consistent is that even your own citations do not support your claims
about airborne fields causing damage.

Chris L Peterson wrote:
> Some interesting reading:
>
> http://ewh.ieee.org/soc/pes/lpdl/archive/coax_cable.pdf
> http://www.lightningsafety.com/nlsi_lhm/lpts.html
> http://www.lightningtech.com/f_sets/laboratory.html
> http://www.ursi.org/Proceedings/ProcGA05/pdf/E04.6(0885).pdf
> http://www.phoenixcon.com/info/articles/PGT_McCurdyREPRINT_FINAL.pdf
> http://64.70.157.146/archive/keitharmstrong/systems_installations5.html#5.2
>
> All of these documents (and thousands more if you want to Google
> "lightning lemp" discuss the nature of indirect lightning damage via
> high electric fields and/or magnetic fields. Consistent in all is the
> idea that lightning rods and building grounding do not provide adequate
> protection to electronic equipment in the building because of indirect

> effects. ...

w_tom

unread,
May 2, 2006, 9:56:47 PM5/2/06
to
Again, did you read your second citation before claiming this as an
example of EMP damage? Lightningsafety.com does not make the claims
you have erroneously made about airborne electromagnetic fields. In
fact, you own second citation:
http://www.lightningsafety.com/nlsi_lhm/lpts.html
describes what is necessary to avoid damage:
> The grounding system must address low earth impedance as well as
> low resistance. .... A single point grounding system is achieved when
> all equipment within the structure(s) are connected to a master bus
> bar which in turn is bonded to the external grounding system at one
> point only. Earth loops and differential rise times must be avoided.

Why single point ground? Why avoid ground loops created by multiple
building grounds, etc? Because they never define damage from
electromagnetic fields. Damage not even defined in numbers. The
National Lightning Safety Institute teaches how to protect humans and
even electronics. What do they discuss? Earthing. They don't even
mention EMP in that citation that Chris Peterson claims demonstrates
EMP damage.

Did you even read your second citation before posting it? Or do you
assume others will not read it? Your second citation - like your first
citation - only describes what I have defined to eliminate observatory
damage.

Your third citation at:
http://www.lightningtech.com/f_sets/laboratory.html
describes various factors in lightning, provide no numbers, AND are
selling a two day course and company products. Yes, a 100' long wire
antenna can have hundreds or even thousand of volts during a nearby
strike. How do we reduce that voltage to near zero? How do we make
that 'induced' transient irrelevant? We connect a NE-2 neon glow lamp
(a tiny indicator bulb found inside lighted switches) to that antenna.
Now that thousands of voltage becomes near zero because we have
conducted some milliamps. So yes, there is the protection from nearby
fields - and it is built into all electronics. Your third citation is
selling a two day course on these simple protection methods - that is
already inside properly designed electronics. This is to you an IEEE
paper? They don't even provide useful numbers? It is a sales brochure
- not a technical citation. Why do you cite an advertisement as a
technical paper? Did you read your third citation before posting it?

Consistent is the idea that earthing do provide adequate protection
to electronic equipment in the building. Consistent is that EMP is
never defined as a typical source of damage. Consistent is that Chris
Peterson's own citations do not even support his assumptions.

Even his very first citation in another post demosntrated why earthing
- the single point ground - is necessary for protection. From his own
testimony - that he removed a ground to obtain better protection - that
is classic of a building whose earthing is inadequate AND suffers from
multiple point grounding. Nearby lightning strikes to earth then can
cause electroncs damage inside that building. Even his own first
citation demosntrates why he can have damage AND why the dead cow
example demonstrated same.

Chris L Peterson wrote:
> Some interesting reading:
>
> http://ewh.ieee.org/soc/pes/lpdl/archive/coax_cable.pdf
> http://www.lightningsafety.com/nlsi_lhm/lpts.html
> http://www.lightningtech.com/f_sets/laboratory.html
> http://www.ursi.org/Proceedings/ProcGA05/pdf/E04.6(0885).pdf
> http://www.phoenixcon.com/info/articles/PGT_McCurdyREPRINT_FINAL.pdf
> http://64.70.157.146/archive/keitharmstrong/systems_installations5.html#5.2
>
> All of these documents (and thousands more if you want to Google
> "lightning lemp" discuss the nature of indirect lightning damage via
> high electric fields and/or magnetic fields. Consistent in all is the
> idea that lightning rods and building grounding do not provide adequate
> protection to electronic equipment in the building because of indirect
> effects. Given the relative ease of protecting power lines, it appears
> that the primary source of damage to electronics from lightning is by
> currents induced by the electric or magnetic fields, including the
> fields from cloud-to-cloud lightning several kilometers distant.

> ...

w_tom

unread,
May 2, 2006, 10:36:55 PM5/2/06
to
The fifth citation is an introductory primer on devices and numbers
in protection. It does not discuss 'system' solutions AND does not
demonstrate that electromagnetic waves cause damage. It does - as
required in any protection solution - discuss earthing:
> One of the key factors to protect any sensitive industrial
> equipment is a low impedance earth ground.

Again nothing discusses EMP damage and EMP protection. It's just
not the typical source of damage. As a primer, it does not provide
sufficient information to create protection. Did you read this? This
is not a citation. Chris Peterson's fifth citaion is a primer -
introduces defintions to first time learners. And it does not even
define the assumed damage from EMP. It does not claim what Chris says
it does. Again, did you read it before posting it?

The sixth citation is a more compreehensive primer that introduces
system level considerations. It repeatedly defines various aspect of
the single point ground. For example:
> SPD protection can also be defeated by the 'earth lift' across
> a structure, but the measures described earlier should deal
> with this for SPDs located at equipotential bonding plates and
> METs.
Equipotential bonding and Main Earthing Terminal? Again
characteristics of single point earthing. Meanwhile EMP has nothing to
do with this citation nor with solutions to avoid damage.

> But when applying SPDs to items of equipment, SPD surge
> currents can cause a significant 'earth-lift' at the equipment
> itself due to the inductance of its protective bonding
> conductor - unless the equipment's chassis is bonded directly
> to a MESH-CBN.
Again discussing problems due to direct electrical transients and not
from electromagnetic fileds. The CBN - common bonding network - just
another example of a single point ground.

And so we have this sixth and final citaiton - and it still does not
define electromagnetic fileds as destructive. Again, I ask if these
citations were read before being cited. They don't support speculation


"of indirect lightning damage via high electric fields and/or magnetic
fields."

The most critical and essential feature of a building wide protection
system is even defined by Chris Peterson's own sixth citation:

http://64.70.157.146/archive/keitharmstrong/systems_installations5.html#5.2
that demonstrates fundamental earthing techniques including bulkhead
and mesh techniques. But again, its about earthing - keeping those
electrical currents outside of the facitlity.

Chris L Peterson wrote:
> Some interesting reading:
>
> http://ewh.ieee.org/soc/pes/lpdl/archive/coax_cable.pdf
> http://www.lightningsafety.com/nlsi_lhm/lpts.html
> http://www.lightningtech.com/f_sets/laboratory.html
> http://www.ursi.org/Proceedings/ProcGA05/pdf/E04.6(0885).pdf
> http://www.phoenixcon.com/info/articles/PGT_McCurdyREPRINT_FINAL.pdf
> http://64.70.157.146/archive/keitharmstrong/systems_installations5.html#5.2
>
> All of these documents (and thousands more if you want to Google
> "lightning lemp" discuss the nature of indirect lightning damage via
> high electric fields and/or magnetic fields. Consistent in all is the
> idea that lightning rods and building grounding do not provide adequate
> protection to electronic equipment in the building because of indirect
> effects. Given the relative ease of protecting power lines, it appears
> that the primary source of damage to electronics from lightning is by
> currents induced by the electric or magnetic fields, including the
> fields from cloud-to-cloud lightning several kilometers distant.

> ...

Chris L Peterson

unread,
May 2, 2006, 11:44:01 PM5/2/06
to
On 2 May 2006 18:13:25 -0700, "w_tom" <w_t...@usa.net> wrote:

> Chris L Peterson - did you even read those papers (and company
>promotion brochures)before you posted them? Let's review your very
>first citation:
> http://ewh.ieee.org/soc/pes/lpdl/archive/coax_cable.pdf
>First document does not "discuss the nature of indirect lightning
>damage via high electric fields and/or magnetic fields." It discusses
>damage due to electricity flowing in earth - another example of why the
>cow (in my previous post) was killed.

Amongst other things, the paper discusses "the correlation between the
horizontal magnetic field and the induced current".

And the other references discuss similar effects, and are all quite
clear on the definition of "induced currents".

You can believe what you want. If you think that lightning can't
inductively produce damaging currents, both by its magnetic field and
its electric field, you are definitely taking a minority position
(possibly a minority of one).

This is a science newsgroup. You should probably take your ideas to a
newsgroup just for others who got their physics degrees from a matchbook
coupon.

Pierre Vandevennne

unread,
May 3, 2006, 6:39:30 PM5/3/06
to
"w_tom" <w_t...@usa.net> wrote in
news:1146612531.2...@e56g2000cwe.googlegroups.com:

> It's not complex. Earthing is also why Ben Franklin's lightning rods
> work. Note repeated references to a simple concept - 'less than 10
> feet'.

Does that mean that Franklin's lightning rods are unable to protect
buildings higher than ten feets?

Just a question...

w_tom

unread,
May 3, 2006, 7:02:28 PM5/3/06
to
'The paper discusses "the correlation between the horizontal magnetic
field and the induced current" ' - the first citation - is same as
when discussing fields from a commercial radio broadcast station. What
do fields in lightning and from the commercial radio station do? Both
cause noise on a radio receiver. There is nothing - no numbers - no
claims - that electromagnetic fields cause hardware damage. Not in any
of your six citations. And that is my point. You claim damage from
electromagnetic fields and yet cannot even cite a science source.
Damage from electromagnetic fields is nothing but speculation. It is
not based in fundamental theory. It is not based in experimental
observation.

Chris saw damage, could not understand why, and therefore only
speculated damage from EM fields. He only speculated AND he provides
no citations to prove his speculations.

Your own citations demonstrated how to avert damage by using what?
What do your own papers discuss to protect electronics? Earthing.
What would protect that cow from death due to a nearby lightning
strike? Halo ground. What is necessary for every building containing
high tech electronics? Ufer grounds or equivalent earthing. What
should be first inspected or installed to terminate damage due to
lightning? Exact same solution corrected in that FL Orange County
emergency response center. Earthing.

That paper provided EM wave equations for fields. Why? Communication
cables must not even suffer interference - noise - from those fields.
Damage is not even considered because such damage is made completely
irrelevant by circuits inside electronics. Again, did you read those
data sheets for interface chips from Maxim Semiconductor, et al?
Interface chips to meet IEC standard for 20,000 volts - without damage.
Don't take my word for it. Review those data sheets.

Why do we earth a building? So that protection inside all
electronics is not overwhelmed.

Damage to electronics is due to direct strikes. Lightning striking a
nearby tree was a direct strike to the dead cow. Chris Peterson
referenced this as GPR. How did we eliminate GPR problems even long
before WWII? Single point earth ground. Halo or Ufer ground should be
standard in every observatory to eliminate electronics damage. A
facility without that most basic solution - that was even standard in
telephone switching centers almost 100 years ago - is all but asking
for electronics damage due to lightning. Earthing is where protection
from lightning begins.

Dan Mckenna

unread,
May 4, 2006, 1:07:40 AM5/4/06
to

From what I gathered, Lightning is not DC where you worry about the
resistance but a complex waveform that requires a low inductance and is
treated as a R.F. problem. Some of the early work on lightning
protection came from measuring the rise time on strikes to the Empire
State Building with low bandwidth cathode ray tube oscilloscopes. This
lead to an underestimation of the current rise times and standards that
did not go far enough in specifying very low inductance paths.

In a metal building lightning rods may not offer additional protection
as the building itself is the lowest impedance structure. For a wooden
building you want to control the attachment point and a rod is used to
do this. One talk I went to on the statistics of lightning strikes in
Tucson presented a strike density plot that had a correlation length of
50 meters. The speaker concluded that even with a structure high enough
to exceed the 70 degree "protection cone" you still had a chance for a
direct strike 50 meters away from the structure. A solar telescope on
Kitt Peak uses a wire antenna that creates a protection perimeter as is
used to protect fuel storage tanks. The grounds I have used are two
types, wet grid... a mesh that is spread out with a watering system
or a set of radial grounds as high frequency current are subject to the
skin effect and a lower impedance is obtained with multiple grounds
designed a a counterpoise like used in high power low frequency
transmitting antennas.

A class of expensive fancy rods known a dissipative system some times
used radioactive sources or many fine points to "drain the field"
preventing or diminishing the strike intensity. A series of tests
debunked this class of devices and demonstrated no advantage over a
blunt rod. (New Mexico Tech ?) The E field of a thunder storm is
centered many thousands if not 10s of thousand feet in altitude and it
seems not to care about the details of the attachment. The path
inductance and current density in the grounding system does however
effect the peak voltages induced into the structure and can cause
secondary breakdowns if the impedance is two high. In addition large
strikes e.g. 250,000 amps will generate secondary discharges off of high
field regions caused by corners which cause large field gradients and
rods at the corners of a structure help to control the current path even
though they are "only" 10s of amps.

All of this is covered in the Lightning book by Uman (sp?)


d.


Pierre Vandevenne

unread,
May 4, 2006, 9:34:12 AM5/4/06
to
Thanks for the thoughful reply Dan - you got me interested in the book
but... ouch, at the price, one can only hope it has protective virtues
by itself!

http://www.amazon.com/gp/product/0521583276/

w_tom

unread,
May 5, 2006, 4:36:39 AM5/5/06
to
Lightning rods containing radioactive materials are called Early
Streamer Emmission (ESE) - a technology even discredited by the NFPA.
Lightning is not stopped, blocked, absorbed, or discharged. Lightning
will create an electrically shortest path to earthborne charges. That
shortest path previously demonstrated by up a cow's hind legs and down
via fore legs - the 'dead cow' example.

Military is even more blunt about ESE (dissipator) technology. From
a 1998 Air Force Instruction 32-1065:
> 14.1. General. Systems must comply with NFPA 780 and AFM 88-9,...
> Early streamer emission systems or charge dissipation systems are
> not permitted.

So that lightning does not find a short path to earth via
electronics, a 'whole house' protector makes a 'less than 10 foot'
connection to earth. That 'ten foot' necessary to make the earthing
connection significantly shorter than paths via electronics. So that a
lightning rod is effective, its connection to earth also must be as
short as possible. IOW if a rod is 30 feet above earth, then its
connection to earth must be significantly more conductive (electrically
shorter) than 30 feet of building (and yes, wood is an electrical
conductor).

Why worry about short conductors? The electric current called
lightning is RF electricity. Not just RF fields. Electricity is most
destructive. Wire impedance means even splices, sharp bends, wires
inside conduit, etc all adversely affect that path to earth.
Electricians who are trained in concepts of wire resistance are
therefore not familiar with wire impedance and RF nature of lightning.
What is sufficient for human safety (low resistance earthing) must be
enhanced for transistor safety (low impedance earthing).

Dan discusses a 70 degree cone of protection. Other industry
professionals define it as 60 degrees. Same concept.

If lightning finds a shorter path to earth via something outside of
that 'cone of protection', then lightning will take that path. There
is no ' "drain the field" preventing or diminishing the strike
intensity." ' Will radioactive rods (ESE) discharge something miles
above? Of course not. Lightning makes a connection to earth borne
charges. Either lightning takes a shorter path via Franklin rods or
'whole house' protector to single point earthing OR lightning takes a
destructive path via building or household electronics. Lightning
builds an electrical conductor - plasma - then discharges the cloud
electrically through that conductor. ESE devices do nothing to stop
that connection.

Wire antenna protection would be same wire found atop long distance
electric transmission towers. This 'catenary' (also called overhead
ground wire) connects lightning to earth so that lightning does not
strike any of three lower AC electric wires; a 'cone of protection'
beneath that catenary. A catenary provides protection similar to
Franklin's air terminal. The term 'counterpoise' comes from same
transmission line protection system - how that tower is earthed. I
believe counterpoise was defined first in a 1929 Westinghouse paper.

Dr Uman from U of Florida has long been considered a pioneer in
lightning protection. For example when a demonstration ESE device was
literally blown off an experiemental FAA building within days of
installation, I believe Dr Uman analyzed that failure. Other papers
such as Dr Mousa in IEEE Transactions on Power Delivery on 4 October
1998 entitled "Applicablity of Lightning Elminiation Devices to
Substatons and Power Lines" states:
> ... the suggestion that lightning can be eliminated has been
> soundly rejected by the scientific community base on an
> evaluation which was quite rigourous.
Dr Mousa then provides figures of those unacceptable ESE devices in a
paper that defines effective protection methods. Dr Mousa then states:
> 5. Charge dissipators will have no effect whatsoever on the frequency
> of lightning strikes to substations and transmision towers ...

Dan also notes a widely discussed NM experiment that demonstrated
blunt lightning rods were more effective than pointed rods. Dissipative
(ESE) devices were not tested. Meanwhile, both pointed and blunt air
terminals remain only as reliable as their earthing. A simple earthing
electrode massively increases effectiveness. Then earthing is
significantly enlarged for an even lesser earthing improvement. High
reliability facilities and AC electric substations would typically
start construction with a massive earthing mesh. This for two reasons:
to make earthing as conductive as possible AND to create equipotential.
Earthing installed when construction first begins.

Direct strikes rarely exist at and above 200,000 amps. Majority of
strikes are 25,000 amps or less. A lurker may never experience a
200,000 amp lightning strike. Therefore household protection minimally
at 50,000 amps, and further enhanced by utility provided protection,
would withstand most every direct strike without damage or degradation
for 10+ years. A life expectancy number that can vary based upon
neighborhood and regional conditions. Such protector costs maybe $1
per protected appliance.

Residential protection is not sufficient for high tech and high
reliability facilities such as telephone switching stations, commerical
broadcasting, and observatories. Therefore significant earthing must
be installed even as footings are poured. Again, concepts of
conductivity and equipotential are addressed when architect plans are
first concieved. Such solutions make GPR irrelevant. If not
installed, then at minimum, the facility requires a kludge enhancement
- halo ground or equivalent. Such facilities may even include
surrounding chaiin link fence into the earthing system.

Earthing to protect FL homes is demonstrated in pictures at:
http://members.aol.com/gfretwell/ufer.jpg
This is only residential protection. What's in your house?

We still don't do this for every home. Yet this and more was
standard where lightning damage was not acceptable in telephone
switching stations - even multiple generations before transistors
appeared. The technology is that well proven.

All solutions are dependent on diverting lightning into paths to
earth that are not destructive. That means a single point earth ground
- minimally an earthing electrode, or an Ufer ground, or a massive mesh
network. Ufer grounds developed originally for and so effective that
direct lightning strikes would even avert munition explosions. Bonding
to earth, as previously demonstrated in FL Orange Country Emergency
Response facilities, curtailed all future damage from lightning -
direct and nearby. Earthing being the most critical component in any
protection system as even demosntrated in 1920s GE and Westinghouse
papers.

Without such earthing, then facility electronics may be damaged even
by a nearby strike as demonstrated by the 'dead cow' example and by
application note from a highly regarded industry benchmark:
http://www.polyphaser.com/ppc_TD1026.aspx

0 new messages