ftp.cse.unsw.edu.au/2014.06.ftp.cse.unsw.edu.au.tar/ftp.cse.unsw.edu.au/pub/doc/misc/lightning/lore

Internet Message Format | 1993-02-23 | 10.4 KB

From usage!metro!seagoon.newcastle.edu.au!news.cs.uow.edu.au!sol.deakin.OZ.AU!munnari.oz.au!bunyip.cc.uq.oz.au!marlin.jcu.edu.au!ccarg Fri Feb  5 17:45:42 AEST 1993
Article: 1243 of aus.computers
Newsgroups: aus.computers
Path: usage!metro!seagoon.newcastle.edu.au!news.cs.uow.edu.au!sol.deakin.OZ.AU!munnari.oz.au!bunyip.cc.uq.oz.au!marlin.jcu.edu.au!ccarg
From: [e-mail] (Alan Gunders)
Subject: Lightning strikes again.
Message-ID: <[e-mail]>
Organization: James Cook University
Distribution: aus
Date: Mon, 1 Feb 93 04:03:03 GMT
Lines: 51

When we had trouble with lightning we did some investigations and found
out a few things.

    1. A spark requires 1000 volts to jump 1 c.m. or 2500 volts per inch.
    (Have a quess at what voltage is required to create a lightning strike)

    2. A single lightning strike has an average of 28,000 volt/amps in it.

    3. Once lightning has struck and an ionized path is esthablished
    other inter cloud discharges can occurr recharging the cloud,
        and then another discharge can follow the path of the first strike.
    Multiple discharges of six strikes are common.

    4.Every piece of metal within 30 Kms of a lightning discharge has
    a measurable voltage induced into it by electromagnetic radiation.
    The closer to the strike the greater the voltage.

    5.Most electrical switches have contacts less than 6 m.m. apart,
    so that it is possible to calculate the voltage required to jump
    that gap.

    6. The most common surge protector is a 350 volt gas arrestor
    which can shunt aprox 20 amps for 3 cycles of power.
    (ie: three cycles of 50 hz about .15 seconds.)

    7. Each computer is connected to a three pin plug which can be
    completely removed from the wall outlet, and even that is not totally
    safe but it is the best we can find.

We had lightning strikes that blew electronic sprinkler timers out from
under the ground as well as blowing every comm port in two buildings and
setting fire to two terminal servers.  If the hit is close enough no
amount of surge protection will be of any help.
During the storm season we disconnect all equipment we can and try to
protect the most vital equipment that we can't.

Telephone lines connected to modems are also a potential hazzard.

P.S. Optic fibre is very nice.


     JAMES COOK UNIVERSITY OF NORTH QUEENSLAND  computer centre

        ++++++++++++++               Ray Belbin     Alan Gunders
       +++++++++++++                  cctec/ccrdb    cctec/ccarg
          ++++  echnicians                @marlin.jcu.edu.au
         ++++  echnicians             077 814 773     077 815 905
        ++++  echnicians              077 814 545     077 814 545
       ++++  echnicians                     FAX [phone]




From usage!sserve!manuel.anu.edu.au!johng Fri Feb  5 17:46:56 AEST 1993
Article: 1241 of aus.computers
Path: usage!sserve!manuel.anu.edu.au!johng
From: [e-mail] (John Gallant)
Newsgroups: aus.computers
Subject: Re: Lightning strikes again.
Date: 1 Feb 1993 23:17:08 GMT
Organization: Australian National University
Lines: 81
Distribution: aus
Message-ID: <[e-mail]>
References: <[e-mail]>
Reply-To: [e-mail]
NNTP-Posting-Host: 150.203.59.4
X-Newsreader: TIN [version 1.1 PL6]


Just thoght I would clear up a few technical errors.

Alan Gunders ([e-mail]) wrote:

: When we had trouble with lightning we did some investigations and found
: out a few things.
:
:     1. A spark requires 1000 volts to jump 1 c.m. or 2500 volts per inch.
:     (Have a quess at what voltage is required to create a lightning strike)
:
:     2. A single lightning strike has an average of 28,000 volt/amps in it.

Which? They aren't the same. Its not a single value either, it lasts a
few hundred microseconds and the current at least shows an exponential
decay.

:
:     3. Once lightning has struck and an ionized path is esthablished
:     other inter cloud discharges can occurr recharging the cloud,
:         and then another discharge can follow the path of the first strike.
:     Multiple discharges of six strikes are common.
:
:     4.Every piece of metal within 30 Kms of a lightning discharge has
:     a measurable voltage induced into it by electromagnetic radiation.
:     The closer to the strike the greater the voltage.

Read "induction" for "radiation". Yep, you don't have to be actually
struck by lightning to experience problems. The earth around a lightning
strike has pretty large currents flowing in it, and pretty large
potential differences exist. I understand that is how animals get killed
by a nearby lightning strike - the potential difference between front
and back legs is enough to electrocute them.

:
:     5.Most electrical switches have contacts less than 6 m.m. apart,
:     so that it is possible to calculate the voltage required to jump
:     that gap.
:
:     6. The most common surge protector is a 350 volt gas arrestor
:     which can shunt aprox 20 amps for 3 cycles of power.
:     (ie: three cycles of 50 hz about .15 seconds.)
:

There are various levels of protection from different kinds of devices.
MOVs are faster but lower energy than gas arrestors and can be placed
after the gas arrestor to further reduce the spike.

:     7. Each computer is connected to a three pin plug which can be
:     completely removed from the wall outlet, and even that is not totally
:     safe but it is the best we can find.

Like someone else said, you can't really protect against a direct hit,
but having all the plugs out (including telephone for faxes and modems)
protects against just about everything else.

:
: We had lightning strikes that blew electronic sprinkler timers out from
: under the ground as well as blowing every comm port in two buildings and
: setting fire to two terminal servers.  If the hit is close enough no
: amount of surge protection will be of any help.
: During the storm season we disconnect all equipment we can and try to
: protect the most vital equipment that we can't.
:
: Telephone lines connected to modems are also a potential hazzard.
:
: P.S. Optic fibre is very nice.
:
:
:      JAMES COOK UNIVERSITY OF NORTH QUEENSLAND  computer centre

This guy speaks from experience. I only have the theory.

I have some more info at home from a Critec seminar, if anyone is
interested, including details of good surge protector design.

--
    John Gallant           [e-mail]  "Every \item command in
Centre for Resource and      ph: [phone]    item_list must have an
 Environmental Studies      fax: [phone]    optional argument".
Australian National University                     Leslie Lamport, LaTeX


From usage!spectrum!amiga Fri Feb  5 17:47:20 AEST 1993
Article: 1244 of aus.computers
Newsgroups: aus.computers
Path: usage!spectrum!amiga
From: [e-mail] (Amiga Utilities)
Subject: Re: Lightning strikes again.
Message-ID: <[e-mail]>
Sender: [e-mail]
Nntp-Posting-Host: plum10.spectrum.cs.unsw.oz.au
Reply-To: [e-mail] (Amiga Utilities)
Organization: University Of New South Wales (UNSW), Sydney, Australia.
References: <[e-mail]> <[e-mail]>
Distribution: aus
Date: Tue, 2 Feb 1993 02:57:05 GMT

In article <[e-mail]> [e-mail] writes:
>
>Just thoght I would clear up a few technical errors.
>

So did I :-)

>Alan Gunders ([e-mail]) wrote:
>
>: When we had trouble with lightning we did some investigations and found
>: out a few things.
>:
>:     1. A spark requires 1000 volts to jump 1 c.m. or 2500 volts per inch.
>:     (Have a quess at what voltage is required to create a lightning strike)
>:

Not quite. It's 10000 Volts/cm, or 1000 Volts/mm.

>I have some more info at home from a Critec seminar, if anyone is
>interested, including details of good surge protector design.

Let's have the surge protector design, PLEASE!!!


     ///    [e-mail]
    ///     The amiga utilities account.
\\\///      University of New South Wales
 \XX/       Sydney, Australia


From usage!sserve!csadfa.cs.adfa.oz.au!pgc Fri Feb  5 17:48:46 AEST 1993
Article: 1249 of aus.computers
Newsgroups: aus.computers
Path: usage!sserve!csadfa.cs.adfa.oz.au!pgc
From: [e-mail] (Phil Clark)
Subject: Re: Lightning strikes again.
Message-ID: <[e-mail]>
Sender: [e-mail]
Organization: Australian Defence Force Academy, Canberra, Australia
References: <[e-mail]>
Distribution: aus
Date: Wed, 3 Feb 1993 01:09:26 GMT

>From article <[e-mail]>, by [e-mail] (Alan Gunders):
> When we had trouble with lightning we did some investigations and found
> out a few things.
>
(Lots deleted)

It would appear that you did not find out quite enough, as you have included
quite a few errors.

It requires about 1kV per mm to arc in normal air.

Lightning is preceeded by a step leader which establishes the ionised path
for the strike that follows. A typical strike is about 18kA to 20kA for the
first stroke with about 1/2 this for the second and third strokes and
occasionaly others. Generally about 50% of stikes  will be at least 18kA,
about 10% will exceed 60kA and approx 1% will exceed 100kA.

Because of the rapid rise time of the stroke (pulse) inductance plays
a major role in where the current goes. A lightning strike is essentially
a constant current source until the charge is released.

The ionised path DOES NOT allow recharging. In fact the reverse is true, it
is the fact that the cloud is insulated that allows the charge to build up.

It is possible to protect against a direct strike, however this requires
special design and usually a costly installation.

When a strike hits the ground, large currents flow in the surrounding area.
These currents also have a short risetime and may be coupled to nearby
cables in the ground. Buries cables anr not necessarily protected. In addition
large potentials can be developed across quite short distances of the
surface, depending on the resistance, inductance, etc. at the time.


Phil Clark [VK1PC]         Department of Computer Science,     Phone:
                           Australian Defence Force Academy,   [phone]
Email:                     Northcott Drive, Campbell,          Fax:
[e-mail]   Canberra, Australia, 2600.          [phone]