4+ mmm as in solid copper wire that is 4 mm in diameter.
That will earth lightning and remain intact. The experiments
recently performed in AZ used blunt rods and sharp pointed
rod. The purpose of a lightning rod is to earth a direct
strike. Lightning preferred earth via blunt rods.

Delicate electronics already has thousands of volts of
internal protection. This defined in manufacturer
semiconductor data sheets, required in IEC and other
standards, and is why we divert lightning to earth so that
electronic internal protection is not overwhelmed.

How well proven is the technology? The £multi-million
telephone switching computer connects to overhead wires
everywhere in town. Do they shutdown for each thunderstorm?
Of course not. They also use same 'well proven even before
WWII' technology. Protection of electronics is that routine
and that well established in virtually every town.

Protection of electronics in that weather station is defined
by the quality of and connection to earth ground. Lightning
that does not find a path to earth via electronics does not
damage that electronics.

Dave Liquorice wrote:
Blunt as in a straight cut or blunt as in terminating in a hemisphere?
Assuming by "rod" you mean something of circular cross section.
...

Please define what you mean by "4+ mm".
...

That is what I said.
...

Delicate electronics does not take kindly to the huge induced voltages
that a lightning strike produces over several hundred yards distant.
Yes, you can build electronics that will survive but that does not
come cheap. In effect the electronics to read the sensors, convert the
data to a form suitable to be sent over a low power RF link cannot be
protected at a price the consumer is prepared to pay.
...

On Sat, 03 Dec 2005 18:43:20 -0500, w_tom wrote:

4+ mmm as in solid copper wire that is 4 mm in diameter.

So about 12mm^2 crossectional area, the normal way of specifying cable
sizes.

The experiments recently performed in AZ used blunt rods and sharp
pointed rod.

You still have not answered the direct question on the defintion of
"blunt".

Delicate electronics already has thousands of volts of internal
protection. This defined in manufacturer semiconductor data sheets,
required in IEC and other standards,

********. *Some* integrated circuits have limited protection from ESD,
without it you can't handle 'em without damage but your average signal
diode, transistor or other components do not.

Do not confuse the CE marking with lightning protection. CE marking is
purely down the abilty of the kit to reject and not produce
interference. On the rejection side the field strengths are minimal
compared to those surrounding a lightning strike.

If electronics is so well protected can you tell me why I have a dead
network card here? It died when there was a lightening strike 300
yards away.

How well proven is the technology? The £multi-million telephone
switching computer connects to overhead wires everywhere in town.
Do they shutdown for each thunderstorm?

No, but when ever there is a storm round here the number of telecoms
vans out and about increases noticeably. And again you are talking
"£multi-million telephone switching computer" (what only one?), that
is also a critcal part of the countries infra structure, not a bit of
cheap consumer electronics.

Lightning that does not find a path to earth via electronics does
not damage that electronics.

Explain my blown network card then please? The strike was several
hundred yards away with no physical connection between it and the
strike.

--
Cheers
Dave. pam is missing e-mail

I am completely befuddled why one cannot tell the difference
between a pointed verses a blunt lightning rod.

Next, what does CE - a safety standard - have in commmon
with something completely different - transient protection?
The UK standard for transient protection is (if I remember)
BS6651. Meanwhile, damage from lightning storms can occur
easeier in the UK. Other nations routinely install and
properly earth a 'whole house' protector on every incoming
phone line - for free. Same technique that also protects UK's
"£multi-million telephone switching computer" from damage
during lightning storms - with inexpensive and so effective
solutions. Phone line protection being so inexpensive and so
effective as to be standard in North America.

So, yes. Without meeting basic earthing requirements even
in BS6651, then UK residents would routinely suffer totally
unnecessary damage. A classic example of homes still not
constructed as if the transistor exists. A classic example of
electronics damage is directly traceable to human failure.

Dave Liquorice would have a dead network card (whose specs
would declare something like 2000 volts protection) because a
transient overwhelmed that 2000 volt NIC internal protection.
Just another example of why 'whole house' protection system is
installed - so that NIC protection is not overwhelmed. But
again, one need not trust me. Voltage necessary to overwhelm
NIC protection is provided on NIC's data sheet. Details
provided in that BS6651 standard. Just another example of why
one should first learn the numbers before wildly assuming
nothing can protect from lightning.

Meanwhile a classic example of semiconductors that meet (in
this case) 8,000 volt and 15,000 volt transients without
damage. But again, one need not trust me. Learn the numbers.
An example of learning numbers before wildly speculating:
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/4301

Fundamental to all this is a so well proven concept.
Protection is only as effective as its earth ground.

Dave Liquorice wrote:
...
You still have not answered the direct question on the defintion of
"blunt".
...

********. *Some* integrated circuits have limited protection from ESD,
without it you can't handle 'em without damage but your average signal
diode, transistor or other components do not.

Do not confuse the CE marking with lightning protection. CE marking is
purely down the abilty of the kit to reject and not produce
interference. On the rejection side the field strengths are minimal
compared to those surrounding a lightning strike.

If electronics is so well protected can you tell me why I have a dead
network card here? It died when there was a lightening strike 300
yards away.
...

No, but when ever there is a storm round here the number of telecoms
vans out and about increases noticeably. And again you are talking
"£multi-million telephone switching computer" (what only one?), that
is also a critcal part of the countries infra structure, not a bit of
cheap consumer electronics.
...

Explain my blown network card then please? The strike was several
hundred yards away with no physical connection between it and the
strike.

On Sun, 04 Dec 2005 18:57:33 -0500, w_tom wrote:

I am completely befuddled why one cannot tell the difference
between a pointed verses a blunt lightning rod.

Read the orginal question, then answer it.

Dave Liquorice would have a dead network card (whose specs
would declare something like 2000 volts protection) because a
transient overwhelmed that 2000 volt NIC internal protection.
Just another example of why 'whole house' protection system is
installed - so that NIC protection is not overwhelmed.

The surge was induced in a few metres of network cable and found its
way to ground through the PCs earthed chassis, after frying the
decoupling Cs and blacking the board a bit. It also took out the
network side of the connected IP camera, but without visible damage,
the camera isn't earthed. That was the only damage, the camera still
worked (on it's serial interface) and the PC suffered no damage
neither did the modem connected to the phone line and PC or the modem
connected to the camera and phone line. Nothing else in the house
suffered. Oh and remember the strike was a couple of hundred yards up
the road.

Please explain how "whole house" protection would have protected this
NIC?

--
Cheers
Dave. pam is missing e-mail

From this distance, I cannot say what was and was not the
earthing path. For example, you may consider concrete
non-conductive. But to lightning, concrete is conductive.
Same could even apply to wall that camera was mounted on.
Incoming on camera network port, through network interface IC,
then to ground via camera mounting screws.

A most typical reason for damage is incoming on AC mains,
into computer motherboard, through modem, and out to earth via
telco installed (for free) 'whole house' protector. Nothing
in motherboard is damaged even though motherboard is in that
path. A most common damage is modem's off hook relay driver
(PNP) transistor. Path that includes an electrical connection
from relay coil to relay wiper. Yes, those electrical parts
(normally considered separate) inside relay are connected
together during the destructive transient.

In your case, let's ask a few questions. That network card
is galvanically isolated. The network cable is electrically
(galvanically) isolated from NIC electronics by a
transformer. And yet that transformer still connected the
transient into NIC. Again, appreciate what one might not
consider connected is indeed connected via paths previously
unknown to the human.

Lightning is not capricious. But it does find paths
previously unknown to humans who must learn from the damage.
A classic example of 'best evidence is the dead body'.

It is possible for a TV and VCR, side by side and
interconnected, to both suffer the same lightning strike. And
yet only one is damaged - the other remains fully functional.
Why? Again, a path from incoming lightning to earth ground
must first be learned.

Even a mouse wire dangling behind desk on baseboard heater
could become, to lightning, a direct connection to earth
ground. So many paths exist which is why protection inside a
building is so complicated and often so futile. Best
protection earths transients before they can enter a building.

If we built homes to protect transistors, then rebar inside
the concrete footings would be the most critical component of
a lightning protection system. It's called Ufer grounding -
originally pioneered to protect ammunition from direct
lightning strikes. That building Ufer ground would provide
the single point earth ground and act as a halo ground; making
earth beneath the entire building equipotential. But since we
still don't plan for transistor protection, then it is
essential for all incoming utility wires to be earthed at a
single point. That earthing be the best earth ground
available. That earthing makes it unnecessary for lightning
to seek earth ground inside a building through household
appliances.

To get through the NIC, there must be a path incoming that
entered and left NIC on network cable and computer ground.
Since whole house grounding was not utilized, then lightning
overwhelmed a typically 2000 volt NIC protection.

One example of a protection system that uses multiple
earthing methods for even better protection is page 14 at:
http://www.leminstruments.com/pdf/LEGP.pdf

Do we need protection that extensive. No. Even an earth
ground rod at the service entrance provides a massive increase
in protection. High reliability facilities add all that
extra earthing just for a little better protection.

Another example is
http://www.epri-peac.com/tutorials/sol01tut.html . Assume a
transient enters on phone line. Now telephone line protector
(NID) connects to building single point ground via cold water
pipe. But pipes may be too long, have solder joints, etc.
IOW connection to earth is not sufficient. Therefore the
transient also seeks earth ground, destructively, via FAX
machine.

Just some examples of how damage can result if transient is
not earthed before entering a building - the importance of a
single point earth ground.

Dave Liquorice wrote:
The surge was induced in a few metres of network cable and found its
way to ground through the PCs earthed chassis, after frying the
decoupling Cs and blacking the board a bit. It also took out the
network side of the connected IP camera, but without visible damage,
the camera isn't earthed. That was the only damage, the camera still
worked (on it's serial interface) and the PC suffered no damage
neither did the modem connected to the phone line and PC or the modem
connected to the camera and phone line. Nothing else in the house
suffered. Oh and remember the strike was a couple of hundred yards up
the road.

Please explain how "whole house" protection would have protected this
NIC?