*TOTALLY* isolating phone from line, electrically?

[Don Bruder]

OK folks, here's a little something that just blew through this disaster
area I call my mind. Tossing it out for the idea of determining
practicality:

Lightning kills computers and computer gear. We all know this to be a
pretty well established fact. We also know that, in most (not all, but a
very large percentage) cases, the lightning damage didn't come from the
power lines, which are, in general, pretty well protected, either by
compliance with well-researched/written building codes, requirements for
grounding a specific way, surge suppressors (both on the pole and within
the house) and so on. Indeed, most lightning damage seems to come into
the building by way of the phone line, which is nowhere near as heavily
"regulated and sheilded" by building practices. Phone wire runs the
lightning right into the house, cooking off whatever is attached
directly, and sometimes nearby items. Usual scenario: phone line takes a
hit, modem fries, takes computer's serial ports (or even more of the
motherboard) with it.

With that in mind, I've just had something resembling a brainstorm, and
want to bounce it off this merry band of electrical lunatics to see if
it's at all practical.

Since the phone line is (or at least, for the sake of this discussion,
I'm "ASS-U-ME"-ing that it is...) the most frequent route for a
lightning hit to follow and cause damage, it seems to me that
electrically isolating the phone line from the house and contents would
be the best route to take. As of right now, I'm *REALLY* hazy on the
details, but the seemingly ideal implementation would be a box (black or
otherwise...) which plugs into the demarcation point (the place where
the wire from the pole connects to your house, if you're not up on
phoneco lingo) then everything else in the house that needs/uses a phone
line plugs into that.

Anybody ever heard of/encountered such a beast?

If one exists, what would you expect/be willing to pay for it, and who
would you look to in order to source it?

If no such thing exists, has anyone got a good explanation for why not?

 

[Russell Miller]

If no such thing exists, has anyone got a good explanation for why not?

I've never heard of one, and I can think of several reasons why not right
off the bat.

Telephones use several different types of signals for different purposes.
They use high voltage, low frequency pulses for ringing, there's an offhook
voltage, and there's an onhook voltage. A device that you would create in
order to do this isolation would need to somehow process those signals,
isolate them (perhaps, no, probably, using optoisolators), then regenerate
them on the house side.

The regeneration would require power.

The power would come from... say it with me... line current.

Which lightning could cause surges in.

No more isolation.

Also, lightning could easily fry the circuit before the isolation, and arc
across the optoisolator pins, frying the internal circuit. So, now,
instead of having one point of entry (assuming the phones aren't powered,
which is a really poor assumption these days), you have two.

The only real way to isolate telephones from the external network is to pull
the plug, or perhaps include a heavy duty switch that gets tripped when
lightning is detected (heavy duty enough that even if it gets a near direct
hit by lightning it won't arc, though the EMP might fry things anyway...)

--Russell

 

[Don Bruder]

I've never heard of one, and I can think of several reasons why not right
off the bat.

Telephones use several different types of signals for different purposes.
They use high voltage, low frequency pulses for ringing, there's an offhook
voltage, and there's an onhook voltage. A device that you would create in
order to do this isolation would need to somehow process those signals,
isolate them (perhaps, no, probably, using optoisolators), then regenerate
them on the house side.

Understood, accepted, and at least *PART* of why I'm wanting commentary.

The regeneration would require power.

That's a given...

The power would come from... say it with me... line current.

Ahhh, but line current from what source? NOTHING says that the phone
line has to be the power source for this box, and in point of fact, I'm
thinking that the device would be *MUCH* more robust if it were run from
mains (with appropriate transformation, rectification, filtering, etc)
or other "non phoneline" power source.

Which lightning could cause surges in.

Perhaps, perhaps not... recall that I'm operating from the premise that,
in most cases, the phone line is going to be the one affected by a
strike, with the power lines almost always being well-enough protected
that anything less than a direct-to-the-wire hit on the wires between
the transformer and your house is *PROBABLY* going to be a non-issue.

No more isolation.

Only true in cases where the phone line itself is the power source. See
above.

Also, lightning could easily fry the circuit before the isolation, and arc
across the optoisolator pins, frying the internal circuit. So, now,
instead of having one point of entry (assuming the phones aren't powered,
which is a really poor assumption these days), you have two.

Errr... Huh? You lost me. I understand the possibility that the hit
could jump the optoisolator, but the rest of it just left me wondering
what it was that you actually said.

The only real way to isolate telephones from the external network is to pull
the plug

I flatly refuse to accept that. "Nobody's done it, so it can't be done"
didn't hold any water when Columbus set sail from Spain, it didn't hold
any water when Kennedy said "let's put a man on the moon", and it
doesn't hold any water today. Only a very few things are impossible, and
even those are questionable.

, or perhaps include a heavy duty switch that gets tripped when
lightning is detected (heavy duty enough that even if it gets a near direct
hit by lightning it won't arc, though the EMP might fry things anyway...)

EMP is, of course, a concern, but at least for my purposes, there's very
little, if anything, I can do about it. If a hit is close enough to
generate sufficient EMP to knock something out, chances are scary-high
that the device that fried is going to be the least of your worries,
since the house is probably burning down around you as you stand there
pondering your fried phone equpiment.

 

[Jerry G.]

That was an excellent explanation!

I was going to suggest the simplest. He should get a cell phone, and let
the problems be with the provider. There are some new technology cell
phones that can be used as a semi-highspeed modem with a computer. These
are very effective for wireless email and general browsing.

It will be his luck that the cell phone would be on the charger, the next
time lightning hits his power line! I can just see it now...

--

A few years ago, I was watching TV in the living room, and my wife was
watching another program in the bed room. There was a loud crash on the
side of the house, that almost put me on my butt! I ran to the bedroom, and
my wife was shaking in fear, and the air conditioner was smoking like it was
on fire! Lightning hit the dammed air conditioner! It was burned up
inside! The outside of the building had all black colour in the bricks near
the air conditioner. The air conditioner was scrapped.

What gets me, is that the air conditioner has a heavy metal box enclosure,
and was grounded through the AC line. How did it get burned up inside???
All the motors were fried, including the compressor. I had the machine
checked anyways to know what failed when it got hit. As a safety
precaution, I changed the AC outlet that it was plugged in to. But, the
conclusion was that lightning hit the air conditioner somehow. There was a
big burn mark on one corner, and some of the metal on the edge of the case
was melted!

After some arguments, I managed to get the insurance company to pay for it.
I found out the hard way that the claim was not worth the effort. I paid
about 3 times for it, just in the increase!


--

Greetings,

Jerry Greenberg GLG Technologies GLG
=========================================
WebPage http://www.zoom-one.com
Electronics http://www.zoom-one.com/electron.htm
=========================================

If no such thing exists, has anyone got a good explanation for why not?

I've never heard of one, and I can think of several reasons why not right
off the bat.

Telephones use several different types of signals for different purposes.
They use high voltage, low frequency pulses for ringing, there's an offhook
voltage, and there's an onhook voltage. A device that you would create in
order to do this isolation would need to somehow process those signals,
isolate them (perhaps, no, probably, using optoisolators), then regenerate
them on the house side.

The regeneration would require power.

The power would come from... say it with me... line current.

Which lightning could cause surges in.

No more isolation.

Also, lightning could easily fry the circuit before the isolation, and arc
across the optoisolator pins, frying the internal circuit. So, now,
instead of having one point of entry (assuming the phones aren't powered,
which is a really poor assumption these days), you have two.

The only real way to isolate telephones from the external network is to pull
the plug, or perhaps include a heavy duty switch that gets tripped when
lightning is detected (heavy duty enough that even if it gets a near direct
hit by lightning it won't arc, though the EMP might fry things anyway...)

--Russell

 

[Russell Miller]

Ahhh, but line current from what source? NOTHING says that the phone
line has to be the power source for this box, and in point of fact, I'm
thinking that the device would be *MUCH* more robust if it were run from
mains (with appropriate transformation, rectification, filtering, etc)
or other "non phoneline" power source.

Line current == mains current in US. What we have here is a failure to
communicate.

Perhaps, perhaps not... recall that I'm operating from the premise that,
in most cases, the phone line is going to be the one affected by a
strike, with the power lines almost always being well-enough protected
that anything less than a direct-to-the-wire hit on the wires between
the transformer and your house is *PROBABLY* going to be a non-issue.

Fair assumption, but only an assumption.

Only true in cases where the phone line itself is the power source. See
above.

True if the prior assumption is correct.

Errr... Huh? You lost me. I understand the possibility that the hit
could jump the optoisolator, but the rest of it just left me wondering
what it was that you actually said.

I was referring to the fact that if the device is mains powered, you have
two failure modes rather than one - two points of entry into the system.

I flatly refuse to accept that. "Nobody's done it, so it can't be done"
didn't hold any water when Columbus set sail from Spain, it didn't hold
any water when Kennedy said "let's put a man on the moon", and it
doesn't hold any water today. Only a very few things are impossible, and
even those are questionable.

Good luck. I don't know of any, and these are good reasons it'll be
difficult.

EMP is, of course, a concern, but at least for my purposes, there's very
little, if anything, I can do about it. If a hit is close enough to
generate sufficient EMP to knock something out, chances are scary-high
that the device that fried is going to be the least of your worries,
since the house is probably burning down around you as you stand there
pondering your fried phone equpiment.

Depends where it hits. Lightning is a tricky bugger and rarely does what
you expect.

--Russell

 

[Russell Miller]

After some arguments, I managed to get the insurance company to pay for
it.
I found out the hard way that the claim was not worth the effort. I paid
about 3 times for it, just in the increase!

I used to live in Toledo, Ohio. One of the strange things about that area
was that occasionally a thunderstorm would come through that was unusually
electrically active. And that, by the way, is an understatement. I mean,
the kind of storm where CG lightning hits all around you (and CLOSE, too)
about once every couple of seconds and the cat and dog would go hiding
under the couch.

I understand a great deal about the mechanics of thunderstorms but I really
would like to know what makes a thunderstorm do that.

Anyway, I remember two such storms. The first storm, I was about 9 or 10
years old. My parents were having some marital issues, and we, for some
reason, left the cat with a family friend while they were hashing them out.
Well, we went and got the cat, and on the way back, one of those
thunderstorms hit. We put the car in the garage, which was right next to
(about four feet from) a telephone pole. I ran to the back door with the
cat, opened the door, went inside, and about ten seconds later lightning
hit the telephone pole outside and made the telephone ring. That was
freaky.

Oddly enough, the telephone survived, and there was no damage to anything,
but I'd say at least 100 volts went through the telephone line.

--Russell

 

[Don Bruder]

Line current == mains current in US. What we have here is a failure to
communicate.

OK, gotcha. We were heading for the same store, just took a different
road to get there. I was reading your "line current" to mean the
phoneline current, since the context of the discussion was phones and
phone lines.

Fair assumption, but only an assumption.

True, it is an assumption. But where else does one start?

For the most part, it seems to be a safe assumption, though. The amount
of lightning absorption/dissipation/other protection on power lines is
positively unreal when you actually start paying attention to what's
there. And it has to be - Power lines are such a huge potential target
for a lightning hit that if they weren't heavily protected, a storm
would take out the juice for miles upon miles of countryside with just a
single hit. I've yet to see a strom that was satisfied with only
dropping a single strike, which in turn means that mega-heavy-duty
lightning protection *MUST* be in place, otherwise, whole sections of
the country would be dark almost constantly. Florida, with its daily
thunderstorms at about 4-5 PM leaps to mind as a prime example - 4-ish
rolls around, thunderclouds start moving in. By 4:30, 5 at the outside,
it's pouring rain, flashing like a disco, and otherwise looking like the
end of the world. By 5:30 or so, the rain is done, the clouds are
almost, if not completely, gone, and the sun is turning the place into
the world's biggest sauna - yet the only time during the year I lived
there that we lost power (not counting 2-3 second flickers) was when one
storm hit several trees on the property, one of which fell on the wires
and snapped off a total of 8 poles, leaving us without power for close
to a week as the power company tried to get gear in that could reset
them. (swampy area - pole-setter trucks kept hitting soft spots and
sinking past the axles before getting to the locations they needed to be
in to get the job done - After 4 tries in as many days, and still not
even halfway done, they finally gave up and dragged some kind of
tank-like military gizmo out of mothballs, and it went in on crawler
tracks to get the job done)

I was referring to the fact that if the device is mains powered, you have
two failure modes rather than one - two points of entry into the system.

Ahhh... comprende. But that goes back to the initial assumption: That
the power lines are *PROBABLY* protected better by the electric company
than any precautions you or I can practically apply once the wire hits
the house. I'm thinking that the risk of failure on the power side of
things is vanishingly small when compared to the danger that anything
connected to the telephone line is exposed to.

Good luck. I don't know of any, and these are good reasons it'll be
difficult.

Heh... You're showing your "glass half empty" streak

I'll take it as a challenge and see what I can come up with. Let ya know
as things progress.

Depends where it hits. Lightning is a tricky bugger and rarely does what
you expect.

That's a definite fact! I'm assuming (Oh gawd, not another
assumption...) that for a strike to generate sufficient EMP to be a
danger to the health of electronic devices, the strike is going to have
to be pretty darn close - as in practically right on top of whatever it
is that gets fried. Otherwise, the inverse square law, and possibly
others, applies to pretty much rule out any damage from more distant
strikes.

<grumble> 7 modems, a serial port, and a computer in 3 years... There's
GOTTA be a way of protecting them that doesn't involve me babysitting
them 24/7, or idiocy like having to unplug them after every use.

 

[Don Bruder]

That was an excellent explanation!

I was going to suggest the simplest. He should get a cell phone, and let
the problems be with the provider. There are some new technology cell
phones that can be used as a semi-highspeed modem with a computer. These
are very effective for wireless email and general browsing.

Great concept, 'cept for one minor detail:
I'm way rural, and on top of that, located in a sort of natural "bowl"
formation that kills cell phone (and TV, and radio) signals. The only
way any kind of signal gets to the house is over a wire, or off a
satellite dish. Over-the-air broadcast signals just don't make it in.
Even if it weren't for the "bowl", cell coverage starts getting
seriously "iffy" about 7-8 miles from the house - that's not really a
topographical/transmission problem, just a lack of any cell towers
closer to us than about 15 miles.

And no, I ain't interested in moving to the shitty^H^H^H^H^H^Hcity so I
can get cell-reception! I happen to LIKE living out here in the
boondocks, far, far away from the idiots and general insanity that city
life forces one to endure.

 

[Russell Miller]

Heh... You're showing your "glass half empty" streak

More like my "latent engineer" streak. Prepare for every eventuality.
That's one reason why I'm still alive, at least so far Especially
considering how many televisions I've had my fingers in while powered up as
a preteen.

I'll take it as a challenge and see what I can come up with. Let ya know
as things progress.

Good luck. Don't get too discouraged if it doesn't work.

<grumble> 7 modems, a serial port, and a computer in 3 years... There's
GOTTA be a way of protecting them that doesn't involve me babysitting
them 24/7, or idiocy like having to unplug them after every use.

Do you have telephone surge protectors connected to them? Those will at
least provide a small measure of protection.

I have a UPS with telephone surge protection. Living in the northern part
of tornado alley, where electrically active storms seem pretty darn common
(but so far less active than the really bad ones I wrote about in another
thread), I'm darned glad I have it.

--Russell

 

[Don Bruder]

More like my "latent engineer" streak. Prepare for every eventuality.
That's one reason why I'm still alive, at least so far Especially
considering how many televisions I've had my fingers in while powered up as
a preteen.

You too, eh? Only mine went well past "pre-teen"... Sometimes you just
*HAVE TO* have the juice on in there to make a diagnosis. Kinda hairy
sometimes, but when it's the only way to do the job...

(Got bit by the suction-cup thingy - What *IS* the proper name for that
thing, anyway???? I've heard it called everything from "anode cap" to
"aquadag", but almost never the same thing twice - exactly ONCE. After
that I learned that the old "one hand in TV set, one hand in hip pocket"
advice was much less inconvenient than writhing on the floor in pain for
several minutes, and not being able to pick up anything with the "bit"
hand for half a day.)

Good luck. Don't get too discouraged if it doesn't work.

Don't be so encouraging!!!!

On the plus side, a google search netted me a phone line isolator
patent, dated 1998, that SOUNDS good. If the beaureaucrat-ese in the
claims has any resemblance to real-world behavior, it should be the
*TOTAL* electrical isolation I'm hoping to find. Now if I can just
figure out who (if anybody) has licensed it and started making/selling
the beasts...

Failing that, I may just break down and buy a copy of the patent (which
claims, although I haven't been able to get 'em to load, to have
schematics and all the info needed for a working prototype) and build
one myself. Don't bother me with "that's illegal". Frankly, I don't
care. Somebody wants to come after me for patent infringement over
building one widget, they're more than welcome - They might even win,
and congratulations to 'em if they do. But the simple fact is you can't
squeeze blood out of a rock.

(Close-enough-to-bankruptcy is pretty much a bulletproof defense against
idiots with more lawyers than brains - If I've got nothing to begin
with, whatcha plan on taking away from me?)

Do you have telephone surge protectors connected to them? Those will at
least provide a small measure of protection.

The first dead modem bought it while connected via one of those. The
second modem exploded (literally - blew chips into fragments, cracked
the case, and blasted popcorn-like capacitor contents all over the
place) while connected through TWO phone-line suppressors. At that
point, I gave up on them as a waste of time, money, and effort.

Fuse - n: The 15 cent part that the only 500 dollar part in your project
will protect from overvoltage/amperage conditions by exploding.

Surge Suppressor - n: The 15 dollar part that will be protected from
voltage spikes, lightning hits, and similar calamaties, by the
multi-hundred dollar dollar computer system attached to it exploding.

 

[Dbowey]

Some of the posted insight included:

AND

I flatly refuse to accept that. "Nobody's done it, so it can't be done"

didn't hold any water when Columbus set sail from Spain, it didn't hold

any water when Kennedy said "let's put a man on the moon", and it
doesn't hold any water today. Only a very few things are impossible, and even

those are questionable.
----
You won't win an argument with fluffy rhetoric. What you are in need of is a
"Protective Connecting Arrangement." It's been done before...... search on
that and just "PCA." PCAs were required by the neurotic Telcos to protect
their facilities from the naughty, bad customer owned equipment. The
"protection" worked both ways. I believe it was in 1984 the FCC began the
Registration Program and mandated the Telcos to quit using PCAs. My first
point is that what you wish to do is not entirely new, but is, in my opinion, a
poor idea.

My next point is that, unless you get the industry to create a new Network
Interface spec (meaning all of the technical parameters of the Network signals
and the Customer Installation signals at the interface, then you MUST use the
interface definitions that now exist. You can find all of them in FCC CFR 47,
Part 68 and also find useful insight in the ANSI Standards for the types of
services in which you have an interest, such as Loop-Start and Ground-Start
Lines.

When you finish with your research on this, doing the latter will cause you to
create a piece of equipment containing all the protective elements than now
exist in residential terminal equipment, so then, lightning can blow hell out
of it instead of blowing the phones and modems. It would need to be customer
owned equipment....... the FCC will not permit the Telco to own it, so it would
neccesarily need to be Registered.

As to working with the industry to define a new, protected interface, do a
search on Committee T1. Committee T1 has Technical Subcommittees (TSC),
attended by manufacturers, Telcos, Special Interest Groups, Interexchange
Carriers, etc. The TSCs have Working Groups who are charged with doing the
technical work. TSC T1E1 writes Network Interface Standards.

Read about them, see who are the contacts, and work with them. They may or may
not agree with you. In any case, these are the people who will eventually
decide if it's worthwhile.

And finally, they like facts. Fluffy dialog won't go far.

Don

 

[Dbowey]

Don Bruder posted:

<grumble> 7 modems, a serial port, and a computer in 3 years... There's

GOTTA be a way of protecting them that doesn't involve me babysitting
them 24/7, or idiocy like having to unplug them after every use.

Have you sorted out whether the surge is from the Power line or the Telephone
line?

Don

 

[Harry Muscle]

OK folks, here's a little something that just blew through this disaster
area I call my mind. Tossing it out for the idea of determining
practicality:

Lightning kills computers and computer gear. We all know this to be a
pretty well established fact. We also know that, in most (not all, but a
very large percen ) cases, the lightning damage didn't come from the
power lines, which are, in general, pretty well protected, either by
compliance with well-researched/written building codes, requirements for
grounding a specific way, surge suppressors (both on the and within
the house) and so on. Indeed, most lightning damage seems to come into
the building by way of the phone line, which is nowhere near as heavily
"regulated and sheilded" by building practices. Phone wire runs the
lightning right into the house, cooking off whatever is attached
directly, and sometimes nearby items. Usual scenario: phone line takes a
hit, modem fries, takes computer's serial ports (or even more of the
motherboard) with it.

With that in mind, I've just had something resembling a brainstorm, and
want to bounce it off this merry band of electrical lunatics to see if
it's at all practical.

Since the phone line is (or at least, for the sake of this discussion,
I'm "ASS-U-ME"-ing that it is...) the most frequent route for a
lightning hit to follow and cause damage, it seems to me that
electrically isolating the phone line from the house and contents would
be the best route to take. As of right now, I'm *REALLY* hazy on the
details, but the seemingly ideal implementation would be a box (black or
otherwise...) which plugs into the demarcation point (the p where
the wire from the connects to your house, if you're not up on
phoneco lingo) then everything el the house that needs/uses a phone
line plugs into that.

Anybody ever heard of/encountered such a beast?

If one exists, what would you expect/be willing to pay for it, and who
would you look to in order to source it?

If no such thing exists, has anyone got a good explanation for why not?

--
Don Bruder - [email protected] <--- Preferred Email - SpamAssassinated.
Hate SPAM? See <http://www.spamassassin.org> for some seriously great info.
I will choose a path that's clear: I will choose Free Will! - N. Peart
Fly trap info pages:

<http://www.sonic.net/~dakidd/Horses/FlyTrap/index.html>

I didn't read the whole thread, but have you looked into whole house surge
protectors. They can be professionally installed into your breaker box and
protect both the power lines and phone lines (and sometimes even the cable
line).

Harry

 

[w_tom]

One example of how CG lightning is created AND, for example,
why 'tornado alley' does not typically suffer so often from
this destructive type of lightning.

Take a cumulus cloud. Let high speed upper atmosphere winds
blow the top of that cloud well ahead. Anything that falls to
earth (rain, hail, etc) would contain electrical charges from
the top of that cloud. Now we have a cloud bottom that is
three miles up and four miles behind the earth charged
region. CG lightning will make a connection from bottom of
cloud to those earth borne charges.

However it will not take the 5 miles path. 5 miles through
non-conductive air is electrically longer than 3 miles down
and 4 miles through earth. And so we have power lines that
make that connection, through power line, into house, through
appliances (especially modems, portable phone base stations,
and faxes), to earth via phone line (because unlike AC
electric lines, the phone lines routinely have 'whole house'
protectors installed). Therein lies an example of how surges
damage electronics.

Much must be cited to contradict so many urban myths.
First, lightning is electricity. Therefore from primary
school science, first a complete electrical circuit must
exist. Before anything is damaged by that surge, first the
surge must pass through everything in that circuit. Too many
myth purveyors will say lightning entered on phone line,
damaged modem, then stopped - in direct contradiction to
science taught in primary school.

Second, that complete circuit means that something must have
both an incoming and outgoing electrical path; else no surge
damage. Why surges do not damage DRAM, for example. A surge
that passes through motherboard, and then modem, to obtain
earth via the phone line does not damage all those other
motherboard semiconductors. Those other semiconductors had an
incoming path but not outgoing path.

Third, lightning has traveled how many miles through
non-conductive air. Is a silly little isolation transformer
going to stop (or absorb) what miles of non-conductive air
could not? Of course not. Any surge protector that claims to
stop, block, absorb, or filter a lightning circuit is simply
scamming the naive. Surge protector as a 'dam' is not
effective. Surge protector as a 'dike' can be effective. But
that means the flood (surge) must have a better path
downstream to earth ground. Surges can be diverted; not
stopped, blocked, filtered, or absorbed.

Summary of effective protection was demonstrated in 1752 by
Franklin and proven in so many 1930s professional papers.
Summary is discussed in "Opinions on Surge Protectors?" on 7
Jul 2003 in the newsgroup alt.certification.a-plus or
http://tinyurl.com/l3m9
Detailed discussion including warranty, UPSes, low voltage,
why computer fails, no protector required for cable, joules,
etc:
"Power Surge" on 29 Sept 2003 in the newsgroup
alt.comp.hardware or
http://tinyurl.com/p1rk

The most critical component of a surge protection system is
earth ground. Many effective systems don't even require a
surge protector. For example, that grounding block (properly
located and earthed) on the CATV line provides the entire
cable lightning protection - no surge protector required or
effective.

Surge 'protectors' are not surge 'protection'. Those are
separate components. A surge 'protector' is only another
component that can make a connection to surge 'protection' -
single point earth ground. Protection is a building wide
system. This being the most important sentence in the entire
post:
A surge protector is only as effective as its earth ground.

Importance (and it is an art) of effective earthing is
discussed among engineers in two threads:
Storm and Lightning damage in the country 28 Jul 2002
Lightning Nightmares!! 10 Aug 2002
http://tinyurl.com/ghgv and http://tinyurl.com/ghgm

And yes, if you thought surge protection was about some
simple, retail, plug-in device ... well they count on you not
knowing any of this - at great profit. Surge protection is
earth ground. To sell those grossly overpriced, undersized,
ineffective products - they avoid all mention of earthing.
Try to find any mention of earthing on any plug-in protector.
No grounding - earthing. No earth ground means no effective
protection. Welcome to a science (and art) that was very well
understood before WWII.

Why was that spot in Toledo so CG lightning active? Start
by learning the geology. Geology (and what we bury in it) is
a major reason for where CG lightning occurs. Again and why -
its all about earthing - not grounding - earthing.

 

[w_tom]

I believe you are confusing the National Electrical Code
with lightning protection. The NEC is written only for the
purpose of human safety. It makes no effort to address
transistor safety. The most common source of destructive
surges - indeed the shortest and most direct path incoming to
electronic appliances - is AC electric.

We still build new homes as if the transistor did not exist
- even 30 years after the transistor became ubiquitous.
However, telephone connections inside the typical house is
made via a 'whole house' surge protector - provided free by
the telco. You can find yours in a box, often the NID,
located (by code) within twelve inches of where the wire
enters your building. And, of course, that surge protector is
only as effective as its earth ground meaning that 12 AWG (or
larger) wire must make a less than 10 foot connection to earth
ground.

Again, the code is only concerned with human safety.
Therefore the code says that ground wire must be less than 20
feet. But for effective transistor safety, that wire must be
less than 10.

Posted elsewhere in this thread is much more information.
Must of which will be in direct contradiction to so many urban
myths - and yet was standard knowledge even before WWII. For
example, your telco connects a $multi-million computer
directly to overhead wires everywhere in town. Do they
disconnect for thunderstorms? Or do they stop service for
three days while they replace the $multi-million computer? Of
course not. They too use simple 'whole house' protection
methods. All incoming wires are first earthed before entering
the building - either by direct connection or via a surge
protector. Earthing - not surge protector - being the 'magic'
solution. Earthing techniques so long and well understood
that we should be building all new homes using Ufer grounds.
Again, a technology probably long older than you - because
effective surge protection was that well understood for that
long.

Benchmark in surge protection is Polyphaser. Their
application notes are considered legendary. What do they
discuss? Their products? Of course not. Polyphaser is
selling effective protection. They discuss earthing ...
extensively. A surge protector is also only as effective as
its earth ground.

Phone line appliances such as phones and modems already have
the galvanic isolation you are seeking - typically rated at
2000 or 5000 volts. So why does damage occur? That
protection that also exists in computer power supplies and
most every other household appliance assumes the building has
a 'whole house' protection system. Something not required by
any electrical codes. If the 'system' does not exist, then an
incoming surge will overwhelm internal appliance protection -
seeking earth ground, destructively, via that appliance.

Even Home Depot sells effective 'whole house' protectors for
residential AC electric - Intermatic EG240RC and IG1240RC or
Siemens QSA2020. But and again, a surge protector is only as
effective as its earth ground which is why those 'whole house'
protectors are effective IF properly earthed - as discussed in
those other threads listed in the other post. Effective
'whole house' protection that costs about $1 per protected
appliance.

 

[Ian Stirling]

You too, eh? Only mine went well past "pre-teen"... Sometimes you just
*HAVE TO* have the juice on in there to make a diagnosis. Kinda hairy
sometimes, but when it's the only way to do the job...

(Got bit by the suction-cup thingy - What *IS* the proper name for that
thing, anyway???? I've heard it called everything from "anode cap" to
"aquadag", but almost never the same thing twice - exactly ONCE. After

Aquadag is the graphite-based paint that covers the majority of the back
of the tube, except a small area around the anode cap, which covers the
anode.

o1You can purchase optoisolated serial cardfs/

http://www.serial-cards.co.uk/
You can also get optoisolators with two RS232 plugs.
And even RS232 - optical fibre when you need hundreds of Kv isolation.

You'd probably need to work out some means of powering the modem.
A transformer cut in half with a bit of plastic/glass in the middle is
common.

 

[Don Bruder]

Don Bruder posted:

Have you sorted out whether the surge is from the Power line or the Telephone
line?

Don

Since at least four of the dead modems showed obvious signs that the
strike that wiped them came in through the phone line (traces from the
phone jack vaporized, so-called "protective elements" directly connected
to the phone jack traces, and so on, all cooked), I'm calling it a
more-than-reasonable hunch. The other three... Well, to the best of my
knowing, a wall-wart SHOULD cook off long before allowing a killer spike
to get to through to whatever it happens to be powering. The warts are
good (two of them in use as I type, one with different connector on the
end powering my discman player, the other topping up a set of Ni-Cds),
but the modems are toast. Best I can say is "make your own assumption".
The cooked serial port card was obviously blasted from the direction of
the modem (only device attached to it) and the computer was a "dunno
how, but it's got burnt traces on the motherboard, so it's toast"
concept.

Based on that info, if you can tell me which line the strike came from,
you're doing better than me.

 

[Spajky]

OK folks, here's a little something that just blew through this disaster
area I call my mind. Tossing it out for the idea of determining
practicality:

Lightning kills computers and computer gear. We all know this to be a
pretty well established fact. We also know that, in most (not all, but a
very large percentage) cases, the lightning damage didn't come from the
power lines, which are, in general, pretty well protected, either by
compliance with well-researched/written building codes, requirements for
grounding a specific way, surge suppressors (both on the pole and within
the house) and so on. Indeed, most lightning damage seems to come into
the building by way of the phone line, which is nowhere near as heavily
"regulated and sheilded" by building practices. Phone wire runs the
lightning right into the house, cooking off whatever is attached
directly, and sometimes nearby items. Usual scenario: phone line takes a
hit, modem fries, takes computer's serial ports (or even more of the
motherboard) with it.

With that in mind, I've just had something resembling a brainstorm, and
want to bounce it off this merry band of electrical lunatics to see if
it's at all practical.

cheap & effective way is IMHO getting a surge protector for mains &
phone line ( possibly for Lan connection too) in single box from which
all my sensitive machinery gets power & other connections .

IMHO it prevents (if house electricity grounding is Ok & properly
done, should be!) almost 99% of surges.

Will not protect against direct or very close lightning hit, but that
are rare if you are not really a "hot spot" for it.

mine saved my equipment few times in this few past years
(I just replace blown fuses few times & than I replace the protection
every few hits - every 2 or 3y).

It works for me & is cheaper than my modem card.

-- Regards, SPAJKY
& visit site - http://www.spajky.vze.com
Celly-III OC-ed,"Tualatin on BX-Slot1-MoBo!"
E-mail AntiSpam: remove ##

 

[Jerry Greenberg]

The only solution I can see would be to buy a bunch of low cost
phones, and a good one for serious work. You can leave the good one
unjacked when not needed. Leave a low cost phone in the wall jack in
case you need to receive a call. If the phone gets blown out, you can
toss it, and then use another spare. Most likely the phone will last
a number of years before it gets blown by lightning anyways. You may
even wear out the phone from use, before it gets hit by lightning
again.

Making all kinds of design changes to the telephone line, and
experiementing would be more costly, and also most likely a great loss
of time in relation to getting a number of good phones. These phones
are all designed to be disposable anyways. If a new phone fails
during the warranty, and there are no visual signs of any catastrophy,
the dealer should exchnage it.

--

I don't blame you for liking to live in the country. I can see your
point. I myself have been brought up and lived in a city all my life.
I like to visit the country often, but would not feel comfortable to
live there full time, considering what I am used to.


Jerry Greenberg
http://www.zoom-one.com

 

[CWatters]

If no such thing exists, has anyone got a good explanation for why not?

In the end the only really good insulator is distance. If lightening can arc
from a cloud to the ground it can jump a few inches and bypass any black box
you can invent. It just takes enough voltage.

Having said that.... perhaps you could build something using fiber optics.
You just need to decide how long a length of fibre you need to seperate the
two ends of your black box. (and solve the power supply problem others have
pointed out).