Can't detect dial tone
Jared Dunn
One is an ISA 56k Zoom Flexmodem, the other a PCI Conexant V.90.
The latter is in an old P-200 system I've cobbled together from various
pieces lying around, the former is in a fairly new Compaq Presario P-500.
Both are running Win98.
They will dial when I turn off the "wait for dial tone" option in dialup
networking, but they won't connect to my ISP. They also both respond to
the diagnostics section in the Modem Control Panel(I don't have the actual
codes they gave back handy, but, if needed, I can get them for you.)
According to the Compaq Diagnostics Utility(which I don't necessarily
trust,) the Conexant Modem is working properly, and thus it's a phone line
problem, but, when I took the system out to work and tried to connect, I
still had the same problems.
Other things I have tried include: Uninstalling and reinstalling the
modems, wiping and reinstalling basically everything networking and
communications-related on both 'puters, including doing a quick-restore
on the Compaq, fiddling around with the IRQ's and Com port settings,
switching the modems between computers and so on and so forth, all to no
avail. I also installed an old 14.4 Cirrus Logic modem I had lying around
in the P-200, and still had the same problems(but, for all I know it's
fried too, or the jumpers could have been wrong, since it isn't plug and
play.)
So, at this point, I'm pretty well flummoxed. This all started after a big
electrical storm, so I'm leaning towards that being the problem. But,
everything else on both systems seems to be working fine, and there is no
overt evidence of lightning damage(no scorch marks, everything else
attached to the phone lines is still ok, etc). So, I just wanted to know
if there's anything else I can try before I go spend $ on two new modems.
Thanks in advance,
J.
-----------------------------------------------------------------------
"Three passions, simple but overwhelmingly strong, have governed my life:
the longing for love, the search for knowledge, and unbearable pity for
the suffering of mankind."
- Bertrand Russell
Jared Dunn +++++ UIUC Bio/Phil major
http://suspensionofdisbelief.org/
icq#5740361 aim:jddunn13
w_tom
line, damages a modem, then halts. We learned electricity in primary school.
To have electric current, a circuit must exist. To be in a circuit, the modem
must have both an incoming and outgoing path. Current is conducted through
everything in that circuit before at least one component fails. Defined as
one path is the phone line. Where is the other path?
Lightning seeks earth ground. It found earth ground through your modem.
The phone line is typically connected to an excellent earth ground - 'whole
house' surge protection. Phone lines are usually the outgoing path.
Therefore no scorch marks on phone lines. The most common incoming path is
the AC electric. Those are the wires higher on utility poles that are struck
more often.
So why was the phone not damaged? What are the incoming and outgoing
paths? Outgoing should be obvious. Where is an incoming path? No incoming
path means no surge damage.
No appliance need be powered to be damaged. In fact, an adjacent surge
protector could have actually contributed to the modem damage in a powered off
computer.
The surge from AC power to phone line would have traveled through a
motherboard ground, into the modem, through dialtone detection circuits, into
a phone line ground. No Dialtone Detected is a most common error message for
this type of surge damage.
How to avoid damage in the future? This is old technology from the 1930s
still little known by those mostly educated by boxes on retail shelves. A
surge must be shunted to earth ground before it enters the building. As noted
earlier, a phone line already has such 'whole house' surge protection.
However we still wire new homes as if the transistor did not exist; AC power
has no shunt to earth ground at the service entrance. No 1000+ joule 50+ KV
'whole house surge protection connected less than 10 feet to central earth
ground. How much for effective surge protection? Less than $50 at Home Depot
- or about $1 per protected appliance.
Same applies to CATV, satellite dish, etc. They too must connect less than
10 feet to central earth ground, either direct or through a 'whole house'
surge protector, before entry into the building.
GoGoGopher
"Jared Dunn" <jdd...@uiuc.edu> wrote in message
news:Pine.GSO.4.31.010711...@ux11.cso.uiuc.edu...
About the only possible thing you could hope for is Divine
intervention and a resurrection for your dead modems.
It sounds like you have tried most everything within reason and you
just don't want to hear what your common sense is telling
you......your modems are *toast*.
I recently suffered your same situation, -ae- lightning strike, modem
problems, diagnostics telling me everything was ok, rest of 'puter
functioning fine, and *no* scorch marks. But, my modem was in reality,
dead as a doornail. The ring and tip aspect (the red and green phone
wires) on the modem circuit had shorted to ground. To add insult to
injury, I have surge protection on top of surge protection....whole
house AND local on my phone lines and power supply.
The point is, when you combine computers with lightning, odd things
sometimes happen that often go un-detected by inferior software-based
diagnostics or are not as obvious as a scorch mark.
If you truly want to protect sensitive electronics, unplug your 'puter
during an electrical storm, INCLUDING the phone and / or cable lines.
I know...not very practical, and downright impossible if you're away,
but that is the only way to (almost) guarantee one hundred percent
protection. Suppressors and shunts, when new and "unused" are fine for
occasional low voltage surges (appx. 200-400 volts) such as when the
power company's auto-fault detector / clear is initiated, or for a
distant lightning strike that has greatly diminished in potential by
the time it reaches you. The trouble with suppressors and shunts is
that they get tired with age and use (as we all do). Because their
ability to function is diminished with each abnormal spike they deal
with, sooner or later they'll let you down, even on a relatively
small spike. As far as a direct or even a close proximity lightning
strike to your power system, forget it...no suppressor in this world
is going to protect your electronics. At that potential energy level
(1,000,000 volts, give or take 500,000), electricity is like a 600
pound gorilla....it's going to go where it wants to, and no sixty
dollar suppressor is going to stop it, regardless of what the
manufacturers claim.
Hope you find a good modem at a good price, I'm looking too.(and full
of questions)
Bob
Jared Dunn
thanks for the education, and sorry for my little display of ignorance.
Kinda odd, I've had various computers connected to phone lines for at
least six years and through many thunderstorms with no ill effects. guess
I was just lucky, and probably due for this little misfortune.
I also might have researched this a bit more before asking if I didn't
have very limited internet access due to the aforementioned problems. Looks
like I'm going bargain-modem hunting! And making sure to unplug the damned
things from the wall next time we have a storm around here.
thanks again,
J.
-----------------------------------------------------------------------
"Three passions, simple but overwhelmingly strong, have governed my life:
the longing for love, the search for knowledge, and unbearable pity for
the suffering of mankind."
- Bertrand Russell
Jared Dunn +++++ UIUC Bio/Phil major
http://suspensionofdisbelief.org/
http://bluishorange.com/gangbang/
icq#5740361 aim:jddunn13
w_tom
once every 8 years. Therefore you were probably due for some damage.
Most people know very little about surges, how surges operate, how surge
protection works, or the frequency of the event. IOW some are so misinformed as
to recommend surge protection as a solution to daily or weekly computer
crashes. Most are so misinformed as to call ineffective, overpriced plug-in
surge protectors as 'quality surge protectors' when those devices don't even
mention critical earth ground let alone make a connection to same.
There is no reason to be embarrassed concerning a lack of surge knowledge.
Most don't possess it. Many emphatically recommend plug-in surge protection
when they don't even know what a surge protector does - nor appreciate the
importance of earth ground. In short, you asked the right question.
Effective surge protection, for everything, not just your modems, is less
than $50 at Home Depot or about $1 per protected appliance. But then it may
have another 8 years before you need protection.
v.Richard
The problem with modems and lightning is different from those
typically encountered via AC power lines. AC power lines may get a
direct lightning hit - and the voltage/surge does seek ground.
Phone lines are balanced - neither of the 2 wires that enters your
home is at ground potential. In fact, if there is a low resistance to
ground anywhere on either of these wires, you will hear a hum on your
phone. Once outside the house, the wiring the phone company uses has a
sheath at ground potential shielding the phone line.
I believe that most problems with modem damage via lightning is not
due to a direct hit on a phone line, but due to abnormal difference of
potential (surge) on the tip/ring wires induced past the ground sheath
of the phone line and the solid-state DAA in the modem. I don't
believe you can protect a phone line with a surge protector that
grounds the phone line, but rather, you need a protector that shorts
the phone line (when triggered) at the point of protection.
A modem with electro-mechanical DAA is less susceptible to damage.
http://modemsite.com/56k/lightning.htm
Aloha,
Richard
http://modemsite.com/
"w_tom" <w_t...@usa.net> wrote in message
news:3B4CEDFE...@usa.net...
w_tom
1) Assumed in the previous post is a linear ground on phone lines. But
both lines are grounded by a non-linear device called a surge protector.
Surge protectors simply shunt a surge from one wire to all others - and
remain open circuit devices when not shunting a surge. Yes, both lines
are isolated from earth ground when in normal operation. But when common
mode voltage increases (lightning), then there is a non-linear impedance
change - created by premise interface surge protection installed routinely
(and for free) by (American) phone companies. IOW the phone line has a
direct connection to earth ground when a surge comes from the pole or from
inside the building.
2) The technology pioneered in the 1930s, installed in telco Central
Offices (COs), and proven so successful for so many decades - shunt to
earth ground before a surge can enter a building. However we still build
homes as if the transistor did not exist. We install no such 'whole
house' protection - then assume concepts not demonstrated by 50+ years of
research. One is this idea that lightning creates a differential mode
surge. If in doubt, then count the repeated references to earth ground
in application notes from the benchmark in surge protection:
http://www.polyphaser.com/ppc_technical.asp
3) Erroneously assumed: the surge is differential mode - a potential
difference between tip and ring. If true, then telcos would not install
two surge protectors - one from each wire to earth ground. Instead they
would install only one surge protector from wire to wire. Telcos don't
install the differential mode surge protection because destructive surges
are common mode - each wire shunted to ground.
Furthermore, how does a circuit from cloud to ground (CG) create a
current flows in opposite directions on two twisted wires? It does not.
Lightning creates a current flow in the same direction on all bundled
wires - that direction being the best path to earth ground - that
transient being called common mode. Lightning seeks earth ground - on one
wire or all wires - it is still called a common mode surge.
4) Let's look at that electromechanical DAA at modemsite.com . How
fast does any electromechanical device operate? Typically 10 milliseconds
is fast. But for 'benefit of doubt', we assume they are phenomenally
fast. 1 millisecond - or 1000 microseconds. A destructive surge is
typically defined at 30 microseconds. IOW 33 consecutive surges have
damaged the modem before the electro-mechanical device even considered
isolating the modem. What kind of protection is that? Ineffective -
which is but another reason why telcos - who must have even higher
reliability - don't use such systems.
Of course some circuit disconnect protectors will monitor that air for a
precursor electric field - then open the phone line for up to 400 msecs.
It is a neat, although expensive idea requiring more complex electronics
including the proper mounting of an antenna. But why? First most
destructive surges enter on the AC electric lines. Second the homeowner
has a pretty penny of other appliances to be damaged. Third, effective
AC line surge protection is installed for $1 per protected appliance -
including the modem, portable phone base station, fax, and answering
machine. All these are connected to AC mains that have no effective surge
protection but are connected to telco lines that already has effective
surge protection. The source of most destructive surges to telco and
other equipment - the AC mains - where no surge protection exists.
5) What do telcos use? Same 'shunt to earth ground' technology that
has performed so effectively since before WWII and that is also installed
in a subscriber's premise interface. However the distance to earth
ground is so critical that telcos place their surge protection as close to
earth ground as possible - to shunt the surge to earth ground ASAP and
before a surge can get near their switching computers.
6) Previously posted:
> I don't believe you can protect a phone line with a surge protector
> that grounds the phone line, but rather, you need a protector
> that shorts the phone line (when triggered) at the point of protection.
First this belief says what professional do routinely to have high
reliability equipment (communication stations, broadcast stations, cell
phone towers, rocket launch facilitates, etc) does not work. It claims
that grounding surges is not effective - even though earth grounding is
critical to surge protection:
http://scott-inc.com/html/ufer.htm
http://lists.contesting.com/_towertalk/199909/0143.html or this
quote from
http://www.telematic.com/index.htm
> Even the best surge protection device is of no use
> if incorrectly installed. Installation, and more specifically
> earthing, is a critical aspect of surge protection.
Second this statement claims that lightning creates differential (normal)
mode surges. Shunting a surge to earth ground is proven in generations of
strikes - including 25 direct strikes annually on Empire State Building FM
and TV transmitters. Lightning creates a common mode surge. Even where
wire is shielded (and most phone wires are not shielded), even then the
shield also contains a common mode surge that would require critical earth
ground.
7) I wish Dr A J Surtee's article in Power Quality magazine
(www.powerquality.com) entitled "Principles of Lightning Protection for
Telecommunications Facilities" was still available. Maybe you could
obtain a copy from the magazine via e-mail. His graphics make this so
clear without all the calculus and electromagnetic field equations.
However a tsunami of facts is in direct conflict with what are only your
beliefs. For lurkers - just count the number of reasons how and why
effective surge protection is installed just in these seven points. The
history, the technology, the sources of information - you don't have
enough fingers - and most likely this is the first you have heard of these
well proven fundamentals.
Summary) Modem destructive surges typically enter on the AC electric
(that has no surge protection because we still wire homes as if the
transistor did not exist) and seek earth grounds, destructively, also
via phone lines. Noted in that original post is that many forget even how
electricity works - and make erroneous conclusions as to what damaged a
modem. They often forget that a surge must first travel through an entire
circuit before any one device can be damaged. Incoming is typically AC
electric which is most easily struck by lightning; outgoing includes the
earth ground on phone lines - and many other earth grounds.
Lightning is a common mode surge; not a differential mode surge - which
is what so many ineffective surge protectors would rather have their
customers not know. A surge protector is only as good as its earth
ground. No earth ground means no effective surge protection.
GoGoGopher
> Warning: you are about to be buried in technical facts and
citations.
>
BLAH, BLAH,BLAH BLAH BLAH...........Agreed, we are buried.
Jared, the important (on topic) thing is, your modems are ruined.
Regardless wether or not the offending voltage entered through the
phone line or the power supply side is immaterial, dead is dead. Good
luck in your (future) modem purchase.
Just remember...... destructive voltage potential can enter your
computer during an electrical storm from many routes, INCLUDING the
phone line (protection *OR* theory on differential vs common mode
pulses)
That is reality, not just something mistakenly interpreted from a
book, so, if you can, unplug your phone line (and others during a
storm to protect yourself.
w_tom
long since proven and repeatedly demonstrated since 1930 - as well as
many decades of my own personal experiments that date back before there
was a PC. The art of surge protection is earth ground - not the surge
protector. As www.telematic.com states explicitly:
> Even the best surge protection device is of no use
> if incorrectly installed. Installation, and more specifically
> earthing, is a critical aspect of surge protection.
Good grounding is critical, as telematic.com states. It can become an
art - but it is the heart of surge protection - contrary to what is sold
in most retail stores. Some spend the money, still have damage, then
simply give up. In one case, a quick examination noted ground lines
bundled together with signal wires. IOW the owner did not follow basic
instructions on grounding - then simply blameds something else, rather
than himself, for the failure. Surge damage to any modem is directly
traceable to human failure.
The first damaged modem reports that proper surge protection is
required. "Fool me once, shame on you (lightning). Fool me twice, shame
on me." No sense being a fool the second time since effective surge
protection can be installed for only $1 per protected appliance.
Franc Zabkar
to keyboard and composed:
> The surge from AC power to phone line would have traveled through a
>motherboard ground, into the modem, through dialtone detection circuits, into
>a phone line ground. No Dialtone Detected is a most common error message for
>this type of surge damage.
Can you offer an explanation for the lightning damage I observed some
time ago?
(1) An internal ISA modem showed the following damage:
- burnt passives, eg capacitors, in the DAA
- damaged Tx/Rx op amps on the datapump side of the transformer
- cooked datapump (too hot to touch, visible hot spot, not
responding to commands)
(2) The motherboard showed no damage other than failure in IRQ 3 logic
(chipset fault)
(3) A telephone attached to the PHONE socket of same modem was not
damaged. PHONE and LINE sockets were wired directly in parallel.
(4) The modem was on-hook and the PC (AT P/S) was switched off.
(5) The telephone service had to be "re-enabled" at the exchange. It
appears that the disturbance had caused some protection circuit
to operate (?).
(6) No other electrical appliances were damaged, with the possible
exception of a portable telephone in another room, but on the
same circuit. This telephone proved to have an O/C AC adaptor,
plus shorted diode and transistors in regulator circuit within
the base unit. I suspect, however, that the telephone failure
was man-made. The owner had attempted to use several replacement
adaptors of the wrong voltage and/or polarity. Quite possibly
it was the line, and not the phone, that was the original fault.
Flat batteries may have also complicated the diagnosis.
-- Franc Zabkar
Please remove one 'g' from my address when replying by email.
w_tom
(which is only rated for 500 V) or some other path? I am not sure how to
understand that portable phone damage. Was it's power supply damaged which is why
the owner may have tried other power units?
The telco repaired something in the CO. Did they also repair in the premise
NID?
This appears to be the less common telephone line strike. For some reason, a
premise interface protector was not functional or overwhelmed. The surge had to
get past the 2000V transformer (or may have overwhelmed that transformer because
premise interface protection did not exist) to pass through the datapump
components. Then proceed to earth ground through the motherboard ground and green
safety ground wire. Also the surge attempted earth ground via the IRQ3 line and
probably through protection diodes on the Intel type Interrupt controller chip
(8253?). This failure mode would be even more likely if the computer was located
close to a grounded circuit breaker box.
Of course without informed visual inspection, I can only widely speculate. As
for the adjacent phone, it probably was exposed to the surge - an incoming path -
but had no outgoing path - therefore no damage.
Franc Zabkar
to keyboard and composed:
> How did the surge get around the transformer? Did it use the off-hook relay
>(which is only rated for 500 V) or some other path? I am not sure how to
>understand that portable phone damage. Was it's power supply damaged which is why
>the owner may have tried other power units?
The owner is an older gentleman. I'm not sure that even he remembers
the sequence of events.
> The telco repaired something in the CO. Did they also repair in the premise
>NID?
No, only at the CO.
> This appears to be the less common telephone line strike.
I have seen an older 2400bps internal modem suffer damage to the same
component on two occasions, specifically an optocoupler and associated
diodes. These caused the modem not to detect ringing. All these
components were on the line side of the hook relay. There were no burn
marks. No other damage occurred, neither to the computer nor to any
electrical appliance within the house.
> For some reason, a
>premise interface protector was not functional or overwhelmed. The surge had to
>get past the 2000V transformer (or may have overwhelmed that transformer because
>premise interface protection did not exist) to pass through the datapump
>components. Then proceed to earth ground through the motherboard ground and green
>safety ground wire. Also the surge attempted earth ground via the IRQ3 line and
>probably through protection diodes on the Intel type Interrupt controller chip
>(8253?). This failure mode would be even more likely if the computer was located
>close to a grounded circuit breaker box.
The motherboard "chipset" consisted of only one chip. The modem
chipset was also basically the one chip, with integrated DSP,
controller, and ISA interface logic, plus support components such as
SRAM, EPROM, and NVRAM. The EPROM appears to have survived.
w_tom
should ever enter the house. However in the days of 2400 baud, there were still many
telephone lines that were not properly surge protected - grossly insufficient earth
ground connections. The surge protector either degraded or the ground was just too far
away to be effective. As long as a surge protectors is sufficiently sized, then surge
protection is completely defined by central earth ground quality. A surge protector is
only as good as its earth ground.
That 2000 V could be incoming from any utility - not just a phone line. Many forget
their primary school electrical lessons. Electricity must have both an incoming and
outgoing path. In the case of that optocoupler, was the surge entering on phone lines
to obtain earth ground elsewhere? Or did it enter on the AC mains (or CATV) to find
earth ground on the phone lines. Most never ask those questions. They quickly assume
lightning entered on the closest wire - the phone line - destroyed the modem - then
stopped. Electricity must pass through everything in the circuit before one circuit
component is damaged. That applies even to the optocoupler which was exposed to
voltages well above what are considered acceptable, today, in any building.
One nice thing about modem damage - it provided information on how to avoid all
future modem damage - if the human remembers how electricity works and that lightning
seeks earth ground... destructively if necessary.
Franc Zabkar
to keyboard and composed:
> A damaged optocouple means at least a 2000 volt transient. Well in excess of what
>should ever enter the house. However in the days of 2400 baud, there were still many
>telephone lines that were not properly surge protected - grossly insufficient earth
>ground connections. The surge protector either degraded or the ground was just too far
>away to be effective. As long as a surge protectors is sufficiently sized, then surge
>protection is completely defined by central earth ground quality. A surge protector is
>only as good as its earth ground.
>
> That 2000 V could be incoming from any utility - not just a phone line. Many forget
>their primary school electrical lessons. Electricity must have both an incoming and
>outgoing path. In the case of that optocoupler, was the surge entering on phone lines
>to obtain earth ground elsewhere? Or did it enter on the AC mains (or CATV) to find
>earth ground on the phone lines. Most never ask those questions. They quickly assume
>lightning entered on the closest wire - the phone line - destroyed the modem - then
>stopped. Electricity must pass through everything in the circuit before one circuit
>component is damaged. That applies even to the optocoupler which was exposed to
>voltages well above what are considered acceptable, today, in any building.
I've taken onboard what you've said, but I can't see how the observed
damage is consistent with a strike on the mains. As the optocoupler
output is connected to the DSP, and as the DSP is earthed via the
building ground, why did the DSP not sustain any damage? Instead, on
two separate occasions the damage seems to have been confined to the
line side of the optocoupler. As far as I can tell, the only ground
connection within the DAA is via two ceramic 150pF, 2kV RF suppression
capacitors.
In any case, why would a surge on the AC mains look for an earth
ground via the phone line when it has access to an electrically closer
earth ground via the mains receptacle?
Is it possible that the optocoupler damage was not caused by lightning
but by an excessive ringing voltage?
Hooda Gest
"Franc Zabkar" <franc...@dinggoblue.net.au> wrote in message
news:3b550d7e...@news.dingoblue.net.au...
Franc, for what it's worth, the only time I suffered damage to a modem was
during a major thunderstorm. Lightning struck the house across the street
and destroyed several appliances (including TVs, a refridgerator, a
microwave, and a couple of phones) and blew out eletrical outlets. The
neighbor to the east of me lost a TV and a phone. The neighbor on the west
lost 2 TVs and his phone. My damage was limited to a USR Courier that was on
protected power but unprotected phone line. (All the rest of my electronics
were/are on surge suppressor units and suffered no harm) The damage symptom
was No Dial Tone. The modem passed all diagnostics, commands all worked, but
the modem would not dial out even with X3 set. The modem did seize the line
but the level was way down.
I believe, in spite of w_tom's detailed explanations, that the surge came
through the phone line.
There is a thing called "difference of potential" that doesn't seem to be
mentioned in those explanations. I believe I was once taught that a lower
voltage would appear to be a "ground" to a sufficiently higher voltage.
Perhaps my memory is poor, though.
Hooda
w_tom
building on two hot AC wires (assuming the neutral being grounded). Wire is not a
conductor to electricity. That outlet ground may or may not be earthed depending, for
example, how far the outlet ground connects . Outlet ground can be farther away from earth
ground than the phone line's ground. It is simply which ground was easier to obtain AND if
the building really has an earth ground. Many older buildings don't even have the NEC
required earth ground (no surge protection would be possible), let alone one sufficient for
lightning. The phone line in older buildings is simply a easier path to earth, especially
in those older homes.
Furthermore, how does a surge get from AC main to outlet ground? In this scenario,
there would be no good path which is why the surge might seek destruction through a 2000 V
optocoupler. Actually the surge path usually is through the easier off-hook or dialtone
detection circuits. But then I am not familiar with your PSU and modem's details.
Optocoupler damage is quite rare because other, weaker bypass paths exist and are usually
used.
The surge, if incoming from the phone line, does not have to pass through the DSP chip
since 1) the optocouple has another pin connected to motherboard ground and 2) DSP inputs
are probably also diode protected providing another connection to ground. Of course
another, simultaneous earth ground path would be through those 150pf capacitors. Not all
components in the surge's path are destroyed.
Excessive ring voltage? It is possible. Phone line surge protection is about 600 volts
differential. But then where did 600 V come from? Maybe from the electric utility 4K
primary? In which case, 1) if phone line protection is the older carbons, then those
devices are failed and should have been replaced anyway, or 2) if the newer semiconductor
type device, then the phone line would probably had a permanent short to require surge
protector replacement, or 3) maybe the phone line surge protection has also been compromised
- all 4KV appeared on the optocoupler. Just a few more things to look at. Is the
building's phone line surge protection still installed or operational?
Rick Collins
> Did the building have a whole house surge suppressor? If not,
then a surge enters the
> building on two hot AC wires (assuming the neutral being grounded).
Wire is not a
> conductor to electricity.
I'm pretty sure you didn't mean that.
v.Richard
device will drop 1/2-volt in forward direction, and block flow in
reverse direction. The device will be rated for current in forward
direction, and maximum voltage in reverse direction.
On-hook Phone line: 48V DC between tip and ring - balanced - neither
lead is at ground potential.
Modem on-hook: A device - either relay (electro-mechanical) or
solid-state (usually opto-electronic) prevents a connection to the
'main' part of the modem. 'Secondary' circuit coupled to phone line at
very high impedance to detect ring.
If a very high difference of potential occurs between tip and ring,
and this difference reaches a solid-state junction, it can be
damaged - often with no visible sign of damage.
A modem that has electro-mechanical DAA is most vulnerable in it's
'secondary' connection to phone line - ring detection, and sometimes
will operate (but not be able to detect ring, or dial-tone, and in
sometimes with impaired performance). A modem with solid-state DAA is
vulnerable everywhere.
I don't know enough to explain scientifically how the 'surge' occurs
with respect to phone lines and thunderstorms - but various evidence
would indicate it is happening - and it can happen independent of any
connection between the modem and AC power/ground: it can happen with
the modem totally disconnected from AC power/ground, and despite
functional telco lightning protection. In this case, the only hope of
protection is a device that 'shorts' the tip and ring (like a MOV) -
and some modems do have such device.
I do not disagree with most of w_tom's writings regarding protection
of AC-powered devices - but, I wouldn't trust his protection for my
modem.
Aloha,
Richard
http://www.modemsite.com/
"Hooda Gest" <Be@One_WithDotCom> wrote in message
news:9j3un5$hi3$1...@slb2.atl.mindspring.net...
w_tom
"difference of potential" so high across the optocoupler - 2000+ V?
Furthermore, it is the "difference of potential" that makes surges incoming on
phone lines so difficult. Phone lines, properly installed, will shunt potential
differences (common mode) greater than (typ.) 320 V to earth ground before that
potential difference can enter a house. If properly installed, then how did a
2000 V potential difference bypass the standard phone line surge protection to
blow out an optocoupler?
One critical evidence is missing from this discussion. What kind of phone
line surge protection existed (if any), was it still properly grounded (by how
long a wire), AND what was the condition of surge protection after modem
damage? Unfortunately most don't even know that premise interface surge
protection exists on phone lines.
Difference of potential is why, for example, only one of two adjacent
appliances is damaged OR why a modem is damaged when the phone is not. Many
will make grossly erroneous claims that one device acted as a fuse to protect
the other - which even flys in the face of how fuses work.
To have "difference of potential", first there must be a current flow through
an entire circuit. For lightning, that circuit is appliances that complete a
connection from cloud to earth. Since some appliances don't have both the
'incoming' and 'outgoing' path, then they cannot have the "difference of
potential". No complete circuit and no current flow means no "difference of
potential" and therefore no surge damage. It is why a modem can be damaged but
a phone may not be damaged.
We refer to ground as the lower potential simply as a frame of reference since
lightning surges are really RF energy - not a DC voltage. Since a circuit is
between cloud and ground AND since ground is physically lower, then we just tend
to call ground a 'lower potential' regardless of whether the CG lightning strike
is positive or negative.
Hooda Gest
circuit is not closed until the phone device (or modem) seizes the line.
However, there will be a difference in potential on that line in regards to
any potential external electrical source. Yes, the "normal" path for that
electrical charge would be toward the CO rather than the home but it
actually travels in both directions.
A sufficient electrical charge will jump across and not just go to ground.
Like it or not, whether it fits with theory or not, my observations and
experience suggest the phone line is the source of the damaging surge more
often than you think.
But I don't wish to argue with you, I've seen enough of your posts to know
you will never be convinced of this.
Hooda
"w_tom" <w_t...@usa.net> wrote in message news:3B55ED5A...@usa.net...
w_tom
> Wire is not a conductor to electricity.
It should have read:
> Wire is not a conductor to lightning.
To complete the though expressed by the next sentence, instead I shall
paraphrase. Wire is a conductor to to DC or 60 Hz electricity. But
wire becomes an electronic component to lightning. That outlet ground is
typically not earthed depending, for example, how far the outlet ground
connects to earth ground. For effective earth grounds, that wire should
be less than 10 feet. A shorter wire means a more effective surge
protector.
How effective is a surge protector for one surge? If sufficiently
sized, then a surge protector's effectiveness is only determined by the
distance to and quality of earth ground.
It is the length of wire, more than anything else, that determines
whether a surge protector is connected to earth ground. It is why the
benchmark in surge protection, Polyphaser, does not have a connection to
earth ground. That would only diminish surge protection. Instead, a
Polyphaser surge protector mounts directly ON earth ground. A short
connection to earth ground is that critical - because wire is not a
conductor to lightning - which is what I intended to say. Thanks to Rick
for a keynote correction.
v.Richard
pass through the DSP chip
> since 1) the optocouple has another pin connected to motherboard
In the case of external modem, there is no connection to motherboard
ground and opto-coupler (except possibly via the RS232 cable), and is
irrelevant with respect to this discussion and phone line surge with
internal or external modem:
The reason for opto-coupler in the first place is the phone line must
remain isolated from motherboard/earth ground. The 'ground' connection
in the opto-coupler is for the light that triggers the 'switch'. The
'switch' is isolated from ground - the relevant damaging voltage being
that between tip and ring without regard to earth ground.
Aloha,
Richard
http://modemsite.com
w_tom
differential mode signal. Significant characteristic: a differential
mode circuit can be shorted anywhere to ground without affecting the
signal. A differential mode signal, theoretically, can have ANY voltage
"potential difference" to ground and not be affected. (In reality,
semiconductors in differential mode circuits have real world limitations -
a 'voltage to ground' maximum rating.)
While a signal is carried differential mode, another signal can be
carried on the same wires in common mode. The differential mode and
common mode signals can co-exist without interference.
A differential mode signal would have currents on each phone wire in
opposite directions. A common mode signal will travel in the same
direction on both wires or may only travel on one wire.
Sound confusing? That is what plug-in surge protector manufacturers
want you to be. They sell on half truths. Many that emphatically
recommend ineffective surge protection don't even understand fundamentals
of differential mode and common mode. And yet those concepts underlie
surge protection.
"I wouldn't trust his protection for my modem" is a gross
misrepresentation since I have only repeated what your phone company has
known and practiced successfully since the 1930s. The concepts are that
old and that well proven. They don't install protective diodes 'tip to
ring'. The premise interface and the CO contain surge protection (diodes,
the carbons, GDTs) from each wire direct to earth ground. Furthermore,
they place the surge protection as far from the transistors and a close to
earth ground as possible. A preferred installation places surge
protection 50 meters from the CO switching computers. That separation
increases surge protection.
Modem manufacturers once placed surge protection between tip and ring
(in the days of 300 and 1200 baud). It did not work because destructive
surges are not differential mode. A common mode surge traveling down both
wires AND is Zero volts 'tip to ring'. That same common mode surge
punches right through the modem - destructively.
Destructive surges are from lightning - common mode electricity that
seeks earth ground. The surge travels down one or both wires in the
same direction seeking earth ground - destructively if necessary. As
noted in Polyphaser's applications notes (which also define these concepts
and therefore repeatedly discuss earth ground) - the single point ground
window is essential to common mode surge protection.
In http://www.polyphaser.com/pdf/PTD1028.pdf
> A discussion and example of inadequate telephone room grounding. ...
> As separate low inductance ground conductor must be routed from the
telco
> protection block to a low resistance, fast transient response ground
system.
Why the fast transient response ground system? Because destructive
surges are common mode.
They don't install 'tip to ring' protection. Effective surge
protection is from each phone wire to a central earth ground because
professionals address destructive surges - common mode surges. They don't
install for differential mode protection.
However if I want to make big profits at your expense, I might sell
differential mode surge protection, hype it into a big brand name, but
only call it surge protection. I would conveniently forget to tell you
that it was not effective for common mode surge protection. Instead I
would just call it a surge protector. Furthmore, I would sell undersized
component that vaporize even when the surge was too small to damage a
computer. That way the uninformed would preach how a surge protector
sacrificed itself to protect his computer.
Again, this is not 'my' surge protection. The IEEE Transactions on
Electromagnetic Compatibility - 4 Nov 1998 is about nothing but lightning
and surge protection. Everything is earth grounds, connections to earth
grounds, the direct strike, and ground loops created in a building by the
direct strike. It would be totally irresponsible to make any contrary
conclusions without first understanding basic concepts such as common mode
vs normal mode, those well proven fundamentals of surge protection, AND
central earth ground.
I am rather appalled that one would admit to insufficient information,
and yet draw a contradictory conclusion anyway - especially when so much
bad information - mostly half truths - is promoted in retail stores such
as CompUSA, Sears, and Staples - which don't sell any effective surge
protection. That retail stuff violates 50+ years of experience - and even
violates principals expressed in the above IEEE papers. That retail stuff
basically protects from surges that do not exist. One need not have any
engineering background to appreciate these concepts. But first one must
understand common mode vs differential mode.
Again, this is not my surge protection. It is the effective surge
protection advocated by professionals. The problem with v.Richard's
scenario: there is no "very high difference of potential occurs between
tip and ring" because destructive surges are common mode. The voltage
potential is high 'tip to ground', high 'ring to ground', and Zero volts
'tip to ring'.
w_tom
routinely installed, for free, on phone lines. Did you know that it exists?
Does your phone line still use the carbons? Why do you ignore protection
routinely install by AT&T, before the breakup, even in the 1950s?
How can you say "the phone line is the source of the damaging surge" and yet
pretend premise interface surge protection does not exist? What are those
things doing if they are not surge protection? Also what kind of phone line
protection was on the neighbor's and your house? Nothing? Of course not. That
would even be a violation of NEC requirements. How does a surge enters through
phone lines and completely ignore existing protection?
w_tom
ground. Pin 7 of 25 pin RS232 connectors is a direct connection ground to
ground. But since it is a common mode surge, then all other RS-232
lines, com port to external modem, also carry this common mode
transient. External modems are well connected to motherboard ground when
examining the connection in AC terms and surge voltages. Even the 1488
and 1489 receivers and drivers complete that connection.
My impression is that I am discussing concepts little understood here.
For example, that off hook relay will conduct current from modem to phone
line; from relay coil to relay contacts. If that concept is not
understood, then one really cannot appreciate the many paths taken by an
RF energy surge called lightning. A 2000+V surge can pass through an
optocoupler. Somehow, others think that optocoupler is perfect
isolation. It is not. Like all electronic components, it has
limitations. For the off-hook relay, that limitation is about 500V. For
optocouplers and the isolation transformer, that is about 2000 V. When
voltages exceed those numbers, then those isolators change to become
conductors.
I suspect some don't appreciate how isolation barriers break down, or
even why a surge can use all RS-232 wires to travel from motherboard into
external modem.
The optocoupler's purpose is irrelevant. The question is why a 2000+V
current could have flowed through the optocoupler.
Hooda Gest
"w_tom" <w_t...@usa.net> wrote in message news:3B570756...@usa.net...
> Please explain why you avoid discussion of the 'whole house' surge
protection
> routinely installed, for free, on phone lines. Did you know that it
exists?
> Does your phone line still use the carbons? Why do you ignore protection
> routinely install by AT&T, before the breakup, even in the 1950s?
I don't ignore it and, as far as I can tell, Bellsouth properly grounded the
phone line installation at my current home. It was the same where I lived
before but that didn't protect the modem that was not run through a phone
line surge protector. The other modem, which was run through a surge
protector, suffered no damage whatsoever. It may have been just luck since
no phones were damaged.
And, let me point out that the telco's primary concern is protecting the
Central Office and not the customer. I know, I have worked in the industry
for 30 years. Yes, they do try to protect the customer premises to some
degree but that isn't their primary concern.
>
> How can you say "the phone line is the source of the damaging surge" and
yet
> pretend premise interface surge protection does not exist? What are those
> things doing if they are not surge protection? Also what kind of phone
line
> protection was on the neighbor's and your house? Nothing? Of course not.
That
> would even be a violation of NEC requirements. How does a surge enters
through
> phone lines and completely ignore existing protection?
First, quit putting words in my mouth. I never "pretend[ed] premise
interface surge protection does not exist." I will say that I do not depend
on it.
A surge enters through protection when the surge is not sufficient to jump
the air gap at the protection point but is strong enough to jump the gap
between Tip and Ring at the line interface.
All electricity follows the path of least resistance.
Hooda
w_tom
incoming path was probably the AC electric. The surge used the phone line
surge protector as the outgoing path to earth ground. Remember, without a
complete circuit, there can be no surge damage. A surge passes through
everything in the circuit before one item can fail.
Note the list of damaged items in you and your neighbors homes. All were
connected to the common incoming surge source - AC electric. Why were phones
not damaged? Where do they have both an incoming and outgoing surge path? You
must have both to complete an electrical circuit - else no damage.
Because a modem was not damaged does not prove effective protection. It is
the nasty little characteristic of surge protection that makes it such an art.
Surge damage is easily proven. But effective surge protection is not always
easily identified. You don't really know if lightning would have damaged the
modem anyway. It could have already found better paths to earth ground
elsewhere.
You can prove that surge protection has failed, but proving that it has worked
is nearly impossible for laymen. However that 1930 research has proven (and is
still repeatedly proven even in IEEE papers) that effective surge protection
shunts the surge to earth ground before it can enter the building. One
excellent example of this is the entire issue of IEEE Transactions on
Electromagnetic Compatibility of 4 Nov 1998. (One article not about surge
protection describes the National Lightning Detection Network).
Telcos not only protect the CO using 'whole house' surge protection, but they
also protect the customer with similar protection in the premise interface.
Correctly noted is that telcos are more interested in CO protection. It is why
CO surge protectors have no connection to earth ground - instead surge
protection is preferably mounted directly ON earth ground. All incoming lines
first connect to good earth ground - either directly or through surge
protectors.
Homes don't get the same protection. Home phone line surge protection is not
as good because there is a green ground wire between the NID and earth ground.
But the NEC (and I believe EIA497) requires protection at the home premise
interface. It is there AND it has been upgraded to semiconductor technology for
more than 15 years - not GDT technology previously described.
But this is basic to all surge protection. A sufficiently sized surge
protector's effectiveness is defined only by its earth ground. That is what
lightning seeks. That path to earth ground defined what was and was not damaged
in you and your neighbor's homes.
The phone lines already have effective 'whole house' surge protection. It is
why destructive surges more often enter via other utilities such as CATV and
especially via AC electricity - that has no 'whole house' surge protection
connected less than 10 feet to earth ground. Phone lines have or are upgraded
to semiconductor based surge protection in the NID. It is why destructive
surges don't come from the phone line AND it is why 'whole house' surge
protection on AC mains provides effective modem protection.
In surge protection, electricity does not travel in the path of least
resistance. It travels in the path of least impedance - a significant
difference and why that wire to earth ground must be less than 10 feet.
Hooda Gest wrote:
> I don't ignore it and, as far as I can tell, Bellsouth properly grounded the
> phone line installation at my current home. It was the same where I lived
> before but that didn't protect the modem that was not run through a phone
> line surge protector. The other modem, which was run through a surge
> protector, suffered no damage whatsoever. It may have been just luck since
> no phones were damaged.
>
> And, let me point out that the telco's primary concern is protecting the
> Central Office and not the customer. I know, I have worked in the industry
> for 30 years. Yes, they do try to protect the customer premises to some
> degree but that isn't their primary concern.
>
> > ...
Hooda Gest
"w_tom" <w_t...@usa.net> wrote in message news:3B572DE8...@usa.net...
> Of course the Bell South surge protector did not protect the modem. The
> incoming path was probably the AC electric. The surge used the phone
line
> surge protector as the outgoing path to earth ground. Remember, without a
> complete circuit, there can be no surge damage. A surge passes through
> everything in the circuit before one item can fail.
>
> Note the list of damaged items in you and your neighbors homes. All
were
> connected to the common incoming surge source - AC electric. Why were
phones
> not damaged? Where do they have both an incoming and outgoing surge path?
You
> must have both to complete an electrical circuit - else no damage.
Go back and reread it... there were phones lost and they were not phones
with electrical supplies.
But thanks for trying on that one.
Yes, most of the damage was done to their electrical appliances, especially
in the house that was struck. But also note that NONE of my electrical
appliances were damaged in any way in spite of the fact that many were not
on any kind of additional protection. My TVs were on surge suppression, as
were my VCRs, computers, and stereos. However, no electric clocks were
protected, no small kitchen appliances were protected, no large kitchen
appliances (fridge, for example), etc. The only devices that were on
protection for the phone lines were one modem and my phones. One modem was
on power surge protection but was not on phone line protection. That is the
modem that was damaged.
Now, you can argue till your fingers fall off, if you wish.
>
> Because a modem was not damaged does not prove effective protection. It
is
> the nasty little characteristic of surge protection that makes it such an
art.
> Surge damage is easily proven. But effective surge protection is not
always
> easily identified. You don't really know if lightning would have damaged
the
> modem anyway. It could have already found better paths to earth ground
> elsewhere.
That's correct. We can only be sure when protection fails, can't we?
>
> You can prove that surge protection has failed, but proving that it has
worked
> is nearly impossible for laymen. However that 1930 research has proven
(and is
> still repeatedly proven even in IEEE papers) that effective surge
protection
> shunts the surge to earth ground before it can enter the building. One
> excellent example of this is the entire issue of IEEE Transactions on
> Electromagnetic Compatibility of 4 Nov 1998. (One article not about surge
> protection describes the National Lightning Detection Network).
>
> Telcos not only protect the CO using 'whole house' surge protection, but
they
> also protect the customer with similar protection in the premise
interface.
> Correctly noted is that telcos are more interested in CO protection. It
is why
> CO surge protectors have no connection to earth ground - instead surge
> protection is preferably mounted directly ON earth ground. All incoming
lines
> first connect to good earth ground - either directly or through surge
> protectors.
You want to explain that one again? Put a ground on that line and you are
now attached to CO ground which is to be kept separate from earth. A telco
office is grounded to earth but the equipment inside is a closed circuit not
attached to that ground. This is done to protect the equipment from
lightning strikes which run through the building's external ground.
> Homes don't get the same protection. Home phone line surge protection
is not
> as good because there is a green ground wire between the NID and earth
ground.
Which is supposed to shunt off strikes and prevent damage to the phones (and
anyone using them) inside the home. However, it is far from perfect and the
standard warning is to hang up when a storm is in the area. Nothing will
protect you from a close or direct strike but the protection will minimize
damage from strikes that are not direct and not close to the property.
However, there will still possibly be a light surge that can damage
sensitive electronics. This is what I bellieve causes the failures in most
cases.
But you are entitled to disagree.
Hooda
dl
>
>Please explain why you avoid discussion of the 'whole house' surge protection
>routinely installed, for free, on phone lines. Did you know that it exists?
>Does your phone line still use the carbons? Why do you ignore protection
>routinely install by AT&T, before the breakup, even in the 1950s?
A number of years ago, the phone company installed a network interface in the
basement. It is at least 6 feet away from anything that could be considered a
good ground. A few years later, the telco decided the network interface had to
be outside, so they installed a different box outside - again, it's a number of
feet from anything that would make a good ground. I see nothing else inline
that may perform the function of lightning/surge suppression. By good ground, I
am refering to a short, LOW INDUCTANCE path to earth. There may be a surge
suppressor on the pole, but again, the true ground is far away.
With the extremely fast rise times asssociated with lightning induced
transients, I can easily see thousands of volt induced common-mode into the
phone lines for enough time to do damage to connected equipment.
dl
w_tom
green wire to central earth ground. But then some installer still don't have a
clue. For example, a telco employee finally reinstalled the overhead drop (that
lay on the ground since last winter). So he cut off all ground wires (including the
correctly reinstalled CATV ground wire) so that all wires were far from the earth
ground ... but they looked nice with clean sharp bends. Sharp bends - just made the
earth ground connection worse. Longer distance to earth ground just made the earth
ground worse. Tie wrapped ground wires to all signal wires - just coupled any
lightning surge from ground wire directly into the house.
It's not my house. But when I see the disaster he has created, it is all I can do
just to walk away and say nothing. Yes, some telco employees still have the
attitude that lightning protection will never happen so why bother. It is very
frustrating to have people too entrenched in 'what they knew' to accept new
technology - albeit new to homes and to installers, but technology that is older
than every reader.
What is a good earth ground? From http://www.polyphaser.com/ppc_technical.asp
> A lightning ground system should be capable of dispersing large amounts
> of electrons from a strike over a wide area with minimum ground potential
> rise. It should be capable of doing this very quickly.
However this is better explained in
http://www.harvardrepeater.org/news/lightning.html
> ... from personal and broadcast experience spanning 30 years, that you
> can design a system that will handle *direct lightning strikes* on a routine
> basis. It takes some planning and careful layout, but it's not hard, nor is it
> overly expensive. ...
> ... you *must* have a single point ground system that eliminates all ground
> loops. And you must present a low *impedance* path for the energy to go.
> That's most generally a low *inductance* path rather than just a low ohm
> DC path. Lightning energy is RF, and incredible voltages can build up across
> very small inductors when we're dealing with thousands of amps of
> surge current.
Somehow, even in the utility industry, this is too radical for many old timer
expert to appreciate. They still insists that wire is a conductor - even to
lightning. And so we have installers who still destroy effective surge protection
by not installing low impedance (not low resistance) earth grounds because they
never had to do it before.
I point out another bad surge protection practice (be it the wire's length, sharp
bends, or not meeting at a common earth ground). The old electrician just says the
same thing - "but it meets [NEC] code and that's all that matters". How dare we
exceed requirement for human safety to also provide transistor safety. 30 years
after the transistor enters the home and we still don't install effective transistor
protection. Go figure.
In your case, to meet NEC requirements and to address transistor safety, all
ground wires should have been routed short (less than 10 feet), direct (no sharp
bends or wire splices nor passing inside conduits or metal pipe), and independent
(not share any other ground wire until all ground wires meet at the central point
earth ground. All grounds must have a common connection to meet NEC requirements.
That common connection must be at the earth ground to make effective surge
protection. A minimal central point ground is the 8 foot copper rod if in
conductive soil (most definitely not sand). However even some electricians cut
that rod in half so as to get two rods for the price of one, or to make the rod
easier to install. Go figure. No wonder we have so much unacceptable surge
damage.
Ironically we build new homes with vastly superior earth ground - and we don't
even use them. The concrete foundation and wire mess in the basement floor makes
some of the best earth grounds we could ever want. ie.
http://scott-inc.com/html/ufer.htm . But we still don't use them. Go figure.
Franc Zabkar
put finger to keyboard and composed:
>
Actually the LED "trigger" is on the line side of the optocoupler. The
phototransistor switch is on the DSP side.
Franc Zabkar
to keyboard and composed:
> Did the building have a whole house surge suppressor? If not, then a surge enters the
>building on two hot AC wires (assuming the neutral being grounded). Wire is not a
>conductor to electricity. That outlet ground may or may not be earthed depending, for
>example, how far the outlet ground connects . Outlet ground can be farther away from earth
>ground than the phone line's ground. It is simply which ground was easier to obtain AND if
>the building really has an earth ground. Many older buildings don't even have the NEC
>required earth ground (no surge protection would be possible), let alone one sufficient for
>lightning. The phone line in older buildings is simply a easier path to earth, especially
>in those older homes.
In Australia we most often use the MEN system (multiple earthed
neutral). AFAIK, the earth wires from each of the circuits are
soldered to a common earth point. This earth common then connects to
the "general mass of earth" via an earth strap on an exposed section
of an underground water pipe. The common earth point is also wired to
the AC neutral via a terminal block in the distribution box.
ex-Moderator-bu...@faked.com
>dl wrote:
Having lost one Courier V.everything & several phones due to a
proximity strike - all our telco cables are underground in the part of
Hong Kong where I live - I insisted that HK Telco fit suitable
protective devices to both my phone lines - which they did.
As further protection, my two Couriers are ONLY connected to line when
the computers are switched on. I have a multiple contact change-over
relay sitting inside the computer cases - powered by the 12 Volt in
the computer - and also fitted with surge suppressors too.
I had to assemble yet another unit like this for my sister in Scotland
after she lost two modems to lightning close by - and she has to
contend with telco cables in the air.... and Brit telco refused to fit
safety devices.
I certainly won't suggest that my engineering will prove 100%
protection - but it certainly provides a bit of peace of mind if a
thunderstorm brews up when I'm not around to disconnect the Couriers
Now I'm on ADSL and seldom have any reason to use the dial-up Couriers
except to check out the very few remaining Hong Kong BBSes that still
linger on - is there life after BBS mass suicide in the face of
Internet?
w_tom
connection for up to 400 milliseconds. 400 milliseconds is under a time limit that the
CO will monitor for disconnect. However, they take your design one step farther. They
ground the incoming phone line so that a surge does not cross open relay contacts.
Lets take a POTS modem for example. The Off hook relay contacts will isolate up to 200
volts. That means some voltage above 200 volts will cross those open relay contacts.
The Off hook relay also has 500 volt isolation between coil and contacts. Again, a surge
greater than 500 volts can cross from coil to relay contacts.
If you did not ground an incoming line, then a surge could cross relay contacts, or
could cross into the relay coil control circuit or computer +12V. It is why an
isolation relay should not just open the contacts but should also shunt the line side to
nearby earth ground either directly or via surge protection.
You seem to understand that even underground wires carry surges into the building which
is why bulkhead (the single point earth ground window) surge protection is installed:
http://www.polyphaser.com/pdf/PTD1028.pdf
UK modem users have a problem. BT once installed surge protection devices. Today they
refuse. As a result, there is a tremendous surge in modem sales when T-storms ravaged
the country on and after 4 July. Many reports were not just modem damage but of
optocoupler damage. Surges must be tremendous and larger when optocouplers are damaged
(2000+V). But then the UK does not have surge protection on phone lines meaning that
surges can be even more destructive.
Junior Tracy
> least six years and through many thunderstorms with no ill effects. guess
> I was just lucky, and probably due for this little misfortune.
Jared, you have indeed been "Mr. Lucky" until your luck ran out in last
thunderstorm.
Either unplug ALL from power and phone wall outlets, or, if you do not
already have one installed, get a good Surge Protector with a high Joule
rating, a low response time, and a low "pass surge to ground" rating or
whatever it is called.
And be sure to get one through which you pass your incoming phone line, like
a Belken Surgemaster.
.
"Jared Dunn" <jdd...@uiuc.edu> wrote in message
news:Pine.GSO.4.31.01071...@ux5.cso.uiuc.edu...
> On Wed, 11 Jul 2001, w_tom wrote:
>
> thanks for the education, and sorry for my little display of ignorance.
> Kinda odd, I've had various computers connected to phone lines for at
> least six years and through many thunderstorms with no ill effects. guess
> I was just lucky, and probably due for this little misfortune.
>
> I also might have researched this a bit more before asking if I didn't
> have very limited internet access due to the aforementioned problems.
Looks
> like I'm going bargain-modem hunting! And making sure to unplug the damned
> things from the wall next time we have a storm around here.
>
> thanks again,
> J.
>
> -----------------------------------------------------------------------
> "Three passions, simple but overwhelmingly strong, have governed my life:
> the longing for love, the search for knowledge, and unbearable pity for
> the suffering of mankind."
> - Bertrand Russell
>
> Jared Dunn +++++ UIUC Bio/Phil major
> http://suspensionofdisbelief.org/
> http://bluishorange.com/gangbang/
> icq#5740361 aim:jddunn13
>
Junior Tracy
number?
What do you use?
"GoGoGopher" <dah...@charter.net> wrote in message
news:tkvrubb...@corp.supernews.com...
>
> "w_tom" <w_t...@usa.net> wrote in message
> news:3B4F7C0B...@usa.net...
> > Warning: you are about to be buried in technical facts and
> citations.
> >
>
> BLAH, BLAH,BLAH BLAH BLAH...........Agreed, we are buried.
> Jared, the important (on topic) thing is, your modems are ruined.
> Regardless wether or not the offending voltage entered through the
> phone line or the power supply side is immaterial, dead is dead. Good
> luck in your (future) modem purchase.
> Just remember...... destructive voltage potential can enter your
> computer during an electrical storm from many routes, INCLUDING the
> phone line (protection *OR* theory on differential vs common mode
> pulses)
> That is reality, not just something mistakenly interpreted from a
> book, so, if you can, unplug your phone line (and others during a
> storm to protect yourself.
>
>
>
>
Junior Tracy
number?
What, if any, do you use?
"w_tom" <w_t...@usa.net> wrote in message news:3B4E9938...@usa.net...
> You are probably statistically in the ballpark. Surges typically occur
about
> once every 8 years. Therefore you were probably due for some damage.
Junior Tracy
and/or lightning problem as it affects the modem and the computer.
We need a specific recommendation on how to avoid the problem.
What surge protector do you use, if any? Make? Model number?
Same request is made to all other readers.
Spare us the long PhD dissertations.
Make specific recommendations.
"v.Richard" <nos...@808hi.invalid> wrote in message
news:bCG37.526$0s2....@typhoon.hawaii.rr.com...
> I take issue with what W_tom says:
Junior Tracy
I am certain your memory is excellent. I don't care about all this theory. I
want protection.
Specifically, how would you recommend I protect my computer and modem in
Tampa, Florida, which is the southwest anchor of the major lightning belt
that runs northeast across the state through Lakeland and Orlando to Daytona
Beach.
This belt has more lightning days and strikes than anywhere in the
contiguous 48 states. Believe 4 corners in New Mexico runs a not close
second.
Any doubters can check the Tampa Electric company's records.
We got lightning!
UPS? Specs? Make? Model?
Surge protector? Specs? Make? Model?
Like you, in 8 years I've lost only one modem to lightning, and nothing
else.Had same symptoms as one you cited.
I attribute this to the fact that the local power company, Tampa Electric,
has its power lines overhead. The highest conductor is the Ground conductor,
which is grounded at every pole, to a copper plate on top of pole, and there
is a ground wire from every such plate down the pole to a buried ground rod.
IMHO this provides a shielding effect from lightning.
No one on the block has lot any TV's or appliances from lightning storms.
w_tom
surge from one wire to all others. How is that surge protection? The surge
is now on all wires still seeking earth ground. If the Belkin was adjacent to
a powered off computer, then the surge on a black hot wire has simply been
shunted (shorted or diverted) to the green safety wire. Now a surge on the
green wire will seek every path to earth ground. One good earth ground is a
phone line, destructively via motherboard and modem. And yes, I have traced
such damage by repairing parts at the IC level due to a surge shunted by
adjacent surge protection. In fact, some chips were actually cracked where a
surge passed through the IC pin - the surge path of destruction was that
obvious.
IOW I have seen experimental evidence in many different situations that is
consistent with the theories so repeatedly publishes since the 1930s. You
don't even need see these scary pictures of more expensive plug-in surge
protectors: http://www.zerosurge.com/truth.htm
How to identify ineffective, overhyped surge protection? No dedicated line
less than 10 feet to earth ground AND, for some reason, they forget to mention
critical earth ground on the box. No earth ground means no effective surge
protection.
How much was that Belkin unit. $20 or $70 to ineffectively protect only
one appliance? What outrageous overpricing? And how many joules in that surge
protection - another embarrassing question?
Having demonstrated ineffective surge protection - and noting that
recommendations were made without knowledge of how surge protection works or of
accurate specifications - we now address effective surge protection.
Phone lines already have effective surge protection if upgraded in the past
15+ years. To many simply assume a surge entered on the phone line -
conveniently forgetting even what was taught about electricity in elementary
school science. Without a complete circuit, there can be no surge damage.
For AC electric, the source of most destructive surges, Home Depot sells
'whole house' surge protection for less than $50 as Intermatic EG240RC or
Siemens QSA2020. Minimal protection must be 1000 joules and 50K amps.
(Because of where it is located, plug-in surge protection must exceed 3000
joules to provide equivalent life expectancy). Home Depot's effective surge
protection for about $1 per protected appliance vs the Belkin at what... $70
per ineffectively protected appliance? The Belkin is outrageously overpriced,
undersized, AND has no earth ground?
There are numerous other 'whole house' surge protectors with even better
specs. None are plug-in type. However a surge protector is only as effective
as its earth ground. Since we have Belkin and other plug-in surge protector
advocates, then this must be repeated again. A surge protector is only as
effective as its earth ground. That plug-in surge protector is simply too far
from any earth ground to be effective. This is defined by basic electrical
characteristics of wire. But to make it easy, we repeat that a surge shunt
path should be less than 10 feet.
Effective surge protection shunts a surge less than 10 feet to earth ground
before a surge can enter the building. Once inside the building, a surge seeks
earth ground destructively via appliances. Effective surge protection, as
demonstrated since the 1930s, shunts a surge to earth ground before it can
enter a building. That is 'whole house' surge protection. That has been
demonstrated since the 1930s. That is why your telco continues operations
uninterrupted everywhere during every thunderstorm. That is what plug-in surge
protection does not perform - effective surge protection.
The benchmark in surge protection is Polyphaser. Anyone who does not know of
Polyphaser should be suspect as having no surge protection knowledge.
Polyphaser application notes are at:
http://www.polyphaser.com/ppc_technical.asp
Polyphaser sells superior surge protection. How can you tell? First, they
don't preach silly warranties. Second, their application notes constantly
demand good earth ground. They sell surge protection but they insist, even
more so, that better earth grounds be installed. IOW more demand better earth
ground than promote their own products. Therefore you know they have superior
products. That is a benchmark of superior surge protection.
Some background from real world testimony -
http://scott-inc.com/html/ufer.htm
http://lists.contesting.com/_towertalk/199909/0143.html
It is no accident that the poster asked for surge protector recommendations -
and I mostly ignored it. He is making a mistake in thinking the overhyped
surge protector represents superior surge protection. But then I posted the
sentence and it did not sink in. A surge protector is only as effective as its
earth ground. Why do I keep mentioning earth ground many more times than I do
surge protectors? Because I am answering the question more accurately. No
earth ground means no effective surge protection. So what is the brand name in
earth grounds? Some people just insist that more money means better quality
because they cannot see real quality underground. A surge protector is only as
good as its earth ground - to beat the horse bloody.
w_tom
Kmart, Sears, Lowes, or Walmart, well, a surge protector is only as good as
its earth ground. Does it have a dedicated connection less than 10 feet to
earth ground. Does it meet the minimal 1000 joules and 50K amps for
residential AC electric? No effective surge protection is sold there.
Previously noted were two products for residential AC electric sold at Home
Depot from less than $50 - although the model numbers were not provided:
Intermatic EG240RC and Siemens QSA2020.
Other possible whole house surge protectors are:
http://members.home.net/kapland/surge.htm
http://www.dale-electric.com/ditek.htm
http://www.deltala.com/prod02.htm
http://www.ch.cutler-hammer.com/surge/products/chsp.html
http://members.tripod.com/~StorminProtection/index-31.html
http://www.mimcv.com/residential.html
I am disappointed in poor specs of GE's THQLSURGE and Sutton Design's
product. Also older Square D products were embarrassingly poor. But Square D
recently markets a complete whole house system that integrates AC electric
with phone, CATV, etc.
Of course the other 'system' component must also be installed. Even the
most trivial surge protector can provide some protection, as I have previously
demonstrated with only a 39 joule Radio Shack MOV. Why? The direct strike
was shunted only 2 feet to earth ground. And that is the point. The surge
protector does not make the surge protection 'system'. In that "PhD
dissertation", the point was repeated - but so often overlooked maybe because
it was not a product on some store shelf - EArth GRound. If you want better
surge protection, your questions should have been how to improve earth
ground. It was not a dissertation. It kept telling you where the heart of
a surge protection lies - underneath earth's surface. It has no make or model
number.
Hooda Gest
you checked for an additional ground cable coming from your telco interface
box down to a ground rod? Next, check for a proper ground in your home
wiring. You can do this with a simple device from Radio Shack. If both of
these check out, all you need is any decent surge suppressor by APC or any
of the others. The cost is anywhere from $15 to more than $50, it's up to
you what you wish to spend.
Hooda
"Junior Tracy" <joh...@mindspring.com> wrote in message
news:9jijcd$brf$1...@slb5.atl.mindspring.net...
w_tom
Visual inspection is a valid test of earth ground. That would be visual inspection of the household central earth ground AND of the earth ground on pole transformers. Connections must be firm. The ground rod must not shake in earth. Ground soil must be conductive. Grounding in sand or gravel material is all but no earth ground. Also inspect earth ground to utility's pole transformer and overhead wire. That pole ground is also critical to effective household surge protection.
Equipment to measure for earth ground- try
http://www.leminstruments.com
, or this $200 device from
http://www.radioshack.com/product.asp?catalog%5Fname=CTLG&category%5Fname=CTLG%5F008%5F002%5F005%5F000&product%5Fid=910%2D5682
(not sold in Radio Shack stores).
These testers will be well beyond the budget and needs of most readers,
only tests the actual earth ground, and does not test for household connections
to the earth ground rod. Effective test for earth ground will
be visual inspection.
As for effective surge protection, the list was posted previously:
Intermatic EG240RC at Home Depot
Siemens QSA2020 at Home Depot
http://members.home.net/kapland/surge.htm
http://www.dale-electric.com/ditek.htm
http://www.deltala.com/prod02.htm
http://www.ch.cutler-hammer.com/surge/products/chsp.html
http://members.tripod.com/~StorminProtection/index-31.html
http://www.mimcv.com/residential.html
http://www.elect-spec.com/wire_in2.htm
That overhead electric utility ground wire, again, is old technology discussed in a Franklin Institute paper published in Sept 1924. Again, effective surge protection techniques - especially the function of critical earth ground, are that old and that well proven.
GoGoGopher
<ex-Moderator-buy/sell-HK...@faked.com> wrote in message
news:3b58fa3e$1...@newsgate.hknet.com...
>
> As further protection, my two Couriers are ONLY connected to line
when
> the computers are switched on. I have a multiple contact
change-over
> relay sitting inside the computer cases - powered by the 12 Volt in
> the computer - and also fitted with surge suppressors too.
>
A very interesting idea, would you please email me any specs. you'd
care to share?
Thanks,
Bob
GoGoGopher
<snip>
> Visual inspection is a valid test of earth ground.
You are so wrong,.... the only valid test of earth ground in an
electrical system is a comparative analysis resistance check between
the entrance ground buss connection to earth. The dinky little Radio
Shack meter you recommend (as an extravagance, no less) is only
applicaplable in monopole earth systems, NOT the tripole systems that
have been required in my area (and most of the rest of the USA....
where do YOU live?) for the past ten years. Being the grounding expert
that you proclaim to be, one would expect you to be aware of this.
<snip>
> That pole ground is also critical to effective
> household surge protection.
<snip>
Why? According to your own proclamations, (over, and over, AND OVER)
any grounding beyond ten feet is totally immaterial concerning surge
protection. Make up your mind.
You also state in a previous post that low cost suppressors only
ground (shunt) to *all* the other "wires" in the electrical
system...... thereby (according to you) providing no path to ground.
Huh? The ground and neutral wires aren't paths to ground?
In another post you state that in "older" systems (without a dedicated
"safety" ground wire) that there is no path to ground in such systems,
again, Huh?.....What do you call the neutral wire of a two wire
system???? Go outside and look at where the neutral wire goes in a two
wire system....... straight to earth at the poll! Granted, the
aforementioned systems do not qualify as the most *effective*
grounding methods concerning surge suppression as would be desirable,
but they are nonetheless GROUNDED, and do provide some measure of
protection. Most people cannot afford the type of protection you
recommend, as in the "pennies per appliance" scenario you suggest for
total service entrance protection. That is total bullshit. Maybe back
in 1930 it would cost pennies, but not in 2001. When was the last time
you called an Electrician out to rewire your service entrance? Most
of us cannot afford to have shunting to ground within the ten foot
requirement you stipulate......unless we drive a ground rod through
the floor next to our computer.
According to your charts and graphs....... the Telco's of the world
provide impenetrable defense from electrical spikes at our homes
(including direct lightning strikes), you obviously haven't dealt with
Ameritech, neverless you say.......unstopped surges MUST originate via
the AC supply,... never through the phone lines........Damage to my
modem due to power surges must pass through my computer's open power
switch (that has an interrupting rating of 10,000 volts), through it's
power supply, across the motherboard, (and all it's associated
peripherals), jump across the PCI (and all it's associated
peripherals), most of which are rated @ 5 volts+/- or less, and
destroy the modem aspect that is rated @ 100 volts (ring circuit)
without damaging any of the computer's circuits? Give me a break!
According to you , you are the only one who understands basic AC-DC
fundamentals in this group.......as you have repeatedly pointed out.
Also, you are the only one privy to the "secrets" of surge protection
originated in the 1930's, forgotten by all, and known to no one but
you. Thank you . I honestly feel you are a very knowledgeable
person.........I also feel you are full of shit (and full of yourself)
The trick to getting your point across is not in impressing your
audience with your vast knowledge,(we all know you're smart) but
rather using one's vast knowledge to explain what you know in terms
that we can understand, minus the bullshit we can all see through.....
(you don't realize this aspect)
Bob