After Lightening Storm - Computer Will Not Turn on

"w_tom" <w_tom1@usa.net> wrote in message
news:4c82cf30-116c-4937-9d59-83e1ee8d69f7@x29g2000prd.googlegroups.com...
On Jul 23, 8:02 pm, "S.Lewis" <Gossa...@interesting.com> wrote:
> Depends on the system. I've seen them where the dial-up modem (card)
> itself
> was only fried; then again I've seen the slot also fried......

We routinely would fixed electronics by identifying damaged parts
for each board, replacing those parts, and never having future
failures. Properly noted is an effect called 'overstress' where parts
fail later. Same analysis is required on aerospace hardware so that
failure never happens. IOW trace excessive currents by analysis,
replace any part that may have exceeded manufacturer specs, and
eliminate overstress failures.

How are modems most often damaged? First surge circuit must be
identified. A most common path through modems is incoming on AC
electric (bypassing protection that is inside all power supplies),
through motherboard, through modem, then out to earth ground via phone
line.

A most common modem failure (the weakest point in that circuit) is a
PNP transistor that drives its off-hook relay.

In another situation, two plug-in protectors earthed a surge
destructively via adjacent and powered off computers. We traced the
surge path to earth ground. Protectors connected a surge into both
motherboards (bypassing the power supply). Outgoing on NIC. Through
network to a third computer. Out that computer to earth ground via a
dialup modem. In this case, various network interface chips and the
modem suffered damage. All parts in those paths replaced. Computers
never failed again. But again, overstress was made irrelevant by
tracing a surge's circuit.

We know some components will not be surge damaged. For example,
memory has an incoming path from motherboard. Where is the outgoing
path? None exists. So memory boards are not surge damaged. Essential
for damage is both an incoming and outgoing path. Without both, then
damage and overstress will not happen.

Damage is not capricious. But surges appear capricious without
sufficient konwledge. Surges do not enter via power supplies - due to
numerious layers of protection. But plug-in protectors can divert a
surge around that supply; directly into a motherboard.

Effective protection has always been to divert energy into earth
before a surge can enter the building. Incoming utility wires (ie
telephone, cable TV) have surge protection connected to earth. Surges
typically do not enter via these utilities. Most destructive surges
enter on AC electric. To have damage, a surge forms a connection to
earth. Outgoing connection to earth include telephone and cable TV.
Anything in that path can suffer damage or overstress.

I think one of the most important things you stated was "to earth". The
grounding of many homes is not what is should be. It is not unreasonable to
add an extended (military standard is 3 meters below ground level) grounding
rod. I've seen may instances where ground was a 6-12 inch piece of rusted
rebar planted in some extremely dry soil. Hardly any ground at all.

Also in some areas the grounding cable you see coming down the side of the
power pole is only bottom grounded, the wire ouny runs to the bottom of teh
pole, about a meter down. In very dry conditions this is also inadéquate
ground to stop a high voltage transient from seeking a more expedicious path
to earth.

On Jul 24, 9:22*am, "Kevin Childers" <kchil...@mail.win.org> wrote:
> I think one of the most important things you stated was "to earth". *The
> grounding of many homes is not what is should be. *It is not unreasonable to
> add an extended (military standard is 3 meters below ground level) grounding
> rod. *I've seen may instances where ground was a 6-12 inch piece of rusted
> rebar planted in some extremely dry soil. *Hardly any ground at all.

Proper earthing means meeting and exceeding post 1990 National
Electrical Code. A 12 inch piece of metal is not earthing. Code
demands at least one 3 meter ground rod. In some locations, that is a
massive and sufficient earthing. In other geologies, that earthing is
not sufficient. So builders exceed what is required by code so that
homes have surge protection. Ufer grounds:
http://members.aol.com/gfretwell/ufer.jpg

Essential is that ground at the base of every transformer.
Discussed above was secondary protection. Homeowners should inspect
their primary surge protection:
http://www.tvtower.com/grounding_and_bonding.html

Properly noted: sometimes this utility pole earthing is not
sufficient. That earthing was upgraded as part of a solution in a
Nebraska radio station:
http://www.copper.org/applications/e.../nebraska.html

On Thu, 24 Jul 2008 07:35:53 -0700 (PDT), w_tom <w_tom1@usa.net>
wrote:

>On Jul 24, 9:22*am, "Kevin Childers" <kchil...@mail.win.org> wrote:
>> I think one of the most important things you stated was "to earth". *The
>> grounding of many homes is not what is should be. *It is not unreasonable to
>> add an extended (military standard is 3 meters below ground level) grounding
>> rod. *I've seen may instances where ground was a 6-12 inch piece of rusted
>> rebar planted in some extremely dry soil. *Hardly any ground at all.
>
> Proper earthing means meeting and exceeding post 1990 National
>Electrical Code. A 12 inch piece of metal is not earthing. Code
>demands at least one 3 meter ground rod. In some locations, that is a
>massive and sufficient earthing. In other geologies, that earthing is
>not sufficient. So builders exceed what is required by code so that
>homes have surge protection. Ufer grounds:
> http://members.aol.com/gfretwell/ufer.jpg
>
> Essential is that ground at the base of every transformer.
>Discussed above was secondary protection. Homeowners should inspect
>their primary surge protection:
> http://www.tvtower.com/grounding_and_bonding.html
>
> Properly noted: sometimes this utility pole earthing is not
>sufficient. That earthing was upgraded as part of a solution in a
>Nebraska radio station:
> http://www.copper.org/applications/e.../nebraska.html


I'm not an electrical engineer nor consider myself knowledgeable in
grounding. I can only say that beside my garage just on the outside
of the electric circuit breaker box in the garage, I have a cable
attached to a copper rod and I think the rod is about 10' or so deep
(not sure exactly how deep so it could be 12). I have to check the
top of the rod sticking outa the ground because I once saw the cable
not attached (don't know if this means I had NO grounding or less??).
I fixed that of course right away but I had already gone thru a storm
or two before I noticed it (it's not something you tend to think about
right away).

w_tom wrote:
> On Jul 23, 8:02 pm, "S.Lewis" <Gossa...@interesting.com> wrote:
>> Depends on the system. I've seen them where the dial-up modem (card) itself
>> was only fried; then again I've seen the slot also fried......
..
Excellent information on surges and surge protection is in a guide from
the IEEE at:
<http://www.mikeholt.com/files/PDF/LightningGuide_FINALpublishedversion_May051.pdf>

And a less technical guide from the NIST at:
http://www.nist.gov/public_affairs/p.../surgesfnl.pdf
..
> How are modems most often damaged? First surge circuit must be
> identified. A most common path through modems is incoming on AC
> electric (bypassing protection that is inside all power supplies),
> through motherboard, through modem, then out to earth ground via phone
> line.
..
Similarly the NIST guide, using US insurance information, indicates
equipment most frequently damaged by lightning is
computers with a modem connection
TVs, VCRs and similar equipment (presumably with cable TV
connections).
All can be damaged by high voltages between power and signal wires.

Currents to earth from a strong surge will lift the 'ground' at a
building above 'absolute' ground. The protection from voltage between
power and signal wires is to keep their ground references together. That
requires a *short* ground wire from the phone/cable entry protectors to
the ground at the power service. An example of a ground wire that is too
long is in the IEEE guide starting pdf page 40. (The guide says in that
case "the only effective way of protecting the equipment is to use a
multiport [plug-in] protector.")
..
> In another situation, two plug-in protectors earthed a surge
> destructively via adjacent and powered off computers. Outgoing on NIC. Through
> network to a third computer.
etc

If you RTFM for any competent plug-in surge suppressor it will say all
interconnected equipment needs to be connected to the same plug-in
suppressor, or interconnecting wires need to go through the suppressor.
External connections, like phone, also need to go through the suppressor.

That is because connecting all wiring through the suppressor prevents
damaging voltages between power and signal wires. The suppressor clamps
the voltage on all wires to the common ground at the suppressor.

Both the IEEE and NIST guides say plug-in suppressors are effective.

--
bud--

And this explains why the surge worked. ;>) ... Ben Myers

On Thu, 24 Jul 2008 14:07:31 -0500, bud-- <remove.budnews@isp.com> wrote:

>w_tom wrote:
>> On Jul 23, 8:02 pm, "S.Lewis" <Gossa...@interesting.com> wrote:
>>> Depends on the system. I've seen them where the dial-up modem (card) itself
>>> was only fried; then again I've seen the slot also fried......
>.
>Excellent information on surges and surge protection is in a guide from
>the IEEE at:
><http://www.mikeholt.com/files/PDF/LightningGuide_FINALpublishedversion_May051.pdf>
>
>And a less technical guide from the NIST at:
>http://www.nist.gov/public_affairs/p.../surgesfnl.pdf
>.
>> How are modems most often damaged? First surge circuit must be
>> identified. A most common path through modems is incoming on AC
>> electric (bypassing protection that is inside all power supplies),
>> through motherboard, through modem, then out to earth ground via phone
>> line.
>.
>Similarly the NIST guide, using US insurance information, indicates
>equipment most frequently damaged by lightning is
> computers with a modem connection
> TVs, VCRs and similar equipment (presumably with cable TV
>connections).
> All can be damaged by high voltages between power and signal wires.
>
>Currents to earth from a strong surge will lift the 'ground' at a
>building above 'absolute' ground. The protection from voltage between
>power and signal wires is to keep their ground references together. That
>requires a *short* ground wire from the phone/cable entry protectors to
>the ground at the power service. An example of a ground wire that is too
>long is in the IEEE guide starting pdf page 40. (The guide says in that
>case "the only effective way of protecting the equipment is to use a
>multiport [plug-in] protector.")
>.
>> In another situation, two plug-in protectors earthed a surge
>> destructively via adjacent and powered off computers. Outgoing on NIC. Through
>> network to a third computer.
>etc
>
>If you RTFM for any competent plug-in surge suppressor it will say all
>interconnected equipment needs to be connected to the same plug-in
>suppressor, or interconnecting wires need to go through the suppressor.
>External connections, like phone, also need to go through the suppressor.
>
>That is because connecting all wiring through the suppressor prevents
>damaging voltages between power and signal wires. The suppressor clamps
>the voltage on all wires to the common ground at the suppressor.
>
>Both the IEEE and NIST guides say plug-in suppressors are effective.

I can guarantee you that a bad modem or network card will not prevent a
computer from powering up. This is totally ludicrous.

James
-------------------------------




It will be one of three places where the machine took a hit:

1. The modem. This seems to be the most common place, and you will
most likely see any of burned, blown or bubbled components or board
traces upon inspection. You should remove it, and if you get lucky,
the computer won't be harmed at all.

2. The network interface (if hooked up directly to an internet
connection device, such as a cable modem or DSL box). If it's on the
motherboard, and the system won't power up, that system board is done.
A network add-in card can be removed much as the modem.

3. The power supply. Power supplies are supposed to shut down
completely and not work if something fails or is damaged inside them.
Whether or not this happens depends upon the quality of the supply,
its age and at least some luck. If the supply didn't shut down
properly upon failure, it could have killed the motherboard.

To diagnose this properly, you will need basic hand tools and a good
look at the expansion cards or motherboard. Testing the power supply
will require a power supply tester, but it's very likely to be fine
especially since the computer was "off" at the time.

William

On Fri, 25 Jul 2008 04:16:41 -0400, "James" <jnipperxxx@nospamfdn.com>
wrote:

>I can guarantee you that a bad modem or network card will not prevent a
>computer from powering up. This is totally ludicrous.
>
>James
>


I think you are right about an external modem but are you sure about a
bad network card if not removed? I don't have this experience that's
why I ask. Of course if I knew I had a bad network card, I'd just
remove it and reboot without it.

On Jul 25, 5:51*am, RnR <rnrte...@gmail.com> wrote:
> On Fri, 25 Jul 2008 04:16:41 -0400, "James" <jnipper...@nospamfdn.com>
> wrote:
>
> >I can guarantee you that a bad modem or network card *will not prevent a
> >computer from powering up. * This is totally ludicrous.
>
> >James
>
> I think you are right about an external modem but are you sure about a
> bad network card if not removed? I don't have this experience that's
> why I ask. *Of course if I knew I had a bad network card, I'd just
> remove it and reboot without it.

The PCs I worked on (with internal dial-up modems) that took a
hit...shorted (or killed) the supply voltage. And most definitely
would not boot with that card in.
The insulation or dielectric breaks-down from the induced surge.
If it burned or exploded components to the point of non-
conduction...then it still could boot.

"James" <jnipperxxx@nospamfdn.com> wrote in message
news:gbSdnVYb1J91FhTVnZ2dnUVZ_judnZ2d@comcast.com. ..
>I can guarantee you that a bad modem or network card will not prevent a
> computer from powering up. This is totally ludicrous.
>
> James
> -------------------------------
>



Uhm, wrong sir. Not uncommon at all to cause a system to appear "dead" (ie -
no POST but with power LEDs).


Stew

On Fri, 25 Jul 2008 11:33:45 -0500, "S.Lewis"
<Gossamer@interesting.com> wrote:

>
>"James" <jnipperxxx@nospamfdn.com> wrote in message
>news:gbSdnVYb1J91FhTVnZ2dnUVZ_judnZ2d@comcast.com ...
>>I can guarantee you that a bad modem or network card will not prevent a
>> computer from powering up. This is totally ludicrous.
>>
>> James
>> -------------------------------
>>
>
>
>
>Uhm, wrong sir. Not uncommon at all to cause a system to appear "dead" (ie -
>no POST but with power LEDs).
>
>
>Stew
>

Stew, I also thought like you because I thought a long time ago I
read something to the effect but without facts I didn't want to say
anything. You just confirmed what I was thinking tho. thanks.