PC shuts down by itself following power surge incident

Discussion in 'Computer Support' started by Jimmy Dean, Mar 24, 2005.

  1. Jimmy Dean

    w_tom Guest

    Did we not learn from WMDs? Feelings should supersede
    facts? Facts even exist in your own town. The $multi-million
    telephone switching computer, connected to overhead wires
    everywhere in town, shuts down with every approaching
    thunderstorms? The expensive radio station stops broadcasting
    when thunderstorms approach? That if protection from direct
    lightning strikes is not possible. How do you 'feel' a
    conclusion when examples around you say otherwise? Protection
    from lightning is routine, proven even before WWII, and
    inexpensive. Provided were even sources of effective
    protection from manufacturers of credibility. And still you
    'feel' otherwise?

    The mistake should be glaringly obvious. You had modem
    damage because the human failed to install effective
    protection. Defined was effective protection. Do you have
    it? Of course not. Therefore modem damage occurred. We
    install protectors to eliminate the most destructive transient
    - lightning. Find a technically responsible source that says
    otherwise - and good luck. There is no fact to justify your
    'feelings'.

    http://www.harvardrepeater.org/news/lightning.html
    Do you also 'feel' quality is found in higher price? Pay
    almost $100 for a product from Monster Cable and get no
    effective protection. Pay about $1 per appliance and get
    protection what has been proven effective from direct
    lightning strikes. You are assuming some kind of relationship
    between price and quality that is promoted by myth.

    Take a $3 retail power strip. Add some $0.10 parts. Sell
    it as a surge protector for $15 or $50. You tell me where
    price demonstrates quality. The more expensive plug-in surge
    protectors are so poor as even demonstrated by insufficient
    joules - a number.

    You have also 'felt' a power supply problem exists. No one
    can know without numbers. And that is the point. Numbers
    prove effective protection is installed for lightning.
    Numbers may also identify your intermittent. If you feel the
    power supply is defective or that "a direct lightning strike
    - I do not believe anything really protects from that", then
    apply to the Academy for the Advancement of Junk Science.

    In a world of reality, damage from lightning is so routinely
    avoided as to be considered human failure. If lightning
    damaged that power supply, then suspect a garbage power supply
    marketed to those who typically have insufficient electrical
    knowledge - computer assemblers. Notice what is provided with
    facts. Numbers. Numbers from a multimeter would also report
    in only two minutes whether a power supply is good or bad.
    More time was spent posting 'feelings' than it takes (only two
    minutes) to prove a power supply good or bad.
     
    w_tom, Mar 27, 2005
    #21
    1. Advertisements

  2. Jimmy Dean

    w_tom Guest

    I am assuming you are in Iran where this question has
    immediate relevance. What we do to avoid lightning damage is
    but some of what is done for NEMP. Best to consult benchmarks
    in surge protection for further information that is beyond the
    scope of this discussion:
    http://www.polyphaser.com/kommerce_productdata.aspx?search_val=emp&x=38&y=5

    Background information:
    http://www.geocities.com/CapeCanaveral/5971/emp.html

    Can protection be installed. Its been done routinely for
    generations. After all, how do you think those nuclear tests
    are performed and electronically measured?
     
    w_tom, Mar 27, 2005
    #22
    1. Advertisements

  3. Jimmy Dean

    Top Guest

    You assume incorrectly. I'm retired from the US Army and was in
    communications. I taught a block of instruction about
    protecting from EMP in a tactical environment. I was never
    convinced that what I taught was really practical. If I am
    expecting a nuclear blast I'm sure I would have time to bury
    signal cables 3 feet under ground. That's a bit simplistic but
    it makes my point.

    Top
     
    Top, Mar 27, 2005
    #23
  4. Jimmy Dean

    w_tom Guest

    Even burying cables underground does not provide transient
    protection. If a facility does not have effective lightning
    protection, then (in simple terms) it will not have effective
    EMP protection. Buried cables are also a source of
    destructive lighting transients as demonstrated by figure in
    this application note from a serious industry professional:

    http://www.erico.com/public/library/fep/technotes/tncr002.pdf

    We know how to protect from lightning since such transients
    occur typically once every 2 sq kilometers per year.
    Telephone switching centers with overhead wires everywhere in
    town suffer direct strikes often and without damage. The
    technology was proven even in 1930s research papers. But
    NEMP? How often does the facility get tested? Experimental
    evidence is difficult to obtain and (again) beyond the scope
    of this thread.

    Early on, research concentrated more on the protectors. But
    as protection has been better understood, earthing and single
    point grounds are more the focus of effective protection.
    What would have protected that modem? 'Whole house' protector
    on AC electric connected short to the buildings single point
    earth ground. Protector is only as effective as its earth
    ground.
     
    w_tom, Mar 27, 2005
    #24
  5. Jimmy Dean

    w_tom Guest

    A primary protection system is installed for AC electric
    shunts to earth. But you should visually confirm this system
    is intact:
    http://www.tvtower.com/fpl.html

    Those are upstream shunts. You also need an earthing shunt
    for your building (layered protection). But no requirement
    exists for shunts on household AC mains; the secondary
    protection system. Three wires carry AC power into a
    building. One (neutral wire) must connect to earth ground for
    human safety. And that is the problem. Code is only
    concerned with human safety; not transistor safety. Without
    maybe enhancing that earthing system and without installing a
    'whole house' protector, then those two incoming AC electric
    (hot) wires have no earthing shunts. They are a perfect path
    for a transient to find earth ground, destructively, via
    household appliances.

    One final point. Inspect that electrician installed earthing
    wire. The code demands earthing for human safety reasons.
    Some electricians don't understand enough to go beyond code
    requirements. They don't understand transistor safety. For
    example, the earthing wire from breaker box to earth ground
    must be routed separate from other non-grounding wires, no
    sharp bends, no splices, and as short as possible. A old
    expression: a neat wiring job (with clean square turns and
    wires bundled together using nylon ty wrapps) is inferior
    earthing. You want a surge to take the shortest, straightest,
    and independent wire path to earth ground. Again, this is not
    required by a 'human safety' driven code; therefore
    electricians do not appreciate the concept. Electricians need
    not wire for transistor safety.

    Bottom line answer is no, there typically is no shunt to
    earth for AC electric. Unfortunately home owner must learn
    these complexities to install what has been essential to
    transistor safety for 30 years.
     
    w_tom, Mar 28, 2005
    #25
  6. Jimmy Dean

    Millimeter Guest

    Sorry to be late in the thread but, what are you on about?



    If this is true then you should be able to define which point in the
    path should fail first.
    Assuming that all adapter cards are in paralell rather than inseries,
    then one would expect that each card would complete a seperate
    circuit.
    Me thinks that the "less than an inch" of clearance between the mobo
    and case would offer little resistance to the "destructive transient"
    in crossing the air barrier to the grounded case.
    I have witnessed a lightening strike on an older dial telephone. In
    lucky cases it will overpower the earpiece and blow the magnet out.
    In severe cases the carbon can actually ignite from the emf generated
    by the lightening.
    In truth, you should go back to 2nd grade and restudy the lightening
    class which you apparently slept through also. The electrons do not
    actually travel from cloud to earth, they travel from earth to cloud.
    It is simply a case that we see the light which is closer to us sooner
    that it appears to travel towards the earth, rather than away from it.

    The telco lines would offer a lesser resistance than an AC line. The
    former operates at a low voltage whereas the AC lines on the street
    would operate in the thousands of volts which could offer more
    competition to the lightening, than would a telco drop.
    So you are suggesting that certain versions of Windows may be aware
    that a lightening incursion had occured?

    I do however agree somewhat with your following testing procedures, if
    one has a multimeter. Otherwise, swapping in "known good parts" is a
    reasonable expectancy for any computer diagnosis. OP is dealing at
    the component level and not at the board level, IMO.


    Hope this helps and doesn't offend,
    Millimeter
     
    Millimeter, Apr 5, 2005
    #26
  7. Jimmy Dean

    w_tom Guest

    I am not sure which electrical principle is confusing the
    issue. For example, in a reference to the surge circuit path:
    In an earliest post, one classic component in the DAA
    section was defined as typically damaged first; after
    lightning has first formed a complete current path from cloud
    to earth. A path that must exist as required by the
    definition of electricity - as taught in 2nd grade science.
    This modem is but one where that DAA component was damaged by
    lightning, replaced, and the modem works fine for years.
    First traced out the surge path. Find component(s) in that
    electric circuit that was damaged or overstressed. Modem (or
    network card, or fax, or credit card machine, etc) then work
    just fine. But first one must locate both the incoming and
    outgoing path of a transient. Shotgun repairing methods don't
    demonstrate why such failures occur. No incoming and outgoing
    paths means no electricity flow - no destructive transient -
    as demonstrated by locating and replacing the individual
    electrical parts.

    We do this type of repair not to save money. We do it to
    learn why failures happen.

    Does electricity go cloud to earth or from earth to cloud?
    Irrelevant as basic circuit theory demonstrates. But then
    lightning also exists in both positive and negative type.
    Both do same damage. Both first form a complete electrical
    circuit from cloud to earth. When excessive current is
    flowing through everything in that circuit, then something
    fails. What fails? Often a component that is trying to
    "stop, block, or absorb" a transient that cannot be "stopped,
    blocked, or absorbed" - lightning. Why? Another answer made
    obvious from basic circuit theory.

    An assumption about which way electrons go is completely
    irrelevant. First, lightning does not generate a DC
    transient. DC transients do not exist. Lightning is AC which
    is why, for example, long wires do not effectively earth
    protectors. A long earthing connection to a protector (ie via
    wall receptacles) means no effective protection. The key word
    in point one is impedance.

    Comments about sleeping through 2nd grade science are
    incendiary as well as irrelevant. Second, claiming that
    electrons travel from cloud to earth assumes the transient is
    some kind of ocean wave that will crash on a beach.
    Electrical transients do no flow as ocean waves - contrary to
    popular myth. But again, that point two was explained in
    above previous posts.

    Even stating "telco lines would offer a lesser resistance"
    suggests insufficient comprehension of basic electrical
    concepts. Wire impedance - not resistance - is essential to
    understand transient protection. But then these are
    transients - and not a DC current flow of electrons from cloud
    to earth. Destructive transients are AC. Just some of the
    basic electrical concepts contradicted in popular myths about
    lightning damage.

    I have no idea what this means:
    I know of nothing in basic electrical circuit theory that is
    defined by that paragraph. What competition? Where in basic
    circuit theory does 'competition' exist?

    Lightning is electricity that seeks earth ground -
    destructively as necessary. To obtain that path, voltage will
    rise, as necessary, to damage anything that might "stop,
    block, or absorb" that electricity flow. In a term from basic
    circuit theory, it is called a current source.

    A transient either will find a destructive path through
    household transistors OR will be harmlessly earthed before
    those transistors can be found - become part of a destructive
    circuit. Posted above were technical reasons why effective
    'whole house' solutions are installed for less money.
    Protectors are only as effective as the earth ground which is
    even why wire impedance (not wire resistance) is so critical
    to effective protection. And why which electrons flow
    direction between cloud and earth is totally irrelevant.
    Protection has always been about earthing as defined in
    numerous earlier posts in this thread starting and after 24
    Mar 2005.
     
    w_tom, Apr 5, 2005
    #27
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.