Computer toasted?

Discussion in 'Computer Support' started by mike4ty4@yahoo.com, Oct 27, 2006.

  1. Guest

    wrote:
    > "pengulin" <> wrote:
    >
    > >Since you have had trouble turning it on - at least on occaision -- in the
    > >past, did you try shorting the pins on the motherboard for the power switch?
    > >You've replaced the power supply, but that doesn't mean the switch couldn't
    > >be bad as well.
    > >

    >
    > To follow up, also test the power supply and see if it works.
    >
    > "While we're at it, you can power up an ATX power supply by shorting
    > the green wire with any black ground wire. "
    > http://www.bluemax.net/techtips/ATXPowerSupplyWiring/ATXPowerSupplyWiring.htm
    > --
    > http://nonk.nonk.info/pages/guys_caught_staring/guy_caught_staring_07.jpg


    You sure that's safe? The integrated ethernet light is on, which means
    the power
    supply is generating a voltage. I even tried connecting the old one
    which works at
    least intermittently, and it still wouldn't start.
     
    , Oct 28, 2006
    #21
    1. Advertising

  2. Guest

    wrote:

    >
    > wrote:
    >> "pengulin" <> wrote:
    >>
    >> >Since you have had trouble turning it on - at least on occaision -- in the
    >> >past, did you try shorting the pins on the motherboard for the power switch?
    >> >You've replaced the power supply, but that doesn't mean the switch couldn't
    >> >be bad as well.
    >> >


    >> To follow up, also test the power supply and see if it works.
    >>
    >> "While we're at it, you can power up an ATX power supply by shorting
    >> the green wire with any black ground wire. "
    >> http://www.bluemax.net/techtips/ATXPowerSupplyWiring/ATXPowerSupplyWiring.htm
    >> --
    >> http://nonk.nonk.info/pages/guys_caught_staring/guy_caught_staring_07.jpg


    >You sure that's safe? The integrated ethernet light is on, which means
    >the power
    >supply is generating a voltage. I even tried connecting the old one
    >which works at
    >least intermittently, and it still wouldn't start.


    Ok, If I were having the trouble your having getting a computer to
    work, I'll put the mother board on a piece of glass (non conductive)
    hook everything up on the table (I have a glass table <G>) and see if
    it works there, change things out until it does. then install it into
    the case.

    As for the ethernet light, there are many voltages a power supply
    supplies.
    --
    http://nonk.nonk.info/pages/guys_caught_staring/guy_caught_staring_07.jpg
     
    , Oct 28, 2006
    #22
    1. Advertising

  3. w_tom Guest

    Well at the very start, you still have no reason to believe the
    original power supply failed. Demonstrated is why we don't replace
    things until first collecting facts. Two minutes with a 3.5 digital
    multimeter would have condemned or exonerated the original supply.
    However, having subjected the system to so much 'surgery', no one can
    be sure even what the original problem was.

    Did CPU overheat? I don't even see the CPU listed. It if was an
    Intel CPU, then no problem.

    Having changes so many things, it is now time to first establish what
    is and is not working. Even an illuminated light does not report a
    good supply.

    Get a meter to learn what is wrong, immediately, and before replacing
    anything. Solve a problem the first time. Procedures to take such
    measurements were detailed previously in
    microsoft.public.windowsxp.general on 7 Jun 2006 entitled "Dead
    computer" at
    http://tinyurl.com/qcvuq

    Numbers from those measurements then mean replies will have better
    credibility.

    Foundation of every computer is its power supply system. Notice the
    word 'system'. Power supply is only one 'system' component. Get those
    numbers. In your case, first numbers from green gray wires are
    important both before and after power switch is pressed.

    Disturbing is a power switch tested with a battery. Without limiting
    current, that battery could have harmed switch contacts. Just another
    reason why computer repairing means a screwdriver and the essential 3.5
    digit multimeter. In fact, that two minute test with a multimeter
    would have identified switch as good - without disconnecting anything -
    without adding more unknowns to the problem by moving wires. Fixing
    things right the first time is not found in 'replace this, then replace
    that' techniques - also called shotgunning. First collect facts
    without changing anything. Takes less time. Finds and fixed the
    problem a first time. Does not end up curing only symptoms. System
    does not fail later due to too many changes and other human created
    failures.

    wrote:
    > You sure that's safe? The integrated ethernet light is on, which means
    > the power
    > supply is generating a voltage. I even tried connecting the old one
    > which works at
    > least intermittently, and it still wouldn't start.
     
    w_tom, Oct 28, 2006
    #23
  4. Guest

    w_tom wrote:
    > Well at the very start, you still have no reason to believe the
    > original power supply failed. Demonstrated is why we don't replace
    > things until first collecting facts. Two minutes with a 3.5 digital
    > multimeter would have condemned or exonerated the original supply.
    > However, having subjected the system to so much 'surgery', no one can
    > be sure even what the original problem was.
    >


    How would I test it with the meter? Would I have to open it up and poke
    around in there, what with all that voltage? That sounds kind of
    dangerous.
    Would it be safe to do that? Should I risk? But replacing the original
    power
    supply did stop the spotaneous reboot problem, so something must have
    been wrong with it.

    > Did CPU overheat? I don't even see the CPU listed. It if was an
    > Intel CPU, then no problem.
    >


    It's an Intel. No overheating problem as far as I can see. The chip
    temperature was only around 118 F/48 C. Before it used to reach
    130 F or more.

    > Having changes so many things, it is now time to first establish what
    > is and is not working. Even an illuminated light does not report a
    > good supply.
    >


    There was not much changed ("surgery" as you call it), in fact, except
    for the CPU cooler and power supply. Putting in the newer power supply
    stopped the spontaneous reboot and the new cooler brought the chip
    temperature down. That is ALL that was replaced.

    > Get a meter to learn what is wrong, immediately, and before replacing
    > anything. Solve a problem the first time. Procedures to take such
    > measurements were detailed previously in
    > microsoft.public.windowsxp.general on 7 Jun 2006 entitled "Dead
    > computer" at
    > http://tinyurl.com/qcvuq
    >
    > Numbers from those measurements then mean replies will have better
    > credibility.
    >
    > Foundation of every computer is its power supply system. Notice the
    > word 'system'. Power supply is only one 'system' component. Get those
    > numbers. In your case, first numbers from green gray wires are
    > important both before and after power switch is pressed.
    >
    > Disturbing is a power switch tested with a battery. Without limiting
    > current, that battery could have harmed switch contacts. Just another
    > reason why computer repairing means a screwdriver and the essential 3.5
    > digit multimeter. In fact, that two minute test with a multimeter
    > would have identified switch as good - without disconnecting anything -
    > without adding more unknowns to the problem by moving wires. Fixing
    > things right the first time is not found in 'replace this, then replace
    > that' techniques - also called shotgunning. First collect facts
    > without changing anything. Takes less time. Finds and fixed the
    > problem a first time. Does not end up curing only symptoms. System
    > does not fail later due to too many changes and other human created
    > failures.
    >


    The switch was one bought at Radio Shack because the original switch
    broke, several years ago, in fact. It was not a computer-specific
    switch,
    it was more general-purpose, and I doubt the battery would therefore
    have
    damaged it.

    But IS IT SAFE to test the power supply in the way you describe?

    > wrote:
    > > You sure that's safe? The integrated ethernet light is on, which means
    > > the power
    > > supply is generating a voltage. I even tried connecting the old one
    > > which works at
    > > least intermittently, and it still wouldn't start.
     
    , Oct 28, 2006
    #24
  5. Guest

    wrote:
    > w_tom wrote:
    > > Well at the very start, you still have no reason to believe the
    > > original power supply failed. Demonstrated is why we don't replace
    > > things until first collecting facts. Two minutes with a 3.5 digital
    > > multimeter would have condemned or exonerated the original supply.
    > > However, having subjected the system to so much 'surgery', no one can
    > > be sure even what the original problem was.
    > >

    >
    > How would I test it with the meter? Would I have to open it up and poke
    > around in there, what with all that voltage? That sounds kind of
    > dangerous.
    > Would it be safe to do that? Should I risk? But replacing the original
    > power
    > supply did stop the spotaneous reboot problem, so something must have
    > been wrong with it.
    >
    > > Did CPU overheat? I don't even see the CPU listed. It if was an
    > > Intel CPU, then no problem.
    > >

    >
    > It's an Intel. No overheating problem as far as I can see. The chip
    > temperature was only around 118 F/48 C. Before it used to reach
    > 130 F or more.
    >


    "before" = "before I installed the new cooler". Idle temp then was
    around 127 F, to be more precise.

    > > Having changes so many things, it is now time to first establish what
    > > is and is not working. Even an illuminated light does not report a
    > > good supply.
    > >

    >
    > There was not much changed ("surgery" as you call it), in fact, except
    > for the CPU cooler and power supply. Putting in the newer power supply
    > stopped the spontaneous reboot and the new cooler brought the chip
    > temperature down. That is ALL that was replaced.
    >
    > > Get a meter to learn what is wrong, immediately, and before replacing
    > > anything. Solve a problem the first time. Procedures to take such
    > > measurements were detailed previously in
    > > microsoft.public.windowsxp.general on 7 Jun 2006 entitled "Dead
    > > computer" at
    > > http://tinyurl.com/qcvuq
    > >
    > > Numbers from those measurements then mean replies will have better
    > > credibility.
    > >
    > > Foundation of every computer is its power supply system. Notice the
    > > word 'system'. Power supply is only one 'system' component. Get those
    > > numbers. In your case, first numbers from green gray wires are
    > > important both before and after power switch is pressed.
    > >
    > > Disturbing is a power switch tested with a battery. Without limiting
    > > current, that battery could have harmed switch contacts. Just another
    > > reason why computer repairing means a screwdriver and the essential 3.5
    > > digit multimeter. In fact, that two minute test with a multimeter
    > > would have identified switch as good - without disconnecting anything -
    > > without adding more unknowns to the problem by moving wires. Fixing
    > > things right the first time is not found in 'replace this, then replace
    > > that' techniques - also called shotgunning. First collect facts
    > > without changing anything. Takes less time. Finds and fixed the
    > > problem a first time. Does not end up curing only symptoms. System
    > > does not fail later due to too many changes and other human created
    > > failures.
    > >

    >
    > The switch was one bought at Radio Shack because the original switch
    > broke, several years ago, in fact. It was not a computer-specific
    > switch,
    > it was more general-purpose, and I doubt the battery would therefore
    > have
    > damaged it.
    >
    > But IS IT SAFE to test the power supply in the way you describe?
    >
    > > wrote:
    > > > You sure that's safe? The integrated ethernet light is on, which means
    > > > the power
    > > > supply is generating a voltage. I even tried connecting the old one
    > > > which works at
    > > > least intermittently, and it still wouldn't start.
     
    , Oct 28, 2006
    #25
  6. w_tom Guest

    wrote:
    > ...
    > But IS IT SAFE to test the power supply in the way you describe?


    You performed major surgery - replaced a power supply. And now you
    ask about safety? Capacitors inside a disconnected power supply can
    store far more dangerous charges? Did you know voltages inside your
    car battery could be more dangerous?

    Your questions were answered in the previous post that included
    pictures and this link http://tinyurl.com/qcvuq . You did major
    surgery. You disconnected wires. Such surgery is not necessary in
    posted procedure / pictures AND such can cause changes that either
    masks a problem or that exponentially complicate the problem. Simply
    disconnecting and reconnecting the power supply connector could have
    eliminated (temporarily) those reboots and not solved the problem.

    How would a meter find the problem? Return to and read the earlier
    post that stated exact how and why the meter does its job. Why did you
    not even view the pictures? Earlier post answered your questions.

    Meanwhile, it could not overheat with heatsink ajar. Overheating
    would have been likely - and occurred so fast that temperature would
    not be noted fast enough. Your numbers only confirm what we knew
    happened when you said, "Intel".
     
    w_tom, Oct 28, 2006
    #26
  7. I downloaded new messages from 24hoursupport.helpdesk on 28 Oct 2006
    03:46:15 -0700, and "w_tom" <> vomited forth this bit
    of wisdom :

    > wrote:
    >> ...
    >> But IS IT SAFE to test the power supply in the way you describe?

    >
    > You performed major surgery - replaced a power supply. And now you
    >ask about safety? Capacitors inside a disconnected power supply can
    >store far more dangerous charges? Did you know voltages inside your
    >car battery could be more dangerous?
    >
    > Your questions were answered in the previous post that included
    >pictures and this link http://tinyurl.com/qcvuq . You did major
    >surgery. You disconnected wires. Such surgery is not necessary in
    >posted procedure / pictures AND such can cause changes that either
    >masks a problem or that exponentially complicate the problem. Simply
    >disconnecting and reconnecting the power supply connector could have
    >eliminated (temporarily) those reboots and not solved the problem.
    >
    > How would a meter find the problem? Return to and read the earlier
    >post that stated exact how and why the meter does its job. Why did you
    >not even view the pictures? Earlier post answered your questions.
    >
    > Meanwhile, it could not overheat with heatsink ajar. Overheating
    >would have been likely - and occurred so fast that temperature would
    >not be noted fast enough. Your numbers only confirm what we knew
    >happened when you said, "Intel".


    I still think it was his power cord.
     
    A Colored Guy, Oct 28, 2006
    #27
  8. Guest

    w_tom wrote:
    > wrote:
    > > ...
    > > But IS IT SAFE to test the power supply in the way you describe?

    >
    > You performed major surgery - replaced a power supply. And now you
    > ask about safety? Capacitors inside a disconnected power supply can
    > store far more dangerous charges? Did you know voltages inside your
    > car battery could be more dangerous?
    >


    Well OK, I'll try it. I read the website.

    > Your questions were answered in the previous post that included
    > pictures and this link http://tinyurl.com/qcvuq . You did major
    > surgery. You disconnected wires. Such surgery is not necessary in
    > posted procedure / pictures AND such can cause changes that either
    > masks a problem or that exponentially complicate the problem. Simply
    > disconnecting and reconnecting the power supply connector could have
    > eliminated (temporarily) those reboots and not solved the problem.
    >
    > How would a meter find the problem? Return to and read the earlier
    > post that stated exact how and why the meter does its job. Why did you
    > not even view the pictures? Earlier post answered your questions.
    >
    > Meanwhile, it could not overheat with heatsink ajar. Overheating
    > would have been likely - and occurred so fast that temperature would
    > not be noted fast enough. Your numbers only confirm what we knew
    > happened when you said, "Intel".


    What could overheat? The CPU? The heatsink was properly fastened and
    the cooler working.
     
    , Oct 28, 2006
    #28
  9. Larry Crites Guest

    You know, if you have to ask that question , you don't know what you're
    doing and need to take your computer to a shop.

    Larry
    Behold Beware Believe

    <> wrote in message
    news:...
    >
    >>

    >
    > How would I test it with the meter? Would I have to open it up and poke
    > around in there, what with all that voltage? That sounds kind of
    > dangerous.
    > Would it be safe to do that? Should I risk? But replacing the original
    > power
    > supply did stop the spotaneous reboot problem, so something must have
    > been wrong with it.
    >




    --
    Posted via a free Usenet account from http://www.teranews.com
     
    Larry Crites, Oct 28, 2006
    #29
  10. Shel-hed Guest

    On 28 Oct 2006 11:28:56 -0700, wrote:


    >Well OK, I'll try it. I read the website.


    You are being trolled. Or this guy is a moron.
    Generally speaking, the more evangelistic or demeaning they are, the more likely
    they are suffering from a huge personality disorder.
     
    Shel-hed, Oct 28, 2006
    #30
  11. Guest

    w_tom wrote:
    > wrote:
    > > ...
    > > But IS IT SAFE to test the power supply in the way you describe?

    >
    > You performed major surgery - replaced a power supply. And now you
    > ask about safety? Capacitors inside a disconnected power supply can
    > store far more dangerous charges? Did you know voltages inside your
    > car battery could be more dangerous?
    >
    > Your questions were answered in the previous post that included
    > pictures and this link http://tinyurl.com/qcvuq . You did major
    > surgery. You disconnected wires. Such surgery is not necessary in
    > posted procedure / pictures AND such can cause changes that either
    > masks a problem or that exponentially complicate the problem. Simply
    > disconnecting and reconnecting the power supply connector could have
    > eliminated (temporarily) those reboots and not solved the problem.
    >
    > How would a meter find the problem? Return to and read the earlier
    > post that stated exact how and why the meter does its job. Why did you
    > not even view the pictures? Earlier post answered your questions.
    >
    > Meanwhile, it could not overheat with heatsink ajar. Overheating
    > would have been likely - and occurred so fast that temperature would
    > not be noted fast enough. Your numbers only confirm what we knew
    > happened when you said, "Intel".


    Well, I've done the tests described on the websites you gave, on the
    new power supply. Haven't done the old one yet, but the new one is the
    one that was in the computer at the time it gave up the ghost.

    -----------
    Checked the power cord, it was OK.

    Power supply test result with voltmeter:

    ---
    The power supply fan spun up OK when the green & black wires on the
    M/B connector were connected as the websites described

    The drive connectors gave 12 V

    The M/B connector:

    Yellow->black wires (12 V) read values around 12.5 V. The websites you
    referred me to said that under 11 V it would not be reliable and under
    10.5 it wouldn't boot.

    Pin #9 had 5 V

    Note: All power supply tests were done without any load connected, the
    thing was disconnected from everything inside the computer.

    The button switch on the case tested OK.
    ---
    -----------

    Also, what do you mean by the CPU could "not" overheat w/ the
    heatsink ajar? To me that would ensure that it WOULD overheat...
    Also the heatsink was NOT ajar, it was in there good and secure,
    and the CPU cooling fan worked OK.
     
    , Oct 28, 2006
    #31
  12. Rich Wilson Guest

    "A Colored Guy" <> wrote in message
    news:...
    >I downloaded new messages from 24hoursupport.helpdesk on 28 Oct 2006
    > 03:46:15 -0700, and "w_tom" <> vomited forth this bit
    > of wisdom :
    >
    >> wrote:
    >>> ...
    >>> But IS IT SAFE to test the power supply in the way you describe?

    >>
    >> You performed major surgery - replaced a power supply. And now you
    >>ask about safety? Capacitors inside a disconnected power supply can
    >>store far more dangerous charges? Did you know voltages inside your
    >>car battery could be more dangerous?
    >>
    >> Your questions were answered in the previous post that included
    >>pictures and this link http://tinyurl.com/qcvuq . You did major
    >>surgery. You disconnected wires. Such surgery is not necessary in
    >>posted procedure / pictures AND such can cause changes that either
    >>masks a problem or that exponentially complicate the problem. Simply
    >>disconnecting and reconnecting the power supply connector could have
    >>eliminated (temporarily) those reboots and not solved the problem.
    >>
    >> How would a meter find the problem? Return to and read the earlier
    >>post that stated exact how and why the meter does its job. Why did you
    >>not even view the pictures? Earlier post answered your questions.
    >>
    >> Meanwhile, it could not overheat with heatsink ajar. Overheating
    >>would have been likely - and occurred so fast that temperature would
    >>not be noted fast enough. Your numbers only confirm what we knew
    >>happened when you said, "Intel".

    >
    > I still think it was his power cord.


    Yep, sounds like exactly what happens when you don't push the power
    connector in far enough, or when there's a loose connection in the cord.

    If we say it enough times, someone might listen!
     
    Rich Wilson, Oct 29, 2006
    #32
  13. w_tom Guest

    I am not sure what was just posted. Was that power supply in the
    computer and connected to motherboard when powered? You cannot jumper
    to green wire to conduct useful measurements. Power supply must be
    connected to rest of its power supply 'system' and must be loaded by
    rest of computer.. Otherwise the test is useless. Otherwise numbers
    tell us noting more than the green illuminated light.

    Looking back at cited posts, my post did not provide numbers - wrong
    post was cited. Those websites provided numbers that are too low and
    not useful. Websites now only provide pictures of how to touch meter
    probes to pins. But meter should be used without disconnecting
    anything inside a computer.

    Essential are three digit voltages for each wire - green, gray,
    purple, red, orange, and yellow. Any voltage below 3.23, 4.87, or 11.7
    means defect. If 12 volts was at 11, the system had long since and
    obviously failed. Furthermore, providing those numbers can provide
    other information. Voltages on green, gray, and purple wire - what
    they do before and when switch is pressed - are useful facts.

    Obvously power cord was connected as indicated (in an earlier post)
    by illuminated green light.

    Heatsink ajar: some CPUs without an attached heatsink literally burn
    - vaporize inside. If the CPU was an AMD and was powered it without
    the heatsink, then CPU would have burned - toasted - become a paper
    weight. Just another example of why major surgery can be hazardous,
    and why we collect facts without changing anything. However, you are
    lucky. You had an Intel chip that contains protection if the heatsink
    is ajar. An Intel CPU will not overheat even without a spinning fan or
    without heatsink firmly attached.

    Meanwhile, testing a power supply without the rest of a power supply
    'system' and without a load reports little that is useful. Jumpering
    that green to black wire only says why the previous green light
    illuminated - that power supply cord works. Power supply must be
    attached as if it was never removed. Originally noted, power supply
    'system' can be confirmed in less than two minutes using only a
    multimeter.

    wrote:
    > Well, I've done the tests described on the websites you gave, on the
    > new power supply. Haven't done the old one yet, but the new one is the
    > one that was in the computer at the time it gave up the ghost.
    >
    > -----------
    > Checked the power cord, it was OK.
    >
    > Power supply test result with voltmeter:
    >
    > ---
    > The power supply fan spun up OK when the green & black wires on the
    > M/B connector were connected as the websites described
    >
    > The drive connectors gave 12 V
    >
    > The M/B connector:
    >
    > Yellow->black wires (12 V) read values around 12.5 V. The websites you
    > referred me to said that under 11 V it would not be reliable and under
    > 10.5 it wouldn't boot.
    >
    > Pin #9 had 5 V
    >
    > Note: All power supply tests were done without any load connected, the
    > thing was disconnected from everything inside the computer.
    >
    > The button switch on the case tested OK.
    > ---
    > -----------
    >
    > Also, what do you mean by the CPU could "not" overheat w/ the
    > heatsink ajar? To me that would ensure that it WOULD overheat...
    > Also the heatsink was NOT ajar, it was in there good and secure,
    > and the CPU cooling fan worked OK.
     
    w_tom, Oct 29, 2006
    #33
  14. Guest

    w_tom wrote:
    > I am not sure what was just posted. Was that power supply in the
    > computer and connected to motherboard when powered? You cannot jumper
    > to green wire to conduct useful measurements. Power supply must be
    > connected to rest of its power supply 'system' and must be loaded by
    > rest of computer.. Otherwise the test is useless. Otherwise numbers
    > tell us noting more than the green illuminated light.
    >


    I cannot even power it up with it connected to the motherboard. When I
    push the switch it does not start. Since it has to go through the
    motherboard,
    I think that's where the problem is. When I jumpered the green to black
    wire like it said, the power supply fan came on, and then I proceeded
    to
    measure the voltages from the wires going into the MB connector. The
    numbers agreed with what was given on the sites -- 12V lines gave
    around 12V (some registered around 12.5 V, is that too high? Could it
    have
    *over*loaded something in the computer? Would that explain the tilting
    killing it?), 5V ones gave 5V, etc. Since the fan was able to start, I
    therefore think the problem is between the switch and the power supply
    --
    the motherboard. It could have overheated perhaps due to lack of a
    running
    case fan, it could have died of old age, etc. If the power supply
    failed so
    bad as to not even run it's fan then why does "hotwiring" it start it?

    > Looking back at cited posts, my post did not provide numbers - wrong
    > post was cited. Those websites provided numbers that are too low and
    > not useful. Websites now only provide pictures of how to touch meter
    > probes to pins. But meter should be used without disconnecting
    > anything inside a computer.
    >



    You have to disconnect a connector to test it. Can you measure through
    insulated wires? What? Strip the insulation off? The website didn't say
    to
    do that...

    > Essential are three digit voltages for each wire - green, gray,
    > purple, red, orange, and yellow. Any voltage below 3.23, 4.87, or 11.7
    > means defect. If 12 volts was at 11, the system had long since and
    > obviously failed. Furthermore, providing those numbers can provide
    > other information. Voltages on green, gray, and purple wire - what
    > they do before and when switch is pressed - are useful facts.
    >


    How do you test that with the darned thing plugged into the motherboard
    and the pins/etc. inaccessible? (you can't connect things to a
    connector
    that is already connected to something. That should be obvious.) You
    have to pull out SOME plug in order to test it. D'oh! And by "long
    since"
    failed, remember that this power supply was brand new... are you
    saying there was a manufacturing defect? Not that such a thing is
    impossible, just is that what you are saying?

    > Obvously power cord was connected as indicated (in an earlier post)
    > by illuminated green light.
    >
    > Heatsink ajar: some CPUs without an attached heatsink literally burn
    > - vaporize inside. If the CPU was an AMD and was powered it without
    > the heatsink, then CPU would have burned - toasted - become a paper
    > weight. Just another example of why major surgery can be hazardous,
    > and why we collect facts without changing anything. However, you are
    > lucky. You had an Intel chip that contains protection if the heatsink
    > is ajar. An Intel CPU will not overheat even without a spinning fan or
    > without heatsink firmly attached.
    >


    And the heatsink is firmly attached & the CPU fan ran, so that is not
    the
    problem. Amd yes, I have seen the video of the AMD chip roasting
    itself.
    There was even another video where it actually _exploded_ and blew
    a hole in the motherboard, although I've been told that that one was a
    hoax.

    > Meanwhile, testing a power supply without the rest of a power supply
    > 'system' and without a load reports little that is useful. Jumpering
    > that green to black wire only says why the previous green light
    > illuminated - that power supply cord works. Power supply must be
    > attached as if it was never removed. Originally noted, power supply
    > 'system' can be confirmed in less than two minutes using only a
    > multimeter.
    >


    I tested all the connections on the M/B connector. I followed the
    instructions
    on the website, which says nothing about plugging it into the MB and
    somehow
    measuring through insulated wires.

    > wrote:
    > > Well, I've done the tests described on the websites you gave, on the
    > > new power supply. Haven't done the old one yet, but the new one is the
    > > one that was in the computer at the time it gave up the ghost.
    > >
    > > -----------
    > > Checked the power cord, it was OK.
    > >
    > > Power supply test result with voltmeter:
    > >
    > > ---
    > > The power supply fan spun up OK when the green & black wires on the
    > > M/B connector were connected as the websites described
    > >
    > > The drive connectors gave 12 V
    > >
    > > The M/B connector:
    > >
    > > Yellow->black wires (12 V) read values around 12.5 V. The websites you
    > > referred me to said that under 11 V it would not be reliable and under
    > > 10.5 it wouldn't boot.
    > >
    > > Pin #9 had 5 V
    > >
    > > Note: All power supply tests were done without any load connected, the
    > > thing was disconnected from everything inside the computer.
    > >
    > > The button switch on the case tested OK.
    > > ---
    > > -----------
    > >
    > > Also, what do you mean by the CPU could "not" overheat w/ the
    > > heatsink ajar? To me that would ensure that it WOULD overheat...
    > > Also the heatsink was NOT ajar, it was in there good and secure,
    > > and the CPU cooling fan worked OK.
     
    , Oct 29, 2006
    #34
  15. w_tom Guest

    Again, disconnect nothing. When connected to the motherboard,
    power switch does nothing. Now we have important facts. Dead bodies
    always tell us useful information. Again, before pressing power
    switch, what are the voltages on gray, green, and purple wires? And
    what does meter demonstrate when power switch is pressed? Without
    those numbers, nobody can provide useful information.

    Again, notice what is the most important fact. The numbers. Purple
    wire must provide a voltage that is above 4.87 when computer appears
    completely dead. A voltage that exists if power cord is connected to
    wall receptacle. A purple wire voltage that does not change when power
    switch is pressed. Green wire is what voltage when computer is not
    powered - before power switch is pressed? It should be more than 2.0
    volts. When power switch is pressed, that voltage must drop down to
    less than 0.8 volts. Does it? And what are those values?

    Gray wire is near zero volts when computer is powered off (and whn
    purple wire has more than 4.87 volts). What does that wire do when
    power switch is pressed? Does it rise above 2.4 volts? Does it stay
    above 2.4 volts? How high does it remain?

    These are useful facts about the entire power supply 'system'. Only
    useful information is what 'system' does - not just the power supply.

    Damage due to overheating is promoted by myths. Heat is a tool to
    find intemittent or to find failures before those failures are
    apparent. Don't worry about heat. Those who grap for solutions
    without first learning facts will blame only what they understand. IOW
    they worry about heat - and then cure symptoms. One case fan is more
    than enough for any system. But those who don't first learn technology
    and the numbers want more fans.

    Why did hotwire (green wire jumper) cause power supply to work?
    Again, we have no idea because the reasons are so long and because the
    power system is not part of a power supply 'system'. Problem may be
    inside power supply. Or problem may be in rest of power supply
    'system'.

    Websites had changed since I last saw them. If they discuss
    disconnecting anything to take a reading, then ignore it. A white
    nylone 20 pin connector to motherboard has holes where each wire
    enters. Push probes into those same holes to get a meter reading.
    Again, no reason to strip any wire or disconnect anything to make
    necessary voltage readings. No reason to disconnect anything until
    facts tell us where to look. Every wire is accessible.

    Again, website was cited mostly for the pictures on how to use probe
    and read a meter. And since then, web sites may have added facts
    erroneous. Do not follow their instructions. Take voltages as I have
    posted. No disconnections.

    wrote:
    > ...
    > I cannot even power it up with it connected to the motherboard. When I
    > push the switch it does not start. Since it has to go through the
    > motherboard,
    > I think that's where the problem is. When I jumpered the green to black
    > wire like it said, the power supply fan came on, and then I proceeded
    > to
    > measure the voltages from the wires going into the MB connector. The
    > numbers agreed with what was given on the sites -- 12V lines gave
    > around 12V (some registered around 12.5 V, is that too high? Could it
    > have *over*loaded something in the computer? Would that explain the
    > tilting killing it?), 5V ones gave 5V, etc. Since the fan was able to start,
    > I therefore think the problem is between the switch and the power supply
    > --
    > the motherboard. It could have overheated perhaps due to lack of a
    > running case fan, it could have died of old age, etc. If the power
    > supply failed so bad as to not even run it's fan then why does
    > "hotwiring" it start it?
    > ...
    >
    > You have to disconnect a connector to test it. Can you measure through
    > insulated wires? What? Strip the insulation off? The website didn't say
    > to do that...
    > ...
    >
    > How do you test that with the darned thing plugged into the motherboard
    > and the pins/etc. inaccessible? (you can't connect things to a
    > connector that is already connected to something. That should be
    > obvious.) You have to pull out SOME plug in order to test it. D'oh! And
    > by "long since" failed, remember that this power supply was brand
    > new... are you saying there was a manufacturing defect? Not that such
    > a thing is impossible, just is that what you are saying?
    > ...
    >
    > And the heatsink is firmly attached & the CPU fan ran, so that is not
    > the problem. Amd yes, I have seen the video of the AMD chip
    > roasting itself. There was even another video where it
    > actually _exploded_ and blew a hole in the motherboard, although
    > I've been told that that one was a hoax.
    > ...
    >
    > I tested all the connections on the M/B connector. I followed the
    > instructions on the website, which says nothing about plugging it
    > into the MB and somehow measuring through insulated wires.
     
    w_tom, Oct 29, 2006
    #35
  16. Meat Plow Guest


    >> You performed major surgery - replaced a power supply. And now you
    >> ask about safety? Capacitors inside a disconnected power supply can
    >> store far more dangerous charges? Did you know voltages inside your car
    >> battery could be more dangerous?


    Caps in a high voltage supply could be dangerous, not a PC power supply.
    The voltage in a car battery won't hurt you. However the current could
    burn you if you shunted the terminals.

    --
    Pierre Salinger Memorial Hook, Line & Sinker, June 2004

    COOSN-266-06-25794
     
    Meat Plow, Oct 29, 2006
    #36
  17. Guest

    w_tom wrote:
    > Again, disconnect nothing. When connected to the motherboard,
    > power switch does nothing. Now we have important facts. Dead bodies
    > always tell us useful information. Again, before pressing power
    > switch, what are the voltages on gray, green, and purple wires? And
    > what does meter demonstrate when power switch is pressed? Without
    > those numbers, nobody can provide useful information.
    >


    How do I test it with everything connected, plug in the motherboard? I
    need to get at that plug to test it! You can't get the meter probes
    into the
    plug!

    > Again, notice what is the most important fact. The numbers. Purple
    > wire must provide a voltage that is above 4.87 when computer appears
    > completely dead. A voltage that exists if power cord is connected to
    > wall receptacle. A purple wire voltage that does not change when power
    > switch is pressed. Green wire is what voltage when computer is not
    > powered - before power switch is pressed? It should be more than 2.0
    > volts. When power switch is pressed, that voltage must drop down to
    > less than 0.8 volts. Does it? And what are those values?
    >


    How can I test it "when the power switch is pushed" with the CONNECTOR
    PLUGGED INTO THE MOTHERBOARD AND THE PINS _INACCESSIBLE_?!?!?!?!

    > Gray wire is near zero volts when computer is powered off (and whn
    > purple wire has more than 4.87 volts). What does that wire do when
    > power switch is pressed? Does it rise above 2.4 volts? Does it stay
    > above 2.4 volts? How high does it remain?
    >
    > These are useful facts about the entire power supply 'system'. Only
    > useful information is what 'system' does - not just the power supply.
    >
    > Damage due to overheating is promoted by myths. Heat is a tool to
    > find intemittent or to find failures before those failures are
    > apparent. Don't worry about heat. Those who grap for solutions
    > without first learning facts will blame only what they understand. IOW
    > they worry about heat - and then cure symptoms. One case fan is more
    > than enough for any system. But those who don't first learn technology
    > and the numbers want more fans.
    >
    > Why did hotwire (green wire jumper) cause power supply to work?
    > Again, we have no idea because the reasons are so long and because the
    > power system is not part of a power supply 'system'. Problem may be
    > inside power supply. Or problem may be in rest of power supply
    > 'system'.
    >


    Read the websites you pointed me to.

    > Websites had changed since I last saw them. If they discuss
    > disconnecting anything to take a reading, then ignore it. A white
    > nylone 20 pin connector to motherboard has holes where each wire
    > enters. Push probes into those same holes to get a meter reading.
    > Again, no reason to strip any wire or disconnect anything to make
    > necessary voltage readings. No reason to disconnect anything until
    > facts tell us where to look. Every wire is accessible.
    >


    The holes already contain wires. How can I get the thick meter probes
    into there? If I force them into there I may damage the connections and
    create more problems. Haven't you looked at a power supply connector?
    The wires go INTO it.

    > Again, website was cited mostly for the pictures on how to use probe
    > and read a meter. And since then, web sites may have added facts
    > erroneous. Do not follow their instructions. Take voltages as I have
    > posted. No disconnections.
    >


    So it's best to damage the connections then and create more problems.
    That doesn't make sense.

    > wrote:
    > > ...
    > > I cannot even power it up with it connected to the motherboard. When I
    > > push the switch it does not start. Since it has to go through the
    > > motherboard,
    > > I think that's where the problem is. When I jumpered the green to black
    > > wire like it said, the power supply fan came on, and then I proceeded
    > > to
    > > measure the voltages from the wires going into the MB connector. The
    > > numbers agreed with what was given on the sites -- 12V lines gave
    > > around 12V (some registered around 12.5 V, is that too high? Could it
    > > have *over*loaded something in the computer? Would that explain the
    > > tilting killing it?), 5V ones gave 5V, etc. Since the fan was able to start,
    > > I therefore think the problem is between the switch and the power supply
    > > --
    > > the motherboard. It could have overheated perhaps due to lack of a
    > > running case fan, it could have died of old age, etc. If the power
    > > supply failed so bad as to not even run it's fan then why does
    > > "hotwiring" it start it?
    > > ...
    > >
    > > You have to disconnect a connector to test it. Can you measure through
    > > insulated wires? What? Strip the insulation off? The website didn't say
    > > to do that...
    > > ...
    > >
    > > How do you test that with the darned thing plugged into the motherboard
    > > and the pins/etc. inaccessible? (you can't connect things to a
    > > connector that is already connected to something. That should be
    > > obvious.) You have to pull out SOME plug in order to test it. D'oh! And
    > > by "long since" failed, remember that this power supply was brand
    > > new... are you saying there was a manufacturing defect? Not that such
    > > a thing is impossible, just is that what you are saying?
    > > ...
    > >
    > > And the heatsink is firmly attached & the CPU fan ran, so that is not
    > > the problem. Amd yes, I have seen the video of the AMD chip
    > > roasting itself. There was even another video where it
    > > actually _exploded_ and blew a hole in the motherboard, although
    > > I've been told that that one was a hoax.
    > > ...
    > >
    > > I tested all the connections on the M/B connector. I followed the
    > > instructions on the website, which says nothing about plugging it
    > > into the MB and somehow measuring through insulated wires.
     
    , Oct 29, 2006
    #37
  18. Meat Plow Guest

    On Sun, 29 Oct 2006 10:36:38 -0800, mike4ty4 Has Frothed:

    > w_tom wrote:
    >> Again, disconnect nothing. When connected to the motherboard, power
    >> switch does nothing. Now we have important facts. Dead bodies always
    >> tell us useful information. Again, before pressing power switch, what
    >> are the voltages on gray, green, and purple wires? And what does meter
    >> demonstrate when power switch is pressed? Without those numbers, nobody
    >> can provide useful information.
    >>
    >>

    > How do I test it with everything connected, plug in the motherboard? I
    > need to get at that plug to test it! You can't get the meter probes into
    > the
    > plug!


    Should be able to get your meter probe to the backside of the plug's pins.

    --
    Pierre Salinger Memorial Hook, Line & Sinker, June 2004

    COOSN-266-06-25794
     
    Meat Plow, Oct 29, 2006
    #38
  19. Guest

    Meat Plow wrote:
    > On Sun, 29 Oct 2006 10:36:38 -0800, mike4ty4 Has Frothed:
    >
    > > w_tom wrote:
    > >> Again, disconnect nothing. When connected to the motherboard, power
    > >> switch does nothing. Now we have important facts. Dead bodies always
    > >> tell us useful information. Again, before pressing power switch, what
    > >> are the voltages on gray, green, and purple wires? And what does meter
    > >> demonstrate when power switch is pressed? Without those numbers, nobody
    > >> can provide useful information.
    > >>
    > >>

    > > How do I test it with everything connected, plug in the motherboard? I
    > > need to get at that plug to test it! You can't get the meter probes into
    > > the
    > > plug!

    >
    > Should be able to get your meter probe to the backside of the plug's pins.
    >


    Won't that damage the wires? I mean I'm forcing something down into
    there.
    How thick/thin should the probes be (diameter of the metal leads that
    you
    stick in)?

    PS. I also tried starting the computer with the OLD power supply and it
    still
    failed to run. So either both are bad or something else is.

    > --
    > Pierre Salinger Memorial Hook, Line & Sinker, June 2004
    >
    > COOSN-266-06-25794
     
    , Oct 29, 2006
    #39
  20. Meat Plow Guest

    On Sun, 29 Oct 2006 11:35:44 -0800, mike4ty4 Has Frothed:

    >
    > Meat Plow wrote:
    >> On Sun, 29 Oct 2006 10:36:38 -0800, mike4ty4 Has Frothed:
    >>
    >> > w_tom wrote:
    >> >> Again, disconnect nothing. When connected to the motherboard, power
    >> >> switch does nothing. Now we have important facts. Dead bodies
    >> >> always tell us useful information. Again, before pressing power
    >> >> switch, what are the voltages on gray, green, and purple wires? And
    >> >> what does meter demonstrate when power switch is pressed? Without
    >> >> those numbers, nobody can provide useful information.
    >> >>
    >> >>
    >> > How do I test it with everything connected, plug in the motherboard? I
    >> > need to get at that plug to test it! You can't get the meter probes
    >> > into the
    >> > plug!

    >>
    >> Should be able to get your meter probe to the backside of the plug's
    >> pins.
    >>
    >>

    > Won't that damage the wires? I mean I'm forcing something down into there.
    > How thick/thin should the probes be (diameter of the metal leads that you
    > stick in)?
    >
    > PS. I also tried starting the computer with the OLD power supply and it
    > still
    > failed to run. So either both are bad or something else is.


    Shouldn't cause any damage. Might use a piece of a paper clip to get down
    inside. Take a close look under decent lighting and look how the plug is
    contructed.

    --
    Pierre Salinger Memorial Hook, Line & Sinker, June 2004

    COOSN-266-06-25794
     
    Meat Plow, Oct 29, 2006
    #40
    1. Advertising

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

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. quiet_table
    Replies:
    1
    Views:
    476
    mark mandel
    Jan 16, 2004
  2. Vinay

    Toasted 20D?

    Vinay, Apr 10, 2007, in forum: Digital Photography
    Replies:
    8
    Views:
    405
    Vinay
    Apr 18, 2007
  3. mike3

    Computer Toasted again???

    mike3, Nov 24, 2008, in forum: Computer Support
    Replies:
    5
    Views:
    492
    §ñühwØ£f
    Nov 25, 2008
  4. Replies:
    4
    Views:
    363
  5. mike3

    Computer Toasted or something else?

    mike3, Mar 8, 2010, in forum: Computer Support
    Replies:
    7
    Views:
    412
    chuckcar
    Mar 8, 2010
Loading...

Share This Page