Re: PC Struck by Lightning?!
 

First, it is routine to run equipment during a thunderstorm
without damage. But that assumes the human has installed
inexpensive solutions and did not spend greater money on
ineffective plug-in solutions.

You don't even know if the power supply is working.
Spinning fans tell you nothing useful other than the power
cord is connected to the wall.

First confirm the computer is built on a properly
functioning foundation. A multimeter confirms power supply
integrity. Then move on to verifying integrity of other
components. Without first confirming power supply integrity,
then all other solutions are but secondary and may be futile.

Meter may find a problem even with the power supply
controller - which is another component of the power supply
system. Power supply is only one component of that system.
But to see this, you need that multimeter and basic knowledge
as found in two previous posts:
"Computer doesnt start at all" in alt.comp.hardware on 10
Jan 2004 at
http://tinyurl.com/2t69q and
"I think my power supply is dead" in alt.comp.hardware on 5
Feb 2004 at
http://tinyurl.com/yvbw9

One does not even remove a single component until the power
cord is remove from wall. Even when powered off, many
components still have electricity if power cord connects to
wall receptacle. Moving any component when electrically hot
is destructive (except where that component is designed
specifically to be removed hot such as a USB cable).

Dave Hardenbrook wrote:
> One of my clients made the mistake of running their system during a
> thunderstorm. This was about three weeks ago, and since then the system
> has gradually degraded in performance -- First it worked okay, then it
> would fail to boot to Windows normally (I thought maybe all that was
> needed was System Restore or Recovery Console), now the system halts at
> POST with the error: "Award Bootblock BIOS ROM checksum error". Is
> there a chance that it's just the CMOS battery, or is it pretty certain
> that lightning did strike and the mobo is fried? And if the mobo is
> toast, are the other components probably gone as well?
>
> --
> Dave

Re: PC Struck by Lightning?!
 

Point 1: a spinning fan tells us nothing useful as correctly
posted. For example, +12 volts could be only +8 volts or what
you called "might not be working well". Either +12 volts is
working "well and right", or it is not working at all. +12
volts at only 8 volts is not working at all. Nada. Fan would
spin. Computer and power supply still don't work. That is
basic electrical knowledge.

Then it gets more interesting. Many power supplies don't
power a fan from +12 volts; as you have assumed. Many power
supplies power the fan from -12 volts. This provides a
minimum load for the -12 volt power supply chip. +12 volts
output might not exist. Meanwhile -12 volts could still be
spinning fan. Just a second basic electrical reason why a
spinning fan tells us nothing useful.


Point 2: informed technicians don't disconnect anything from
computer when doing analysis. Standard procedure - don't
change anything before collecting the facts - just like a
crime scene. Power supply integrity is verified without
disconnecting a power supply; that being common knowledge for
electronic techs. A power supply is measured with the 3.5
digit multimeter while connected to a computer. This also so
that the rest of that power supply 'system' can be verified.
Statements about testing a supply without load are totally
irrelevant and imply poor debugging habits.

Again, there is no such thing as a power supply working if
it is not working "well" or "right". Supply must output
sufficient voltage which is why that 3.5 digit multimeter is
necessary - as tool as essential as a screwdriver. Totally
erroneous assumption that a spinning fans reports something
useful. A spinning fan only suggests AC power is available
from a wall receptacle. Multiple reasons for why that
spinning fan reports nothing useful were provided with basic
electronic knowledge.

Point 3 for the Original Poster: first, measure those power
supply voltages as described in previous posts. Even if
recorded numbers means nothing, still, more informed posters
can now provide informed replies. Currently every post is
nothing more than wild speculation because we don't have those
power supply voltages. Replies can only be as good as the
information provided - the numbers.

Second, the CMOS battery is also verified using that
multimeter. The typical 3 volt CMOS battery should measure 3
volts or more. If it measures 2.8 volts, it is still not a
reason for system problems. Again, numbers for the meter
identify problems immediately with bad repair habits such as
shotgunning. Either that battery is clearly defective OR it
is not. No reason to first buy a new battery only to guess a
battery might be defective.

Barry Watzman wrote:
> Re: "Spinning fans tell you nothing useful other than the power cord is
> connected to the wall."
>
> Not true. The fans run from the +12 volt supply, on the secondary side
> of the switcher. So if they work, then at some leve, the power supply
> is working. Might not be working "well" or "right", but it's working.
>
> Also, many supplies will not work unless at least some outputs have some
> level of load connected. You cannot, usually, measure the voltages with
> the power supply not connected to a load.