Dead computer :-( Any ideas?

  • Thread starter Thread starter Larc
  • Start date Start date
L

Larc

When I tried to turn on my secondary computer last night, nothing happened.
LEDs were lit on the built-in LAN socket as usual, so it was getting power. It
had been running with no problems. Both the PSU and RAM checked out OK in
another system. Video is built in, so no separate graphics card.

The only change I've made recently was setting up a specific key for keyboard
startup of the system. I started it that way several times before there was a
brief power outage a few nights ago. The first time I tried to start it after
that outage was last night when it wouldn't do anything. I did short the CMOS
to revert to default setup... twice, but no joy. Of course, the computer should
still have started with the power button on the box even if it couldn't be
started from the keyboard. It was plugged into a surge protector that didn't
trip off due to the outage, and everything else plugged into that protector is
working.

- MSI 845G Max motherboard with P4 2G CPU and one 512MB stick of Crucial PC2100
RAM.

Any ideas before I tearfully mumble last rites over the motherboard?

Larc



§§§ - Please raise temperature of mail to reply by e-mail - §§§
 
When I tried to turn on my secondary computer last night, nothing happened.
LEDs were lit on the built-in LAN socket as usual, so it was getting power. It
had been running with no problems. Both the PSU and RAM checked out OK in
another system. Video is built in, so no separate graphics card.

The only change I've made recently was setting up a specific key for keyboard
startup of the system. I started it that way several times before there was a
brief power outage a few nights ago. The first time I tried to start it after
that outage was last night when it wouldn't do anything. I did short the CMOS
to revert to default setup... twice, but no joy.
Click to expand...

Did you pull the power cord out before you shorted the CMOS?
ATX powered systems cannot fully be cleared with power to the mother
board as there is a residual current applied to the CMOS.Remove power
cable/clear CMOS and try again.
HTH :)

Of course, the computer should
still have started with the power button on the box even if it couldn't be
started from the keyboard. It was plugged into a surge protector that didn't
trip off due to the outage, and everything else plugged into that protector is
working.

- MSI 845G Max motherboard with P4 2G CPU and one 512MB stick of Crucial PC2100
RAM.

Any ideas before I tearfully mumble last rites over the motherboard?

Larc



§§§ - Please raise temperature of mail to reply by e-mail - §§§
Click to expand...



--
Free Windows/PC help,
http://www.geocities.com/sheppola/trouble.html
email shepATpartyheld.de
Free songs download,
http://www.soundclick.com/bands/8/nomessiahsmusic.htm
 
| Did you pull the power cord out before you shorted the CMOS?
| ATX powered systems cannot fully be cleared with power to the mother
| board as there is a residual current applied to the CMOS.Remove power
| cable/clear CMOS and try again.
| HTH :)

Thanks, but the power cord wasn't connected either time I cleared the CMOS.

Larc



§§§ - Please raise temperature of mail to reply by e-mail - §§§
 
| Did you pull the power cord out before you shorted the CMOS?
| ATX powered systems cannot fully be cleared with power to the mother
| board as there is a residual current applied to the CMOS.Remove power
| cable/clear CMOS and try again.
| HTH :)

Thanks, but the power cord wasn't connected either time I cleared the CMOS.

Larc
Click to expand...

Pull all PCI cards out and if you have two or more RAM sticks try each
separately?
What BIOS is it?



--
Free Windows/PC help,
http://www.geocities.com/sheppola/trouble.html
email shepATpartyheld.de
Free songs download,
http://www.soundclick.com/bands/8/nomessiahsmusic.htm
 
| Pull all PCI cards out and if you have two or more RAM sticks try each
| separately?
| What BIOS is it?

Thanks again, Shep, but I don't have any PCI cards in that unit. There is only
one stick of RAM and it checked out OK when I put it in another computer and ran
MemTest86.

BIOS is AMI. I had flashed MSI's latest version 1.6 for the 845G Max, dated
December 17, 2002.

Larc



§§§ - Please raise temperature of mail to reply by e-mail - §§§
 
| Pull all PCI cards out and if you have two or more RAM sticks try each
| separately?
| What BIOS is it?

Thanks again, Shep, but I don't have any PCI cards in that unit. There is only
one stick of RAM and it checked out OK when I put it in another computer and ran
MemTest86.

BIOS is AMI. I had flashed MSI's latest version 1.6 for the 845G Max, dated
December 17, 2002.

Larc
Click to expand...

Format a blank floppy disk.Copy a known good original bios.rom to it.
Rename the file to
amiboot.rom
Re-boot the system and leave it for half an hour.
You may get lucky and it's just a corrupted BIOS.





--
Free Windows/PC help,
http://www.geocities.com/sheppola/trouble.html
email shepATpartyheld.de
Free songs download,
http://www.soundclick.com/bands/8/nomessiahsmusic.htm
 
You say there was power becuase the lights on the LAN were on. This does not
alway mean there is good power.

JaMMeR
 
| You say there was power becuase the lights on the LAN were on. This does not
| alway mean there is good power.

Thanks, but the PSU works perfectly in another system. Nothing I've tried has
helped the problem so far, so I'm getting more convinced that something is wrong
with the motherboard.

Larc



§§§ - Please raise temperature of mail to reply by e-mail - §§§
 
He posted correctly. Just because the supply works in
another system does not mean a supply is functioning
properly. Just because lights work does not mean the power
supply provides sufficient voltage. The fastest solution
involves not complicating the problem by removing and
installing components. First get facts before trying to fix
anything. That means the all so important 3.5 digit
multimeter. Procedure summarized in "Testing power supply" in
newsgroup alt.comp.hardware.homebuilt starting 3 Sept 2003 or
http://tinyurl.com/2gjet

If not obvious, three functional components to your problem -
power supply, motherboard, and switch. Nothing yet says any
of those components are good or bad. Yes, your power supply
could still be defective which is why multimeter numerical
reports on purple and green (as well as red, yellow, and
orange) wires are so important. Numbers that must agree with
limits (and not be in lowest quarter of those limits) in
http://www.hardwaresite.net/faqpowersupply.html

Is something wrong with the motherboard? That summarized
procedure would answer in minutes - without buying and
replacing motherboards. Every good computer tech first uses
his 3.5 digit multimeter for this problem - to identify the
reason for failure before fixing anything.
 
Is something wrong with the motherboard? That summarized
procedure would answer in minutes - without buying and
replacing motherboards. Every good computer tech first uses
his 3.5 digit multimeter for this problem - to identify the
reason for failure before fixing anything.
Click to expand...

You're high on something. If you could use a multimeter to spot a problem
with a power supply, you wouldn't NEED the multimeter to know that the power
supply was bad. Most power supply problems that lock up a system or cause
it to reboot are voltage drops that are too fast to even register on a
multimeter. If a voltage problem is constant enough to be seen by a
multimeter, your "bad power" symptoms are likely to be so severe that there
will be NO DOUBT that your power supply is bad, even if you don't have a
multimeter handy. -Dave
 
Multimeter will report potential power supply problems
before they even appear as computer crashes. That includes
problems due to too much load and problems due to failing
electrolytic capacitors.

How many decades of design does Dave C have? Clearly none
(typical of computer experts) which explains why he insults
and then cites some transient as if it cannot be detected.
Multimeter readings, as defined in previous post, will find
power supply problems before problems become apparent to the
user. This from one who had been building this stuff before
PCs even existed.
 
w_tom said:
How many decades of design does Dave C have? Clearly none
(typical of computer experts) which explains why he insults
and then cites some transient as if it cannot be detected.
Multimeter readings, as defined in previous post, will find
power supply problems before problems become apparent to the
user. This from one who had been building this stuff before
PCs even existed.
Click to expand...

Well aren't you da man. As usual, you come off sounding like a Class A
jackass. I'll bet you're just a lady killer at all the cocktail parties.
 
w_tom said:
Multimeter will report potential power supply problems
before they even appear as computer crashes. That includes
problems due to too much load and problems due to failing
electrolytic capacitors.

How many decades of design does Dave C have? Clearly none
(typical of computer experts) which explains why he insults
and then cites some transient as if it cannot be detected.
Multimeter readings, as defined in previous post, will find
power supply problems before problems become apparent to the
user. This from one who had been building this stuff before
PCs even existed.
Click to expand...


Ummm, if you even remotely suspect that a power supply is bad, replace it.
That is, unless you want to look really stupid when that "good" power supply
blows up on you a week or so after you checked it with a multimeter. A
multimeter is a good tool for curiosity sake, only after-the-fact. That is,
after a power supply starts throwing sparks, you can look at the outputs if
you want to, for entertainment value, to see how far out of whack they are.
Most power supply problems are so marginal that you won't be able to tell
anything useful with a simple multimeter. If the voltages are so far off
that a multimeter will show it to be bad, you shouldn't NEED the multimeter
to know that the power supply is bad. You can disagree with me all you want
to, but that won't change the fact that I'm right. -Dave
 
Back
Top