«

"alice" <> wrote in message
news: ups.com...
> On Mar 13, 3:30 pm, Ghostrider <-...@fitron.142> wrote:
>> alice wrote:
>> > I just opened up my PC case to take off the CPU fan and heatsink,
>> > cleaned it, put it back on, and now the whole thing will not power on.
>> > The power light comes on for about 5 seconds, and the CPU fan starts
>> > to spin, then it just turns off.
>> > The reason I cleaned the fan in the first place is because the machine
>> > was overheating and randomly turning off frequently.
>> > Is there any thing I can do at this point, and if I need to replace
>> > parts, what should I start with?
>>
>> Five seconds might not seem like too long but there is a good chance
>> that the CPU got fried. This would occur for the lack of heat-conducting
>> paste between the CPU heat slug and the surface of the heatsink. Let's
>> hope not but all of the other troubleshooting ideas have not succeeded.
>
> So if it was the lack of paste, would adding it now fix anything? I
> would assume not. Should I be looking for a new processor? It seems as
> though it was failing anyway, shutting down randomly for months now.

I still feel it is a power supply issue. Not enough wattage to power three
drives along with everything else. It finally gave up the ghost.

On Mar 13, 4:13 pm, "lobo" <el_l...@nowhere.net> wrote:
> "alice" <a...@fearofdolls.com> wrote in message
>
> news: ups.com...
>
>
>
>
>
> > On Mar 13, 3:30 pm, Ghostrider <-...@fitron.142> wrote:
> >> alice wrote:
> >> > I just opened up my PC case to take off the CPU fan and heatsink,
> >> > cleaned it, put it back on, and now the whole thing will not power on.
> >> > The power light comes on for about 5 seconds, and the CPU fan starts
> >> > to spin, then it just turns off.
> >> > The reason I cleaned the fan in the first place is because the machine
> >> > was overheating and randomly turning off frequently.
> >> > Is there any thing I can do at this point, and if I need to replace
> >> > parts, what should I start with?
>
> >> Five seconds might not seem like too long but there is a good chance
> >> that the CPU got fried. This would occur for the lack of heat-conducting
> >> paste between the CPU heat slug and the surface of the heatsink. Let's
> >> hope not but all of the other troubleshooting ideas have not succeeded.
>
> > So if it was the lack of paste, would adding it now fix anything? I
> > would assume not. Should I be looking for a new processor? It seems as
> > though it was failing anyway, shutting down randomly for months now.
>
> I still feel it is a power supply issue. Not enough wattage to power three
> drives along with everything else. It finally gave up the ghost.- Hide quoted text -
>
> - Show quoted text -

I did try unplugging the drives, one at a time, and each time it would
still not power up.

On Mar 13, 7:02 pm, "alice" <a...@fearofdolls.com> wrote:
> I'm no longer in front of it, and I can't remember the wattage, but I
> want to say 350?

There is little to zero you can see or fix only by visual inspection
- which is why so many have you shotgunning with zero progress.
What's worse, you tried to fix something without first establishing
what was wrong - you removed CPU heatsinks. So what we have is maybe
two suspects - the problem now is exponentially more complex.

(BTW, if holes in the heatsink were not completely clogged, then
heat would not be the reason for problems. )

Start by establishing what is and is not. Collect facts. Don't
swap parts. Don't disconnect anything. First discover what you do
have.

The reason a 3.5 digit multimeter is sold even in K-mart? It's for
people with even junior high school education. You need that
multimeter and a screwdriver - the two most essential tools. In two
minutes we define that entire power supply 'system' either good or bad
with almost no doubts. The procedure is posted in "When your computer
dies without warning....." starting 6 Feb 2007 in the newsgroup
alt.windows-xp at:
http://tinyurl.com/yvf9vh

Taking those voltage numbers from red, orange, and yellow wires when
accessing many other peripherals (multitasking - increasing the load)
will make those numbers even more informative. Without those numbers,
you help can only say "try this and try that" rather than do what is
demonstrated on a TV 'CSI': follow the evidence.

Meanwhile, other useful information may have been stored in the
system (event) logs. Information from those logs also means future
replies will be more informative.

But having changed and removed so much, then you may have one or now
multiple problems. You have exponentially complicated the solution.
Above is step one in a process of stepping through the problem;
breaking the problem down into parts and analyzing those parts; KISS.

If you keep swapping stuff, you may only make the problem
unsolvable. Exactly why shotgunning is done only by those without.

Get the $20 meter that is sold in K-mart, Radio Shack, Lowes, Wal-
mart, Sears, or Home Depot. Provide those numbers. Then get responses
that say exactly what is good or bad AND what to do next for a
definitive solution - no more shotgunning.

On Mar 13, 7:03 pm, "alice" <a...@fearofdolls.com> wrote:
> So if it was the lack of paste, would adding it now fix anything? I
> would assume not. Should I be looking for a new processor? It seems as
> though it was failing anyway, shutting down randomly for months now.

I don't even see where the CPU manufacturer is listed. If an Intel,
then CPU completely protects itself. It AMD and with a lesser
connection to heatsink, then most AMD CPUs would suffer no damage.

Don't worry for now about removing old heatsink paste. The existing
stuff would have been sufficient. If the heatsink was properly
manufactured, then even no paste was sufficient. Why do others
recommend removing and replacing the paste? They confuse complete
failure with what is only a mildly less conductive thermal connection.

Do unplug drives and all that other stuff completely unrelated.
That is classic shotgunning. You are making the problem exponentially
more complex. Even if a car mechanic did what you were doing, then
he would be fired. Posted above from this author is what you do.
Stop wildly speculating. That other post is how one uses principles
even from junior high science to step through the problem - break up
the exponential complication - see a problem that cannot be solved by
visual inspection or shotgunning.

On Mar 13, 4:54 pm, "w_tom" <w_t...@usa.net> wrote:
> On Mar 13, 7:03 pm, "alice" <a...@fearofdolls.com> wrote:
>
> > So if it was the lack of paste, would adding it now fix anything? I
> > would assume not. Should I be looking for a new processor? It seems as
> > though it was failing anyway, shutting down randomly for months now.
>
> I don't even see where the CPU manufacturer is listed. If an Intel,
> then CPU completely protects itself. It AMD and with a lesser
> connection to heatsink, then most AMD CPUs would suffer no damage.
>
> Don't worry for now about removing old heatsink paste. The existing
> stuff would have been sufficient. If the heatsink was properly
> manufactured, then even no paste was sufficient. Why do others
> recommend removing and replacing the paste? They confuse complete
> failure with what is only a mildly less conductive thermal connection.
>
> Do unplug drives and all that other stuff completely unrelated.
> That is classic shotgunning. You are making the problem exponentially
> more complex. Even if a car mechanic did what you were doing, then
> he would be fired. Posted above from this author is what you do.
> Stop wildly speculating. That other post is how one uses principles
> even from junior high science to step through the problem - break up
> the exponential complication - see a problem that cannot be solved by
> visual inspection or shotgunning.

Thanks for the advice. I will go get a multimeter when I have a
chance.
Here are the facts-
The PC has been occassionally turning off suddenly, with no warning,
for a couple of months. The event logs show nothing out of the
ordinary as far as I can tell. No errors.
I checked the temp. of the system in the BIOS screen and it was
getting above 80 C. It was recommended to me to check and clean the
CPU fan and heatsink. They were both very dirty and clogged with dust.
To clean them, I removed them both, then put them back.
After that, the PC would only turn on for a few seconds.
At that point, I tried another working PSU. Same thing. Then I tried
unplugging drives. Same thing. Then I tried a different CPU fan from
another PC. Same thing.
I'll let you know what I find with the meter.

On Mar 13, 5:14 pm, "alice" <a...@fearofdolls.com> wrote:
> On Mar 13, 4:54 pm, "w_tom" <w_t...@usa.net> wrote:
>
>
>
>
>
> > On Mar 13, 7:03 pm, "alice" <a...@fearofdolls.com> wrote:
>
> > > So if it was the lack of paste, would adding it now fix anything? I
> > > would assume not. Should I be looking for a new processor? It seems as
> > > though it was failing anyway, shutting down randomly for months now.
>
> > I don't even see where the CPU manufacturer is listed. If an Intel,
> > then CPU completely protects itself. It AMD and with a lesser
> > connection to heatsink, then most AMD CPUs would suffer no damage.
>
> > Don't worry for now about removing old heatsink paste. The existing
> > stuff would have been sufficient. If the heatsink was properly
> > manufactured, then even no paste was sufficient. Why do others
> > recommend removing and replacing the paste? They confuse complete
> > failure with what is only a mildly less conductive thermal connection.
>
> > Do unplug drives and all that other stuff completely unrelated.
> > That is classic shotgunning. You are making the problem exponentially
> > more complex. Even if a car mechanic did what you were doing, then
> > he would be fired. Posted above from this author is what you do.
> > Stop wildly speculating. That other post is how one uses principles
> > even from junior high science to step through the problem - break up
> > the exponential complication - see a problem that cannot be solved by
> > visual inspection or shotgunning.
>
> Thanks for the advice. I will go get a multimeter when I have a
> chance.
> Here are the facts-
> The PC has been occassionally turning off suddenly, with no warning,
> for a couple of months. The event logs show nothing out of the
> ordinary as far as I can tell. No errors.
> I checked the temp. of the system in the BIOS screen and it was
> getting above 80 C. It was recommended to me to check and clean the
> CPU fan and heatsink. They were both very dirty and clogged with dust.
> To clean them, I removed them both, then put them back.
> After that, the PC would only turn on for a few seconds.
> At that point, I tried another working PSU. Same thing. Then I tried
> unplugging drives. Same thing. Then I tried a different CPU fan from
> another PC. Same thing.
> I'll let you know what I find with the meter.- Hide quoted text -
>
> - Show quoted text -

I forgot to mention, it's an AMD Athlon.

On Mar 13, 5:14 pm, "alice" <a...@fearofdolls.com> wrote:
> On Mar 13, 4:54 pm, "w_tom" <w_t...@usa.net> wrote:
>
>
>
>
>
> > On Mar 13, 7:03 pm, "alice" <a...@fearofdolls.com> wrote:
>
> > > So if it was the lack of paste, would adding it now fix anything? I
> > > would assume not. Should I be looking for a new processor? It seems as
> > > though it was failing anyway, shutting down randomly for months now.
>
> > I don't even see where the CPU manufacturer is listed. If an Intel,
> > then CPU completely protects itself. It AMD and with a lesser
> > connection to heatsink, then most AMD CPUs would suffer no damage.
>
> > Don't worry for now about removing old heatsink paste. The existing
> > stuff would have been sufficient. If the heatsink was properly
> > manufactured, then even no paste was sufficient. Why do others
> > recommend removing and replacing the paste? They confuse complete
> > failure with what is only a mildly less conductive thermal connection.
>
> > Do unplug drives and all that other stuff completely unrelated.
> > That is classic shotgunning. You are making the problem exponentially
> > more complex. Even if a car mechanic did what you were doing, then
> > he would be fired. Posted above from this author is what you do.
> > Stop wildly speculating. That other post is how one uses principles
> > even from junior high science to step through the problem - break up
> > the exponential complication - see a problem that cannot be solved by
> > visual inspection or shotgunning.
>
> Thanks for the advice. I will go get a multimeter when I have a
> chance.
> Here are the facts-
> The PC has been occassionally turning off suddenly, with no warning,
> for a couple of months. The event logs show nothing out of the
> ordinary as far as I can tell. No errors.
> I checked the temp. of the system in the BIOS screen and it was
> getting above 80 C. It was recommended to me to check and clean the
> CPU fan and heatsink. They were both very dirty and clogged with dust.
> To clean them, I removed them both, then put them back.
> After that, the PC would only turn on for a few seconds.
> At that point, I tried another working PSU. Same thing. Then I tried
> unplugging drives. Same thing. Then I tried a different CPU fan from
> another PC. Same thing.
> I'll let you know what I find with the meter.- Hide quoted text -
>
> - Show quoted text -

I forgot to mention, it's an AMD Athlon.

On Mar 13, 7:03 pm, "alice" <a...@fearofdolls.com> wrote:
> So if it was the lack of paste, would adding it now fix anything? I
> would assume not. Should I be looking for a new processor? It seems as
> though it was failing anyway, shutting down randomly for months now.

I don't even see where the CPU manufacturer is listed. If an Intel,
then CPU completely protected itself. If AMD and with a lesser
connection to heatsink, then most AMD CPUs would suffer no damage.

Don't worry for now about removing old heatsink paste. The existing
stuff would have been sufficient. If the heatsink was properly
manufactured, then even no paste was sufficient. Ignore the hype.
Install 'follow the evidence'. What you are doing violates everything
even demonstrated in CSI. Worse, by disconnecting drives, et al, you
have making the problem exponentially more complex. That is what
shotgunning does.

Posted above from this author is what to do. Stop wildly
speculating. Other post is how one 'follows the evidence' - break up
that now exponential complication. Find a problem that cannot be
solved by visual inspection or shotgunning. Only then do we fix
anything.

On Mar 13, 4:43 pm, "w_tom" <w_t...@usa.net> wrote:
> On Mar 13, 7:02 pm, "alice" <a...@fearofdolls.com> wrote:
>
> > I'm no longer in front of it, and I can't remember the wattage, but I
> > want to say 350?
>
> There is little to zero you can see or fix only by visual inspection
> - which is why so many have you shotgunning with zero progress.
> What's worse, you tried to fix something without first establishing
> what was wrong - you removed CPU heatsinks. So what we have is maybe
> two suspects - the problem now is exponentially more complex.
>
> (BTW, if holes in the heatsink were not completely clogged, then
> heat would not be the reason for problems. )
>
> Start by establishing what is and is not. Collect facts. Don't
> swap parts. Don't disconnect anything. First discover what you do
> have.
>
> The reason a 3.5 digit multimeter is sold even in K-mart? It's for
> people with even junior high school education. You need that
> multimeter and a screwdriver - the two most essential tools. In two
> minutes we define that entire power supply 'system' either good or bad
> with almost no doubts. The procedure is posted in "When your computer
> dies without warning....." starting 6 Feb 2007 in the newsgroup
> alt.windows-xp at:
> http://tinyurl.com/yvf9vh
>
> Taking those voltage numbers from red, orange, and yellow wires when
> accessing many other peripherals (multitasking - increasing the load)
> will make those numbers even more informative. Without those numbers,
> you help can only say "try this and try that" rather than do what is
> demonstrated on a TV 'CSI': follow the evidence.
>
> Meanwhile, other useful information may have been stored in the
> system (event) logs. Information from those logs also means future
> replies will be more informative.
>
> But having changed and removed so much, then you may have one or now
> multiple problems. You have exponentially complicated the solution.
> Above is step one in a process of stepping through the problem;
> breaking the problem down into parts and analyzing those parts; KISS.
>
> If you keep swapping stuff, you may only make the problem
> unsolvable. Exactly why shotgunning is done only by those without.
>
> Get the $20 meter that is sold in K-mart, Radio Shack, Lowes, Wal-
> mart, Sears, or Home Depot. Provide those numbers. Then get responses
> that say exactly what is good or bad AND what to do next for a
> definitive solution - no more shotgunning.

I've never used a multimeter before, is there anything else I should
know to avoid hurting myself?

"alice" <> wrote in message
news: oups.com...
>I just opened up my PC case to take off the CPU fan and heatsink,
> cleaned it, put it back on, and now the whole thing will not power on.
> The power light comes on for about 5 seconds, and the CPU fan starts
> to spin, then it just turns off.
> The reason I cleaned the fan in the first place is because the machine
> was overheating and randomly turning off frequently.
> Is there any thing I can do at this point, and if I need to replace
> parts, what should I start with?
>


Alice

Several possibilities come to mind why your PC is acting the way it is.

1 You have put the heatsink on backwards and it is not seated correctly on
the CPU face.
2 The lack of thermal paste between HS & CPU is causing the CPU to heat up.
3 CPU has popped out of the socket.
4 It was just the time that the motherboard was going to die.
5 You did this with the power supply plugged into the wall and fried
something.
6 ...

You'll note I don't mention power supply, your description of what happened
makes it highly unlikely it is power supply related.

The most productive course you can take is to unplug everything from the box
(power, keyboard, mouse...) disconnect everything from the motherboard
except the power supply, CPU/Heat Sink and speaker.
Then remove Heat Sink and CPU and carefully reseat them (take appropriate
antistatic measures) with a minimal amount of thermal paste between the Heat
Sink and CPU (the paste is there to fill the microsocopic voids between HS &
CPU so improving heat transfer from CPU to Heat Sink)

At this time finish any dedusting needed, then check the tightness of
motherboard mounts and other case screws.

Now with this minimal configuration, plug the mains in and see if the
computer starts and stays on.
It will most likely start beeping because there is no RAM, this is fine as
it show's the motherboard and CPU are booting and performing their startup
process's correctly.

If the minimal configuration starts then add the disconnected parts one at a
time until you find what is causing the PC to shut down so quickly and
replace repair as appropriate. Do shutdown and disconnect the mains leads
between each item added.

If the minimum configuration does 'not' start then you will have to arrange
to try the CPU in another known good motherboard - or- take it to a tech
with the skills to diagnose hardware problems.

Best
Paul.