CPU ?

  • Thread starter Thread starter philo
  • Start date Start date
P

philo

Right now I'm working on an XP-2200 based machine.
The "customer" had replaced the cpu's heatsink and "thinks" he might
have "damaged" the cpu by overheating. (Too much thermal compound.)
He said that after he replaced the heatsink the machine started to crash a
lot
and that he ended up with a lot of corrupted data. (He's a musician and a
lot of his music files became unplayable.)

He took the machine into a professional repair service and they said they
could
only get the machine to run with stability when they clocked to down to
1.3mhz...so they replaced the cpu and now the machine is running OK.

The whole thing seems a little odd to me...I'm wondering if such a thing is
really
possible.

At any rate I've performed WD's full diagnostic on both the harddrives
and they passed. However I did find some *.aif files that only brought up
a media player error message. *Plus* one file that had *no* name (just a
generic
icon in windows explorer) and size of 18 megs...but when checking properties
showed size of zero. The file could not be deleted or named...so I had to
move all the data out of that folder and delete the entire folder. Prior to
that I
had run chkdsk /f but it found *no* errors.

Philo
 
philo said:
The whole thing seems a little odd to me...I'm wondering if such a thing is
really
possible.
Click to expand...

P4s are meant to slow down when overheating and be "undamagable", but
anything is possible
At any rate I've performed WD's full diagnostic on both the harddrives
and they passed. However I did find some *.aif files that only brought up
a media player error message. *Plus* one file that had *no* name (just a
generic
icon in windows explorer) and size of 18 megs...but when checking properties
showed size of zero. The file could not be deleted or named...so I had to
move all the data out of that folder and delete the entire folder. Prior to
that I
had run chkdsk /f but it found *no* errors.
Click to expand...

Most of the time you can delete these in a command window by using the 8.3
filename.

Michael Culley
 
Right now I'm working on an XP-2200 based machine.
The "customer" had replaced the cpu's heatsink and "thinks" he might
have "damaged" the cpu by overheating. (Too much thermal compound.)
Click to expand...

That sounds very unlikely. Very crude compound or a huge tube or tub of
the stuff might've separated and be so thick that it could interfere with
heat transfer, but any typical compound would be squished out immediately
when the 'sink was placed on a small core like an XP2200, and whatever
excess remained would be thinned further by the overheating CPU and be
even more likely to ooze out within an on/off cycle or two.
He said that after he replaced the heatsink the machine started to crash a
lot
and that he ended up with a lot of corrupted data. (He's a musician and a
lot of his music files became unplayable.)
Click to expand...

I suspect the heatsink was improperly mounted, perhaps backwards or with a
clip that didn't have a center tab to keep the 'sink from being pressed
against the "ledge" of the socket over the lever, cam area, and the core
wasn't fully in contact with the CPU due to it's base not being parallel
to the surface. If such a situation existed then an overapplication of
compound might've actually saved the chip from frying, providing a very
poor interface instead of just air on a large portion of the core.

A crashing system could certainly corrupt data->files without any further
CPU "defect/damage" than that.

He took the machine into a professional repair service and they said they
could
only get the machine to run with stability when they clocked to down to
1.3mhz...so they replaced the cpu and now the machine is running OK.
Click to expand...

I wonder, did they remove and reinstall the heatsink or just turn it on
and seeing it instable, proceeded to underclock it, drawing the wrong
conclusion or else *wanting* to find a reason to replace the CPU, sell a
new one (or was it a warranty replacement?).
The whole thing seems a little odd to me...I'm wondering if such a thing is
really
possible.
Click to expand...

For the CPU to be damaged then only work at 1.3GHz?
There might be an extremely slim chance that heat could strain the
connections and keeping the chip at lower temp is keeping it working, but
that's only a theory, no proof that this has ever happened, and that "slim
chance" is so slim it's not really worth considering, it'd be just as
likely that the CPU would fail regardless of whether the heatsink was
swapped, also unlikely.
At any rate I've performed WD's full diagnostic on both the harddrives
and they passed. However I did find some *.aif files that only brought up
a media player error message. *Plus* one file that had *no* name (just a
generic
icon in windows explorer) and size of 18 megs...but when checking properties
showed size of zero. The file could not be deleted or named...so I had to
move all the data out of that folder and delete the entire folder. Prior to
that I
had run chkdsk /f but it found *no* errors.

Philo
Click to expand...

Hopefully the CPU was retained so it can be further tested.
 
Hopefully the CPU was retained so it can be further tested.
Click to expand...

thanks for the reply...
i do not have the cpu in question and i somehow wonder if the shop really
replaced it...
at any rate the person whose machine it is...is only concerned with getting
it working properly.

here is where i am thus far:
the master harddrive is a WD 40 gig
and the slave is a WD 80 gig...
and as i've stated...both of them pass WD's full diagnostic test...

however if i try to defrag the 80 gig drive in the machine (tried all
combinations
of master/slave ide0, ide1) no matter what i do...
as soon as i try to run defrag...the machine goes into a solid lockup!

however if i put the WD80 in another machine, i can defrag it or
if i put another drive in the machine i'm working on...i can also defrag
it...

so there is a problem only if the 80gig drive is in the machine i'm working
on.
to get the guy up and running, i may just replace the drive or perhaps try
it on a pci...ide controller card.

philo
 
thanks for the reply...
i do not have the cpu in question and i somehow wonder if the shop really
replaced it...
at any rate the person whose machine it is...is only concerned with getting
it working properly.

here is where i am thus far:
the master harddrive is a WD 40 gig
and the slave is a WD 80 gig...
and as i've stated...both of them pass WD's full diagnostic test...

however if i try to defrag the 80 gig drive in the machine (tried all
combinations
of master/slave ide0, ide1) no matter what i do...
as soon as i try to run defrag...the machine goes into a solid lockup!
Click to expand...

Perhaps there's some flag that's marked the drive as bad, needs reset?
That's just a theory, I don't know.
however if i put the WD80 in another machine, i can defrag it or
if i put another drive in the machine i'm working on...i can also defrag
it...

so there is a problem only if the 80gig drive is in the machine i'm working
on.
to get the guy up and running, i may just replace the drive or perhaps try
it on a pci...ide controller card.

philo
Click to expand...

That's an odd situation, I suspect WinXP has some kind of configuration
problem relative to that drive... which I don't know how to fix.

This 80GB drive was the OS drive?
If the drive had data corruption (or even if the other drive was the OS
drive) it may have OS file corruption too... Possibly reinstalling the OS
is necessary, or from a backup or perhaps system restore might be the
first attempt... don't know.
 
That's an odd situation, I suspect WinXP has some kind of configuration
problem relative to that drive... which I don't know how to fix.

This 80GB drive was the OS drive?
If the drive had data corruption (or even if the other drive was the OS
drive) it may have OS file corruption too... Possibly reinstalling the OS
is necessary, or from a backup or perhaps system restore might be the
first attempt... don't know.
Click to expand...


it was just a "lucky" boot...
the next few times i tried the machine with a different slave drive...
running defrag caused a lockup.

what i did was just remove the 40 gig drive and the 80 gig drive
then reinstall XP on a single 120gig drive...
i'm now transferring the data...(luckily i saved about 99% of it)

btw: during the reinstall of XP...i had a momentary power failure *exactly*
at the reboot point! wow talk about luck.
anyway...i now have my workbench on a UPS...don't know why i never bothered
before !

philo
 
Back
Top