How to tell if laptop CPU is overheating?

[Guest]

I replaced the processor in my IBM Thinkpad A31 today, going from
a 1.8Ghz to 2.4Ghz Pentium 4-M.

When I replaced it, I noticed a layer of thermal grease that had
been placed on the processor (I assume for a close contact with the
heatsink, which was placed just above it).

Unfortunately, I didn't have any of this grease on hand, so I smeared
what I could that was left over from the old chip to the new. I am
not certain that it was enough to form a good seal, or even breach
the gap between the chip and the heatsink.

I think used the computer for about an hour, before suddenly worrying
that I might be damaging the processor unknowingly.

My question is, how likely is it that I would have damaged my processor
if, at worst, there was NO contact between the heatsink and the
processor? How about if it was just metal-on-metal, without the
grease?

And how would I *know* that there was a problem? Sudden weird system-level
errors? (I was running a Windows install at the time I decided to shut
it down, so Windows itself wasn't yet running). Would the machine
simply shut down? Would I get any kind of beeping or warning? I suppose
that last part depends on the machine itself... but I wonder if this
model processor provides thermal information that the BIOS (or other
part of the machine) can use to shut itself down.

Anybody know these sorts of things out there? IBM's website has nothing
about it.

- Tim

--

 

[Ron Reaugh]

Use MobileMeter:
http://www.geocities.co.jp/SiliconValley-Oakland/8259/release/0310/mm0310.zip

If CPU temp is below 70C then all is cool(totally intended).

As CPU temp gets ~85C or above then the CPU internally will self throttle
and slow and stay at ~85C until CPU load is reduced.

 

[Ben Jackson]

Unfortunately, I didn't have any of this grease on hand, so I smeared
what I could that was left over from the old chip to the new. I am
not certain that it was enough to form a good seal, or even breach
the gap between the chip and the heatsink.

If you didn't get a good seal the chip probably wouldn't have lasted
more than a few seconds. It's possible that the pentium-M automatically
shifts to a slower speed when it's hot, but that isn't guaranteed to
save it. Those dies are tiny and burn a lot of watts. The number that
sticks in my mind is 40wa.

 

[Andrew]

: I replaced the processor in my IBM Thinkpad A31 today, going from
: a 1.8Ghz to 2.4Ghz Pentium 4-M.

: When I replaced it, I noticed a layer of thermal grease that had
: been placed on the processor (I assume for a close contact with the
: heatsink, which was placed just above it).

: Unfortunately, I didn't have any of this grease on hand, so I smeared
: what I could that was left over from the old chip to the new. I am
: not certain that it was enough to form a good seal, or even breach
: the gap between the chip and the heatsink.

: I think used the computer for about an hour, before suddenly worrying
: that I might be damaging the processor unknowingly.

: My question is, how likely is it that I would have damaged my processor
: if, at worst, there was NO contact between the heatsink and the
: processor? How about if it was just metal-on-metal, without the
: grease?

: And how would I *know* that there was a problem?

Try Motherboard Monitor:

http://www.pcworld.com/downloads/file_description/0,fid,7309,00.asp

It displays temperature settings from the BIOS in Windows, so you can
at least see what the operating temperature is. Too bad you didn't
install this before switching CPU chips so you could compare before
and after temps...

But as the other guy said, under 70C is probably acceptable, though
under 60C would be even better. Your laptop would probably get really
hot if it was 70C...

Andrew
--
----> Portland, Oregon, USA <----
*******************************************************************
----> http://www.bizave.com <---- Photo Albums and Portland Info
----> To Email me remove "MYSHOES" from email address
*******************************************************************

 

[Guest]

Use MobileMeter:
http://www.geocities.co.jp/SiliconValley-Oakland/8259/release/0310/mm0310.zip

If CPU temp is below 70C then all is cool(totally intended).

As CPU temp gets ~85C or above then the CPU internally will self throttle
and slow and stay at ~85C until CPU load is reduced.

So for the Pentium 4-M series, it's unlikely I could have damaged
the chip in this case, you think?

- Tim

--

 

[Guest]

If you didn't get a good seal the chip probably wouldn't have lasted
more than a few seconds. It's possible that the pentium-M automatically
shifts to a slower speed when it's hot, but that isn't guaranteed to
save it. Those dies are tiny and burn a lot of watts. The number that
sticks in my mind is 40wa.

Wow, I didn't know they so much required the heat-sink. This was a
Pentium 4-M, NOT the Pentium M, by the way.

- Tim

--

 

[Guest]

: I replaced the processor in my IBM Thinkpad A31 today, going from
: a 1.8Ghz to 2.4Ghz Pentium 4-M.

: When I replaced it, I noticed a layer of thermal grease that had
: been placed on the processor (I assume for a close contact with the
: heatsink, which was placed just above it).

: Unfortunately, I didn't have any of this grease on hand, so I smeared
: what I could that was left over from the old chip to the new. I am
: not certain that it was enough to form a good seal, or even breach
: the gap between the chip and the heatsink.

: I think used the computer for about an hour, before suddenly worrying
: that I might be damaging the processor unknowingly.

: My question is, how likely is it that I would have damaged my processor
: if, at worst, there was NO contact between the heatsink and the
: processor? How about if it was just metal-on-metal, without the
: grease?

: And how would I *know* that there was a problem?

Try Motherboard Monitor:

http://www.pcworld.com/downloads/file_description/0,fid,7309,00.asp

It displays temperature settings from the BIOS in Windows, so you can
at least see what the operating temperature is. Too bad you didn't
install this before switching CPU chips so you could compare before
and after temps...

But as the other guy said, under 70C is probably acceptable, though
under 60C would be even better. Your laptop would probably get really
hot if it was 70C...

I'll never know what it was before, and I couldn't have tried it anyway,
because I don't have Windows installed on the thing (I'd blanked the disk
when the last chip was in it).

I *do* have a copy of BartPE, so I could boot to that and run one of these
programs from a floppy. Or boot my Linux CD and see if there are any
good motherboard utilities for Linux. Worth a try... I think I'll try it
right now! Thanks for the advice!

- Tim

--

 

[Ron Reaugh]

ip

So for the Pentium 4-M series, it's unlikely I could have damaged
the chip in this case, you think?

Unlikely, but theoretically possible if the CPU has no/little thermal
contact with the heatsink.

 

[Tony Hill]

I replaced the processor in my IBM Thinkpad A31 today, going from
a 1.8Ghz to 2.4Ghz Pentium 4-M.

When I replaced it, I noticed a layer of thermal grease that had
been placed on the processor (I assume for a close contact with the
heatsink, which was placed just above it).

Unfortunately, I didn't have any of this grease on hand, so I smeared
what I could that was left over from the old chip to the new. I am
not certain that it was enough to form a good seal, or even breach
the gap between the chip and the heatsink.

There should be no 'gap', or at least not one that you could notice
with the naked eye. The idea of the thermal compound is to fill the
microscopic air pockets caused by imperfections in the metal. The two
pieces should be touching, and touching rather firmly at that.

I think used the computer for about an hour, before suddenly worrying
that I might be damaging the processor unknowingly.

My question is, how likely is it that I would have damaged my processor
if, at worst, there was NO contact between the heatsink and the
processor?

If there was NO contact, your computer would have crashed within the
first minute of you using it. Most likely the chip would still have
survived with no damage, but that certainly isn't a sure thing.

How about if it was just metal-on-metal, without the
grease?

Highly unlikely that you would damage the chip, the P4 has thermal
protection built that prevents damage in such situations. Of course,
it could be throttling itself, causing the system to run much slower
than it should.

And how would I *know* that there was a problem? Sudden weird system-level
errors? (I was running a Windows install at the time I decided to shut
it down, so Windows itself wasn't yet running). Would the machine
simply shut down?

If it REALLY overheats, yes it would probably shut down.

Would I get any kind of beeping or warning? I suppose
that last part depends on the machine itself...
Yup.

but I wonder if this
model processor provides thermal information that the BIOS (or other
part of the machine) can use to shut itself down.

Yes it does.

 

[Compfix]

I'll never know what it was before, and I couldn't have tried it anyway,
because I don't have Windows installed on the thing (I'd blanked the disk
when the last chip was in it).

I *do* have a copy of BartPE, so I could boot to that and run one of these
programs from a floppy. Or boot my Linux CD and see if there are any
good motherboard utilities for Linux. Worth a try... I think I'll try it
right now! Thanks for the advice!

- Tim

FWIW. I would go to a computer shop and get some thermal compound(arctic
silver is good - about 6UKP a tube), clean both surfaces thoroughly and put
some of the fresh material on - it wont need a lot. Then you will not have
these worries, or cook your processor

Bob

 

[Guest]

FWIW. I would go to a computer shop and get some thermal compound(arctic
silver is good - about 6UKP a tube), clean both surfaces thoroughly and put
some of the fresh material on - it wont need a lot. Then you will not have
these worries, or cook your processor

That is indeed what I'm going to do. The thing is, I want
to find out *how likely* it is that my processor is already
compromised by my having used it with the tiny amount of
the old goop on it. So I'm looking for either info, or
a DOS/boot-time based program that will monitor my
cpu's temperature, since I don't have any OS installed
on that machine right now.

But thanks for the advice!

- Tim

--

 

[Robert Redelmeier]

That is indeed what I'm going to do. The thing is, I want
to find out *how likely* it is that my processor is already
compromised by my having used it with the tiny amount of
the old goop on it. So I'm looking for either info, or
a DOS/boot-time based program that will monitor my cpu's
temperature, since I don't have any OS installed on that

Boot into BIOS setup. There's usually a "PC Health" screen
that shows temperatures live.

-- Robert

 

[Guest]

Boot into BIOS setup. There's usually a "PC Health" screen
that shows temperatures live.

Hmm.. I looked at all my BIOS screens and couldn't find one.
You'd think IBM would have put that into their BIOS.

- Tim

--

 

[Guest]

[ snippage ]

Thanks Tony! That was very helpful!

- Tim

--

 

[glen herrmannsfeldt]

That is indeed what I'm going to do. The thing is, I want
to find out *how likely* it is that my processor is already
compromised by my having used it with the tiny amount of
the old goop on it. So I'm looking for either info, or
a DOS/boot-time based program that will monitor my
cpu's temperature, since I don't have any OS installed
on that machine right now.

The goal is to fill all the air space, as air is a poor
heat conductor. Other than that, less is better than
more. Like in many other cases, too much of a good thing
isn't good.

We can't tell from here how much you actually put in, though.

-- glen

 

[Guest]

The goal is to fill all the air space, as air is a poor
heat conductor. Other than that, less is better than
more. Like in many other cases, too much of a good thing
isn't good.

We can't tell from here how much you actually put in, though.

Just a thin layer, from what was left over from the previous chip.
I applied it with my finger, however, so purists might point to the
fact that skin oils, etc. would get in the way of conductivity.
I'll be doing it the right way with the new stuff (I got a tube
of Arctic Silver last night at CompUSA).

- Tim

--

 

[The little lost angel]

Hmm.. I looked at all my BIOS screens and couldn't find one.
You'd think IBM would have put that into their BIOS.

I've noticed the PC Health screen in a laptop, almost all the desktop
have it. So it could just be some sort of trend where laptop
manufacturers don't want to frighten users when they see 70C in a
laptop and compare it to the 40C in their desktop Pppp

I don't think you really need to worry about it. My P4-M (the real
mobile chip, not the P4 converted to mobile use, I can't keep them
sorted out, P4-M or P4 Mobile?) 1.7G chews about 18W doing office work
and around 40W or so when I deliberately load it up. At these kind of
power levels, I think even a lousy contact would not be too major a
problem, the thermal throttling would have sufficient time to kick in
before things get hhhhhhot PpPppp


--
L.Angel: I'm looking for web design work.
If you need basic to med complexity webpages at affordable rates, email me
Standard HTML, SHTML, MySQL + PHP or ASP, Javascript.
If you really want, FrontPage & DreamWeaver too.
But keep in mind you pay extra bandwidth for their bloated code