Mobile Pentium - to Heatsink contact -- now tight?

  • Thread starter Thread starter Guest
  • Start date Start date
G

Guest

I installed a new chip in my Thinkpad A31, wherein the fan/heatsink
sits atop the chip. Although the hardware guide comes with a list
of torque specs for the screws involved, I don't own a torque screwdriver,
so I just tightened them a bit beyond snug.

Anyone ever have a problem where they tightened such an assembly
to such a point that the resulting pressure on the processor
caused it to fail?

- Tim

--
 
Spammay said:
I installed a new chip in my Thinkpad A31, wherein the fan/heatsink
sits atop the chip. Although the hardware guide comes with a list
of torque specs for the screws involved, I don't own a torque
screwdriver, so I just tightened them a bit beyond snug.

Anyone ever have a problem where they tightened such an assembly
to such a point that the resulting pressure on the processor
caused it to fail?

- Tim

The torque spec is probably for calibrating the assembly machine which
can exert enough torque to lock the screws into the threads, or strip
the threads, or break the head from the screw, or pull the board apart.
Only the IBM Department of Screw Torque Specifications for P4 Central
Processing Chip Heat Sinks in Notebook Computers would know for sure.
"Snug" is a good target when doing it by hand. There is some stress
relief built in to many laptop heat sink assemblies (did you insert the
screws inside a brass insert containing a spring in the aluminum heat
sink?) to compensate the force for thermal expansion so that bottoming
the screws is OK, and once you feel the screw bottom out, there is not
much point in tightening any further past "just a tiny bit", i.e.,
"snug".

Q
 
The torque spec is probably for calibrating the assembly machine which
can exert enough torque to lock the screws into the threads, or strip
the threads, or break the head from the screw, or pull the board apart.
Only the IBM Department of Screw Torque Specifications for P4 Central
Processing Chip Heat Sinks in Notebook Computers would know for sure.

Wow, I want a job in THAT department! :-)
"Snug" is a good target when doing it by hand. There is some stress
relief built in to many laptop heat sink assemblies (did you insert the
screws inside a brass insert containing a spring in the aluminum heat
sink?) to compensate the force for thermal expansion so that bottoming
the screws is OK, and once you feel the screw bottom out, there is not
much point in tightening any further past "just a tiny bit", i.e.,
"snug".

You're quite right, there WERE brass inserts that the screw went through.
Since the machine is working fine now, I'm assuming All Is Well.
Hopefully the Artic Silver I applied is also doing a bang-up job of
heat-transfer, and my computer will live a little longer with it's
new chip.

Thanks much!

- Tim

--
 
Quaoar said:
The torque spec is probably for calibrating the assembly machine which
can exert enough torque to lock the screws into the threads, or strip
the threads, or break the head from the screw, or pull the board apart.
Only the IBM Department of Screw Torque Specifications for P4 Central
Processing Chip Heat Sinks in Notebook Computers would know for sure.
"Snug" is a good target when doing it by hand. There is some stress
relief built in to many laptop heat sink assemblies (did you insert the
screws inside a brass insert containing a spring in the aluminum heat
sink?) to compensate the force for thermal expansion so that bottoming
the screws is OK, and once you feel the screw bottom out, there is not
much point in tightening any further past "just a tiny bit", i.e.,
"snug".

Q
The torque spec is probably to get enough pressure to assure good
contact, but not so much that the chip is cracked or the solder balls
are crushed. Seems unlikely, although possible, that those screws
aren't put in by hand perhaps with a power screwdriver.

del cecchi.
 
Quaoar said:
The torque spec is probably for calibrating the assembly machine which
can exert enough torque to lock the screws into the threads, or strip
the threads, or break the head from the screw, or pull the board
apart.

No. The torque numbers are there for the service technicians. Most screws on
an IBM Thinkpad are coated (black screws), and only meant to be used once.
The correct torque is listed to avoid the screws being overtorqued by the
service people when re-assembling the notebook to avoid breaking the platics
part or to have the torque too low so that screws can come loose.

For uncoated screws like the one of an IBM A31 heatsink, without torque
screw driver I'd recommend screwing in with very low force until the screw
head reaches the component surface, then adding another 90 degree, and
securing the screw with liquid torque seal.

Benjamin
 
Spammay said:
I installed a new chip in my Thinkpad A31, wherein the fan/heatsink
sits atop the chip. Although the hardware guide comes with a list
of torque specs for the screws involved, I don't own a torque
screwdriver, so I just tightened them a bit beyond snug.

That should do fine.
Anyone ever have a problem where they tightened such an assembly
to such a point that the resulting pressure on the processor
caused it to fail?

You can't damage the CPU by overtightening the screws. The CPU PCB can
withstand a lot of pressure, much more than the bolts in the case that will
break loose much earlier.

Benjamin
 
That should do fine.


You can't damage the CPU by overtightening the screws. The CPU PCB can
withstand a lot of pressure, much more than the bolts in the case that will
break loose much earlier.

Thanks for all the information -- sounds like you really know your stuff!
The chip seems to be doing fine, and keeping relatively cool (although I have
noticed that every so often it gets up to 68 degrees C, and the fan runs for awhile,
and it gets back down to 40 or so. Funny laptops.

- Tim

--
 
Back
Top