Passive heatsink for slot 1 PII/PIII?

[Tim Green]

Has anyone ever seen a passive heatsink like this for a Slot1 PII/PIII
processor:

http://timgreen.bei.t-online.de/72_1.jpg

I found it on Ebay and the seller says it comes out of a Compaq or Dell
computer but doesn't have any more information. I'm building a PC-based
video-recorder and I'd like it to be as quiet as possible. But I suspect
that it would be too much to hope that I can cool a 933 MHz PC with this
little piece of anodized aluminum -- particularly since I read somewhere
that the 933 was one of the hotter PIIIs.

Any ideas?

 

[JT]

Has anyone ever seen a passive heatsink like this for a Slot1 PII/PIII
processor:

http://timgreen.bei.t-online.de/72_1.jpg

I found it on Ebay and the seller says it comes out of a Compaq or Dell
computer but doesn't have any more information. I'm building a PC-based
video-recorder and I'd like it to be as quiet as possible. But I suspect
that it would be too much to hope that I can cool a 933 MHz PC with this
little piece of anodized aluminum -- particularly since I read somewhere
that the 933 was one of the hotter PIIIs.

Any ideas?

These passive heatsinks relied on a duct going to a rear case fan. You
still need to move air accross them to keep the processor cool. They will
work with a larger, slower, quieter fan that still moves a lot of air, but
they don't eliminate the need for a fan. They just move where the fan
should be. I don't think they would be adequate for a 933 P3

 

[Tim Green]

I don't think they would be adequate for a 933 P3

Sigh...I thought so, but you can dream. The original fan on the P3 has
gotten pretty noisy and it would have been nice to eliminate the noise
completely. It's difficult to cool Slot 1 processors quietly because you
can't really mount a 60mm or 80mm fan on them...

 

[w_tom]

Most cooling for a heatsink is accomplished by moving air
slowly over the heatsink. Massive increases in airflow only
result in minor cooling improvement. Better designed PII and
PIII systems (including Dells and Compaqs) used a duct to
concentrate airflow into the ATX power supply across heatsink
fins. Therefore no 'noisy' heatsink fan was necessary.

To better answer your question, one must first provide some
specs - especially the essential "degree C per watt" spec.
Without that number, then only testing will obtain a number
and therefore CPU temperature.

That heatsink probably came with a ducted power supply so
that sufficient, low volume airflow was directed over that
heatsink. Some other examples of passive heatsinks. Use
these examples to help understand what is possible and
equivalent. Minimally acceptable products always provide that
'degree C per watt' number:
http://209.58.132.52/products/microp/desktop.shtml

 

[WooduCoodu]

the 400mhz p2 that was in my old dell r400 had a very similar looking
heatsink and was passively cooled, no ducts or anything. not sure how it was
attached to the processor, perhaps a thermal adhesive, but there was a
special bracket that held the heatsink to the motherboard.

 

[~misfit~]

Has anyone ever seen a passive heatsink like this for a Slot1 PII/PIII
processor:

http://timgreen.bei.t-online.de/72_1.jpg

I found it on Ebay and the seller says it comes out of a Compaq or Dell
computer but doesn't have any more information. I'm building a PC-based
video-recorder and I'd like it to be as quiet as possible. But I suspect
that it would be too much to hope that I can cool a 933 MHz PC with this
little piece of anodized aluminum -- particularly since I read somewhere
that the 933 was one of the hotter PIIIs.

I have an identical heatsink attached to a PII-350 that came from a Dell.

I doubt it would cool your 933, even on the PII it was near a fan.

Compaq PCs used an entirely different heatsink made by Foxcon.

 

[kony]

Has anyone ever seen a passive heatsink like this for a Slot1 PII/PIII
processor:

http://timgreen.bei.t-online.de/72_1.jpg

I found it on Ebay and the seller says it comes out of a Compaq or Dell
computer but doesn't have any more information. I'm building a PC-based
video-recorder and I'd like it to be as quiet as possible. But I suspect
that it would be too much to hope that I can cool a 933 MHz PC with this
little piece of anodized aluminum -- particularly since I read somewhere
that the 933 was one of the hotter PIIIs.

Any ideas?

Of all the things I'd call that, "little" isn't one of them... It
looks quite large, like the steel clips on the bottom of it, slide on
to a slot-1 board's plastic heatsink retention bracket. This is a
completely separate piece(s) of plastic from the slot-1 retention
bracket. If your board doesn't have the heatsink bracket, and the
heatsink doesn't come with it (or your board doesn't have the holes
for it), the heatsink may put excessive strain on the slot.

I might actually have a duct for one of those, but I"m unsure if it
fits that particular 'sink or something similar, and of course it also
needs a power supply that can accept it, and the slot on the board has
to be in exactly the right spot... these OEM 'sinks are best left in
the original system because it's just too much work to adapt another
system to use one.

Anyway, given a small drill bit, a couple of tiny bolts, I'm sure you
could mount a fairly large but quiet fan on that heatsink (as well as
many others) but that particular 'sink looks like it might be too
tall/wide, that having the fan on the end (parallel to the face of the
CPU) might block one or more memory slots on the typical slot 1 board.

What about the 'sink it's currently using? No way to strap a
different fan on there? Aluminum heatsinks are pretty easy to drill
holes into for bolt-mounting a fan, or take a few bent pieces of metal
(like electrical tie-down straps) to bridge the gap between where you
can get a good mounting-point on the 'sink, and the mounting points on
the fan of your choice. Maybe I'm not being clear, but just be
creative, there are tons of ways to strap a fan onto a heatsink.


Dave

 

[Timothy Daniels]

"w_tom" posted a dead link:

Minimally acceptable products always provide that
'degree C per watt' number:
http://209.58.132.52/products/microp/desktop.shtml

"The page cannot be found"


*TimDaniels*

 

[Timothy Daniels]

the 400mhz p2 that was in my old dell r400 had a very similar looking
heatsink and was passively cooled, no ducts or anything. not sure how
it was attached to the processor, perhaps a thermal adhesive, but there
was a special bracket that held the heatsink to the motherboard.

I probably have a "passive" heatsink identical to yours, but it's near
the exhaust fan - which creates a constant draft through the cooling fins.
It's "passive" in the sense that no dedicated fan is necessary, but it does
rely on a forced air flow as opposed to convection or radiation.
(OT question: Which form of cooling do communication satellites use?)


*TimDaniels*

 

[WooduCoodu]

i don't have this computer anymore but i do remember that it only had one
rear case fan and the heatsink was on the opposite side of slot processor
from the fan so i doubt there was much airflow across it with the processor
between it and the fan.

 

[w_tom]

Web sites. They come and they go. Here is another that
will provide examples of Passive CPU heatsinks - and exists:
http://tinyurl.com/ou58

 

[w_tom]

Heatsinks are rated typically with an airflow of 200 LFM.
To a human hand, that would appear to be no airflow.
Quadrupling that airflow only doubles the cooling abilities of
that heatsink. And so it goes. Massive increases in air flow
has diminishing improvement in cooling.

Cooling is a discussion about bottlenecks. Heat bottlenecks
are most often where medium (material) changes. CPU to
thermal compound. Thermal compound to heatsink. Heatsink to
air. The science (or art) of heatsink design is found in
minimizing these bottlenecks. When more of a heatsink
directly contacts CPU (instead of thermal compound), then
another bottleneck is addressed. Massive increases in airflow
does not necessarily address a thermal bottleneck.

Ideal cooling would be a fan not mounted on heatsink. BTW,
all CPUs have a maximum permitted weight. Fan instead
mounted on sound retardent material so that air blows across
heatsink. Low noise. Acceptable amount of cooling -
depending, of course, on CPU wattage and other parameter
measured by the "degree C per watt" specification.

Better computer designs did these mathematical calculations,
then confirmed that design with experiments. It is why better
designed PIIs and PIIIs never required a CPU fan - and all
that noise. How many watts in that processor. Therein lies
an important number before a passive heatsink design is even
considered.

 

[Muttley]

Is the processor still under warranty?
IIRC it's three years for a retail boxed CPU.
Might be able to get a replacement fan assy from Intel.

Had to do this for a noisy fan on a P3-450 a fair while ago (while still
under warranty) and got a replacement shipped at nil cost.

John S.

 

[Tim Green]

First of all thanks to everyone for the input. The background to the
airflow/thermal border problems should probably be getting more coverage
nowadays, but then that would cause massive problems for computer
manfacturers. I think what we're really saying is that either most current
processors (with the exception of VIA and Transmeta/Crusoe) are massively
overclocked, or it's time for fluid cooling as standard... )

What about the 'sink it's currently using? No way to strap a
different fan on there?
...Maybe I'm not being clear, but just be
creative, there are tons of ways to strap a fan onto a heatsink.

That's clearly the way I'm going to be going. My first thought was the path
of least resistance --if there had been a workable passive solution it would
have been ideal.

Basically what I'm building is a Linux-based multimedia centre with two
TechnoTrend satellite DVB TV cards (one Nexus-S hardware MPEG for recording,
one plain vanilla Nova-S for watching while recording) that's going to
replace the satellite receiver, DVD player, CD player and VCR, with the
additional advantage of full support for MP3, DivX etc. Plus it's going to
be networked to my office computer so I can control recordings, updates etc.
from there, while the normal "media" controls will be handled in the living
room from a remote.

 

[spodosaurus]

Has anyone ever seen a passive heatsink like this for a Slot1 PII/PIII
processor:

http://timgreen.bei.t-online.de/72_1.jpg

I found it on Ebay and the seller says it comes out of a Compaq or Dell
computer but doesn't have any more information. I'm building a PC-based
video-recorder and I'd like it to be as quiet as possible. But I suspect
that it would be too much to hope that I can cool a 933 MHz PC with this
little piece of anodized aluminum -- particularly since I read somewhere
that the 933 was one of the hotter PIIIs.

Any ideas?

I've seen large heat sinks mounted on old Dell PII 233's...BUT they have
a system fan pulling the air past them at the back of the case, along
with the PSU fan!

--

Are you registered as a bone marrow donor? You regenerate what you
donate. You are offered the chance to donate only if you match a person
on the recipient list. Visit www.marrow.org or call your local Red Cross
and ask about registering to be a bone marrow donor.

spam trap: replace shyah_right! with hotmail when replying

 

[Timothy Drouillard]

I HAVE seen that heat sink before, but I can't remember which system I saw
it on. Could very well have been a Dell or a Compaq.

I have a Dell Dimension XPS B1000R with a 1gig slot 1 CPU that uses a
passive heat sink on the CPU but Dell mounts a large fan on the rear of the
case then uses a duct to direct the airflow across the CPU.