(NVIDIA) Fan-Based-Heatsink Designs are probably wrong. (suck, don't blow ! heatfins direction)

[Tim Williams]

But also very probably wrong. The volume of air going through the heat
sink is proportional to the amount of cooling you get for a given design
so there is a definite bias towards not blocking off half the free air
flow. I would guess at something more like allowing 2/3 to 3/4 of free
airflow as about the best depending on the exact heatsink geometry. It
could easily be higher - easy enough to do the experiment.

Indeed, and add to that the fact that fans are not "resistive" air
sources. The peak power point (pressure * flow) occurs at a pressure of
about 25% of maximum (fully blocked) pressure. You can't operate very far
from this condition or your flow will be too slow.

If you include dynamic pressure (mass flow), fans are even more nonlinear.
Next time you have a squirrel cage type laying around, hook it up and play
with it. Put your hand over the outlet. You'll find the velocity is
great until about 1-2 diameters away, where you start feeling the force of
ram air. Within about 0.5 to 0.25 diameters, pressure is maximum, because
flow hasn't gone to zero yet, meanwhile static pressure is building. Put
your hand all the way up to block it, and static pressure goes to maximum,
but velocity goes to zero, so the power you're feeling drops sharply.

BTW, I use the example of a squirrel cage because they provide more
pressure, making a more illustrative example. Regular axial fans do as
well, and manufacturers typically provide comparable graphs.

Tim

 

[Flasherly]

But also very probably wrong. The volume of air going through the heat
sink is proportional to the amount of cooling you get for a given design
so there is a definite bias towards not blocking off half the free air
flow. I would guess at something more like allowing 2/3 to 3/4 of free
airflow as about the best depending on the exact heatsink geometry. It
could easily be higher - easy enough to do the experiment.

I suspect the perfect shape for an optimum heatsink is rather more
complex than the typical fins we get but the designs used at present are
good enough and much easier to engineer. Heat pipes have helped
enormously with the latest generation of quiet heatsinks.

It is a sobering thought that high performance CPUs often have a heat
output per unit area that exceeds the tip of a soldering iron.

Nope, no theta modules presently marketed, just a couple squirrel
cages and extant circular Zalman takes from a fairy good run given a
premium to pricing structures. Mass seems the byword, now-a-days,
massive as a restriction only limited by standardization among case
manufacturers. Had one recently, the typical behemoth of 7- to 1155
sockets, I picked for a proverberial song & dance, which barely missed
a nonstandard case construction I do own, within designer case
specifications by a mere 1/4", (yes, I simply had to measure it), as
opposed to a standard, however exact fit, such as Rosewill's
understudy of Antec cases. Excepting the fan -- I'd as well be
veritably ecstatic over results obtained within a reality of present
heatwick technology -- as it is, the size of an exterior case fan
ported and packaged to that CPU HS is at best, safer to defer for a
project to rewire its connection off the MB current draw and onto a PS
lead, proper.

As mentioned aside by similar instances of a P4 or AMD X2, a benefit
not only set to 107F (and lower, I reside at as low as 100F respective
to ambient temperatures), is a backtest of their efficiency to
approach flash computational processes conceivably closer to these
"soldering tips" of 130F, for as much in as least time possible then
to regain steady state 107F operational status. It's my own personal
theory, fwtw, that case designs importantly conducive to achieving
such good results, are at much an impasse the last generation of P4s
encountered before heading into nonlinear modes of augmented,
multicore processing. Extensively drilled, variously meshed and
screened, the approach has effectively advanced to a breadboard
construct from a standpoint of free air.

 

[krw]

Right.

I've bought about every heatsink or fan imaginable, given and within,
as another mentioned, an axiomatic engineering construct concept --
'if it's done right [in the first place], it's time to [attempt] an
improvement, [if and while not in need of entirely new construction
concepts]';- time simply follows, technologically speaking, to march
upon and then past any R&D Dept., in failure aptly to communicate, if
at least not uniquely, then an underlying implementation of adequacy
pertaining to key hard and software, constructional elements coming
in, daily, across so broad a field as computers. Where, specifically,
I see you for a fit is at a coincident juncture of heatsink fins and
augmented airflow, both being common terms to common acceptance for
pragmatic practice. The argument you have placed to ally yourself as
well follows true to the same axiom: that being one, effectively, of a
quest for perpetual motion: for so long as the key component of design
efficiency is established, widely employed to a common basis of
underlying industrial acceptance, the cost factor, then, is
effectively one which requires of me nothing more, out of my pocket,
than to advance a better sense of encouragement that such benefit, you
have chosen to propose, indeed, is of worthier consideration. Stop by
anytime if you'd like to talk, John. My office is at the top of the
stairs.

Are you really John Fields?

Or DimBulb. It's sometimes hard to tell them apart.

 

[TheGlimmerMan]

Or DimBulb. It's sometimes hard to tell them apart.

You and Larkin are both idiots.

 

[krw]

You and Larkin are both idiots.

From you, AlwayWrong, that's some compliment.

 

[Quadibloc]

What you want is immersion cooling:

I did a Google search after my post, and found that all the references
to the use of mineral oil were to immersion cooling. However, I have
also seen it noted that mineral oil could damage some parts of present-
day computers.

I was thinking in terms of avoiding immersion, but using a forced
flow.

John Savard

 

[Jasen Betts]

Indeed, and add to that the fact that fans are not "resistive" air
sources. The peak power point (pressure * flow) occurs at a pressure of
about 25% of maximum (fully blocked) pressure. You can't operate very far
from this condition or your flow will be too slow.

If you include dynamic pressure (mass flow), fans are even more nonlinear.
Next time you have a squirrel cage type laying around, hook it up and play
with it. Put your hand over the outlet. You'll find the velocity is
great until about 1-2 diameters away, where you start feeling the force of
ram air. Within about 0.5 to 0.25 diameters, pressure is maximum, because
flow hasn't gone to zero yet, meanwhile static pressure is building. Put
your hand all the way up to block it, and static pressure goes to maximum,
but velocity goes to zero, so the power you're feeling drops sharply.

BTW, I use the example of a squirrel cage because they provide more
pressure, making a more illustrative example. Regular axial fans do as
well, and manufacturers typically provide comparable graphs.

I've observed this too, I first observed it playing with a squirrel-cage fan
driven by a 1200W series universal motor (Electrolux ) The universal motor
made it more obvious because these motors respond more to load changes.

Modern cooling fans often have a pulse output that indicates fan speed, but
monitoring software seems to only alarm on a low speed, where a high speed
should possibly also be alerted as it may indicate a clogged heatsink.

The fans with a PWM input would be harder to handle as the pwm to
speed relationship would need to be compared

 

[David]

All of you are responding to a known troll and idiot, Skyduck Farting.

 

[Flasherly]

I wonder if CPU chip layouts include hot-spot distribution, like
putting the hottest bits into the corners or something.

I wouldn't as generally implemented within a more direct approach
ostensibly to remove added abstractions through dynamic modeling
practices as gate clocking and fetch latches contingent upon sensor
relays. Labeled under a safety badge to ensure another preventative
layer against failure conditions, the intent is established in
advantage to saving the core processor when a $2 heatsink or fan
malfunctions or improperly is manipulated outside provisional intents
by which they're packaged and distributed.

 

[Flasherly]

Word salad. You must be AlwaysWrong.

Lex parsimoniae -- irrespective of older processors I do own, to have
attributed heat as randomness to an ordering of chip density cannot
either wrong or right, whether suspect or implicit in indulgence, much
as application [more correctly] negates relevancy by dint of simple
apparency;- well, almost. . .I did elect not to turn on heat
throttling in my CMOS.

 

[Skybuck Flying]

"Jeff Liebermann" wrote in message

I have this theory that the fins of a heat sink should reduce a fan's
free-flow rate by 50% for optimum heat transfer.

"
On the original assertion, that it's better to suck than to blow,
methinks that's wrong.
"

I will draw a more detailed drawing for you what I ment with "suck". There
is also some "blow" involved.

Side view of proposed heat sink design by skybuck:


+--------------------------------------+
<out----------- airflow -------------in FAN or CASE FAN
<------------- airflow ---------------
|^|^|^|^|^|^|^|^|^|^|^|^| <- suckage effect going up
|S|S|S|S|S|S|S|S|S|S|S|S|
|U|U|U|U|U|U|U|U|U|U|U|U| <- heatfins
|C|C|C|C|C|C|C|C|C|C|C|C|
|K|K|K|K|K|K|K|K|K|K|K|K|K|
+--------------------------------------+

By blowing air over the heat fins as proposed this will hopefully create a
suck effect, sucking any dust out from between the heatfins

I do see some problems with this design... the tunnel will be small.... and
a big fan will have trouble blowing air into it... maybe a small one will be
enough... low rpm hopefully.

Bye,
Skybuck.

 

[Skybuck Flying]

"Timothy Daniels" wrote in message

[.............] On the original assertion, that it's better to suck than
to blow,
methinks that's wrong. .....

"
Given the same mass/sec flow of air over the fins of a heatsink,
the best heat transfer is by blowing due to the greater turbulence -
which disturbs the boundary layer of air that lies in contact with
the fins and puts more flowing air in direct contact with the surface
of the fins. In the case where the fins rise up away from the source
of the heat, it's best to blow downward from the ends of the fins
toward the source of the heat. IOW, the air should move in a
direction opposite to the heat flow.

This principle is not only used in heat transfer systems, but also in
biological systems in oxygen transfer through membranes - as in
fish gills where the blood moves across the gill membrane in a
direction opposite to the flow of water. The basis of this principle
lies in the finite heat (or gas) capacity of a fluid and that greatest
heat (or gas) flow occurs as a linear function of the difference of
temperature (or gas concentration) between 2 bodies. Apply a little
calculus, and the principle of opposing flows results. This design
principle was recently seen when I opened up the case of a friend's
PC to clean it out: The cooling fins for the CPU rose up from the CPU,
and the cooling fan blew air down along the fins toward the CPU.
Obviously, the designer had paid attention during college freshman
physics.
"

I'd love to see simulation that actually includes dust particles and hair to
see how much effectiveness remains for this theory.

I suspect the simulation software used at the time did not include these
factors, and therefore all designs might be totally wrong for dusty/hairy
environments.

Bye,
Skybuck.

 

[Skybuck Flying]

wrote in message

Note that heat transfer by volume isn't usually the goal, so much as
minimum temperature is. In a counterflow setup, the hottest part of the
heatsink is cooled by the hottest air. If you flip it around, the hottest
part of the heatsink gets cooled by the coolest air, achieving the highest
heat flux for a given surface area and temperature difference -- more
power density, at some expense to mass flow and pumping loss. You might
avoid this, for example, if you had to use pure nitrogen (or helium, for
that matter) for some process, minimizing the gas flow to keep operating
cost down.

"
Why not use compressed/expanded air for this purpose ? Using a piston
compressor to compress the air to a few bars, the air gets quite hot,
then let it go through a heat exchanger to get rid of most of the heat
and cool the pressurized air closer to ambient temperature.

Let the air expand to normal ambient pressure and the air temperature
is now well below ambient temperature and let it flow through
semiconductor heatsinks to the environment.

To avoid problems with dust and condensation, a closed loop might make
sense, but of course, now the heat exchanger would also have to
dissipate the heat from the semiconductor. However, the heat exchanger
can be remotely located and it can have much higher temperatures than
the semiconductors, getting rid of the heat into the environment would
be easier.
"

I like this idea of a closed air system very much...

Maybe a case which is build entirely out of "heatsinks" or something... to
get rid of as much heat from inside the case to the outside...
without actually sucking in any dust/hair.

Bye,
Skybuck.

 

[Skybuck Flying]

Also perhaps a vaccuum would be created on the suck sections.

So then on the bottom little holes would need to be made to create little
openings to let air in...

So then it starts to seems a little bit more like the blown through
design... but this would be
some kind of hybrid design.

Some blow through and some suckage

Hopefully dust won't be sucked in from those tiny little holes... or at
least a whole lot less then the other designs...
otherwise it would be pointless.

Bye,
Skybuck.

 

[Skybuck Flying]

Well, if this is the case, if it's the case of maximizing contact area then
here is an idea for a chip/gpu:

The gpu is cut up into many tiny little pieces.

The tiny little pieces are distributed over the entire graphics card.

Tiny little heatsinks which are larger then the gpu piece are stuck on top
of it.

This should maximize the area a bit more... better distribution of heat.

Since it's a parallel chip consisting out of multiple cores... it should be
possible to cut up those cores and distribute them
across the graphics card...

Added benefit is also more lanes towards all tiny little cores... for more
bandwidth and more memory lookup power.

These tiny little gpu pieces could by stuck between capcitators... or maybe
even on top of them... or vice versa...

Not sure if that's a good idea... or where to best place them... but some
spreading out seems nice.

If this would be any better than current situation remains to be seen...
current heatsinks also pretty massive
across the graphics board... so maybe it don't matter, or just very
little...

Or maybe it does matter... maybe having everything on a small little area
prevents optimal heat transfer...

Thus cutting the chip up into multiple pieces might make it better.

Maybe the entire chip design should be more like a building with windows in
it... and blow air directly through the chip... instead of an additional
heatsink.

Bye,
Skybuck.

 

[Skybuck Flying]

"John Larkin" wrote in message

wrote in message


"
Why not use compressed/expanded air for this purpose ? Using a piston
compressor to compress the air to a few bars, the air gets quite hot,
then let it go through a heat exchanger to get rid of most of the heat
and cool the pressurized air closer to ambient temperature.

Let the air expand to normal ambient pressure and the air temperature
is now well below ambient temperature and let it flow through
semiconductor heatsinks to the environment.

To avoid problems with dust and condensation, a closed loop might make
sense, but of course, now the heat exchanger would also have to
dissipate the heat from the semiconductor. However, the heat exchanger
can be remotely located and it can have much higher temperatures than
the semiconductors, getting rid of the heat into the environment would
be easier.
"

I like this idea of a closed air system very much...

"
It's been done, with better working fluids. You have one in your
kitchen.
"

Yeah in case such a special case does not exist, a next best thing might
simply be a mini/tiny refrigator and place the entire pc inside of it...

My fridge actually has small little holes on the back side... so some cables
could go through it...

But it's a scary idea... electronics and moist.... hmm I'll have to look
into this somemore...

For now biggest drawback could be noise of fridge.... or maybe fridge can't
handle the pc heat at all...

Hmm..

Bye,
Skybuck.

 

[Martin Brown]

It's not. Look at a heatsink theta-versus-air-flow curve. There's also
cooling at zero flow.

Convective and radiative cooling only. And not much of the latter if the
thing is made of shiny metal. Proportional to flow plus a small additive
constant isn't exactly rocket science. In any forced air cooling design
worth its salt airflow cooling totally dominates.

I don't think the experiment is easy. With the same fan, you'd have to
vary the airflow resistance, like the pin or fin density or something,
and keep the remaining geometry the same.

PWM to vary the power provided to the fan so the work done by the fan is
on moving the air through it and to first order the output velocity
field scales fairly well apart from deep inside the heatsink where there
is or should be some turbulence.

It occurs to me in this discussion that the performance of a standard
rectangular vaned heatsink might be improved by putting diagonal wires
through the vanes at say 45 degrees to generate vortex wakes that mix up
the laminar flow after the first or second set of vanes.

This would matter in a case like deciding between two heatsinks that
are the same overall dimensions but have different fin densities,
cooled by the same fan. Heatsink data sheets give you half the
information you need (theta vs flow) but not the other half
(backpressure vs flow).

Measuring rpm of the fan against power supplied will give you a decent
proxy for back pressure and many PC fans are so equipped.

I wonder if CPU chip layouts include hot-spot distribution, like
putting the hottest bits into the corners or something.

Perhaps but I would guess only by coincidence since the main output
drive buffers are probably physically close to the edge of the die.

Regards,
Martin Brown