David Maynard said:
Nope and you should have read it all before knee jerk replying.
"Nope" is not angument, Maynard.
I am not "assuming" anything. You are.
So what's your point?
Local turbulence is already taken care of by the part itself.
Just try to get 'laminar' flow over one. It's almost, if not, impossible.
Is this your "OFF/ON" theory of turbulence? Turbulence occurs
in degrees just as laminar flow does.
Is "simply untrue" the extent of your reasoning?
Is "just a fact" the extent of your reasoning?
It won't have laminar flow over the part no matter how hard you try to do it. The 'turbulence' don't need your 'help' on the other
side of the case.
If by "other side of the case" you mean "after contacting multiple
parts", the turbulence will have been reduced by drag against the
parts. If by "other side of the case" you mean "during the time of
transiting free space within the case", any turbulence generated
upstream will benefit the cooling of parts that it eventually impinges.
Is "Nope" the extent of your reasoning?
Which isn't "6 inches" away.
Transiting 6" probably takes the air one to two seconds.
So what makes you think turbulence will die out in that
period of time?
Is saying "useless waste of energy" the extent of your argument?
We're not talking about airfoils and lift.
Boundary layers and the effects of air flow apply universally,
especially in subsonic regimes.
Hey, pal. The extra fan ain't 'free energy'. Nor is it there for 'turbulence.
Who said turbulence was "free"? What an internal fan produces is
both turbulence and re-circulation. It does NOT increase bulk air flow.
Kornball has called re-circulated air "pre-heated" and "used" air -
which obviously isn't so bad if most PCs, especially of the gaming
variety - use internal fans.
You should ponder your own words, there, "inevitable turbulence" when looking at the physical components.
Read again. I wrote "Since turbulence can be produced by
SHARP-EDGED HOLES IN FLAT PLATES, the opportunity
arises to use the inevitable turbulence" That means that
the punched holes at the front of the case produce lots of
turbulence.
And right in that turbulence, bathing in it, is where the
hard drive is put by Dell instead of other equally
available positions. It uses this "cheap" turbulence
to cool the hard drive.
Try looking up the 'big boy' stuff and see what they say about intake filter turbulence. Hint, you won't find a single one
bragging at how much it creates but, rather, how little.
Big Boy Stuff, if it's available to YOU, is for the Little Boys.
"Stuff", if it's available to YOU, will tout whatever will
impress YOU about their engineering and design prowess,
and it will not reveal any clever techniques which will
sound non-intuitive to Little Boys. Furthermore, all one
has to do is to inspect the product of the Big Boys,
and Dell, being a "Big Boy", utilizes turbulence in its
designs rather than trying to minimize it.
You put the (hot) parts in front of it because it's cool intake air with lots of flow before it gets dispersed into the whole case
volume.
"Before it gets DISPERSED into the whole case volume".
Hmmm... is that the "Little Balls" theory of air dispersal?
The intake air gets "DISPERSED"? First, it's concentrated,
then it gets "DISPERSED"? I thought you were designing
for laminar flow across the case. How would intake air get
dispersed?
Hey, I pointed out many times that there is free and direct
flow of intake air across an open cavity at the bottom of
Dell's case, and the hard drive would get the same air
anywhere in that cavity, but that some of the turbulence
would have been diminished by contact with the cavity's
walls. So Dell puts the hard drive flat against and close to
the turbulenvce generated by the passage of air through the
intake holes.
First you claim that turbulence is a waste of energy, now you
say it's "cheap". Which is it?
Nope. Airflow is maximum at the intake.
And what goes in doesn't come out? <LOL>
Where does the air disappear to?
Is there an Air Destroyer in all your designs?
Too much drag from the ducting and tunnels. Try it and you'll find out.
Tell it to Apple Computers. I'm sure they're waiting to hear
from you.
More drag and wasted energy.
If you say so.
You aren't going to get 'laminar flow' so stop worrying about it. What you do is cut down on the wasted turbulence as much as
possible.
And how does one "cut down on the wasted turbulence"?
How does one even cut down on turbulence if not by drag
and viscosity - which would be a "useless waste of energy"?
You and kornball have these ideas about proper air flow
but no means to achieve what you recommend.
That's a nice "if" but nothing you've talked about will improve 'heated parts' turbulence one bit.
Put the heated parts near turbulent areas -
such as right behind vent holes instead of 2 or 3 or 4 or 6 inches
downstream. Make hot surfaces and surfaces just upstream of
them rough or undulating. (The dimples in a golf ball actually
*reduce* the drag on the golf ball by inducing turbulence). Design
heatsinks with labyrinth air passages rather than straight smooth
air paths. Install internal fans. Put small vanes in areas where
air must pass to induce a swirl. Do anything to put the air in
motion instead of calmly transiting the case.
You're still thinking *way* too macro. The heat of a part will, itself, create boundary layer turbulence. You can't avoid it
unless using super high velocities that scrub past it. For 'problem' parts the designer might include surface 'bumps', or some
other mechanism, but the point is there's nothing you have to do about it 'in the case' except get a reasonable amount of air in
and out.
The heat of a piston crown is very hot. Yet it is protected from
the heat of combustion flame by its boundary layer of unburned
gases.
Surface bumps are good for turbulence, even surface pits.
Little vanes are good, too. The problem with little vanes is
that the interior of a PC case varies considerably from one
case to another due to hardware options and cabling that
only a homebuilder can plan or implement detailed air flow.
But you guys are the homebuilders, aren't you?
As for easy ways to implement turbulence, internal fans are
easy to implement.
Look at the CFM specifications for any forced air heatsink. You
will not see a 'turbulence' factor. It is strictly CFM.
http://www.thermaflo.com/crosscut.shtml
"Turbulent air breaks the stagnant air boundary layers
around the pins and, as a result, enhances the heat sink's
thermal performance."
http://www.frostytech.com/articleview.cfm?articleID=2001
"To induce turbulence within the fins and improve thermal
transmission between the air and metal, Thermalright have
modified the aluminum fins by adding "proprietary bent winglets."
In other words, the leading and trailing edges along either side
of the aluminum fins are bent 15° up and 15° down respectively."
Nope. It's more velocity, nothing else.
Velocity without any change in bulk air flow. That's what turbulence
accomplishes, and that's what fans accomplish. You get to choose,
as a designer, which device to use.
But it isn't 'inconsequential'.
What are the consequences, then?
They do despite the problem of it being preheated.
So you agree that re-circulation by fans is useful despite
the fact that no new air is being brought in to increase the
bulk air flow rate.
A fan INCREASES the rate of air flow - more cubic inches of
air per minute - without an increase in bulk air flow rate. That
implies that some of the air passes through the fan more than
once. Turbulence does the same thing to put more air molecules
in contact with a heated surface than laminar flow would. Dismiss
the commonality as "semantics" if you want, but the desirable
effect on cooling is the same. All that is different is that one takes
an active input of energy (the fan), and the other uses some of the
energy of the air flow.
Entropy is the "randomness" of a system - which tends
to increase with time. Why does that make a difference
to where turbulence was generated? All that is important
is that the turbulence impinge the object to be cooled.
Your '6 inch away' turbulence might as well be on the moon.
AHHH! The heart of the matter - kornball's Molasses
Theory of Air Flow. You believe turbulence to be short-lived.
You believe turbulence to be a micro effect, effective only
over microscopic distances. Have you ever stirred your
coffee? Blown a smoke ring? Driven your car through
the rain? Blown out a candle? Does the turbulence disappear
immediately? You defy everyday experience!!
For a 1 inch long surface, like an IC, that 1 inch diameter 'turbulence' is dern near laminar and at the micro scale, where it
matters, it's downright gargantuan.
And that "gargantuan" swirl or eddy increases the velocity
of the air passing along the surface of the IC - thinning the
boundary layer. I can see that you belive that turbulence
is "jiggling of the molecules", whereas turbulence can actually
be planetary and galaxial in size. There is considerable study
into the boundary layer and turbulence of the plasma flow
past our planet from the sun. That is not "micro". And it's
effects are not "micro".
And gone almost as soon as they pass through.
Any pilot will tell you that vortices strong enough to upset
an aircraft can last on the order of minutes. Just blow a
smoke ring to see that. You defy everyday experience to
bolster your argument - such as it is.
A smoke ring is not 'turbulence'. It's the shape left after it's creation.
Any vortex, by definition, is not part of bulk air flow. It is
a description of translational energy BESIDE bulk air flow.
Call is a "shape", or call it increased kinetic enegy, the
effect is to bring more molecules of fluid per unit time into
contact with the surface its passing over.
Is "Nope" the extent of your reasoning?
What makes you think they last, other than 'smoke rings'? Which, if it really was so turbulent as you think would vanish
immediately.. from the turbulence disrupting the shape.
Ask any pilot or air controller about the persistance of turbulence
and then get back to us.
If you want to muster the power to blow a hurricane through the
thing then you could afford all that turbulence you like so much.
A degree of turbulence is all around us - no hurricane is necessary.
A degree of laminar flow is all around us as well. In cooling, though,
turbulence aids bulk air flow by scrubbing down through the
boundary layer that is around all objects to bring the passing fluid
into closer contact with the underlying object.
Is "Bad assumption" the extent of your argument?
I've told you through this whole thing.
All you've been doing is making terse denials.
You haven't explained your reasoning.
It was to take advantage of the cold air coming in where it's of
maximum velocity.
And the velocity reduces further into the case? Isn't it
"Air Out = Air In"? Where does the air go in the cases
that you design?
Yes, right at the surface. Everything else is a waste of energy.
That says *nothing* about WHERE the turbulence gets
generated. It could be generated anywhere upstream
of the part to be cooled. What is important is that it
IMPINGE the part - that it GETS TO the part.
No, your point has been that any and all turbulence,
regardless of where, is a 'good thing' and that's simply not correct.
No, the turbulence must IMPINGE the part to be cooled
to have an effect on the cooling of that part. I have been
saying that consistently throughout this thread.
Your links are all surface micro turbulence examples.
Back up that claim with details from the links, please.
Tell us what you consider to be a "micro" example
and what is a "macro" situation, and then tell us why
they don't interact.
Yes, and that isn't 6 inches away nor at the intake vents.
That's to reduce contact with the walls of the ducting
and with parts not needing cooling. Otherwise, the
turbulence could be generated anywhere.
Which is the thermal surface.
Wrong.
Is "Wrong" the extent of your argument?
Then why don't you just use all that natural 'turbulence'
mother nature put in the atmosphere, since it matters not 'where'?
Because the ducting would be too large.
No, the point is that the turbulence of use is micro turbulence at the surface boundary layer and that your macro 'turbulence',
because it's 'turbulent' only when compared to the case volume, or the room, or the planet, looks just like your dreaded 'laminar
flow' at the micro surface level.
Any velocity of a passing fluid helps to thin the boundary layer.
That velocity may be that of the bulk fluid flow, or it may be
the bulk fluid flow PLUS that of any turbulence. The action of
the two is better for cooling.
Not enough un preheated air flow/
What makes you think the air they got was "pre-heated"?
I think it's a good idea. It's just that your understanding of why is incorrect.
What is "incorrect" about my understanding?
"Smooth air flow" doesn't have turbulence, and you've
admitted turbulence was good for cooling.
The problem is your interpretation of it.
Interpretation of what? Could you be more specific?
What is to be "interpreted", and how have I done
that incorrectly?
*TimDaniels*
