Fans in Series and Fans in Parallel

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

The following web page, posted by David Maynard, helps
to clear up whether two fans in series and parallel flow can
flow more air than a single fan of the same size:
http://www.comairrotron.com/engineering_notes_02.asp

The answer happens to be what one would expect
intuitively - "Yes, they can, but not twice as much."

From the paragraph entitled "Series and Parallel Operation":

"Combining fans in series of [sic] parallel can achieve
the desired airflow without greatly increasing the
system package size or fan diameter. Parallel operation
is defined as having two or more fans blowing together
side by side. The performance of two fans in parallel
will result in doubling the volume flow, but only at
free delivery. As figure 4 shows, when a system curve
is overlaid on the parallel performance curves, the
higher the system resistance, the less increase in flow
results with parallel fan operation. Thus, this type of
application should only be used when the fans can operate
in a low impedance near free delivery."

"Series operation can be defined as using multiple fans
in a push-pull arrangement. By staging two fans in series,
the static pressure capability at a give airflow can be
increased, but again, not to double at every flow point,
as Figure 5 displays. In series operation, the best results
are achieved in systems with high impedance."

"In both series and parallel operation, particularly with
multiple fans (5, 6, 7, etc.) certain areas of the combined
performance curve will be unstable and should be avoided.
This instability is unpredictable and is a function of the
fan and motor construction and the operating point. For
multiple fan installations, Comair Rotron strongly recommends
laboratory testing of the system."

It points out that for high impedance (i.e. air resistance) cases,
the impedance rises at approximately the square of the flow rate
increase, in line with turbulent flow principles of aerodynamics.
This would lead one to seek ways to reduce the impedance before
resorting to a larger fan or fans in parallel. And, of course, one
way to reduce air impedance is to use "round" IDE cables instead
of flat cables in situations using many ATA/ATAPI drives, and to
remove perforated fan grills. Interestingly, in cases having high
impedance, the article says that two fans in series are more effective
than two fans in parallel.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

<been there done that> wrote:
> [......]
> In a high impedance case, the series "pushing in fan" helps lower
> impedance (functionally). It would be more efficient to punch a
> bigger hole in the front though
>
> [.......]
>
> Yep, the key sentence in that webpage is:
>
> "Thus, before increasing the number of fans in a systems, or
> attempting to increase the air volume using a larger fan, the system
> should be analyzed for possible reduction in the overall resistance to
> airflow. "
>
> [........]
> A typical tower case is pretty low impedance inside except when the
> manufacturer pokes those tiny holes for fan grills.


Yup. I removed the wire grill from the case fan and I snipped away
the perforated grill over the power supply fan. And then, because there
is an intake along the bottom of the front bezel, I put the case on 3/4"
of plywood with the front overhanging the edge of the plywood so the
incoming air wouldn't have to sweep across the desk as it entered the
intake. Now the greater air flow makes more noise as it rushes out the
back of the case. Darn!

*TimDaniels*

 

[Spajky]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Thu, 3 Jun 2004 17:50:34 -0700, "Timothy Daniels"
<TDaniels@NoSpamDot.com> wrote:

> Yup. I removed the wire grill from the case fan and I snipped away
>the perforated grill over the power supply fan. And then, because there
>is an intake along the bottom of the front bezel, I put the case on 3/4"
>of plywood with the front overhanging the edge of the plywood so the
>incoming air wouldn't have to sweep across the desk as it entered the
>intake. Now the greater air flow makes more noise as it rushes out the
>back of the case. Darn!

have you tried some other mods like I did or closing the gap between
the fan support & case hole; see my site under comp/mods ...
--
Regards, SPAJKY ®
& visit my site @ http://www.spajky.vze.com
"Tualatin OC-ed / BX-Slot1 / inaudible setup!"
E-mail AntiSpam: remove ##

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"Spajky" wrote:
> have you tried some other mods like I did or closing the gap between
> the fan support & case hole; see my site under comp/mods ...


Nope. There is no gap. I think that some modeling clay
to make the fan venturi continuous with the case and maybe
some kind of horn to let the air decelerate gradually would
help with the exit turbulence, but the rushing sound now can
be rendered inaudible by simply putting a pillow against the
wall behind the case to absorb the sound.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

> A typical tower case is pretty low impedance inside except
> when the manufacturer pokes those tiny holes for fan grills.

o Fan grills comprise ~84% of the resistance in a case
---- Intel published a few test reports somewhere
o Using round wire grills can drop that figure considerably
---- so improving /realised/ airflow in that enclosure
---- and also reducing airflow noise experienced

This applies to the intake grills & the intake cutout in the bottom
front of the plastic fascia - invariably 1/3rd the area of an 80mm fan.

Whilst the Apple Dual G5 used a low free-air %age perforated
aluminium fascia, it does use the entire fascia area as the air inlet.
The exhaust is differently perforated - 68-86% free-air honeycomb.

Dual G5 looks nice anodised black - doesn't help airflow
--
Dorothy Bradbury
www.stores.ebay.co.uk/panaflofan for fans, books & other items
http://homepage.ntlworld.com/dorothy.bradbury/panaflo.htm (Direct)

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"Dorothy Bradbury" wrote:
> > A typical tower case is pretty low impedance inside except
> > when the manufacturer pokes those tiny holes for fan grills.
>
> o Fan grills comprise ~84% of the resistance in a case
> ---- Intel published a few test reports somewhere
> o Using round wire grills can drop that figure considerably
> ---- so improving /realised/ airflow in that enclosure
> ---- and also reducing airflow noise experienced
>
> This applies to the intake grills & the intake cutout in the bottom
> front of the plastic fascia - invariably 1/3rd the area of an 80mm fan.
>
> Whilst the Apple Dual G5 used a low free-air %age perforated
> aluminium fascia, it does use the entire fascia area as the air inlet.
> The exhaust is differently perforated - 68-86% free-air honeycomb.


I imagine intake is a tricky thing and that it's done primarily by
cut-'n-try. For air-cooled engines (i.e. motorcycles and light
aircraft) racers find that turbulence helps a lot in scrubbing
through the boundary layer on the cooling fins to get at the heat.
They find that air velocity alone is not enough. It needs turbulence.
And all those holes of various sizes that I see on the front of my
Dell's case not only let in air, they also make turbulanece of various
vortex sizes. I mean, why didn't they make a short venturi or
"lips" around one big hole? Why did they put a load of seemingly
randomly spaced and randomly sized holes in sheet metal? I
strongly suspect it was for the turbulence. And the trade-off
between low intake resistance and turbulence is probably quite
an art.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Timothy Daniels wrote:
> "Dorothy Bradbury" wrote:
>
>>>A typical tower case is pretty low impedance inside except
>>>when the manufacturer pokes those tiny holes for fan grills.
>>
>>o Fan grills comprise ~84% of the resistance in a case
>>---- Intel published a few test reports somewhere
>>o Using round wire grills can drop that figure considerably
>>---- so improving /realised/ airflow in that enclosure
>>---- and also reducing airflow noise experienced
>>
>>This applies to the intake grills & the intake cutout in the bottom
>>front of the plastic fascia - invariably 1/3rd the area of an 80mm fan.
>>
>>Whilst the Apple Dual G5 used a low free-air %age perforated
>>aluminium fascia, it does use the entire fascia area as the air inlet.
>>The exhaust is differently perforated - 68-86% free-air honeycomb.
>
>
>
> I imagine intake is a tricky thing and that it's done primarily by
> cut-'n-try. For air-cooled engines (i.e. motorcycles and light
> aircraft) racers find that turbulence helps a lot in scrubbing
> through the boundary layer on the cooling fins to get at the heat.
> They find that air velocity alone is not enough. It needs turbulence.
> And all those holes of various sizes that I see on the front of my
> Dell's case not only let in air, they also make turbulanece of various
> vortex sizes. I mean, why didn't they make a short venturi or
> "lips" around one big hole? Why did they put a load of seemingly
> randomly spaced and randomly sized holes in sheet metal? I
> strongly suspect it was for the turbulence. And the trade-off
> between low intake resistance and turbulence is probably quite
> an art.
>
> *TimDaniels*

The two are not related. For one, the relative air velocity over the
air-cooled motor fins is a heck of a lot higher, which is why they're
worried about boundary layers. The air slip streams across the boundary
layer with the boundary layer acting as an insulator.

The case inlet isn't what's being cooled, so turbulence there is of no use
at all, and the airflow inside the case is so distributed as to be
minuscule by comparison to the motor fins on a hog going 60 MPH down the
freeway.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"David Maynard" sputtered:
> Timothy Daniels wrote:
> > I imagine intake is a tricky thing and that it's done primarily by
> > cut-'n-try. For air-cooled engines (i.e. motorcycles and light
> > aircraft) racers find that turbulence helps a lot in scrubbing
> > through the boundary layer on the cooling fins to get at the heat.
> > They find that air velocity alone is not enough. It needs turbulence.
> > And all those holes of various sizes that I see on the front of my
> > Dell's case not only let in air, they also make turbulanece of various
> > vortex sizes. I mean, why didn't they make a short venturi or
> > "lips" around one big hole? Why did they put a load of seemingly
> > randomly spaced and randomly sized holes in sheet metal? I
> > strongly suspect it was for the turbulence. And the trade-off
> > between low intake resistance and turbulence is probably quite
> > an art.
> >
> > *TimDaniels*
>
> The two are not related. For one, the relative air velocity over the
> air-cooled motor fins is a heck of a lot higher, which is why they're
> worried about boundary layers. The air slip streams across the
> boundary layer with the boundary layer acting as an insulator.


The lower the velocity, the thicker the layer of stagnant air -
the boundary layer - that exists at the surface of any object
immersed in air, and thus turbulence is even *more*
important for cooling at low flow velocities, not less
important.


> The case inlet isn't what's being cooled,


Of course not. You haven't considered carefully enough.
The turbulence produced by air entering at a sharp hole
in a sheet of metal is what produces the turbulence which
in turn acts to cool components such as HDD casings and
motherboard/PCI board components. A much smoother
flow, one with a minimum of air touching the edges of the
inlet, would be achieved with one big hole with a smooth
bell-shaped intake like those found on racing carburettors.
But Dell didn't go that route. I wonder why.


> so turbulence there is of no use at all, and the airflow inside
> the case is so distributed as to be minuscule by comparison
> to the motor fins on a hog going 60 MPH down the freeway.


Racing motorcycles and light aircraft speeds are similar -
around 160 mph - so a "hog going 60 MPH down the freeway"
is quite irrelevant. What *is* relevant is the importance of
turbulence in buffeting away whatever boundary layer there is
that is attached to any component that produces heat. At the
low air velocities of a PC case, turbulence greatly assists in
punching down through the relatively thick boundary layer,
and the greater flow resistance that results from the production
of that turbulence is considered by cooling designers to be a
reasonable price to pay.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Mon, 7 Jun 2004 23:31:53 -0700, "Timothy Daniels"
<TDaniels@NoSpamDot.com> wrote:


>> The two are not related. For one, the relative air velocity over the
>> air-cooled motor fins is a heck of a lot higher, which is why they're
>> worried about boundary layers. The air slip streams across the
>> boundary layer with the boundary layer acting as an insulator.
>
>
> The lower the velocity, the thicker the layer of stagnant air -
> the boundary layer - that exists at the surface of any object
> immersed in air, and thus turbulence is even *more*
> important for cooling at low flow velocities, not less
> important.
>

Turbulent air intake on a system case has only a negative effect, that of
reducing intake flow and causing noise. IF the flow rate were same
turbulent and (less so) then the turbulence would help, but in this case
the turbulence coincides to a reduction in flow rate that is more
significant.




>
>> The case inlet isn't what's being cooled,
>
>
> Of course not. You haven't considered carefully enough.
> The turbulence produced by air entering at a sharp hole
> in a sheet of metal is what produces the turbulence which
> in turn acts to cool components such as HDD casings and
> motherboard/PCI board components. A much smoother
> flow, one with a minimum of air touching the edges of the
> inlet, would be achieved with one big hole with a smooth
> bell-shaped intake like those found on racing carburettors.
> But Dell didn't go that route. I wonder why.

Hardly. It's a sharp hole because it's stamped out of the metal wall,
simply the quick and cheap way to do it.

>
>
>> so turbulence there is of no use at all, and the airflow inside
>> the case is so distributed as to be minuscule by comparison
>> to the motor fins on a hog going 60 MPH down the freeway.
>
>
> Racing motorcycles and light aircraft speeds are similar -
> around 160 mph - so a "hog going 60 MPH down the freeway"
> is quite irrelevant. What *is* relevant is the importance of
> turbulence in buffeting away whatever boundary layer there is
> that is attached to any component that produces heat. At the
> low air velocities of a PC case, turbulence greatly assists in
> punching down through the relatively thick boundary layer,
> and the greater flow resistance that results from the production
> of that turbulence is considered by cooling designers to be a
> reasonable price to pay.

no.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"kony" opined:
>
> Turbulent air intake on a system case has only a negative effect,
> that of reducing intake flow and causing noise.


And that's probably why Dell and other major brand PCs
put a plastic fascia over the intake - to block the noise of the
intake turbulence.


> IF the flow rate were same turbulent and (less so) then the
> turbulence would help, but in this case the turbulence coincides
> to a reduction in flow rate that is more significant.
> [...........]
> It's a sharp hole because it's stamped out of the metal wall,
> simply the quick and cheap way to do it.


If low intake resistance to airflow were the prime consideration,
why didn't the manufacturers make just one big hole - which
would minimize turbulence? Why didn't they make one big
hole with the fascia providing curved "lips" to reduce intake
turbulence even more? They seem to gone out of their way to
make the air turbulent while still covering the holes with a fascia
that reflects the noise of intake back into the case.

Remember that there are a whole bunch of components that
need cooling beside the heatsinks. There are multiple boards
with chips and passive components that need cooling, there are
hard drives as well that need cooling, and these components
don't have thin little fingers that stick up in the airflow much
less have a dedicated fan. What they need is turbulence - to
cut down through the boundary layer - and airflow to carry the
heated air away. And Dell and the other major manufacturers
who can afford to have whole-system designers have designed
case intakes that gurantee turbulence and fascias that reduce
the noise that turbulence causes. Otherwise, they would have
put just a big circular hole in front with curved "lips" at the edges
to maximize airflow and minimize turbulence.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Depends on when the Dell was designed:
o Early Dells like HP, Compaqs or "Branded Iron" were designed by testing
---- physical prototype, measurements made, design adjustments
o Later machines are only prototyped at the late stages of design
---- quick 2D finite network model of the rough case/component layout
-------- get a quick appreciation of any hot-spots & dead airflow space area
---- post-it brainstormed designs quickly evaluated
-------- later design features on critical path identified
---- actual design done in CAD, pushed to flotherm 3D for fea analysis
-------- design proven based on component characteristic curves etc
---- final prototype basically to confirm model
-------- not to confirm design re changes late in design stage re sunk cost

The branded PCs are getting better - or rather they are designing closer to
the design envelope more and so maximising their margin as they can specify
everything that goes inside it whereas a generic assembler is more restricted.

Re engine cooling, a surprising factor depends on the cooling:
o Air v Water cooled
o Pebbled (*) or smooth fins involved in motorcycles

Most car engines have a pebbled finish on components to boost surface area
and disrupt the boundary layer, which is effective enough on a vehicle. On a
heatsink you can sometimes see ribbed heatsinks or cross-cut vertical fins.

(*) Pebbled - can't recall the exact name for the process used.

Cooling of components is based on m/sec, ie, cfm thro your cooling duct.
Generally the problem with PC heatsinks is simply one of cost balance.

One thing that would make a noticeable benefit is the machining out of the
heatsink fins under a fan's hub - either completely or in a stepped pattern.
Instead they tend to use a full copper base if needed & fins press-fit etc,
with the larger effect that of removing heated air from the PC case. That is
one aim of BTX with a blow-thro design to stop CPU cooler recirculating
air - altho it is a weak spec re dual-CPU and graphics card cooling needs.

Where boundary layer becomes an issue on heatsinks is where you have very
high fin densities - such as skived-fin heatsinks used in 1U applications. This
is best resolved with very high air velocities - eg, twin 40x28mm Sanyo fans
(15cfm) and eventually 50x40x38mm 56dB(A) twin-motor/rotor fans. This
problem can arise in desktop heatsinks where the design is an interrupted
slotted-fin - slotted fins which alternately obstruct one another along the
heatsink airflow and so present a substantial static resistance & boundary
layer if the actual cfm is insufficient. They're not particularly popular due to
airflow noise/fan noise.

Case perforation is quite expensive - and somewhat limited:
o Standard generic case perforation is actually just punching
---- by that I mean the standard punching machines
o Perforation like a Dual G5 or Dell is a separate extra process step
---- an actual precision perforation machine is used
---- much more expensive for volume use than the general punchers

Generic cases tend to use front-bottom-fascia inlets, branded tend to
use a slightly better design. Branded also tend to use super high flow &
noise fans which can be very cheap - yet allow use in Saudi 45oC+ etc.
Thus they can get away with the lowest number of fans = more margin,
but sometimes it goes (very) wrong in terms of end user noise perception.

Classic example being the G4 cube which had a superb huge heatsink with
heatpipes, blow-thro design pretty much... and then a 120mm 150cfm fan
that was 57dB(A)+. Similar on the 60mm PSU fans too re intake side noise
from putting a 60mm fan (nasty choice blade:hub size) in the short end of a
long box, which does wonders for acoustic transmission of fan noise. Another
is Dell, with some 8xxx series having JMC/JAC ultra-high noise level fans,
and even an NMB B66 or B86 which are custom-only super-high noise as
well as drawing over 10-18 watts in some cases, quite OTT really.

Temperature controlled fans rarely work, even when well matched to
user-perception v temperature-environment there is always lag re load.
New current controlled fans for CPU/PSU are the eventual solution.
--
Dorothy Bradbury
www.stores.ebay.co.uk/panaflofan for fans, books & other items
http://homepage.ntlworld.com/dorothy.bradbury/panaflo.htm (Direct)

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"Dorothy Bradbury" commented:
> Re engine cooling, a surprising factor depends on the cooling:
> o Air v Water cooled
> o Pebbled (*) or smooth fins involved in motorcycles


Most motorcycle engines until relatively recently were air
cooled, as are virtually all current light aircraft. Some
motorcycle racers bead-blasted the cylinder fins and the
overhead cam housings to maximize surface roughness.
One of the Japanese factory racing teams experimented
with smooth versus rough fins in the 70s, and they found
that the rough fins cooled better. The writer reporting this
said that the reason was due to increased turbulence, not
due to increased surface area.

Notice that for racing motorcycles, nice smooth air intakes
for the carbs or injectors is sought, but that a heat exchanger
(i.e. "radiator") in the turbulent wake of the front wheel and
forks is just fine - no placement of the heat exchanger in
smooth air is attempted.

For liquid cooling (not necessarily for automotive use),
heat exchangers can have ridges inside the coils which
put a swirl in the liquid so that increased turbulence puts
more of the liquid in direct contact with the walls of the
coil for better heat exchange, although a smooth-walled
coil would have less flow resistance.


> Temperature controlled fans rarely work, even when well
> matched to user-perception v temperature-environment there
> is always lag re load. New current controlled fans for CPU/PSU
> are the eventual solution.


Interesting. Are there any motherboards or PCs sold now
that have the current controlled fans?

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Tue, 8 Jun 2004 11:39:32 -0700, "Timothy Daniels"
<TDaniels@NoSpamDot.com> wrote:

>"kony" opined:
>>
>> Turbulent air intake on a system case has only a negative effect,
>> that of reducing intake flow and causing noise.
>
>
> And that's probably why Dell and other major brand PCs
> put a plastic fascia over the intake - to block the noise of the
> intake turbulence.

Possibly, but it seems as likley that it's to create a finished look to
the case.


>> IF the flow rate were same turbulent and (less so) then the
>> turbulence would help, but in this case the turbulence coincides
>> to a reduction in flow rate that is more significant.
>> [...........]
>> It's a sharp hole because it's stamped out of the metal wall,
>> simply the quick and cheap way to do it.
>
>
> If low intake resistance to airflow were the prime consideration,
> why didn't the manufacturers make just one big hole - which
> would minimize turbulence?

If creating turbulence was the prime consideration, why stamp round holes
instead of some shape that maximizes turbulence, something more like a
star than a circle or square?


> Why didn't they make one big
> hole with the fascia providing curved "lips" to reduce intake
> turbulence even more? They seem to gone out of their way to
> make the air turbulent while still covering the holes with a fascia
> that reflects the noise of intake back into the case.

That's just it, they did NOT go out of their way, that is the most
significant factore, that they used the easier all-around method of
creating air intake passages. Keep in mind that this is not such a
signficant turbulence to begin with, the air is at relatively low velocity
and the turbulence is hardly relevant by the time the air finds it's way
to most interior components... it is trivial compared to the turbulence
from a fan... If they REALLY wanted turbulence so much (for example,
Dell) they'd certainly have put a fan on all CPU heatsinks instead of the
rear fan w/duct.


>
> Remember that there are a whole bunch of components that
> need cooling beside the heatsinks. There are multiple boards
> with chips and passive components that need cooling, there are
> hard drives as well that need cooling, and these components
> don't have thin little fingers that stick up in the airflow much
> less have a dedicated fan. What they need is turbulence - to
> cut down through the boundary layer - and airflow to carry the
> heated air away. And Dell and the other major manufacturers
> who can afford to have whole-system designers have designed
> case intakes that gurantee turbulence and fascias that reduce
> the noise that turbulence causes. Otherwise, they would have
> put just a big circular hole in front with curved "lips" at the edges
> to maximize airflow and minimize turbulence.
>
>*TimDaniels*

Again, no.
You are putting more thought into your reply than they did in creating the
air intake. The turbulence from the tiny holes in the front wall of a
case are trivial to cooling anything not IMMEDIATELY behind those holes...
and if they'd REALLY wanted turbulence they'd have gone to more elaborate
measures to create it, not the opposite you're suggesting to take extra
measures to reduce it. Quite simply, they did it the inexpensive and easy
way because it doesn't make any significant difference. Turbulence
created on a heatsink surface is in the immediate vicinity of the
(heatsink, heat), not many inches to over a foot away. Exhaust fan
stamped grills are also quite often of similar design, does that mean they
meant to create turbulence in air exiting the case as well?

Threre is a point of diminishing return where some theory of what might
work *best* to a limited extent, when countered with the expected result,
just isn't worthwhile to implement. There are plenty of ways a minor temp
reduction might be seen with a more elaborate chassis, but none of them
signficant enough to implement.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"kony" wrote:
> "Timothy Daniels" wrote:
>
> > And that's probably why Dell and other major brand PCs
> > put a plastic fascia over the intake - to block the noise of the
> > intake turbulence.
>
> Possibly, but it seems as likley that it's to create a finished look to
> the case.


Obviously, a fascia looks better than a flat case with holes
punched in it. But why not let the air enter straight in instead
of making a right angle if minimizing air restriction were the
paramount design consideration? By covering the punched
intake holes with a sound-reflective surface (i.e. the fascia),
much of the noise is reflected back toward the intake holes
instead heading straight for our ears. That this requires the
air to make a right-angle turn, and to suffer greater air flow
resistance thereby, appears to be tolerable to the designers.



> > If low intake resistance to airflow were the prime consideration,
> > why didn't the manufacturers make just one big hole - which
> > would minimize turbulence?
>
> If creating turbulence was the prime consideration, why stamp
> round holes instead of some shape that maximizes turbulence,
> something more like a star than a circle or square?


Because square punches wear out at the corners faster than round
punches? Also, the corner edge acts as a constriction, causing a
higher resistance/turbulance ratio, i.e. making it more costly in an air
resistance sense than a curved edge. Remember that a smooth
venturi would also restrict air and add resistance without adding
turbulence, so just adding resistance doesn't imply adding turbulence.
The designer's goal is to add turbulence at the lowest cost in the
form of resistance, i.e. circular holes are more efficient in generating
turbulence while minimizing resistance.


> > Why didn't they make one big
> > hole with the fascia providing curved "lips" to reduce intake
> > turbulence even more? They seem to gone out of their way to
> > make the air turbulent while still covering the holes with a fascia
> > that reflects the noise of intake back into the case.
>
> That's just it, they did NOT go out of their way, that is the most
> significant factore, that they used the easier all-around method of
> creating air intake passages.


Making dozens of various sized holes variously placed
at the front of the case is easier than making one big hole?
Tell it to the accountants.


> Keep in mind that this is not such a signficant turbulence
> to begin with, the air is at relatively low velocity


That is exactly why turbulence is valuable - the air
is at low velocity and the boundary layer is allowed
to remain a thick blanket on the warm components.


> and the turbulence is hardly relevant by the time the
> air finds it's way to most interior components...


How many seconds do you figure it takes for the
air to "find" its way to most interior components?
Two seconds, three seconds, five seconds?
How long do you think the turbulence persists?
Clue: Think of a smoke ring.


> it is trivial compared to the turbulence from a fan...


Really!? How much shear is there to air
passing past a fan blade as to air passing past
the sharp edges of multiple holes in sheet metal?


> If they REALLY wanted turbulence so much (for example,
> Dell) they'd certainly have put a fan on all CPU heatsinks
> instead of the rear fan w/duct.


Dell cleverly provided the turbulence by the holes
punched in the front of its metal cases, and for my
Pentium II based PC, there is no fan for the CPU
heatsink. There is only a single fan for the entire
case - at the exit. For those CPUs needing heatsink
fans, increased air velocity from the dedicated fan
provides the turbulence at the surfaces of the heatsink.
Notice that the shape of the heatsink "fingers" are not
aerodynamicly shaped which would minimize air flow
resistance. They have right-angle corners to increase
turbulence.


> You are putting more thought into your reply than they
> did in creating the air intake.


I'm sure you've just insulted hundreds of professional
case designers. The more I study the design of my
Dell PC's case and fascia, the more I admire the
skills of the designer(s) to maximize cooling with
the minimum expenditure for fans and the minimum
release of noise. That they were able to cool all
the components - which includes multiple optical
drives and two magnetic hard drives and everything
else with one 92mm exit fan in the back without
making the PC sound like an air conditioner or a
vacuum cleaner is a testimony to their skills.


> The turbulence from the tiny holes in the front wall of a
> case are trivial to cooling anything not IMMEDIATELY
> behind those holes...


Again, visualize a smoke ring. Swirling air (turbulence)
can act for a long time and travel a great distance.
Try sweeping your hand through a column of rising smoke
and time how long it takes for things to settle down to
a single narrow column again. If you're familiar with
aviation, think of how many minutes, how many miles
of separation are needed between aircraft to avoid
being thrown out of control by the turbulence generated
by the aircraft in front. To think that turbulence
disappears in the time it takes air to travel through a
PC case is to think of air as molasses.


> and if they'd REALLY wanted turbulence they'd have gone
> to more elaborate measures to create it, not the opposite


They get turbulence with simple circular holes. Why spend
more money? A penny saved is a penny more of profit.
Spend more time thinking like a bean counter.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"kony" <spam@spam.com> wrote in message
news:b19cc01quhpn1aecifjd2bcarle7jcrohr@4ax.com...
> On Tue, 8 Jun 2004 11:39:32 -0700, "Timothy Daniels"
> <TDaniels@NoSpamDot.com> wrote:
>
>>"kony" opined:
>>>
>>> Turbulent air intake on a system case has only a negative effect,
>>> that of reducing intake flow and causing noise.
>>
>>
>> And that's probably why Dell and other major brand PCs
>> put a plastic fascia over the intake - to block the noise of the
>> intake turbulence.
>
> Possibly, but it seems as likley that it's to create a finished look to
> the case.
>
>
>>> IF the flow rate were same turbulent and (less so) then the
>>> turbulence would help, but in this case the turbulence coincides
>>> to a reduction in flow rate that is more significant.
>>> [...........]
>>> It's a sharp hole because it's stamped out of the metal wall,
>>> simply the quick and cheap way to do it.
>>
>>
>> If low intake resistance to airflow were the prime consideration,
>> why didn't the manufacturers make just one big hole - which
>> would minimize turbulence?
>
> If creating turbulence was the prime consideration, why stamp round holes
> instead of some shape that maximizes turbulence, something more like a
> star than a circle or square?
>
>
>> Why didn't they make one big
>> hole with the fascia providing curved "lips" to reduce intake
>> turbulence even more? They seem to gone out of their way to
>> make the air turbulent while still covering the holes with a fascia
>> that reflects the noise of intake back into the case.
>
> That's just it, they did NOT go out of their way, that is the most
> significant factore, that they used the easier all-around method of
> creating air intake passages. Keep in mind that this is not such a
> signficant turbulence to begin with, the air is at relatively low velocity
> and the turbulence is hardly relevant by the time the air finds it's way
> to most interior components... it is trivial compared to the turbulence
> from a fan... If they REALLY wanted turbulence so much (for example,
> Dell) they'd certainly have put a fan on all CPU heatsinks instead of the
> rear fan w/duct.
>
>
>>
>> Remember that there are a whole bunch of components that
>> need cooling beside the heatsinks. There are multiple boards
>> with chips and passive components that need cooling, there are
>> hard drives as well that need cooling, and these components
>> don't have thin little fingers that stick up in the airflow much
>> less have a dedicated fan. What they need is turbulence - to
>> cut down through the boundary layer - and airflow to carry the
>> heated air away. And Dell and the other major manufacturers
>> who can afford to have whole-system designers have designed
>> case intakes that gurantee turbulence and fascias that reduce
>> the noise that turbulence causes. Otherwise, they would have
>> put just a big circular hole in front with curved "lips" at the edges
>> to maximize airflow and minimize turbulence.
>>
>>*TimDaniels*
>
> Again, no.
> You are putting more thought into your reply than they did in creating the
> air intake. The turbulence from the tiny holes in the front wall of a
> case are trivial to cooling anything not IMMEDIATELY behind those holes...
> and if they'd REALLY wanted turbulence they'd have gone to more elaborate
> measures to create it, not the opposite you're suggesting to take extra
> measures to reduce it. Quite simply, they did it the inexpensive and easy
> way because it doesn't make any significant difference. Turbulence
> created on a heatsink surface is in the immediate vicinity of the
> (heatsink, heat), not many inches to over a foot away. Exhaust fan
> stamped grills are also quite often of similar design, does that mean they
> meant to create turbulence in air exiting the case as well?
>
> Threre is a point of diminishing return where some theory of what might
> work *best* to a limited extent, when countered with the expected result,
> just isn't worthwhile to implement. There are plenty of ways a minor temp
> reduction might be seen with a more elaborate chassis, but none of them
> signficant enough to implement.

One unanswered question remains, which I hope someone would answer:
Why did they not just create a big hole? Is it not cheaper to create just
one big hole than to create many small ones in the case?

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"Navid" asked:
> One unanswered question remains, which I hope someone
> would answer: Why did they not just create a big hole?
> Is it not cheaper to create just one big hole than to create
> many small ones in the case?


If there were just one big hole, there would be
little turbulence as the air contacts the initial
components , and those components without
dedicated fans and/or heatsinks wouldn't get
adequate cooling. In my PC's case, the initial
component without a heatsink is the primary HDD.
By putting the HDD directly behind all those
punched holes, the designers assured that the
face of the HDD would be hit by turbulent air
and would thus be cooled properly. If there
were just one big hole, only the edges of the HDD
would be cooled as the air tumbled around it to
get past the HDD, and the face of the HDD
would be left to overheat.

Here's an analogy: Do you stir your coffee as
you blow on it to cool it? Or do you just blow
on it? The stirring adds turbulence to the liquid
to continually distribute it and run it past the cooling
air. Turbulence in air does a similar thing - it distributes
the heated air at the surface of the component so that
it can be swept away and replaced by the cooler
free-flowing air.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Tue, 8 Jun 2004 16:17:29 -0700, "Timothy Daniels"
<TDaniels@NoSpamDot.com> wrote:

<snip>

> That is exactly why turbulence is valuable

Nope, you're assuming that the effect of turbulence in immediate vicinity
of a heatsink for a chip will translate into significant effect in a
different situation.



> Dell cleverly provided the turbulence by the holes
> punched in the front of its metal cases,

Sorry, there was nothing "clever" about it, holes have been around for a
long time and used in cases too, not only by Dell.


> ... and for my
> Pentium II based PC, there is no fan for the CPU
> heatsink. There is only a single fan for the entire
> case - at the exit. For those CPUs needing heatsink
> fans, increased air velocity from the dedicated fan
> provides the turbulence at the surfaces of the heatsink.
> Notice that the shape of the heatsink "fingers" are not
> aerodynamicly shaped which would minimize air flow
> resistance. They have right-angle corners to increase
> turbulence.

Again you just don't get it. The heatsink shape is beneficial towards
creating turbulence, and helps, but that does not directly relate to the
effect of turbulence from chassis air intake holes.

Also you seem to ignore that the shape of the heatsink, like the holes in
a chassis, are simply what was cheap/easy to machine, or in the case of
the Dell heatsink, extrude and cross-cut. The thought was to provide
adequate cooling at low cost (and low noise as implemented).

>
>
>> You are putting more thought into your reply than they
>> did in creating the air intake.
>
>
> I'm sure you've just insulted hundreds of professional
> case designers. The more I study the design of my
> Dell PC's case and fascia, the more I admire the
> skills of the designer(s) to maximize cooling with
> the minimum expenditure for fans and the minimum
> release of noise. That they were able to cool all
> the components - which includes multiple optical
> drives and two magnetic hard drives and everything
> else with one 92mm exit fan in the back without
> making the PC sound like an air conditioner or a
> vacuum cleaner is a testimony to their skills.

Yes they're skilled, and with a good model of the system they can
strategically place the holes such that air flows where it's needed.
Unfortunately you still don't understand that the edges of the holes are
simply what results from the cheap/easy method of stamping out those
holes, not an advanced theory about turbulence.


> Again, visualize a smoke ring. Swirling air (turbulence)
> can act for a long time and travel a great distance.
> Try sweeping your hand through a column of rising smoke
> and time how long it takes for things to settle down to
> a single narrow column again. If you're familiar with
> aviation, think of how many minutes, how many miles
> of separation are needed between aircraft to avoid
> being thrown out of control by the turbulence generated
> by the aircraft in front. To think that turbulence
> disappears in the time it takes air to travel through a
> PC case is to think of air as molasses.

I've not argued that turbulence doesn't help in cooling, but rather that
the holes in the front of the case do not provide _significant_ difference
in cooling.

>
>
>> and if they'd REALLY wanted turbulence they'd have gone
>> to more elaborate measures to create it, not the opposite
>
>
> They get turbulence with simple circular holes. Why spend
> more money? A penny saved is a penny more of profit.
> Spend more time thinking like a bean counter.


I guess it's never going to sink in so I"ll just post this last thought
and give up on this thread:

Dell did not pioneer stamped holes. Stamped holes existed before anything
gave a thought about turbulence and it's affect on boundary layers.
Stamped holes are present regardless of whether there is ANY cooling issue
at all... that's what happens when fabricating something out of sheet
metal. It's not an advanced or clever design, it's simply a case frame
made out of sheet metal. That it creates turbulence, even that the
turbulence may aid in cooling to a minor, usually insignficant amount, is
a by-product of standard manufacturing methods that preceeded, have
nothing to do with designing a case.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Why didn't they make just one big hole?
One big hole would have lower air resistance and
more air would flow through the case per second.
One big hole would produce less induction noise.
One big hole would be easier to manufacture.
All the holes are now in the lower half of the front
of the case and they all bathe the HDDs first and
then the rest of the case. No apportioning of air
to various components is involved. Why didn't they,
at least in front of the HDDs, make just one big hole?

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Tue, 08 Jun 2004 22:31:35 GMT, "Navid" <nospam@nospam.invalid> wrote:


>One unanswered question remains, which I hope someone would answer:
>Why did they not just create a big hole? Is it not cheaper to create just
>one big hole than to create many small ones in the case?
>


By creating many small holes the air intake can be proportioned so hotter
areas receive greater percentage of air, at least if the case was designed
properly. In other cases it's a matter of reducing metal, cost and
weight, but keeping structural integrity high.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Tue, 8 Jun 2004 17:29:33 -0700, "Timothy Daniels"
<TDaniels@NoSpamDot.com> wrote:

>Why didn't they make just one big hole?

See my prior post in this thread.


>One big hole would have lower air resistance and
>more air would flow through the case per second.

No, it cannot be assumed based upon only a vague idea like "big".
Although the holes create turbulence, having ALL air intake in same may
also have other constrictions that are reduced by the multiple air entry
points.


>One big hole would produce less induction noise.
>One big hole would be easier to manufacture.

Would it?
Based upon what?
Simply the idea that there are multiple holes?

Multiple holes will spread out the air flow, more even distribution.

>All the holes are now in the lower half of the front
>of the case and they all bathe the HDDs first and
>then the rest of the case. No apportioning of air
>to various components is involved. Why didn't they,
>at least in front of the HDDs, make just one big hole?

But is there apportioning to different areas of the drive(s) behind it?
All that several holes would need do to be better at cooling a large
object like a drive, is to be placed further apart than diameter of the
single (alternative) hole, all else being equal. Also it's possible the
smaller holes will allow less noise out due to low frequency wavelengths,
but I think the real reason may be simpler than that, that there are
several holes because it creates a better appearance, even if that
appearance is usually hidden.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"kony" wrote:
> "Timothy Daniels" wrote:
>
> >Why didn't they make just one big hole?
>
> See my prior post in this thread.


Your prior post was rambling and illogical.
Again, why didn't they make just one big hole?


> >One big hole would have lower air resistance and
> >more air would flow through the case per second.
>
> No, it cannot be assumed based upon only a vague
> idea like "big".


One big hole would have a greater area-to-edge-length
ratio than a bunch of smaller holes, thus there would be
less energy lost to shearing of the air which sweeps past
the edges of the hole(s). Less shearing of the air means
less resistance and greater air flow. But lo! The designers
opted for a bunch of smaller holes. How come? Do
they want less air flow?


> Although the holes create turbulence, having ALL air intake
> in same may also have other constrictions that are reduced
> by the multiple air entry points.


I have NO idea what you meant to write, much less what
you meant to think. Try again, please.


> >One big hole would produce less induction noise.
> >One big hole would be easier to manufacture.
>
> Would it?
> Based upon what?
> Simply the idea that there are multiple holes?


One big hole would produce less shearing of the intake
air due to the lower area-to-edge-length ratio. That
translates to less noise. And one big hole would require
less in tooling and tooling maintenance than that required
to punch a bunch of smaller holes.


> Multiple holes will spread out the air flow, more even
> distribution.


Is there a need for even distribution? Yes? Well, I have
the answer for *that* - more turbulence. Turbulence
helps with distribution of velocities as well as cutting
through boundary layers.


> >All the holes are now in the lower half of the front
> >of the case and they all bathe the HDDs first and
> >then the rest of the case. No apportioning of air
> >to various components is involved. Why didn't they,
> >at least in front of the HDDs, make just one big hole?
>
> But is there apportioning to different areas of the drive(s)
> behind it?


Behind what?


> All that several holes would need do to be better at cooling
> a large object like a drive, is to be placed further apart than
> diameter of the single (alternative) hole, all else being equal.


Why?


> Also it's possible the smaller holes will allow less noise out
> due to low frequency wavelengths,


Low frequencies are less affected by the size of an obstacle
than high frequencies. That means that all the material between
the multiple holes reflects and blocks high frequencies *more*
than low frequencies - the opposite of your hypothesis.


> but I think the real reason may be simpler than that, that there are
> several holes because it creates a better appearance, even if that
> appearance is usually hidden.


And who cares about a hole's appearance if the hole can't be seen?

C'mon. Get some sleep.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"Timothy Daniels" <TDaniels@NoSpamDot.com> wrote in message
newsYWdnZ_Jm44wwlvdRVn-gQ@comcast.com...
> "Navid" asked:
>> One unanswered question remains, which I hope someone
>> would answer: Why did they not just create a big hole?
>> Is it not cheaper to create just one big hole than to create
>> many small ones in the case?
>
>
> If there were just one big hole, there would be
> little turbulence as the air contacts the initial
> components , and those components without
> dedicated fans and/or heatsinks wouldn't get
> adequate cooling. In my PC's case, the initial
> component without a heatsink is the primary HDD.
> By putting the HDD directly behind all those
> punched holes, the designers assured that the
> face of the HDD would be hit by turbulent air
> and would thus be cooled properly. If there
> were just one big hole, only the edges of the HDD
> would be cooled as the air tumbled around it to
> get past the HDD, and the face of the HDD
> would be left to overheat.
>
> Here's an analogy: Do you stir your coffee as
> you blow on it to cool it? Or do you just blow
> on it? The stirring adds turbulence to the liquid
> to continually distribute it and run it past the cooling
> air. Turbulence in air does a similar thing - it distributes
> the heated air at the surface of the component so that
> it can be swept away and replaced by the cooler
> free-flowing air.
>
> *TimDaniels*

This is a very interesting thread.
I was half way through cutting the grid on my case when I read your post
about the requirement of it for turbulence!
Seriously, I understand what you suggest the job of the small holes is in
the intake fan path.
But, why do you suggest the exhaust fan has those small holes in its way?
How do those help?
Thanks,
Navid

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"Navid" asked:
>
> This is a very interesting thread.
> I was half way through cutting the grid on my case
> when I read your post about the requirement of it
> for turbulence!
> Seriously, I understand what you suggest the job
> of the small holes is in the intake fan path.
> But, why do you suggest the exhaust fan has those
> small holes in its way?
> How do those help?


I can think of no reason for the grid over the exhaust
fans other than to keep out fingers. Some have also
theorized that it is to reduce escape of RF interference,
in other words, to meet FCC requirements. In my case,
I have removed the wire grill that was over the power
supply exhaust fan, and I have snipped away the grid of
holes that covered the case exhaust fan, and my nearby
AM/FM radios haven't shown any increase in noise.
But for all I know, all the cordless phones on the block
have stoppped working. What I do know is that there
is more air flow, especially out of the case fan exit. The
*possible* downside of greater case exit flow *might*
be less air flow through the power supply due to lower
air pressure in the case, but the air coming out of the
power supply is barely warm, so no biggie. And, as
far as I know, power supply cooling is not designed
to take into account the slight variance in case air pressure
due to varying case exhaust air flow rates. To do so
would be designing too close to the limits.

*TimDaniels*

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

kony wrote:

<snip>

>
> Dell did not pioneer stamped holes. Stamped holes existed before anything
> gave a thought about turbulence and it's affect on boundary layers.
> Stamped holes are present regardless of whether there is ANY cooling issue
> at all... that's what happens when fabricating something out of sheet
> metal. It's not an advanced or clever design, it's simply a case frame
> made out of sheet metal. That it creates turbulence, even that the
> turbulence may aid in cooling to a minor, usually insignficant amount, is
> a by-product of standard manufacturing methods that preceeded, have
> nothing to do with designing a case.

Give him time. Sooner or later he's got to realize that whatever minor
'turbulence' he thinks is being generated by the case holes is being beat
to hell and back by the fan blades immediately behind them.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Timothy Daniels wrote:

<snip>

Believe whatever you want.

However, the fact is, a desire to 'create' turbulence hasn't got a thing to
do with it.