sucks or blows, which is better?

[ToolPackinMama]

I guess "coupled" is a relative thing.

That's the spirit.

 

[Gianmaria Fontana di Sacculmino]

Il giorno Mon, 04 Oct 2004 22:23:13 -0700, ric <[email protected]> ha
scritto:

*Exactly* what do they claim (and where do they claim it?)

http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/cooling_guide.pdf

 

[Gianmaria Fontana di Sacculmino]

The CORRECT way to orient case fans is for fans in the front of the case to
suck in and for fans in the rear of the case to blow out.

It is correct without a front fan as well.

 

[ric]

http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/cooling_guide.pdf

That URL doesn't even discuss backpressure, CFM, or the effect a front
panel fan has on rear fan CFM. Those are the three topics of which you
claimed AMD disagreed with my facts.

I'll ask again: What and where does AMD disagree with what I've said about
backpressure, CFM, and the effect a front panel fan has on rear fan CFM?

 

[ric]

All I said was 'sufficiently'.


It depends on the case and it will have no effect at all on what I like to
humorously call the 'swiss cheese' models.

Even a *little* help in moving air from front to rear will reduce the
backpressure that the rear fan sees. I put a very low RPM fan in the
front of my case. It increased the CFM of the rear fan enough to give
me a 5 degree C reduction of internal ambient, and added NO noise (it
was well below the PSU/CPU/rear fan noise ceiling.)

If there is no other resort then it can help, if the case isn't punched
full of holes everywhere, but it isn't necessarily the best place to look
for improvement if one cares about noise.

Now you wanna enter noise into the equation?

OK. For that, use NO front or rear fans, a big thick copper heatsink on
the CPU, and a 10"-14" low RPM fan mounted in the cover blowing down on
everything.

Very cool, and the only noise is the PSU fan.

 

[David Maynard]

David Maynard wrote:




Even a *little* help in moving air from front to rear will reduce the
backpressure that the rear fan sees. I put a very low RPM fan in the
front of my case. It increased the CFM of the rear fan enough to give
me a 5 degree C reduction of internal ambient, and added NO noise (it
was well below the PSU/CPU/rear fan noise ceiling.)

I'm glad it worked for you, be it because the rear fan was undersized,
system airflow unbalanced, or whatever. But if someone with a proper active
rear exhaust draws from this exchange the notion that adding a silent front
case fan is going to give them an additional 5C drop in case temperature
they're, most likely, going to be disappointed.

Now you wanna enter noise into the equation?

I didn't know it was ever 'out'.

OK. For that, use NO front or rear fans, a big thick copper heatsink on
the CPU, and a 10"-14" low RPM fan mounted in the cover blowing down on
everything.

Very cool, and the only noise is the PSU fan.

Unfortunately, no, it won't be 'very cool' because a 'large' HSF on the CPU
just blows it around in the case and if you don't remove the heat from the
case a internal 'big blower' just makes it's evenly hot inside. And you
can't get the CPU 'very cool' if you're blowing hot case air on it.

 

[ric]

Unfortunately, no, it won't be 'very cool' because a 'large' HSF on the CPU
just blows it around in the case and if you don't remove the heat from the
case a internal 'big blower' just makes it's evenly hot inside. And you
can't get the CPU 'very cool' if you're blowing hot case air on it.

Read more carefully. I didn't write HSF on the CPU. I wrote "a big thick
copper heatsink" on the CPU. No fan. The only fan (besides in the PSU) is
the 10"-14" low RPM fan mounted in the case cover. It blows *cool* room
ambient air onto the copper HS and the rest of the case, and any warm air
is pushed out the front *or* the back (or both.) This works great, but
mounting and filtering the large fan on the cover isn't easy.

 

[David Maynard]

David Maynard wrote:




Read more carefully. I didn't write HSF on the CPU. I wrote "a big thick
copper heatsink" on the CPU. No fan. The only fan (besides in the PSU) is
the 10"-14" low RPM fan mounted in the case cover. It blows *cool* room
ambient air onto the copper HS and the rest of the case, and any warm air
is pushed out the front *or* the back (or both.) This works great, but
mounting and filtering the large fan on the cover isn't easy.

OK. I misinterpreted what you meant with the non existent fan technology.

 

[Gianmaria Fontana di Sacculmino]

That URL doesn't even discuss backpressure, CFM, or the effect a front
panel fan has on rear fan CFM. Those are the three topics of which you
claimed AMD disagreed with my facts.

I don't know about your "facts", I simply copy here what AMD states:
"a front cooling fan does not seem to be essential. In fact, in some
extreme situations, testing showed these fans to be recirculating hot
air rather than introducing cool air".

I'll ask again: What and where does AMD disagree with what I've said about
backpressure, CFM, and the effect a front panel fan has on rear fan CFM?

Probably I misinterpreted your arguments about fans technologies, back
pressures and so on.

 

[ric]

OK. I misinterpreted what you meant with the non existent fan technology.

Non existent? What's non existent?

 

[ric]

I don't know about your "facts", I simply copy here what AMD states:
"a front cooling fan does not seem to be essential. In fact, in some
extreme situations, testing showed these fans to be recirculating hot
air rather than introducing cool air".

Heh, I wish they would have explained their "...in some extreme situations..."
remark. The only situation that could lead to what they described is in a
case with no front fan air vents. Such a case surely is is not on AMD's
approved list. <g>

And there is a difference between providing maximum efficiency and being
"essential."

 

[David Maynard]

Gianmaria Fontana di Sacculmino wrote:




Heh, I wish they would have explained their "...in some extreme situations..."
remark. The only situation that could lead to what they described is in a
case with no front fan air vents.

That would do it but it isn't the only situation. While not common now,
cases a few years back some times had the front case fan mounting such that
the fan was little more than simply 'hung in mid air' at the front of the
case (extremely common in old AT cases). By that I mean, not flush against
any wall or other air director, but still behind the front panel, of
course. And, in those circumstances, fans are real dern good at sucking
their own exhaust back into their own intake and with no case wall,
baffeling, ducting, or anything else, there's nothing to cause it to draw
in air so it just stirs the warm air inside the case.

 

[ric]

That would do it but it isn't the only situation. While not common now,
cases a few years back some times had the front case fan mounting such that
the fan was little more than simply 'hung in mid air' at the front of the
case (extremely common in old AT cases). By that I mean, not flush against
any wall or other air director, but still behind the front panel, of
course. And, in those circumstances, fans are real dern good at sucking
their own exhaust back into their own intake and with no case wall,
baffeling, ducting, or anything else, there's nothing to cause it to draw
in air so it just stirs the warm air inside the case.

I agree that while the above scenario might be horrible for drawing cool air
into the case, if it is moving *any* air towards the rear of the case it will
reduce the static pressure seen by the rear fan and make that fan operate
closer to its zero static pressure CFM. So, while not the greatest fan
placement, it's not a total waste.

 

[David Maynard]

David Maynard wrote:




I agree that while the above scenario might be horrible for drawing cool air
into the case, if it is moving *any* air towards the rear of the case it will
reduce the static pressure

No, it won't, because the 'pressure' it's creating on one side is
counteracted by the *negative* pressure it's creating on the other. And
without anything to keep those two pressure regions separated it just sucks
it's own exhaust right back into it's own intake, after a brief radial
excursion of a few inches from the kinetic energy imparted by the fan blades

I get the feeling you'd argue there was a 'benefit' even if just one
freaking extra molecule happened to wander toward the rear of the case

seen by the rear fan and make that fan operate
closer to its zero static pressure CFM. So, while not the greatest fan
placement, it's not a total waste.

Actually, the ones I described *are* a total waste with respect to case temp.

 

[ric]

No, it won't, because the 'pressure' it's creating on one side is
counteracted by the *negative* pressure it's creating on the other. And
without anything to keep those two pressure regions separated it just sucks
it's own exhaust right back into it's own intake, after a brief radial
excursion of a few inches from the kinetic energy imparted by the fan blades

Re-read what I said. If, as I said, it is moving some air toward the rear
of the case, then it does create a benefit.

I get the feeling you'd argue there was a 'benefit' even if just one
freaking extra molecule happened to wander toward the rear of the case

And I get the feeling that you would argue unless I agreed with you 100%.
I spent a good three months with a large after-market computer products
company studying case internal airflow patterns and fan efficiency. The
effects of a front fan on rear fan efficiency was one of the main areas
of concern. The effects are often understated. This thread is a good example.

 

[David Maynard]

David Maynard wrote:




Re-read what I said. If, as I said, it is moving some air toward the rear
of the case, then it does create a benefit.

I read what you said but it isn't useful to speculate 'if' when it doesn't,
so I explained the 'why' it doesn't.

And I get the feeling that you would argue unless I agreed with you 100%.
I spent a good three months with a large after-market computer products
company studying case internal airflow patterns and fan efficiency.

How was 'efficiency' defined? Airflow per watt consumed? Airflow per Db?
Airflow per dollar cost? How much air one could push through a fixed hole
pattern? Anyone try using a wire grill instead of a punched hole vent? Or
other means to improve ventilation besides more fans?

I'm just asking because it isn't clear.

The
effects of a front fan on rear fan efficiency was one of the main areas
of concern. The effects are often understated. This thread is a good example.

It's your insistence, even in the face of clear examples, it just 'always
does', regardless, that I dispute.

 

[ric]

I read what you said but it isn't useful to speculate 'if' when it doesn't,
so I explained the 'why' it doesn't.

The problem with your argument is that even a poorly placed front fan
often moves air toward the rear of the case.

How was 'efficiency' defined? Airflow per watt consumed? Airflow per Db?
Airflow per dollar cost? How much air one could push through a fixed hole
pattern? Anyone try using a wire grill instead of a punched hole vent? Or
other means to improve ventilation besides more fans?

As far as the front fans effect on the rear fan goes, efficiency was defined
as how close the rear fans CFM was to its published (zero static pressure)
specification [with and without the front fans assistance.]

It's your insistence, even in the face of clear examples, it just 'always
does', regardless, that I dispute.

I don't remember ever stating that it "always does" ("it" being the front
fan moving some air towards the rear.) What I have said is that even with
a poorly placed front fan "it" usually does. If you have a Message ID where
I said "it" always does, I'd like to see it.

 

[David Maynard]

David Maynard wrote:




The problem with your argument is that even a poorly placed front fan
often moves air toward the rear of the case.

And after explaining that, in the example given, you have a fan creating
equal, but opposite, pressures neutralizing the effect of each other you
base that assertion on what?

How was 'efficiency' defined? Airflow per watt consumed? Airflow per Db?
Airflow per dollar cost? How much air one could push through a fixed hole
pattern? Anyone try using a wire grill instead of a punched hole vent? Or
other means to improve ventilation besides more fans?

As far as the front fans effect on the rear fan goes, efficiency was defined
as how close the rear fans CFM was to its published (zero static pressure)
specification [with and without the front fans assistance.]

I see.

It's an interesting criteria but I'm not sure it's the best one for case
cooling.

I don't remember ever stating that it "always does" ("it" being the front
fan moving some air towards the rear.) What I have said is that even with
a poorly placed front fan "it" usually does. If you have a Message ID where
I said "it" always does, I'd like to see it.

You can quibble with how you've avoided saying it so you can say you
haven't said it but no matter what example crops up you always manage to
make the same claim, regardless. To me that's 'always'. But I'll entertain
your 'not always' example of when it doesn't 'help'.

 

[ric]

And after explaining that, in the example given, you have a fan creating
equal, but opposite, pressures neutralizing the effect of each other you
base that assertion on what?

Actual testing with various front fan placements in multiple cases.

As far as the front fans effect on the rear fan goes, efficiency was defined
as how close the rear fans CFM was to its published (zero static pressure)
specification [with and without the front fans assistance.]

I see.

It's an interesting criteria but I'm not sure it's the best one for case
cooling.

As far as studying the effect of a front fan on a rear fans CFM, what
definition would you choose?

You can quibble with how you've avoided saying it so you can say you
haven't said it but no matter what example crops up you always manage to
make the same claim, regardless. To me that's 'always'. But I'll entertain
your 'not always' example of when it doesn't 'help'.

LOL! All I can relate are the results of the tests. In those tests, very few
front fan locations resulted in no effect on rear fan CFM. In those few cases,
other factors (such as harness placement) were factors as well. Sorry if this
disturbs you.

 

[David Maynard]

David Maynard wrote:




Actual testing with various front fan placements in multiple cases.

Unless you mounted a fan like in the example then that experience doesn't
apply.

As far as the front fans effect on the rear fan goes, efficiency was defined
as how close the rear fans CFM was to its published (zero static pressure)
specification [with and without the front fans assistance.]

I see.

It's an interesting criteria but I'm not sure it's the best one for case
cooling.

As far as studying the effect of a front fan on a rear fans CFM, what
definition would you choose?

I said "for case cooling," not simply how to make a rear case fan run as if
it were mounted in free air, so my criteria would be more closely related
to airflow vs noise generated vs component cost and in light of what the
cooling requirements were and what other means for improving cooling might
be possible. For example, if I were looking at a case with punched fan
grills I might be more inclined to improve the punch pattern, or change
them to wire grills, than add a fan. I also might be more inclined to use a
more powerful rear fan, depending on where we're starting from, than add
another fan to the front. And if we were looking at a case design, as
opposed to retrofitting an existing one, I might be more inclined to
increase the fan size, or put the 'second' fan, if needed, in the rear,
rather than add one to the front.

I have no problem with running a fan at 'less that its free air speed' if
it's quiet, cost effective, and sufficiently cools the system so I don't
see that as a 'goal'. I can imagine situations where other requirements so
constrain things that it might become a 'strategy' for increasing airflow,
but not a 'goal' of its own.

LOL! All I can relate are the results of the tests. In those tests, very few
front fan locations resulted in no effect on rear fan CFM. In those few cases,
other factors (such as harness placement) were factors as well. Sorry if this
disturbs you.

It doesn't disturb me. I'm just pointing out that your case test doesn't
necessarily apply to every case and fan mount ever made.

Btw, just curious, do you know if the folks who ran the test used it on
production units?