92mm vs. 90mm Case Fan

On 5/1/2011 6:52 AM, Ben Myers wrote:
On Apr 30, 11:16 pm, wrote:
Ben Myers wrote:
On Apr 30, 11:30 am, wrote:
I'm looking to upgrade the case fan in my Studio XPS 8100.

The existing case fan is a Foxconn PVA092G12M. There's scant information
available from a Google search; this is either a 90mm fan or a 92mm fan.
My question is: Does it matter whether I use 90mm or 92mm? It's 2 lousy
millimeters.

I'm only asking because there are hardly any 90mm fans on Newegg - and
none with 3 pins, but there are tons of 92mm models.

Daddy

I don't understand "upgrade". A fan that blows more cubic feet per
minute? Or simply a replacement fan for one that does not work as
well any more? If the latter, just get another new one per its Dell
part number... Ben Myers

Lol ... I knew somebody would eventually ask me "why are you even doing
this?" And the answer is: I, uh, don't really have a good reason. I
guess I'm trying for a little more quiet. I replaced Dell's stock video
card, and that made my computer noticeably quieter. And case fans are
cheap and easy to install, so I figured give it a shot. The Foxconn
blows 45 CFM. If I can find something real quiet that blows as much air,
I'll try it.

My next project is an upgraded power supply. In a larger sense, all this
is preparation for building my first ever computer.

Daddy

I forgot that the Studio XPS 8100 is a tower computer, not a laptop!
Your reasoning makes sense for a tower or desktop. It would be at
least mildly deranged to try to replace a fan with anything other than
a stock item for a LAPTOP.

But pay attention to the connectors and wires for fan replacement in a
Dell... Ben Myers

i believe that the 90mm vs 92mm issue is simply due to how they are
measured, either by the outer dimension or by the mounting holes. in
other words as far as i know they are the same and mount the same.
also, i think the three pin fan connector on the 8100 is industry
standard with the white wire on pin three. however may dell case fans
are not wired the same as generic fans resulting in and error at start
up about the fan not being detected. the size is not critical but the
the wiring is.

Christopher Muto wrote:
On 5/1/2011 6:52 AM, Ben Myers wrote:
On Apr 30, 11:16 pm, wrote:
Ben Myers wrote:
On Apr 30, 11:30 am, wrote:
I'm looking to upgrade the case fan in my Studio XPS 8100.

The existing case fan is a Foxconn PVA092G12M. There's scant
information
available from a Google search; this is either a 90mm fan or a 92mm
fan.
My question is: Does it matter whether I use 90mm or 92mm? It's 2
lousy
millimeters.

I'm only asking because there are hardly any 90mm fans on Newegg - and
none with 3 pins, but there are tons of 92mm models.

Daddy

I don't understand "upgrade". A fan that blows more cubic feet per
minute? Or simply a replacement fan for one that does not work as
well any more? If the latter, just get another new one per its Dell
part number... Ben Myers

Lol ... I knew somebody would eventually ask me "why are you even doing
this?" And the answer is: I, uh, don't really have a good reason. I
guess I'm trying for a little more quiet. I replaced Dell's stock video
card, and that made my computer noticeably quieter. And case fans are
cheap and easy to install, so I figured give it a shot. The Foxconn
blows 45 CFM. If I can find something real quiet that blows as much air,
I'll try it.

My next project is an upgraded power supply. In a larger sense, all this
is preparation for building my first ever computer.

Daddy

I forgot that the Studio XPS 8100 is a tower computer, not a laptop!
Your reasoning makes sense for a tower or desktop. It would be at
least mildly deranged to try to replace a fan with anything other than
a stock item for a LAPTOP.

But pay attention to the connectors and wires for fan replacement in a
Dell... Ben Myers

i believe that the 90mm vs 92mm issue is simply due to how they are
measured, either by the outer dimension or by the mounting holes. in
other words as far as i know they are the same and mount the same. also,
i think the three pin fan connector on the 8100 is industry standard
with the white wire on pin three. however may dell case fans are not
wired the same as generic fans resulting in and error at start up about
the fan not being detected. the size is not critical but the the wiring
is.

I read elsewhere that the measurement (e.g., 92mm) is taken from the
center of one screw to the center of either adjacent screw (not
diagonally adjacent.) In any event, I am presuming that the tolerances
here do not measure down to one or two millimeters, so I will buy 90mm
or 92mm, whichever is the more suitable fan for me.

The case fan in my Studio XPS 8100 uses a three pin connector. This
means it reports its RPMs to the BIOS, but the fan speed cannot be
controlled by the BIOS; that's where a fourth wire comes in.

Something I've learned about replacing parts in an OEM computer: Most
consumers who buy individual parts (beyond RAM or a hard drive) are
either building their own gaming PC or doing substantial modding and/or
overclocking, so replacement parts are largely marketed to this crowd:
Power supplies have long connectors, case fans have bright LEDs, RAM has
heat spreaders, etc. If all you want is to kick up your performance a
notch or two, prepare to be overwhelmed. Anisotropic filtering? Sheesh!

Daddy

"Daddy" wrote in message
...
Christopher Muto wrote:
On 5/1/2011 6:52 AM, Ben Myers wrote:
On Apr 30, 11:16 pm, wrote:
Ben Myers wrote:
On Apr 30, 11:30 am, wrote:
I'm looking to upgrade the case fan in my Studio XPS 8100.

The existing case fan is a Foxconn PVA092G12M. There's scant
information
available from a Google search; this is either a 90mm fan or a 92mm
fan.
My question is: Does it matter whether I use 90mm or 92mm? It's 2
lousy
millimeters.

I'm only asking because there are hardly any 90mm fans on Newegg -
and
none with 3 pins, but there are tons of 92mm models.

Daddy

I don't understand "upgrade". A fan that blows more cubic feet per
minute? Or simply a replacement fan for one that does not work as
well any more? If the latter, just get another new one per its Dell
part number... Ben Myers

Lol ... I knew somebody would eventually ask me "why are you even doing
this?" And the answer is: I, uh, don't really have a good reason. I
guess I'm trying for a little more quiet. I replaced Dell's stock video
card, and that made my computer noticeably quieter. And case fans are
cheap and easy to install, so I figured give it a shot. The Foxconn
blows 45 CFM. If I can find something real quiet that blows as much
air,
I'll try it.

My next project is an upgraded power supply. In a larger sense, all
this
is preparation for building my first ever computer.

Daddy

I forgot that the Studio XPS 8100 is a tower computer, not a laptop!
Your reasoning makes sense for a tower or desktop. It would be at
least mildly deranged to try to replace a fan with anything other than
a stock item for a LAPTOP.

But pay attention to the connectors and wires for fan replacement in a
Dell... Ben Myers

i believe that the 90mm vs 92mm issue is simply due to how they are
measured, either by the outer dimension or by the mounting holes. in
other words as far as i know they are the same and mount the same. also,
i think the three pin fan connector on the 8100 is industry standard with
the white wire on pin three. however may dell case fans are not wired
the same as generic fans resulting in and error at start up about the fan
not being detected. the size is not critical but the the wiring is.

I read elsewhere that the measurement (e.g., 92mm) is taken from the
center of one screw to the center of either adjacent screw (not diagonally
adjacent.) In any event, I am presuming that the tolerances here do not
measure down to one or two millimeters, so I will buy 90mm or 92mm,
whichever is the more suitable fan for me.

The case fan in my Studio XPS 8100 uses a three pin connector. This means
it reports its RPMs to the BIOS, but the fan speed cannot be controlled by
the BIOS; that's where a fourth wire comes in.

Something I've learned about replacing parts in an OEM computer: Most
consumers who buy individual parts (beyond RAM or a hard drive) are either
building their own gaming PC or doing substantial modding and/or
overclocking, so replacement parts are largely marketed to this crowd:
Power supplies have long connectors, case fans have bright LEDs, RAM has
heat spreaders, etc. If all you want is to kick up your performance a
notch or two, prepare to be overwhelmed. Anisotropic filtering? Sheesh!

Daddy

I replaced the case - processor fan in my xps 8100 with this one from
Amazon.
Sorry for the large text just pasted it from their site.
Perfect fit, just cut the wires and used the old MB plug that lets the fan
sensor work.
Wire color matches, so it was no trouble.
Sam

Antec TriCool 92mm DBB Case Fan with 3-Spd Switch 3/4-Pin & 3-Pin Monitoring

Hi!

The case fan in my Studio XPS 8100 uses a three pin connector.
This means it reports its RPMs to the BIOS, but the fan speed
cannot be controlled by the BIOS; that's where a fourth wire comes
in.

Actually, it can. The age-old trick (from the time that PCs started to
include hardware capable of such things) is to drive the fan from a
pulse-width-modulation controller. This works by varying the ratio of
on time to off time...longer "on" pulses let the fan spin faster,
while longer off pulses slow it down. Do this frequently enough and
the illusion of "smooth" fan speed control can be maintained across
the board.

Dell doesn't do that *in most cases*. Instead, the Dell fans typically
have a thermal sensing bulb poking out somewhere around the hub. This
means the fan itself--and not any other part of the computer--controls
its own operating speed.

The newer four wire fans do things a little differently. The first
three wires do as they've always done and serve to provide voltage
input and tachometer output signals. The fourth wire actually accepts
input data from the fan controller on the computer's mainboard and
then the fan's electronics figure out how best to arrive at the
desired rotational speed.

The newest Dell systems (basically anything that is built around a BTX
motherboard design) forgo the thermal sensing bulb and instead gather
temperature from some point on the motherboard. Exactly where that is
would be rather hard to say--apparently Dell uses custom chips (the
function itself is embedded in the LPCIO IC) and apparently the
internal workings of such chips are Great Secrets. (I actually asked
SMsC about these chips and was told that they are protected under a
nondisclosure agreement. Even their own employees cannot look up
information on them. Evidently Dell--or whoever does the logic design
of Dell's motherboards--thinks that a part used in basically every
type of PC made in the recent past--is really something to keep a
tight lid on. Whatever.)

There are no obvious thermal sensors on the board. Most boards don't
even seem to have places for them--the Dimension E521 board is a
notable exception to this. It has solder spots for sensors.

People who have managed to get some data on the SMsC LPCIO chips that
Dell uses say that hardware monitoring and fan control aren't
supported. That leaves me to guess that Dell uses the core temperature
from the CPU to decide how hot the rest of the system is and then
adjusts the fan speed accordingly, possibly by just writing the
desired speed control data to the fan's control electronics over the
fourth wire.

I'd really like to know for sure. I've been contacted by people within
Dell in times past--I wish someone would pick up on one of my posts
along this subject and tell me what I'd like to know without all the
red tape.

William

William R. Walsh wrote:
Hi!

The case fan in my Studio XPS 8100 uses a three pin connector.
This means it reports its RPMs to the BIOS, but the fan speed
cannot be controlled by the BIOS; that's where a fourth wire comes
in.

Actually, it can. The age-old trick (from the time that PCs started to
include hardware capable of such things) is to drive the fan from a
pulse-width-modulation controller. This works by varying the ratio of
on time to off time...longer "on" pulses let the fan spin faster,
while longer off pulses slow it down. Do this frequently enough and
the illusion of "smooth" fan speed control can be maintained across
the board.

Dell doesn't do that *in most cases*. Instead, the Dell fans typically
have a thermal sensing bulb poking out somewhere around the hub. This
means the fan itself--and not any other part of the computer--controls
its own operating speed.

The newer four wire fans do things a little differently. The first
three wires do as they've always done and serve to provide voltage
input and tachometer output signals. The fourth wire actually accepts
input data from the fan controller on the computer's mainboard and
then the fan's electronics figure out how best to arrive at the
desired rotational speed.

The newest Dell systems (basically anything that is built around a BTX
motherboard design) forgo the thermal sensing bulb and instead gather
temperature from some point on the motherboard. Exactly where that is
would be rather hard to say--apparently Dell uses custom chips (the
function itself is embedded in the LPCIO IC) and apparently the
internal workings of such chips are Great Secrets. (I actually asked
SMsC about these chips and was told that they are protected under a
nondisclosure agreement. Even their own employees cannot look up
information on them. Evidently Dell--or whoever does the logic design
of Dell's motherboards--thinks that a part used in basically every
type of PC made in the recent past--is really something to keep a
tight lid on. Whatever.)

There are no obvious thermal sensors on the board. Most boards don't
even seem to have places for them--the Dimension E521 board is a
notable exception to this. It has solder spots for sensors.

People who have managed to get some data on the SMsC LPCIO chips that
Dell uses say that hardware monitoring and fan control aren't
supported. That leaves me to guess that Dell uses the core temperature
from the CPU to decide how hot the rest of the system is and then
adjusts the fan speed accordingly, possibly by just writing the
desired speed control data to the fan's control electronics over the
fourth wire.

I'd really like to know for sure. I've been contacted by people within
Dell in times past--I wish someone would pick up on one of my posts
along this subject and tell me what I'd like to know without all the
red tape.

William

You analysis is very interesting, as usual.

This is just anecdotal - based on my observations - but I do not sense
that my case fan is varying in speed. But then, I don't really rock out
my hardware in the first place.

Daddy

On Wed, 04 May 2011 15:01:41 -0400, Daddy wrote:

William R. Walsh wrote:
Hi!

The case fan in my Studio XPS 8100 uses a three pin connector.
This means it reports its RPMs to the BIOS, but the fan speed
cannot be controlled by the BIOS; that's where a fourth wire comes
in.

Actually, it can. The age-old trick (from the time that PCs started to
include hardware capable of such things) is to drive the fan from a
pulse-width-modulation controller. This works by varying the ratio of
on time to off time...longer "on" pulses let the fan spin faster,
while longer off pulses slow it down. Do this frequently enough and
the illusion of "smooth" fan speed control can be maintained across
the board.

Dell doesn't do that *in most cases*. Instead, the Dell fans typically
have a thermal sensing bulb poking out somewhere around the hub. This
means the fan itself--and not any other part of the computer--controls
its own operating speed.

The newer four wire fans do things a little differently. The first
three wires do as they've always done and serve to provide voltage
input and tachometer output signals. The fourth wire actually accepts
input data from the fan controller on the computer's mainboard and
then the fan's electronics figure out how best to arrive at the
desired rotational speed.

The newest Dell systems (basically anything that is built around a BTX
motherboard design) forgo the thermal sensing bulb and instead gather
temperature from some point on the motherboard. Exactly where that is
would be rather hard to say--apparently Dell uses custom chips (the
function itself is embedded in the LPCIO IC) and apparently the
internal workings of such chips are Great Secrets. (I actually asked
SMsC about these chips and was told that they are protected under a
nondisclosure agreement. Even their own employees cannot look up
information on them. Evidently Dell--or whoever does the logic design
of Dell's motherboards--thinks that a part used in basically every
type of PC made in the recent past--is really something to keep a
tight lid on. Whatever.)

There are no obvious thermal sensors on the board. Most boards don't
even seem to have places for them--the Dimension E521 board is a
notable exception to this. It has solder spots for sensors.

People who have managed to get some data on the SMsC LPCIO chips that
Dell uses say that hardware monitoring and fan control aren't
supported. That leaves me to guess that Dell uses the core temperature
from the CPU to decide how hot the rest of the system is and then
adjusts the fan speed accordingly, possibly by just writing the
desired speed control data to the fan's control electronics over the
fourth wire.

I'd really like to know for sure. I've been contacted by people within
Dell in times past--I wish someone would pick up on one of my posts
along this subject and tell me what I'd like to know without all the
red tape.

William

You analysis is very interesting, as usual.

This is just anecdotal - based on my observations - but I do not sense
that my case fan is varying in speed. But then, I don't really rock out
my hardware in the first place.

If you have the Dell utilities (which I can get to by doing F12 on
startup, and choosing utilities), one or more of them varies the fan
speed. It says "running at slow speed", "running at high speed" and
you can hear the difference. At least I can.

Daddy