Hi Terry
Sorry it's taken so long for me to reply but I've been working on a project
similar to the one you describe and thought it would be better until I had
some results to share before responding.
> Unless you are heavily into games or other multimedia, I
> would claim that
> a 100W power supply is more than adequate for *most* PC applications.
> Consider:
> <iMac DVD, HP e-vectra, desktop replacement laptops, K6 300 system>
Yes, I agree with you entirely. Most PCs which don't have lots of hard
drives and/or expansion cards would indeed run quite adequately on a 100W
PSU. In fact I was recently working on a HP Vectra VLi8 (low profile case)
which came with a teeny 90W PSU! Sadly it still had a fan in it.
> Obviously you're not going to run a pair of Voodoo cards and
> a bunch of
> 10,000 rpm disks off of a 100W supply, and I've heard that Athlon
> processors are particularly power hungary, but I believe that
> if you are
> willing to sacrifice some performance, you *can* build a
> fan-free system.
It would be great to hear if someone had managed to do this. I'm working on
this at the moment in my spare time but it's actually quite difficult!
> Also, Quantum has recently introduced some low-noise, low-power --and
> somewhat low performance-- disks (LCT15, not to be confused
> with the LCT10 series).
Yes, I think they should be available mid-August from what I've heard. On
paper as far as I recall they are 1dB(A) quieter than the LCT10 series.
However this doesn't exactly fill me with excitement because I have a friend
with an LCT10 and he says it's not even vaguely quiet! (it's better in a
SilentDrive though). In my experience, hard drive manufacturers are
notoriously optimistic with the dB(A) figures they quote for their drives. I
think they must have different ways of measuring them.
> I'm interested in taking such a disk, and a 100W
> convection-cooled p/s,
> and maybe a celeron 566 underclocked to 500, with the best
> passive heat
> sink I can find, and seeing what kind of temps result.
>
> If anyone has any thoughts on that, let me know...
The PC I'm building at the moment is based on a Celeron II 600. Sadly
because Intel clocklock the multiplier on all their CPUs nowadays (and have
done so for quite a long time now), underclocking is not really an option.
Overclocking techniques now rely almost entirely on increasing the system
bus speed and therefore indirectly the CPU frequency, since the CPU
multiplies the bus speed by its fixed multiplier to get its own clock speed.
However this is a little risky since by using this method you are also
overclocking all your expansion cards, including EIDE interface. I've heard
stories of fried SCSI cards and corrupted hard drives as a result of
overclocking the system bus too much although as long as you don't go beyond
75MHz (normally 66) I believe you should be safe.
Anyway I'm getting off the subject. The bottom line is that over or
underclocking is very difficult. I can't even underclock my project system
by reducing the bus speed lower than 66MHz because the (Asus) BIOS doesn't
allow it! So the Celeron runs at 600MHz and there it will stay. I know I
could have bought a 566MHz chip but my supplier didn't have any in stock,
and to buy a 533MHz would run the risk of them sending me an original
non-FCPGA .25 micron Celeron which would have run *much* hotter! The FC-PGA
533 and 566 CPUs do have slightly reduced heat outputs than the 600 but not
by much.
My experiments so far have shown that with *two* Molex Radial Fin coolers
mounted passively double-decker style, the maximum temperature reached by
the CPU in an open-air case is 71 degrees C. I believe Intel's maximum rated
temperature is specified at 70. My result could possibly be bettered by
using a more expensive thermal compound but then I still have to put the
sides on the case! :( Even though I'm using a 100% aluminium Coolermaster
ATC-200 case with a vent at the top, the temperature inside will undoubtedly
increase with the sides on.
By the way, the best way of measuring the maximum CPU temperature seems to
be to press DEL (or whatever) to enter the BIOS setup at boot time and let
it sit there. This way, no clever "software cooling" techniques are in place
and the processor will be running at its hottest. The BIOS should also have
a screen which displays the current CPU temperature on the screen which is
handy.
Software cooling is where the operating system or a program running within
the operating system deliberately places the processor in a special "idle"
mode which uses very little power when the computer isn't doing anything.
Windows NT, Windows 2000 and (as far as I can tell) Windows ME seem to be
the only Microsoft OS's to do this without additional software (CPUIdle is
the famous one and there are others).
The other problem is finding a convection cooled PSU. Other than
disconnecting the fan in a standard PSU and hoping for the best (not
recommended!) I'm a bit stuck. If anyone knows of a source for fanless ATX
PSUs please let me know. The ones my existing supplier provides are
extremely quiet but not actually silent.
Any other thoughts anyone has would of course be extremely interesting.
Kind regards
Glenn
--
Glenn Garrett http://www.quietpc.com/glenn@... PC noise elimination
Hi Art,
Have you tried it yet? How did it go?
Sami
> The idea of mounting a watercooler on a harddisk and packing the
> whole
> assembly in plenty of foam rubber had occurred to me too. I don't
> know
> what kind of rubber would be the best, however. I guess it
shouldn't
> be too massive, or it would still transfer vibrations to the
chassis.
>
> There's only one way to find out, I suppose. I'll try it this
weekend.
>
> Bye,
>
> Art
Hi
I came over this link with a new atx case that is going to be
manufactured... aluminum and made for being quiet and cool...
http://w1.317.telia.com/~u31707990/
Regards,
Thomas
Hi Allan,
> Just been eyeing up the new A7V Duron motherboard from ASUS. It mentions a
> new feature called Auto Fan Off but gives no further details. Can anyone
> shed any light on this feature?
This simply means the board will switch off the CPU fan when placed in
suspend mode. Most ASUS boards have this feature, as do most Giga-Byte
boards. The PSU fan will still be running, though.
A more interesting feature is Suspend-to-RAM (ACPI S3). This maintains the
current state of the computer in RAM while it is almost completely powered
down (PSU fan included). It still draws a small amount of current to
maintain the RAM image. When it's time to resume, the PC boots up into its
previous state very quickly.
Cheers,
Shayne Lennox
slennox@...
At 13:04 2000-07-22 +0100, you wrote:
>Seems to me that the real solution is to channel the heat outside the case
>to large radiant fins.
It seems like the Calm System will use this solution. They say it will be
available in September:
http://www.cnssystem.com/product.html
Regards,
Tomas Risberg
Just been eyeing up the new A7V Duron motherboard from ASUS. It mentions a
new feature called Auto Fan Off but gives no further details. Can anyone
shed any light on this feature?
Allan Toombs
organic_acid@...
'Knowledge is better than wonder'
Water cooling seems a bit OTT from my viewpoint. During the week I was
thinking about how quality hi-fi has been dealing with cooling and noise for
years and has never needed fans and Kryotech style solutions. We're talking
big hot valves and transistors here too.
Seems to me that the real solution is to channel the heat outside the case
to large radiant fins. Imagine a slot board with the chip twisted ninety
degrees so that it can put a large aluminium sink out of one of the
expansion slots. Think about it, there's generally a PCI slot spare and the
back of a PC always has space around it due to projecting VDU and power
cables.
Of course there's no sign of any motherboard or case manufacturers doing
this. Anyone on the list see a problem with my idea?
Allan Toombs
organic_acid@...
'From the eye to the brain is only an inch or two,
From the eye to the heart is only slightly farther.'
Hi,
--- In Silent-PC@egroups.com, "Sami Rantala" <sami.rantala@e...>
wrote:
> Has anyone experience in water cooling? I've heard that
> it may be more silent (depending on the pump) than air cooling.
I used the Senfu watercooling kit (reviewed on many overclockers
websites by now) to replace the Cool-it Dude fan on my Athlon 700. It
works just fine killing noise, but check out
http://nfs.hwextreme.com/burn.shtml for a prime example of the risks
involved in using this experimental equipment. I replaced the
standard
pump with a bigger one, rated to last several years in constant use.
Temperature monitoring remains essential in unattended operation, so
I've rigged my system to power down if the temperature exceeds 45 C.
Note: Current CPU-temp is only 26 C :-)
Sadly the power supply fan (in the standard A-Open HQ-08) turned out
to be the next most irritating noise source :-[
The idea of mounting a watercooler on a harddisk and packing the
whole
assembly in plenty of foam rubber had occurred to me too. I don't
know
what kind of rubber would be the best, however. I guess it shouldn't
be too massive, or it would still transfer vibrations to the chassis.
There's only one way to find out, I suppose. I'll try it this weekend.
Bye,
Art
Has anyone experiense in water cooling? I've heard that
it may be more silent (depending on the pump) than air cooling.
For example one could build silencing box for harddrive and use
water cooling to cool it. It could handle even the faster and hotter
drives (silent drive can handle on 5400rpm ones).
Any ideas, experiences, etc.?
Sami
*** Message re-sent without digital signature. My apologies that you
received two copies, especially if the first one was unreadable.
Hi Sami
I don't know anyone who has obtained good results from lining the case
with sound absorption material, although this may help if you have a
case with a particularly poor panel fit where the panels are
vibrating/rattling. If you're going to try it, I would recommend you
ensure the PC does not overheat and don't use a product which produces
static, such as carpet containing nylon - this can damage integrated
circuits inside the PC.
Vibrations sent into the case from the hard drive can make PCs very
loud. The SilentDrive enclosure product which my company sells greatly
reduces the noise from hard drives but even so, vibrations from the
drive motor and head servo mechanism are made louder when the
SilentDrive or the drive itself are mounted directly to the chassis.
One way around this is not to use mounting screws (use foam cushions,
for example). Obviously that the PC should not be transported while the
drive is mounted like this, and also be aware that the hard drive may
overheat since it cannot use the PC chassis as a heatsink. However, it's
worth doing for a short period of time just to see what difference it
makes to the noise. If you have a very cool-running drive, you may be
able to get away with it.
Kind regards
Glenn
--
Glenn Garrett http://www.quietpc.com/glenn@... PC noise elimination
> -----Original Message-----
> From: Sami Rantala [mailto:sami.rantala@...]
> Sent: Wednesday, 12 July 2000 11:00
> To: Silent-PC@egroups.com
> Subject: [Silent-PC] sound absorption lining, etc.
>
>
> Hi,
> Does anyone know how effective different ways of silencing are?
> (Yes, I know: get silent power supply, fans and hd ;)
>
> For example, how much I can expect to gain from using sound
> absorption
> lining? Or making sure the harddrive doesn't cause case to vibrate?
>
> Sami
>
>
>
> --------------------------------------------------------------
> ----------
> Life's too short to send boring email. Let SuperSig come to
> the rescue.
> http://click.egroups.com/1/6137/1/_/_/_/963395981/
> --------------------------------------------------------------
> ----------
>
> To unsubscribe from this group, send an email to:
> Silent-PC-unsubscribe@egroups.com
>
>
>
Hi Sami
I don't know anyone who has obtained good results from lining the case
with sound absorption material, although this may help if you have a
case with a particularly poor panel fit where the panels are
vibrating/rattling. If you're going to try it, I would recommend you
ensure the PC does not overheat and don't use a product which produces
static, such as carpet containing nylon - this can damage integrated
circuits inside the PC.
Vibrations sent into the case from the hard drive can make PCs very
loud. The SilentDrive enclosure product which my company sells greatly
reduces the noise from hard drives but even so, vibrations from the
drive motor and head servo mechanism are made louder when the
SilentDrive or the drive itself are mounted directly to the chassis.
One way around this is not to use mounting screws (use foam cushions,
for example). Obviously that the PC should not be transported while the
drive is mounted like this, and also be aware that the hard drive may
overheat since it cannot use the PC chassis as a heatsink. However, it's
worth doing for a short period of time just to see what difference it
makes to the noise. If you have a very cool-running drive, you may be
able to get away with it.
Kind regards
Glenn
--
Glenn Garrett http://www.quietpc.com/glenn@... PC noise elimination
> -----Original Message-----
> From: Sami Rantala [mailto:sami.rantala@...]
> Sent: Wednesday, 12 July 2000 11:00
> To: Silent-PC@egroups.com
> Subject: [Silent-PC] sound absorption lining, etc.
>
>
> Hi,
> Does anyone know how effective different ways of silencing are?
> (Yes, I know: get silent power supply, fans and hd ;)
>
> For example, how much I can expect to gain from using sound
> absorption
> lining? Or making sure the harddrive doesn't cause case to vibrate?
>
> Sami
>
>
>
> --------------------------------------------------------------
> ----------
> Life's too short to send boring email. Let SuperSig come to
> the rescue.
> http://click.egroups.com/1/6137/1/_/_/_/963395981/
> --------------------------------------------------------------
> ----------
>
> To unsubscribe from this group, send an email to:
> Silent-PC-unsubscribe@egroups.com
>
>
>
Hi,
Does anyone know how effective different ways of silencing are?
(Yes, I know: get silent power supply, fans and hd ;)
For example, how much I can expect to gain from using sound
absorption
lining? Or making sure the harddrive doesn't cause case to vibrate?
Sami
Terry,
I see one problem with the Fong Kai approach. They use the duct to
pull out air. So they really aren't bringing the cold (relatively)
air
from outside to the CPU. I think the temperature difference makes
quite a difference.
I'm thinking about getting a system with duron CPU and newest graphic
card, so the heat is going to be a problem if i want to keep it quiet.
Sami
--- In Silent-PC@egroups.com, Terry Gray <gray+silent@w...> wrote:
> Sami,
> The latest ATX power supply spec is designed to encourage single-fan
> solutions. I've only seen one hobbyist case that takes advantage
of
it,
> however. There is a Fong Kai case where the power supply sits on a
> bracket that is hinged above the CPU location on the motherboard.
The
> intent is to have the power supply fan pull air into the power
supply
> housing (and eventually out the back of the case) from directly
above the
> CPU heatsink, thereby generating enough airflow around the cpu to
keep it
> cool without needing its own fan.
>
> Obviously whether this works depends on how much heat your cpu
dissipates.
>
> Fong Kai makes another case where there is a plastic fan duct
pulling air
> from above the CPU out the back of the case, to either supplement or
> replace the cpu heatsink fan... again depending on how much
power/heat
> you're dealing with.
>
> -teg
>
> On Tue, 11 Jul 2000 sami.rantala@e... wrote:
>
On Thu, 6 Jul 2000, Thomas Grov wrote:
> >I'm interested in taking such a disk, and a 100W convection-cooled p/s,
> >and maybe a celeron 566 underclocked to 500, with the best passive heat
> >sink I can find, and seeing what kind of temps result.
>
> Very interesting project, but I don't think the 566 Celeron can be
> underclocked because of the clocklock. Only bus speed can be altered, and
> it's already at 66Mhz.
>
> Anyone know differently ?
Thomas,
I'm not sure (though lots of gamers like the Celery 566 because it is
apparently easy to *over*clock it...) however, looking at the reference
Glenn Garrett provided:
http://users.erols.com/chare/elec.htm
it appears that the celeron 566 is *the* choice for a fan-free system,
even if you can't *under*clock it.
According to the data at that URL, the C II 566 uses only a little over
half the power dissipation of any other contender. Even the new Duron uses
nearly twice the power (but it also has considerably higher performance.)
Note that the 500MHz and below Celerons were made with the .25 micron
process and they consume twice as much power as the 566 which is made
using the .18 micron process. 533s were made on both lines, so you have
to look for a 533A to get the .18 version.
-teg
Sami,
The latest ATX power supply spec is designed to encourage single-fan
solutions. I've only seen one hobbyist case that takes advantage of it,
however. There is a Fong Kai case where the power supply sits on a
bracket that is hinged above the CPU location on the motherboard. The
intent is to have the power supply fan pull air into the power supply
housing (and eventually out the back of the case) from directly above the
CPU heatsink, thereby generating enough airflow around the cpu to keep it
cool without needing its own fan.
Obviously whether this works depends on how much heat your cpu dissipates.
Fong Kai makes another case where there is a plastic fan duct pulling air
from above the CPU out the back of the case, to either supplement or
replace the cpu heatsink fan... again depending on how much power/heat
you're dealing with.
-teg
On Tue, 11 Jul 2000 sami.rantala@... wrote:
> Hi,
> Here's a thought about fan noise.
>
> We have this small fast and probably noisy fan on cpu. And we also
> have often extra case fan. Now can't we combine these two and get an
> large silent fan, which pushes loads of air and guide the air flow to
> the heat spots with pipes or ducts. This fan could be even downgraded
> a bit (lower voltage).
>
> This fan could be as large as 120mm. I realise it wouldn't fit ready
> made places in cases, but we can always make more holes to the case.
> Extra bonus with this approach is that the air is cooler outside the
> case so if the duct is properly made it would mean that CPU gets more
> and cooler air at lower noise levels.
>
> Sami Rantala
Hi,
Here's a thought about fan noise.
We have this small fast and probably noisy fan on cpu. And we also
have often extra case fan. Now can't we combine these two and get an
large silent fan, which pushes loads of air and guide the air flow to
the heat spots with pipes or ducts. This fan could be even downgraded
a bit (lower voltage).
This fan could be as large as 120mm. I realise it wouldn't fit ready
made places in cases, but we can always make more holes to the case.
Extra bonus with this approach is that the air is cooler outside the
case so if the duct is properly made it would mean that CPU gets more
and cooler air at lower noise levels.
Sami Rantala
> Are you sure you are not able to buy a Decibel computer from us in Sweden?
> They have a web page in English, so I guess they also want customers
> outside of Sweden.
Yes, you were right, it is possible to order their PCs from other countries.
Thanks for the suggestion!
Regards,
Hrvoje Nezic
>
> Tomas Risberg
>I'm interested in taking such a disk, and a 100W convection-cooled p/s,
>and maybe a celeron 566 underclocked to 500, with the best passive heat
>sink I can find, and seeing what kind of temps result.
Very interesting project, but I don't think the 566 Celeron can be
underclocked because of the clocklock. Only bus speed can be altered, and
it's already at 66Mhz.
Anyone know differently ?
Regards,
Thomas Grov
On Mon, 3 Jul 2000, Glenn Garrett wrote:
> So far, power supplies without fans have yet to make an appearance in the
> world of PCs, simply because the heat they generate requires active air
> extraction. PSUs without fans are available, but they are only rated at low
> wattages (around 100W), which makes them impractical for most PC
> applications.
Glenn,
I'm curious about this statement.
Unless you are heavily into games or other multimedia, I would claim that
a 100W power supply is more than adequate for *most* PC applications.
Consider:
o The iMac DVD was recently reengineered to be completely fan-free, and
while the PowerPC allegedly uses less power than either Pentiums or
Athlons, the Apple engineers had to contend with the heat from the CRT
as well.
o The new HP e-vectra (with I think a celeron 566) claims max power
use of 40 watts (alas, it is not completely fan free).
o Even desktop replacement laptops use less (albeit with mobile cpus, but
then they also power the display.
o I have a K6 300MHz system that uses less than 60W.
Obviously you're not going to run a pair of Voodoo cards and a bunch of
10,000 rpm disks off of a 100W supply, and I've heard that Athlon
processors are particularly power hungary, but I believe that if you are
willing to sacrifice some performance, you *can* build a fan-free system.
Also, Quantum has recently introduced some low-noise, low-power --and
somewhat low performance-- disks (LCT15, not to be confused with the LCT10
series).
I'm interested in taking such a disk, and a 100W convection-cooled p/s,
and maybe a celeron 566 underclocked to 500, with the best passive heat
sink I can find, and seeing what kind of temps result.
If anyone has any thoughts on that, let me know...
-teg
Thanks for all the help.
I have abandoned the thought of making a fanless pc, but instead intend to
use as quiet parts as possible (and since the computer is for my brother
and not for me, it's a suitable guinea pig :). My current plan looks like
this:
Case:
Addtronics Miditower 3891A
Powersupply:
Channel Well Technology 235, 250 or 300 watt
Motherboard:
DFI K6XV3+/66 (I have another DFI board which has the nice function of
turning off the cpu fan when it is below 25C)
Harddrive:
Fujitsu IDE ATA-66 15,3GB 5400RPM (MPF3153AT-T 512 KB cache)
CPU:
AMD K6-2 533 (Any idea where I might buy the mobile K6-2 ?)
CPU fan:
Noise Control Socket 7, 12db(A)
Video card:
Matrox G200
Case fan (which i might remove if not needed):
Papst 8412NGL, 33m3/hour, 80x80mm, 12db(A)
CD-ROM:
Undecided yet. But probably a Pioneer, since I am very happy with my 24x
and 32x Pioneer.
Any thoughts on how powerful powersupply I should buy ? Are the powerful
ones also more noisy ?
Regards,
Thomas Grov
At 09:44 2000-06-30 +0200, Hrvoje Nezic wrote:
>This is really impressing, but unfortunately, Whisper and Decibel computers
>are available only in Sweden. (Oh, you lucky Swedes :) ).
As I have got it the Whisper system sadly isn't available any more, since
the Lap Power company has been put in to liquidation. Hopefully someone
wants to lincense their solutions.
Are you sure you are not able to buy a Decibel computer from us in Sweden?
They have a web page in English, so I guess they also want customers
outside of Sweden.
Tomas Risberg
Hi Thomas
I'm also striving toward a silent PC but as long as you've got a hard drive
it's never going to happen! The only way to achieve total silence is to have
no moving parts...sadly both hard drives and fans have moving parts!
Your approach to the CPU is very good if I may say so. If you haven't
already found it, a full list of CPUs and the power they take is available
at this web address:
http://users.erols.com/chare/elec.htm
Sadly I don't have much optimism that a 100% passive cooler is going to be
an option. The fastest CPU I've ever seen with a passive cooler (and no
other cooling, from say a ducted case fan) is a Pentium 166, (rated at 14.5W
max). I don't know what OS you are planning to use, but unless it's Windows
3.11 or earlier, a PC based on that processor is going to be slow!
The PC I'm writing this on has an Athlon processor which I'm underclocking
from 600MHz to 550MHz, and also running at a lower core voltage (1.40v
rather than 1.60v). This makes it run *much* cooler, and with an efficient
heatsink (Thermaltake Golden Orb), I can run a fan very slowly and the
processor stays just about cool enough - around 53 degrees C with the PC
case on.
The key to successful experimentation is to use a motherboard which supports
temperature sensing, and keep an eye on it! Programs such as Motherboard
Monitor will sit in your Windows system tray and give you a constant readout
of the CPU temperature, which is invaluable when you are first trying
different things.
http://members.brabant.chello.nl/~a.vankaam/mbm/
If the motherboard also allows you to over/underclock then so much the
better. In practice you will not notice a performance decrease by slightly
underclocking most modern CPUs, unless you use the PC for highly graphical
games (which normally have loud sound effects anyway, so you probably won't
mind if the PC generates noise!).
Quiet hard drives are essential if the PC is to run quietly, and they can
further be quietened by putting them in an enclosure (Molex make a product
called the SilentDrive which is designed for this task).
So far, power supplies without fans have yet to make an appearance in the
world of PCs, simply because the heat they generate requires active air
extraction. PSUs without fans are available, but they are only rated at low
wattages (around 100W), which makes them impractical for most PC
applications.
My company specialises in noise reduction solutions for PCs, but I must
emphasise that to date they are just that - noise reduction - not noise
elimination! As part of my work, I am constantly researching ways to make
PCs quieter, whether these comprise techniques, products or both, so I
greatly look forward to other list members' comments. I hope I will be able
to contribute and help wherever I can.
http://www.quietpc.com/
Looking on the bright side though, the amount of PC noise which can be
(relatively easily) eliminated is very high, and although not 100% silent,
it is certainly better to have a quiet computer than one in which no effort
has been made to reduce the noise.
I think the tide is turning and the challenge of making PCs run quietly is
becoming more interesting and appealing then the traditional aim of making
them run flat out as fast as possible. It's also kinder on the wallet!
Kind regards
Glenn
--
Glenn Garrett http://www.quietpc.com/glenn@... PC noise elimination
> -----Original Message-----
> From: Thomas Grov [mailto:aeon2@...]
> Sent: Monday, 03 July 2000 07:56
> To: Silent-PC@egroups.com
> Subject: [Silent-PC] Building a quiet computer
>
> This is my first post to the list.
>
> I am building a silent computer for my brother sometime
> within a month or
> so (Project 'Shhh') and thought I would post my thoughts so far:
>
> I am thinking that if I got a cpy that produces little heat,
> and use a
> large heatsink (with thermal compound, and perhaps lapping
> the cpu) I would
> not need a cpu fan. In a midi/tower case should the power
> supply push air
> out or suck it in on the cpu heatsink if I don't use a cpu fan ?
>
> The cpu's I am considering are:
>
> AMD K6-2 500, 2.2V, 20.75W
> AMD K6-2 533, 2.2V, 20.75W
> AMD Duron 600, 1.5V, 22.7W
> Intel Celeron 500, 2.0V, 27W
> Intel Celeron 533, 2.0V, 28,3W
>
> I realise that performance will be superior on the Duron, but
> since this is
> a early production run, I am not sure I will go for it, but am rather
> leaning towards the K6-2.
>
> If it runs to hot to be fanless (even after underclocking if
> necessary), I
> will but a quiet 12dba cpu fan instead.
>
> I am also buying a super quiet Fujitsu harddrive (5400 udma/66).
>
>
> Any thoughts on my plan so far ? Are there any cases you recommend ?
>
>
> Regards,
>
> Thomas
>
>
> --------------------------------------------------------------
> ----------
> Need a credit card?
> Instant Approval and 0% intro APR with Aria!
> http://click.egroups.com/1/6034/1/_/_/_/962607343/
> --------------------------------------------------------------
> ----------
>
> To unsubscribe from this group, send an email to:
> Silent-PC-unsubscribe@egroups.com
>
>
>
Good questions Thomas, your idea is a sound one but needs your to experiment
cautiously to get things right. I've seen most of the Intel Slot chips
running without fans in Compaq's recently departed Deskpro EP series. The
key thing Compaq used was a big roomy case and a massive aluminium plate
heat sink that projected above the slot in a vicious looking 'antler' series
of fins.
If you're after a roomy desktop case then the field is really limited. I
found getting 3x 5.1/4 bays difficult and ended up with a Vasco model. (Call
me old fashioned but I like my motherboard to lie horizontal, just in case I
need to whip the cover off and prod the inside.)
If you're after a tower case then avoid the 'midi' size ones which are hot
and cramped. If you can look for a wider 'server' style case which often
have more 'headroom' from the chip's point of view.
As regards chips the coolest are 0.18 micron designs at the lowest clock
speeds (good news for the wallet too). The Intel flip chip may have an edge
as the silicon is topside and directly cooled by the fan.
Finally when you get your chip and heatsink together make sure your
motherboard's BIOS has temperature monitoring either in BIOS or through a
Windows utility (I think Linux has one too). You can then power up and
observe the rising temperature and check it is within limits. You can then
satisfy yourself that the chip is not cooking. If things are too hot then
consider one of the new radial fans that run at lower speeds (hence less
noise).
Allan Toombs
organic_acid@...
'Let it go Louie, let it go.'
This is my first post to the list.
I am building a silent computer for my brother sometime within a month or
so (Project 'Shhh') and thought I would post my thoughts so far:
I am thinking that if I got a cpy that produces little heat, and use a
large heatsink (with thermal compound, and perhaps lapping the cpu) I would
not need a cpu fan. In a midi/tower case should the power supply push air
out or suck it in on the cpu heatsink if I don't use a cpu fan ?
The cpu's I am considering are:
AMD K6-2 500, 2.2V, 20.75W
AMD K6-2 533, 2.2V, 20.75W
AMD Duron 600, 1.5V, 22.7W
Intel Celeron 500, 2.0V, 27W
Intel Celeron 533, 2.0V, 28,3W
I realise that performance will be superior on the Duron, but since this is
a early production run, I am not sure I will go for it, but am rather
leaning towards the K6-2.
If it runs to hot to be fanless (even after underclocking if necessary), I
will but a quiet 12dba cpu fan instead.
I am also buying a super quiet Fujitsu harddrive (5400 udma/66).
Any thoughts on my plan so far ? Are there any cases you recommend ?
Regards,
Thomas
[received the following message from Vicky Parra of Noise Pollution
Clearinghouse <npc@...> about an M.I.T. site on Quiet and
Silent PC's. -DS]
-----Original Message-----
From: npc@... [mailto:npc@...]
Sent: Tue, 27 Jun 2000 13:51:07 -0400
To: quiet@...
Subject: quiet computers
> I've recently corresponded with an MIT student (Piotr Mitros,
> pmitros@...) who has the best information I've seen yet posted on
> his webpage, dealing with both premade PC's and modifying existing noisy
> parts or systems: http://pmitros.mit.edu/silence.html
>
> Just wanted to let you know in case you're interested!
>
> Take care,
>
> Vicky Parra, NPC
> These are interesting facts, but it's not that impressing.
>
> Both the LapPower Whisper System and the recently introduced CapTech
> Decibel are much more impressing regarding noise emissions. Whisper with a
> maximimum of about 24 dBA and now Decibel with 16.5 dBA noise. One has also
> to remember that an increase with about 6-10 dBA counts for the impression
> of doubling the noise emission.
This is really impressing, but unfortunately, Whisper and Decibel computers are
available only in Sweden. (Oh, you lucky Swedes :) ).
> How where Fujitsu-Siemens tests performed? Did they use the ISO 7779 test
> standard?
I don't know, perhaps there is more information on the PC Professional website,
but I don't speak German.
Regards,
Hrvoje Nezic
At 10:20 2000-06-28 +0200, you wrote:
>For the comparative test, the German-Japanese computer giant submitted a
design
>solution with the emphasis on good ergonomics, in the shape of its brand-new
>Scenic xS. In the noise tests, the machine posted by far the lowest values of
>all the candidates, notching up a sensationally low noise level of 25.8 dB(A)
>and a very good rating of 30.4 dB(A) during operation.
These are interesting facts, but it's not that impressing.
Both the LapPower Whisper System and the recently introduced CapTech
Decibel are much more impressing regarding noise emissions. Whisper with a
maximimum of about 24 dBA and now Decibel with 16.5 dBA noise. One has also
to remember that an increase with about 6-10 dBA counts for the impression
of doubling the noise emission.
How where Fujitsu-Siemens tests performed? Did they use the ISO 7779 test
standard?
What's interesting though is the interest in these matters from such a big
company as Fujitsu-Siemens.
Tomas Risberg
Scenic xS
For the comparative test, the German-Japanese computer giant submitted a design
solution with the emphasis on good ergonomics, in the shape of its brand-new
Scenic xS. In the noise tests, the machine posted by far the lowest values of
all the candidates, notching up a sensationally low noise level of 25.8 dB(A)
and a very good rating of 30.4 dB(A) during operation. Not only did this
performance gain it the "Editor's Choice" accolade, but also meant it produced
just half the sound level registered by the noisiest test candidate. The
time-based noise emission curve was impressive too, demonstrating a high degree
of uniformity.
These top-class results are the fruits of a carefully thought-out case, designed
for optimum airflow, plus an electronically regulated fan, the speed of which is
continuously optimized by means of sensors: During less compute-intensive
applications, the fan shuts off altogether. According to the manufacturer, the
hard disk installed in the test machine is to be replaced later on by
fluid-bearing models from its own MPE/AT-series, with storage capacities
starting from 10 GB, which promise to be even quieter. The 3300-deutschmark
machine, which boasts an Intel PIII/667 and 17-inch monitor, takes a less
revolutionary approach with its case design:
The top of the Scenic-xS Desktop can rapidly be opened, without the need for
tools, by pressing a button on the back of the case, providing simple access to
the components. The drives and the hard disks are bolted in position, and
removing the motherboard demands a certain amount of effort.
Fujitsu Siemens also puts a high priority on security, an important aspect for
workstation machines. Like most of its rivals, the desktop can be secured by
means of a case lock, but one really noteworthy feature is the smartcard reader
integrated into the front panel - something that none of the other test machines
offer as standard
Ergonomics 70 % 5 points (maximum)
Service 15% 4 points
Equipment 10% 4 points
Performance 5 % 3 points
PC Professional 5/2000
Dear list members,
This is the first message on the Silent-PC mailing list.
After I created this list I was waiting some time to allow
people to join. Every now and than I receive notifications
about new people joining the list, and this makes me happy
because it proves there was a need for such a discussion forum.
Let me say just a few words about myself (and you are encouraged
to do the same).
I am a software engineer and thus I spend a lot of time working
on PC, but noise really bothers me and makes me tired, so I am
naturally very interested in silent PCs.
I am not a hardware man, so I am much more inclined to buy a silent
PC than to make it myself. Right now I intend to buy one of Siemens
silent PCs: SCENIC 620 or SCENIC xS-1171. Perhaps someone
on this list has some experience with these PCs?
Best regards,
Hrvoje Nezic