Re: AHHHH, NICE (quiet) PC FOR $47

Discussion in 'NZ Computing' started by Craig Sutton, Aug 11, 2003.

  1. Craig Sutton

    Craig Sutton Guest


    > (**) These slide back about 4mm, then lift away.
    >
    > Google on "quiet PC initiative", there are a number of tricks you can use,
    > but the best way to start is to optimise noise on the fans or use CPUs
    > which don't need noisy fans in the first place.
    >

    I have a p3-450 with stock fan I want to lower the noise it makes any
    quickfix? hopefully a $0 DIY one?
     
    Craig Sutton, Aug 11, 2003
    #1
    1. Advertising

  2. Craig Sutton

    Evil Bastard Guest

    On Mon, 11 Aug 2003 12:48:42 +0000, Uncle StoatWarbler wrote:

    > On Mon, 11 Aug 2003 18:02:01 +1200, Evil Bastard wrote:
    >
    >> Quite simply, a wooden case enclosure and lined it with Melamine
    >> Acoustic Ceiling Foam.

    >
    > You can get the same effect by sticking acoustic foam/matting on the
    > inside of your case panels(*) and mounting the hard drive screws through
    > grommets.


    Doubt if it would be quite the same, because of (1) the psu fan,
    and (2) the emanation of sound out the back.

    With my enclosures (yes, I've got the workstation running whisper-quiet
    too now - I had just enough foam left over), they overhang the back,
    creating a kind of 'tunnel' of about 10cm, which sucks quite a lot of the
    PSU fan.

    > Just bear in mind that case manufacturers expect to lose 4-5C via the
    > case, so the fans have to work harder or the innards will get warmer.


    No problem - they'll still be pretty quiet.

    > Google on "quiet PC initiative", there are a number of tricks you can use,
    > but the best way to start is to optimise noise on the fans or use CPUs
    > which don't need noisy fans in the first place.


    You're referring to the somewhat noisier AMD CPUs, which I am using.

    But CPU fan is not the only noise source. Don't forget about disk drives,
    video card fan and PSU fan.

    Anyway, I've now got the workstation and server sitting in enclosures,
    together making about 1/3 of the noise (by human perception) than the
    server alone was making without the enclosure - and the server was the
    quieter of the two.

    Conclusion - The enclosure as described gives the best silencing for the
    buck. I have to focus my attention to be able to hear any computer noise
    now.

    I just wish I'd done it years ago - I'm sure the constant barrage of
    noise has not been good for my concentration.

    Cheers
    EB
     
    Evil Bastard, Aug 11, 2003
    #2
    1. Advertising

  3. Craig Sutton

    Craig Sutton Guest

    "Uncle StoatWarbler" <> wrote in message
    news:p...
    > On Mon, 11 Aug 2003 22:57:40 +1200, Craig Sutton wrote:
    >
    > > I have a p3-450 with stock fan I want to lower the noise it makes any
    > > quickfix? hopefully a $0 DIY one?

    >
    > Easy fix #1: Get a bigger, slower fan. It'll still move just as much air.
    >
    > Blade tip noise is the key. faster fans make more noise .noise level is
    > proportional to the cube of the blade speed, while air moved is linearly
    > proportional.


    I am thinking just pulling the cpu out of the socket cleaning the dust and
    crap out of it that should improve things.
    Maybe i should over clock it up to 540mhz and it will blow the crap out on
    its own.
     
    Craig Sutton, Aug 11, 2003
    #3
  4. Thus spake Craig Sutton:
    > I am thinking just pulling the cpu out of the socket cleaning the dust and
    > crap out of it that should improve things.


    That'll improve cooling but most CPU fans aren't temperature sensitive
    so it won't help noise.

    Cheap solution, whack a small resistor in the +12V. YOu can generally
    drop the noise by at least half.

    Experiment a little after deciding the max temp you're happy with. I'm
    happy with 50 C idle.

    --
    aaronl at consultant dot com
    http://homepages.visp.co.nz/~aaronlawrence
    ...Gross Ignorance: 144 times worse than ordinary ignorance.
     
    Aaron Lawrence, Aug 11, 2003
    #4
  5. Thus spake Uncle StoatWarbler:
    > No, I'm referring to things like the Joshua/Jerico CPUs, which dn't need a
    > fan at all.


    The what??

    I heard of the VIA C3...
    --
    aaronl at consultant dot com
    http://homepages.visp.co.nz/~aaronlawrence
    ...Gross Ignorance: 144 times worse than ordinary ignorance.
     
    Aaron Lawrence, Aug 11, 2003
    #5
  6. On Mon, 11 Aug 2003 18:02:01 +1200, Evil Bastard wrote:

    > Quite simply, a wooden case enclosure and lined it with Melamine
    > Acoustic Ceiling Foam.


    You can get the same effect by sticking acoustic foam/matting on the
    inside of your case panels(*) and mounting the hard drive screws through
    grommets.

    Just bear in mind that case manufacturers expect to lose 4-5C via the
    case, so the fans have to work harder or the innards will get warmer.


    (*) Assumes individually removable Aopen style panels(**), of course. You
    can't do this with one piece covers or the foam gets trashed as you slide
    them on/off.

    (**) These slide back about 4mm, then lift away.

    Google on "quiet PC initiative", there are a number of tricks you can use,
    but the best way to start is to optimise noise on the fans or use CPUs
    which don't need noisy fans in the first place.
     
    Uncle StoatWarbler, Aug 11, 2003
    #6
  7. On Mon, 11 Aug 2003 22:57:40 +1200, Craig Sutton wrote:

    > I have a p3-450 with stock fan I want to lower the noise it makes any
    > quickfix? hopefully a $0 DIY one?


    Easy fix #1: Get a bigger, slower fan. It'll still move just as much air.

    Blade tip noise is the key. faster fans make more noise .noise level is
    proportional to the cube of the blade speed, while air moved is linearly
    proportional.

    Good cases will have a well defined airflow which allows air to be drawn
    in and across hot components, then out the back. Most PC cases have no
    thought given to airflow at all and the paths taken are effectively random
    - which means more (noisy) auxilliary fans stirring things up over local
    hotspots.
     
    Uncle StoatWarbler, Aug 11, 2003
    #7
  8. On Mon, 11 Aug 2003 23:03:31 +1200, Evil Bastard wrote:

    > With my enclosures (yes, I've got the workstation running whisper-quiet
    > too now - I had just enough foam left over), they overhang the back,
    > creating a kind of 'tunnel' of about 10cm, which sucks quite a lot of the
    > PSU fan.


    There are a number of "silent fan" PSUs now, with variable speeds, etc.

    >> Google on "quiet PC initiative", there are a number of tricks you can use,
    >> but the best way to start is to optimise noise on the fans or use CPUs
    >> which don't need noisy fans in the first place.

    >
    > You're referring to the somewhat noisier AMD CPUs, which I am using.


    No, I'm referring to things like the Joshua/Jerico CPUs, which dn't need a
    fan at all.

    > But CPU fan is not the only noise source. Don't forget about disk drives,


    Disk drive noise is combatted by going to 7200rpm (yes, seriously!), and
    using acoustic isolation mounts. Almost all the noise you hear is
    amplified by the chassis acting as a sounding board.

    > video card fan


    Bigger, slower fans. Get 120MM (or larger) case fans, run them on 5V instead of 12V
    and they're almost inaudible. For video card fans, use 80mm fans at 5V, etc.

    > and PSU fan.


    See comment above. Check out Enermax EG365AX-VE (W) FCA series and related PSUs

    > Conclusion - The enclosure as described gives the best silencing for the
    > buck. I have to focus my attention to be able to hear any computer noise
    > now.


    You've done it your way, but that's not suitable for a lot of home users
    or offices, especially seeing as putting a PC on the floor is the worst
    possible location you can imagine (under desk is the absolute pits, lots
    of stirred up dust every time someone moves a chair).

    Noise _is_ an issue, people are starting to realise this with home
    theatres, etc and other things requiring PCs in bedrooms or lounges.
     
    Uncle StoatWarbler, Aug 11, 2003
    #8
  9. Craig Sutton

    T.N.O Guest

    "Aaron Lawrence" wrote
    | Experiment a little after deciding the max temp you're happy with. I'm
    | happy with 50 C idle.

    Really? wow, I wouldn't be... My cpus are maxed 24/7 and I wouldnt be happy
    with them going over 60 ever... although they have reached 65 on hot days,
    but then I just turn the fans up(rheostat)
     
    T.N.O, Aug 12, 2003
    #9
  10. Craig Sutton

    Murray Symon Guest

    On Mon, 11 Aug 2003 23:56:33 +1200, Aaron Lawrence wrote:

    > Thus spake Craig Sutton:
    >> I am thinking just pulling the cpu out of the socket cleaning the dust
    >> and crap out of it that should improve things.

    >
    > That'll improve cooling but most CPU fans aren't temperature sensitive so
    > it won't help noise.
    >
    > Cheap solution, whack a small resistor in the +12V. YOu can generally drop
    > the noise by at least half.
    >
    > Experiment a little after deciding the max temp you're happy with. I'm
    > happy with 50 C idle.


    In my experience a series resistor will often not allow the fan to start
    due to startup current draw - a zener diode or voltage regulator will work
    better. The last thing you want is to throttle it too much so that there
    is a risk of it not starting some day.

    Murray
     
    Murray Symon, Aug 12, 2003
    #10
  11. "Craig Sutton" <> wrote in message
    news:bh7sr9$66c$...
    >
    >
    > > (**) These slide back about 4mm, then lift away.
    > >
    > > Google on "quiet PC initiative", there are a number of tricks you can

    use,
    > > but the best way to start is to optimise noise on the fans or use CPUs
    > > which don't need noisy fans in the first place.
    > >

    > I have a p3-450 with stock fan I want to lower the noise it makes any
    > quickfix? hopefully a $0 DIY one?
    >


    Switch it off ;-)
     
    Fred the Fish, Aug 12, 2003
    #11
  12. Thus spake Uncle StoatWarbler:
    > On Mon, 11 Aug 2003 23:56:33 +1200, Aaron Lawrence wrote:
    >
    > > Cheap solution, whack a small resistor in the +12V. YOu can generally
    > > drop the noise by at least half.

    >
    > That adds more heat inside the case. Powering the fan off 5V is easier.


    Not much - a couple of watts. With CPU's putting out 20-100W it's not
    significant.

    --
    aaronl at consultant dot com
    http://homepages.visp.co.nz/~aaronlawrence
    ...Gross Ignorance: 144 times worse than ordinary ignorance.
     
    Aaron Lawrence, Aug 12, 2003
    #12
  13. Thus spake Murray Symon:
    > In my experience a series resistor will often not allow the fan to start
    > due to startup current draw - a zener diode or voltage regulator will work
    > better. The last thing you want is to throttle it too much so that there
    > is a risk of it not starting some day.


    True, you can't go down to the minimum that will run. You have to go
    back up seeing where it consistently starts, then go a bit further than
    that for safety.

    But the best bet would be a large parallel capacitor so that it gets a
    big boost at startup.

    --
    aaronl at consultant dot com
    http://homepages.visp.co.nz/~aaronlawrence
    ...Gross Ignorance: 144 times worse than ordinary ignorance.
     
    Aaron Lawrence, Aug 12, 2003
    #13
  14. Thus spake T.N.O:
    > Really? wow, I wouldn't be... My cpus are maxed 24/7 and I wouldnt be happy
    > with them going over 60 ever...


    Well mine don't go over 56...

    They are rated to better than 80C

    --
    aaronl at consultant dot com
    http://homepages.visp.co.nz/~aaronlawrence
    ...Gross Ignorance: 144 times worse than ordinary ignorance.
     
    Aaron Lawrence, Aug 12, 2003
    #14
  15. On Tue, 12 Aug 2003 17:56:22 +1200, pete wrote:

    > They run really
    > hot - a
    > 3M laser thermometer records 38 dec C at the top centre of the power supply
    > after 20 minutes powered on, in a room with ambient temp of 19 deg C.


    Trust me: 38C is _not_ hot in a PSU.
     
    Uncle StoatWarbler, Aug 12, 2003
    #15
  16. "M" <> wrote in message
    news:t4b_a.11019$...
    >
    > "Uncle StoatWarbler" <> wrote in message
    > news:p...
    > > On Wed, 13 Aug 2003 01:58:40 +1200, Aaron Lawrence wrote:
    > >
    > > >> That adds more heat inside the case. Powering the fan off 5V is

    easier.
    > > >
    > > > Not much - a couple of watts. With CPU's putting out 20-100W it's not
    > > > significant.

    > >
    > > Every bit helps. You're not just adding extra heat in the case, you're

    > also
    > > working the PSU harder.
    > >
    > > This gets kinda hammered in when you're doign system rooms. If you save
    > > 1kW in system power consumption, you save as much again in air

    conditioner
    > > bills and more still in maintenance costs.
    > >

    >
    > Let's see now,
    >
    > Yer average fan pulls
    >
    > 12V * 100mA = 1.2W
    >
    > Let's just say that we stick a resistor in there that drops the current by
    > half
    >
    > 12V * 50mA = 0.6W
    >
    > That looks like less energy dissipation inside the case and a lower load

    on
    > the PSU to me.
    >
    > The temperature will obviously rise due to the lower airflow.
    >
    > Please don't design my machine room.


    Resistors work by converting that extra energy to heat, it doesn't just
    disappear. The total draw would still be 1.2W, 0.6W of which is converted
    into heat and 0.6W of which goes to the fan.

    Cheers,
    Nicholas Sherlock
     
    Nicholas Sherlock, Aug 12, 2003
    #16
  17. On Tue, 12 Aug 2003 20:00:54 +1200, T.N.O wrote:

    > "Uncle StoatWarbler" wrote
    > | That adds more heat inside the case. Powering the fan off 5V is easier.
    >
    > not all fans will start on 5v... I have found 7 to work 99% of the time.


    I've only found old fans do that - and I did test the ones I bought before
    deploying 'em.

    Get good fans, like Papst, etc. They're worth the extra bucks.
     
    Uncle StoatWarbler, Aug 12, 2003
    #17
  18. Craig Sutton

    M Guest

    "Nicholas Sherlock" <> wrote in message
    news:bhbggd$pn2$...
    > "M" <> wrote in message
    > news:t4b_a.11019$...
    > >
    > > "Uncle StoatWarbler" <> wrote in message
    > > news:p...
    > > > On Wed, 13 Aug 2003 01:58:40 +1200, Aaron Lawrence wrote:
    > > >
    > > > >> That adds more heat inside the case. Powering the fan off 5V is

    > easier.
    > > > >
    > > > > Not much - a couple of watts. With CPU's putting out 20-100W it's

    not
    > > > > significant.
    > > >
    > > > Every bit helps. You're not just adding extra heat in the case, you're

    > > also
    > > > working the PSU harder.
    > > >
    > > > This gets kinda hammered in when you're doign system rooms. If you

    save
    > > > 1kW in system power consumption, you save as much again in air

    > conditioner
    > > > bills and more still in maintenance costs.
    > > >

    > >
    > > Let's see now,
    > >
    > > Yer average fan pulls
    > >
    > > 12V * 100mA = 1.2W
    > >
    > > Let's just say that we stick a resistor in there that drops the current

    by
    > > half
    > >
    > > 12V * 50mA = 0.6W
    > >
    > > That looks like less energy dissipation inside the case and a lower load

    > on
    > > the PSU to me.
    > >
    > > The temperature will obviously rise due to the lower airflow.
    > >
    > > Please don't design my machine room.

    >
    > Resistors work by converting that extra energy to heat, it doesn't just
    > disappear. The total draw would still be 1.2W, 0.6W of which is converted
    > into heat and 0.6W of which goes to the fan.
    >
    > Cheers,
    > Nicholas Sherlock
    >


    Is that the Microsoft theory of electronics or have you been taking lessons
    from someone in Raumati ?

    When you put a resistor in series with something you increase the total
    resistance of the load and thus reduce the current and power.

    Yes, the resistor does dissipate heat. Let's assume for the sake of clarity
    that the resistor drops the voltage across the fan to 6V. This makes the pow
    er dissipation in the resistor 6V * 0.05A = 0.3 W, the same as that of the
    fan, making a total of 0.6W.

    M
     
    M, Aug 12, 2003
    #18
  19. Craig Sutton

    lily Guest

    Nicholas Sherlock wrote:

    > "M" <> wrote in message
    > news:t4b_a.11019$...
    >
    >>"Uncle StoatWarbler" <> wrote in message
    >>news:p...
    >>
    >>>On Wed, 13 Aug 2003 01:58:40 +1200, Aaron Lawrence wrote:
    >>>
    >>>
    >>>>>That adds more heat inside the case. Powering the fan off 5V is

    >
    > easier.
    >
    >>>>Not much - a couple of watts. With CPU's putting out 20-100W it's not
    >>>>significant.
    >>>
    >>>Every bit helps. You're not just adding extra heat in the case, you're

    >>
    >>also
    >>
    >>>working the PSU harder.
    >>>
    >>>This gets kinda hammered in when you're doign system rooms. If you save
    >>>1kW in system power consumption, you save as much again in air

    >
    > conditioner
    >
    >>>bills and more still in maintenance costs.
    >>>

    >>
    >>Let's see now,
    >>
    >>Yer average fan pulls
    >>
    >> 12V * 100mA = 1.2W
    >>
    >>Let's just say that we stick a resistor in there that drops the current by
    >>half
    >>
    >> 12V * 50mA = 0.6W
    >>
    >>That looks like less energy dissipation inside the case and a lower load

    >
    > on
    >
    >>the PSU to me.
    >>
    >>The temperature will obviously rise due to the lower airflow.
    >>
    >>Please don't design my machine room.

    >
    >
    > Resistors work by converting that extra energy to heat, it doesn't just
    > disappear. The total draw would still be 1.2W, 0.6W of which is converted
    > into heat and 0.6W of which goes to the fan.
    >
    > Cheers,
    > Nicholas Sherlock
    >
    >

    Only if it is a parallel (shunt) resistor.
     
    lily, Aug 15, 2003
    #19
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Phil-on-a-hill

    Need a nice quiet PSU

    Phil-on-a-hill, Aug 9, 2003, in forum: Computer Support
    Replies:
    2
    Views:
    414
    Jimmy Dean
    Aug 10, 2003
  2. colinco

    Re: AHHHH, NICE (quiet) PC FOR $47

    colinco, Aug 11, 2003, in forum: NZ Computing
    Replies:
    0
    Views:
    460
    colinco
    Aug 11, 2003
  3. T.N.O

    Re: AHHHH, NICE (quiet) PC FOR $47

    T.N.O, Aug 11, 2003, in forum: NZ Computing
    Replies:
    3
    Views:
    299
    Evil Bastard
    Aug 12, 2003
  4. Giuen
    Replies:
    0
    Views:
    1,509
    Giuen
    Sep 12, 2008
  5. Replies:
    0
    Views:
    472
Loading...

Share This Page