What About Watts?

Discussion in 'NZ Computing' started by Tony McKee, Apr 15, 2006.

  1. Tony McKee

    Tony McKee Guest

    Yo, Folks.

    Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.

    http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502

    Does that 500 watts mean that as soon as you plug it into the wall, the PSU
    will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    watts is a measure of the PSU's 'potential' depending on what cards, etc.
    you have drawing off power at any given time?

    Forgive me if the question seems naive, but I am naive when comes to the
    technicals of electricity. I prostrate myself before the tender mercies of
    the august minds resident in this NG, and humbly beg your enlightenment. ;-)

    Cheers, Tony McKee

    --

    ---
    I am a part of all that I have met... yet all experience is but an arch
    Wherethro' gleams that untravel'd world whose margins fade
    Forever and forever... 'ere I move.

    ===-- Ulysses --===
     
    Tony McKee, Apr 15, 2006
    #1
    1. Advertising

  2. Tony McKee wrote:
    > Yo, Folks.
    >
    > Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >
    > http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >
    > Does that 500 watts mean that as soon as you plug it into the wall, the PSU
    > will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    > watts is a measure of the PSU's 'potential' depending on what cards, etc.
    > you have drawing off power at any given time?


    The latter.

    There will be a very small draw <10W when them machine is plugged in but
    "turned off", and the power supply itself will be somewhere from 50-85% efficient.

    The 500W rating is what the supply can output, it will draw more from the mains
    when running at it's rated capacity.
     
    Mark Robinson, Apr 15, 2006
    #2
    1. Advertising

  3. On Sun, 16 Apr 2006 01:16:05 +1200, Mark Robinson
    <2tod.net> wrote:

    >Tony McKee wrote:
    >> Yo, Folks.
    >>
    >> Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >>
    >> http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >>
    >> Does that 500 watts mean that as soon as you plug it into the wall, the PSU
    >> will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    >> watts is a measure of the PSU's 'potential' depending on what cards, etc.
    >> you have drawing off power at any given time?

    >
    >The latter.
    >
    >There will be a very small draw <10W when them machine is plugged in but
    >"turned off", and the power supply itself will be somewhere from 50-85% efficient.
    >
    >The 500W rating is what the supply can output, it will draw more from the mains
    >when running at it's rated capacity.


    But with any modern PC, if you leave it on 24/7, then your monthly
    power bill will suffer. Lets say it draws an average of 70 W from the
    mains. That is 0.07 kW. Over 24 hours, that is 1.68 kW hr = 1.68
    units of electricity to pay for. Over a month, that is about 50.4
    units. At my rate of $0.1566 per unit, that is $7.89. Over a year,
    that is about $96.09.

    Now, take a dual core 2 Gibyte of RAM PC with 10,000 rpm drives and a
    super video card, and have it running CPU intensive work all the time
    (say Seti running in the background). That could draw on average as
    much as 300 W. Per month, that works out as $33.82 and per year
    $411.83. That is enough to have to budget for.

    Now, what if your household has 3 PCs, and you have updated them all
    with nice new modern hardware?
     
    Stephen Worthington, Apr 15, 2006
    #3
  4. Tony McKee

    thingy Guest

    Tony McKee wrote:
    > Yo, Folks.
    >
    > Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >
    > http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >
    > Does that 500 watts mean that as soon as you plug it into the wall, the PSU
    > will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    > watts is a measure of the PSU's 'potential' depending on what cards, etc.
    > you have drawing off power at any given time?
    >
    > Forgive me if the question seems naive, but I am naive when comes to the
    > technicals of electricity. I prostrate myself before the tender mercies of
    > the august minds resident in this NG, and humbly beg your enlightenment. ;-)
    >
    > Cheers, Tony McKee
    >


    500watts is the sum of its outputs, if it is say 80% efficient 625 watts in.

    The PSU outputs different voltages each with its own line(cable), so max
    output is probably effectively reached when one of those lines gets to
    its max wattage out which will depends on what is in the computer....so
    in reality it may be say 350watt as say the +5v reached its max output
    of say 100watt....try and draw anymore and the pc becomes unstable as
    the +5v voltage drops....

    Off could still be a handful of watts, initial boot is probably amongst
    the worst condition, followed by hours of gaming....

    regards

    Thing
     
    thingy, Apr 15, 2006
    #4
  5. Tony McKee

    thingy Guest

    Stephen Worthington wrote:
    > On Sun, 16 Apr 2006 01:16:05 +1200, Mark Robinson
    > <2tod.net> wrote:
    >
    >
    >>Tony McKee wrote:
    >>
    >>>Yo, Folks.
    >>>
    >>>Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >>>
    >>>http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >>>
    >>>Does that 500 watts mean that as soon as you plug it into the wall, the PSU
    >>>will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    >>>watts is a measure of the PSU's 'potential' depending on what cards, etc.
    >>>you have drawing off power at any given time?

    >>
    >>The latter.
    >>
    >>There will be a very small draw <10W when them machine is plugged in but
    >>"turned off", and the power supply itself will be somewhere from 50-85% efficient.
    >>
    >>The 500W rating is what the supply can output, it will draw more from the mains
    >>when running at it's rated capacity.

    >
    >
    > But with any modern PC, if you leave it on 24/7, then your monthly
    > power bill will suffer. Lets say it draws an average of 70 W from the
    > mains. That is 0.07 kW. Over 24 hours, that is 1.68 kW hr = 1.68
    > units of electricity to pay for. Over a month, that is about 50.4
    > units. At my rate of $0.1566 per unit, that is $7.89. Over a year,
    > that is about $96.09.
    >
    > Now, take a dual core 2 Gibyte of RAM PC with 10,000 rpm drives and a
    > super video card, and have it running CPU intensive work all the time
    > (say Seti running in the background). That could draw on average as
    > much as 300 W. Per month, that works out as $33.82 and per year
    > $411.83. That is enough to have to budget for.
    >
    > Now, what if your household has 3 PCs, and you have updated them all
    > with nice new modern hardware?


    Also not all PSUs are equal, some can have an efficiency as low as 50% v
    85%..........

    So buying a higher efficiency PSU can pay for itself in as little as a
    year....

    regards

    Thing
     
    thingy, Apr 15, 2006
    #5
  6. On Sun, 16 Apr 2006 00:48:04 +1200, Tony McKee wrote:

    > Does that 500 watts mean that as soon as you plug it into the wall, the PSU
    > will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    > watts is a measure of the PSU's 'potential' depending on what cards, etc.
    > you have drawing off power at any given time?


    No - that means it can DELIVER 500 watts. It will draw much more than
    that, depending on how efficient it is.


    Have A Nice Cup of Tea

    --
    "When a company starts fighting over IP, it's a
    sign they've lost the real battle, for users."
     
    Have A Nice Cup of Tea, Apr 15, 2006
    #6
  7. Tony McKee

    Brian Dooley Guest

    On Sun, 16 Apr 2006 00:48:04 +1200, "Tony McKee"
    <> wrote:

    >Yo, Folks.
    >
    >Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >
    >http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >
    >Does that 500 watts mean that as soon as you plug it into the wall, the PSU
    >will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    >watts is a measure of the PSU's 'potential' depending on what cards, etc.
    >you have drawing off power at any given time?
    >
    >Forgive me if the question seems naive, but I am naive when comes to the
    >technicals of electricity. I prostrate myself before the tender mercies of
    >the august minds resident in this NG, and humbly beg your enlightenment. ;-)
    >
    >Cheers, Tony McKee


    500W is the amount of power the PSU can deliver to a resistive
    load without cooking up.

    The equation describing this is P=IV, ie 500=Ix240 (in NZ).

    Thus the current drawn from your household supply is 500/240
    amps.

    Roughly 2 amps.

    Nothing else is relevant.

    Your normal wall socket should romp in.
    --

    Brian Dooley

    Wellington New Zealand
     
    Brian Dooley, Apr 15, 2006
    #7
  8. Tony McKee

    thingy Guest

    Brian Dooley wrote:
    > On Sun, 16 Apr 2006 00:48:04 +1200, "Tony McKee"
    > <> wrote:
    >
    >
    >>Yo, Folks.
    >>
    >>Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >>
    >>http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >>
    >>Does that 500 watts mean that as soon as you plug it into the wall, the PSU
    >>will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    >>watts is a measure of the PSU's 'potential' depending on what cards, etc.
    >>you have drawing off power at any given time?
    >>
    >>Forgive me if the question seems naive, but I am naive when comes to the
    >>technicals of electricity. I prostrate myself before the tender mercies of
    >>the august minds resident in this NG, and humbly beg your enlightenment. ;-)
    >>
    >>Cheers, Tony McKee

    >
    >
    > 500W is the amount of power the PSU can deliver to a resistive
    > load without cooking up.
    >
    > The equation describing this is P=IV, ie 500=Ix240 (in NZ).
    >
    > Thus the current drawn from your household supply is 500/240
    > amps.
    >
    > Roughly 2 amps.
    >
    > Nothing else is relevant.
    >
    > Your normal wall socket should romp in.


    No, very wrong...if you have a PC that runs 24/7 then the efficiency of
    the PSU has a big effect on your electricity bill.

    If it runs a few hours a week, no, 4 or 5 hours a day, yes possibly.

    If so 100watts...

    4h x 365d x 0.100kw /0.50% = 171 kw x .17cents = $49.64 a year

    or

    0.85% = $29.20 a year.

    Depending on your payback criteria.....you could save up to $20 a
    year....between PSUs.

    This is a worst case scenario...ie 85% is about the best with 50% about
    the worst....plug in your own numbers....

    tomshardware.com has articles on PSU tests...

    As electricity goes up in cost the savings will get bigger, over the
    life of a PSU say 3 years saving $70 to $80 is possible...

    So, IMHO, if a better PSU is say $10~20 more then it is worth
    considering....

    There are lots of variations on this,

    If you are a gamer and are running games for that 4 hours then your PSU
    is probably up in the 200~300w bracket.....so instead of $20 a year it
    could be as much as $60 a year....justifying a high end PSU on savings,
    let alone stability.

    regards

    Thing
     
    thingy, Apr 16, 2006
    #8
  9. Tony McKee

    shannon Guest

    thingy wrote:
    > Brian Dooley wrote:
    >> On Sun, 16 Apr 2006 00:48:04 +1200, "Tony McKee"
    >> <> wrote:
    >>
    >>
    >>> Yo, Folks.
    >>>
    >>> Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >>>
    >>> http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >>>
    >>> Does that 500 watts mean that as soon as you plug it into the wall,
    >>> the PSU
    >>> will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    >>> watts is a measure of the PSU's 'potential' depending on what cards,
    >>> etc.
    >>> you have drawing off power at any given time?
    >>>
    >>> Forgive me if the question seems naive, but I am naive when comes to the
    >>> technicals of electricity. I prostrate myself before the tender
    >>> mercies of
    >>> the august minds resident in this NG, and humbly beg your
    >>> enlightenment. ;-)
    >>>
    >>> Cheers, Tony McKee

    >>
    >>
    >> 500W is the amount of power the PSU can deliver to a resistive
    >> load without cooking up.
    >>
    >> The equation describing this is P=IV, ie 500=Ix240 (in NZ).
    >>
    >> Thus the current drawn from your household supply is 500/240
    >> amps.
    >>
    >> Roughly 2 amps.
    >>
    >> Nothing else is relevant.
    >>
    >> Your normal wall socket should romp in.

    >
    > No, very wrong...if you have a PC that runs 24/7 then the efficiency of
    > the PSU has a big effect on your electricity bill.
    >
    > If it runs a few hours a week, no, 4 or 5 hours a day, yes possibly.
    >
    > If so 100watts...
    >
    > 4h x 365d x 0.100kw /0.50% = 171 kw x .17cents = $49.64 a year
    >
    > or
    >
    > 0.85% = $29.20 a year.
    >
    > Depending on your payback criteria.....you could save up to $20 a
    > year....between PSUs.
    >
    > This is a worst case scenario...ie 85% is about the best with 50% about
    > the worst....plug in your own numbers....
    >
    > tomshardware.com has articles on PSU tests...
    >
    > As electricity goes up in cost the savings will get bigger, over the
    > life of a PSU say 3 years saving $70 to $80 is possible...
    >
    > So, IMHO, if a better PSU is say $10~20 more then it is worth
    > considering....
    >
    > There are lots of variations on this,
    >
    > If you are a gamer and are running games for that 4 hours then your PSU
    > is probably up in the 200~300w bracket.....so instead of $20 a year it
    > could be as much as $60 a year....justifying a high end PSU on savings,
    > let alone stability.
    >
    > regards
    >
    > Thing
    >
    >
    >
    >


    Why would an expensive switchmode power supply use less electricity than
    a cheap one ?
     
    shannon, Apr 16, 2006
    #9
  10. On Sun, 16 Apr 2006 14:04:20 +1200, shannon wrote:

    > Why would an expensive switchmode power supply use less electricity than
    > a cheap one ?


    It's not a case of price per se.

    It's that some designs are more efficient than others. It is a question of
    efficiency. Some designs waste more electricity than others.

    The better designs can attract higher retail prices because they are...
    more efficient and use less electricity delivering the rated supply.


    Have A Nice Cup of Tea

    --
    1/ Migration to Linux only costs money once. Higher Windows TCO is forever.
    2/ "Shared source" is a poison pill. Open Source is freedom.
    3/ Only the Windows boxes get the worms.
     
    Have A Nice Cup of Tea, Apr 16, 2006
    #10
  11. Tony McKee

    thingy Guest

    shannon wrote:
    > thingy wrote:
    >
    >> Brian Dooley wrote:
    >>
    >>> On Sun, 16 Apr 2006 00:48:04 +1200, "Tony McKee"
    >>> <> wrote:
    >>>
    >>>
    >>>> Yo, Folks.
    >>>>
    >>>> Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >>>>
    >>>> http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >>>>
    >>>> Does that 500 watts mean that as soon as you plug it into the wall,
    >>>> the PSU
    >>>> will start drawing 500 watts worth of leckie? Or, as I suspect, the 500
    >>>> watts is a measure of the PSU's 'potential' depending on what cards,
    >>>> etc.
    >>>> you have drawing off power at any given time?
    >>>>
    >>>> Forgive me if the question seems naive, but I am naive when comes to
    >>>> the
    >>>> technicals of electricity. I prostrate myself before the tender
    >>>> mercies of
    >>>> the august minds resident in this NG, and humbly beg your
    >>>> enlightenment. ;-)
    >>>>
    >>>> Cheers, Tony McKee
    >>>
    >>>
    >>>
    >>> 500W is the amount of power the PSU can deliver to a resistive
    >>> load without cooking up.
    >>>
    >>> The equation describing this is P=IV, ie 500=Ix240 (in NZ).
    >>>
    >>> Thus the current drawn from your household supply is 500/240
    >>> amps.
    >>>
    >>> Roughly 2 amps.
    >>>
    >>> Nothing else is relevant.
    >>>
    >>> Your normal wall socket should romp in.

    >>
    >>
    >> No, very wrong...if you have a PC that runs 24/7 then the efficiency
    >> of the PSU has a big effect on your electricity bill.
    >>
    >> If it runs a few hours a week, no, 4 or 5 hours a day, yes possibly.
    >>
    >> If so 100watts...
    >>
    >> 4h x 365d x 0.100kw /0.50% = 171 kw x .17cents = $49.64 a year
    >>
    >> or
    >>
    >> 0.85% = $29.20 a year.
    >>
    >> Depending on your payback criteria.....you could save up to $20 a
    >> year....between PSUs.
    >>
    >> This is a worst case scenario...ie 85% is about the best with 50%
    >> about the worst....plug in your own numbers....
    >>
    >> tomshardware.com has articles on PSU tests...
    >>
    >> As electricity goes up in cost the savings will get bigger, over the
    >> life of a PSU say 3 years saving $70 to $80 is possible...
    >>
    >> So, IMHO, if a better PSU is say $10~20 more then it is worth
    >> considering....
    >>
    >> There are lots of variations on this,
    >>
    >> If you are a gamer and are running games for that 4 hours then your
    >> PSU is probably up in the 200~300w bracket.....so instead of $20 a
    >> year it could be as much as $60 a year....justifying a high end PSU on
    >> savings, let alone stability.
    >>
    >> regards
    >>
    >> Thing
    >>
    >>
    >>
    >>

    >
    > Why would an expensive switchmode power supply use less electricity than
    > a cheap one ?


    tomshardware has some good articles.

    Some of the fundementals are,

    The design, a cheap design will give the output you expect (or actually
    it might not), but it will "throw away" more of the input to give you
    the output.

    Construction, cheap components have wide tolerances, giving variations
    in output.

    Cooling, some companies will just throw in any cheap fan, this can be
    oversized for the job, so you pump more air through it than needed.

    Component layout, a well designed layout will keep all the components at
    the right temp and minimise the air pressure differential through the
    unit, so a smaller fan is needed, or the fan can run at a lower speed...

    Fan efficiency can vary hugely depending on its size and foil shape.
    Cheap fans have a guessed at fan foil shape, good fans have some
    computer design behind them, also these are quieter.

    regards

    Thing
     
    thingy, Apr 16, 2006
    #11
  12. Tony McKee

    shannon Guest

    thingy wrote:
    > shannon wrote:
    >> thingy wrote:
    >>
    >>> Brian Dooley wrote:
    >>>
    >>>> On Sun, 16 Apr 2006 00:48:04 +1200, "Tony McKee"
    >>>> <> wrote:
    >>>>
    >>>>
    >>>>> Yo, Folks.
    >>>>>
    >>>>> Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.
    >>>>>
    >>>>> http://www.ascent.co.nz/ProductSpecification.aspx?ItemID=345502
    >>>>>
    >>>>> Does that 500 watts mean that as soon as you plug it into the wall,
    >>>>> the PSU
    >>>>> will start drawing 500 watts worth of leckie? Or, as I suspect, the
    >>>>> 500
    >>>>> watts is a measure of the PSU's 'potential' depending on what
    >>>>> cards, etc.
    >>>>> you have drawing off power at any given time?
    >>>>>
    >>>>> Forgive me if the question seems naive, but I am naive when comes
    >>>>> to the
    >>>>> technicals of electricity. I prostrate myself before the tender
    >>>>> mercies of
    >>>>> the august minds resident in this NG, and humbly beg your
    >>>>> enlightenment. ;-)
    >>>>>
    >>>>> Cheers, Tony McKee
    >>>>
    >>>>
    >>>>
    >>>> 500W is the amount of power the PSU can deliver to a resistive
    >>>> load without cooking up.
    >>>>
    >>>> The equation describing this is P=IV, ie 500=Ix240 (in NZ).
    >>>>
    >>>> Thus the current drawn from your household supply is 500/240
    >>>> amps.
    >>>>
    >>>> Roughly 2 amps.
    >>>>
    >>>> Nothing else is relevant.
    >>>>
    >>>> Your normal wall socket should romp in.
    >>>
    >>>
    >>> No, very wrong...if you have a PC that runs 24/7 then the efficiency
    >>> of the PSU has a big effect on your electricity bill.
    >>>
    >>> If it runs a few hours a week, no, 4 or 5 hours a day, yes possibly.
    >>>
    >>> If so 100watts...
    >>>
    >>> 4h x 365d x 0.100kw /0.50% = 171 kw x .17cents = $49.64 a year
    >>>
    >>> or
    >>>
    >>> 0.85% = $29.20 a year.
    >>>
    >>> Depending on your payback criteria.....you could save up to $20 a
    >>> year....between PSUs.
    >>>
    >>> This is a worst case scenario...ie 85% is about the best with 50%
    >>> about the worst....plug in your own numbers....
    >>>
    >>> tomshardware.com has articles on PSU tests...
    >>>
    >>> As electricity goes up in cost the savings will get bigger, over the
    >>> life of a PSU say 3 years saving $70 to $80 is possible...
    >>>
    >>> So, IMHO, if a better PSU is say $10~20 more then it is worth
    >>> considering....
    >>>
    >>> There are lots of variations on this,
    >>>
    >>> If you are a gamer and are running games for that 4 hours then your
    >>> PSU is probably up in the 200~300w bracket.....so instead of $20 a
    >>> year it could be as much as $60 a year....justifying a high end PSU
    >>> on savings, let alone stability.
    >>>
    >>> regards
    >>>
    >>> Thing
    >>>
    >>>
    >>>
    >>>

    >>
    >> Why would an expensive switchmode power supply use less electricity
    >> than a cheap one ?

    >
    > tomshardware has some good articles.
    >
    > Some of the fundementals are,
    >
    > The design, a cheap design will give the output you expect (or actually
    > it might not), but it will "throw away" more of the input to give you
    > the output.
    >


    How exactly ?
    Where will it go ?
    Cheap power supplies don't dissipate any more heat than expensive ones,
    the dissipation of the chopper Mosfets is about the same regardless of
    the build quality, they are likely to have better protection circuits
    and lower noise fans and power factor correction though, the reasons for
    choosing a quality PSU are protection and reliability and low noise, not
    power consumption.


    The power used by a PC is not related to the power supply wattage
    rating, but to the devices that it supplies, the CPU, the cards, the
    drives etc.
    A PC with a 500W power supply will draw the same as a PC with a 200W
    power supply with the same components.
    A PC does not draw the same current 24/7.
     
    shannon, Apr 16, 2006
    #12
  13. Tony McKee

    MarkH Guest

    shannon <> wrote in news:4441d8b0$:

    > thingy wrote:
    >>
    >> The design, a cheap design will give the output you expect (or
    >> actually it might not), but it will "throw away" more of the input to
    >> give you the output.
    >>

    >
    > How exactly ?
    > Where will it go ?
    > Cheap power supplies don't dissipate any more heat than expensive
    > ones


    Are you sure about this? My 480W Thermaltake PSU seems to run much cooler
    than the cheaper 300W one that it replaced, despite me upgrading to
    components that draw more current and also despite the Thermaltake PSU
    running the fan slower and noticeably less airflow coming out of the PSU.
    The conclusion I would draw is that the dearer PSU is more efficient and
    dissipates less heat.



    --
    Mark Heyes (New Zealand)
    See my pics at www.gigatech.co.nz (last updated 5-September-05)
    "The person on the other side was a young woman. Very obviously a
    young woman. There was no possible way she could have been mistaken
    for a young man in any language, especially Braille."
    Maskerade
     
    MarkH, Apr 16, 2006
    #13
  14. Tony McKee

    Mercury Guest

    >> The design, a cheap design will give the output you expect (or actually
    >> it might not), but it will "throw away" more of the input to give you the
    >> output.
    >>

    >
    > How exactly ?


    The efficiency of the implemented design. This can easily be analogised to
    car engine efficiency. The output horsepower of engines differs greatly even
    when the design of the engines are extremely close with resulting huge
    differences in fuel consumption. How is that? Overhead cam shafts? Fuel
    injection? Turbo chargers? Many of these car engine design features also
    have analogous implementations in PSU's!

    > Where will it go ?


    Excess Heat

    > Cheap power supplies don't dissipate any more heat than expensive ones,


    They very often do.

    > the dissipation of the chopper Mosfets is about the same regardless of the
    > build quality,


    For a given spec MOSFET, with a specific duty cycle and equal efficiency of
    delivery of power to the MOSFETS and then through to the output stages,
    feeback loops, load change reaction times etc..... I'd suggest reviewing
    your comments after you have done a bit more reading. But when a MOSFET is
    OFF it is OFF, when a MOSFET is ON it is ON.

    > they are likely to have better protection circuits


    Better Caps that run with least stress, within design specs and as cool as
    possible are possibly a little more important (and irrelevant to this topic)
    since once a PSU is in it is either correct and run until [bad] caps takes
    it out with a bang or something else goes wrong. The protection circuits are
    largely standard, but there are always the cheapo PSU's that let everyone
    down.

    > and lower noise fans


    The higher the efficiency of the PSU, the less heat there is to dissipate so
    the lower CFM rating of fan can be used - the fans themselves use next to no
    power and so their loading is not an issue, but lower CFM rating PSU's are
    often quieter although I would rather have a good PSU fan (fanless is
    largely a waste of time as there has to be a fan somewhere) than a short cut
    shyte one that fails to circulate and stresses the capacitors (or HDD or
    anything else) to an early grave... Lower CFM fans are generally quieter
    too... So here we have a benefit - less CFM needed so room to move on the
    fan ratings with the potential for either Quiet or Cool or Both (thats what
    the switch on the back of one of my high efficiency PSU's is for).

    > and power factor correction though, the reasons for choosing a quality PSU
    > are protection and reliability


    I'll not argue with those two points as Major points!

    > and low noise, not power consumption.


    It is being mandated as a new design requirement - that the PSU efficiency
    must exceed a certain %in to out requirement (forget the spec name, but its
    a modernised version of the Energy Star initiative and applies to PSU's).

    As has already been said, duck on over to tomshardwareguide.com and
    anandtech.com to have a look - there have been several articles of late
    which illustrate these points very well. The benefits come in not just with
    large SLI systems running 24x7, but also systems on standby where there can
    be massive differences when on standby when expressed as % improvement over
    a don't care selection.


    2 cents

    HTH
     
    Mercury, Apr 16, 2006
    #14
  15. On Sun, 16 Apr 2006 17:40:00 +1200, shannon wrote:

    > How exactly ?
    > Where will it go ?
    > Cheap power supplies don't dissipate any more heat than expensive ones,
    > the dissipation of the chopper Mosfets is about the same regardless of
    > the build quality, they are likely to have better protection circuits
    > and lower noise fans and power factor correction though, the reasons for
    > choosing a quality PSU are protection and reliability and low noise, not
    > power consumption.


    Actually, the wasted wattage is dissipated as heat inside the PSU.

    It's all to do with how the PSU converts alternating current at 240 volts
    into direct current at +5, -5, +12, and -12 volts - each supplied with
    sufficient current so that the circut boards won't be trying to draw so
    much that the voltage end up dropping.

    The more efficient PSU will not waste so much of the cycle of the
    alternating current when trying to rectify it into direct current.

    If you want to learn more about this, then I suggest you get a book on
    electricity, specifically one that explains the differences between
    alterating and direct current, and how to convert between one and the
    other. :eek:)


    Have A Nice Cup of Tea

    --
    Fink: "The Linux market is growing 30% to 35% a year."
     
    Have A Nice Cup of Tea, Apr 16, 2006
    #15
  16. Tony McKee

    Tony McKee Guest

    "Stephen Worthington" <34.nz56.remove_numbers> wrote in
    message news:...

    > But with any modern PC, if you leave it on 24/7, then your monthly
    > power bill will suffer. Lets say it draws an average of 70 W from the
    > mains. That is 0.07 kW. Over 24 hours, that is 1.68 kW hr = 1.68
    > units of electricity to pay for. Over a month, that is about 50.4
    > units. At my rate of $0.1566 per unit, that is $7.89. Over a year,
    > that is about $96.09.


    Well, the new rig won't be running 24/7 - half that perhaps. Applying the
    above tablature and its assumed 70 W draw to our Meridian rate of $0.1712
    per unit over 12 hours results in about $50 per annum. That's a 'running
    cost benchmark' I can present to Wifey with a completely honest face.
    Thank-you! ;-)

    > Now, take a dual core 2 Gibyte of RAM PC with 10,000 rpm drives and a
    > super video card, and have it running CPU intensive work all the time
    > (say Seti running in the background). That could draw on average as
    > much as 300 W. Per month, that works out as $33.82 and per year
    > $411.83. That is enough to have to budget for.
    >
    > Now, what if your household has 3 PCs, and you have updated them all
    > with nice new modern hardware?


    That one's easy. Even I know the answer: You live on baked beans for a
    **very** long time. ;-)


    Cheers, Tony McKee

    --

    ---
    I am a part of all that I have met... yet all experience is but an arch
    Wherethro' gleams that untravel'd world whose margins fade
    Forever and forever... 'ere I move.

    ===-- Ulysses --===
     
    Tony McKee, Apr 16, 2006
    #16
  17. Tony McKee

    Tony McKee Guest

    "Tony McKee" <> wrote in message
    news:9Z50g.13300$...

    > Yo, Folks.
    >
    > Let's say you buy an Enermax Liberty 500 watt PSU for your new PC rig.


    ....snippetty snip...

    Thanks for the great replies to this one, Folks - much appreciated. It's all
    going into my 'advice folder' - your writings will live on. You're a
    brilliant crew. Dunno what the likes o' me would do without the likes o'
    you. Diamonds in the mine...


    Cheers, Tony McKee

    --

    ---
    I am a part of all that I have met... yet all experience is but an arch
    Wherethro' gleams that untravel'd world whose margins fade
    Forever and forever... 'ere I move.

    ===-- Ulysses --===
     
    Tony McKee, Apr 16, 2006
    #17
  18. Tony McKee

    Matty F Guest

    thingy wrote:

    > Brian Dooley wrote:


    >> Your normal wall socket should romp in.

    >
    >
    > No, very wrong...if you have a PC that runs 24/7 then the efficiency of
    > the PSU has a big effect on your electricity bill.


    The power that is used by the computer is free in winter whenever
    you have a heater on. So subtract 10% to 30% off the cost for that.
     
    Matty F, Apr 16, 2006
    #18
  19. Tony McKee

    shannon Guest

    Mercury wrote:
    >>> The design, a cheap design will give the output you expect (or actually
    >>> it might not), but it will "throw away" more of the input to give you the
    >>> output.
    >>>

    >> How exactly ?

    >
    > The efficiency of the implemented design. This can easily be analogised to
    > car engine efficiency. The output horsepower of engines differs greatly even
    > when the design of the engines are extremely close with resulting huge
    > differences in fuel consumption. How is that? Overhead cam shafts? Fuel
    > injection? Turbo chargers? Many of these car engine design features also
    > have analogous implementations in PSU's!
    >
    >> Where will it go ?

    >
    > Excess Heat
    >
    >> Cheap power supplies don't dissipate any more heat than expensive ones,

    >
    > They very often do.
    >
    >> the dissipation of the chopper Mosfets is about the same regardless of the
    >> build quality,

    >
    > For a given spec MOSFET, with a specific duty cycle and equal efficiency of
    > delivery of power to the MOSFETS and then through to the output stages,
    > feeback loops, load change reaction times etc..... I'd suggest reviewing
    > your comments after you have done a bit more reading. But when a MOSFET is
    > OFF it is OFF, when a MOSFET is ON it is ON.
    >
    >> they are likely to have better protection circuits

    >
    > Better Caps that run with least stress, within design specs and as cool as
    > possible are possibly a little more important (and irrelevant to this topic)
    > since once a PSU is in it is either correct and run until [bad] caps takes
    > it out with a bang or something else goes wrong. The protection circuits are
    > largely standard, but there are always the cheapo PSU's that let everyone
    > down.
    >
    >> and lower noise fans

    >
    > The higher the efficiency of the PSU, the less heat there is to dissipate so
    > the lower CFM rating of fan can be used - the fans themselves use next to no
    > power and so their loading is not an issue, but lower CFM rating PSU's are
    > often quieter although I would rather have a good PSU fan (fanless is
    > largely a waste of time as there has to be a fan somewhere) than a short cut
    > shyte one that fails to circulate and stresses the capacitors (or HDD or
    > anything else) to an early grave... Lower CFM fans are generally quieter
    > too... So here we have a benefit - less CFM needed so room to move on the
    > fan ratings with the potential for either Quiet or Cool or Both (thats what
    > the switch on the back of one of my high efficiency PSU's is for).
    >
    >> and power factor correction though, the reasons for choosing a quality PSU
    >> are protection and reliability

    >
    > I'll not argue with those two points as Major points!
    >
    >> and low noise, not power consumption.

    >
    > It is being mandated as a new design requirement - that the PSU efficiency
    > must exceed a certain %in to out requirement (forget the spec name, but its
    > a modernised version of the Energy Star initiative and applies to PSU's).
    >
    > As has already been said, duck on over to tomshardwareguide.com and
    > anandtech.com to have a look - there have been several articles of late
    > which illustrate these points very well. The benefits come in not just with
    > large SLI systems running 24x7, but also systems on standby where there can
    > be massive differences when on standby when expressed as % improvement over
    > a don't care selection.
    >
    >
    > 2 cents
    >
    > HTH


    The most efficient power supplies are not the most expensive, if you
    take a look yourself at the sites you quote you will find that all the
    power supplies are about 70%
    The higher costs appears to be due to branding marketing and presentation.
    Quality capacitors will make a difference to the ripple specs, not the
    efficiency
    Its what you attach to your psu that makes the difference in consumption.
    What your system draws in standby is a function of those parts also, not
    the power supply.
     
    shannon, Apr 16, 2006
    #19
  20. On a pleasant day while strolling in nz.comp, a person by the name of
    shannon exclaimed:
    > The most efficient power supplies are not the most expensive,


    Of course, not *necessarily* but the more expensive supplies at the same
    wattage tend to be more efficient, because that reduces noise, power
    consumption and increases reliability.

    > take a look yourself at the sites you quote you will find that all the
    > power supplies are about 70%


    Maybe at that site, but there are plenty of good PSUs that do better
    than that.

    > The higher costs appears to be due to branding marketing and presentation.


    Sorry, but you're wrong. There is an element of this, as in everything,
    but there are many expensive PSUs that perform better.

    > Its what you attach to your psu that makes the difference in consumption.


    Of course, that makes the lion's share of the difference, but there ARE
    big differences in consumer PSU efficiencies. As others said, as bad as
    50% in really bad ones, to 90% in really good ones.

    --
    aaronl at consultant dot com
    For every expert, there is an equal and
    opposite expert. - Arthur C. Clarke
     
    Aaron Lawrence, Apr 16, 2006
    #20
    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. Ciger

    385 Watts PSU $20 rebate

    Ciger, Nov 21, 2005, in forum: Hardware
    Replies:
    7
    Views:
    1,064
    unholy
    Nov 23, 2005
  2. ConceptZone
    Replies:
    5
    Views:
    962
  3. m.jawatts

    mrs watts

    m.jawatts, Sep 13, 2004, in forum: Computer Support
    Replies:
    0
    Views:
    575
    m.jawatts
    Sep 13, 2004
  4. DeserTBoB
    Replies:
    3
    Views:
    3,246
    DeserTBoB
    Jan 23, 2007
  5. watts/sec and the new Sunpaks

    , Mar 4, 2006, in forum: Digital Photography
    Replies:
    22
    Views:
    855
Loading...

Share This Page