Is there a chart of some other reference?
Thanks,
Steven
ste...@main.nc.us - preferred email
ste...@923-tree.com - alternates
s...@efn.org
> Steven Lee wrote in message <6e4n0d$11n$1...@haus.efn.org>...
> >The subject says it well. How many BTUs does a "typical" computer,
> >monitor, laser printer, UPS generate?
> >
> >Is there a chart of some other reference?
> >
> >Thanks,
> >
> >Steven
> >
> >ste...@main.nc.us - preferred email
> >
> >ste...@923-tree.com - alternates
> >s...@efn.org
--
FWIW;BBB
>Aire Kool wrote:
>>
>> Steven , If you can determine how many Watts your equipment uses you can
>> multiply that number by 3.412 and this will tell you the answer in BTU's per
>> hour.
>> Gord J.
>>
>But that's not the whole story, Gord. Even running W-95, the computer
>uses at least some of that energy in a usable form. The fan pushes air,
>the fax pushes paper, the monitor excites phosphor,and so on.
> Not sure what the ratio is, but not all the wattage is released as
>heat.
But, unless the fan pushes air all the way, out of theconditioned space, the
energy in the air movement (as it slows down, by friction) is converted back to
heat. The same effect occurs with pushing paper(along a horizontal axis),
sound, and light emitted from CRT phosphor (photons are bound to be
absorbed(converted to heat) by some atom in the room/house). The net result
is almost all(>99.8%) energy used is converted to heat.
Tim Keating
ktcn...@netrunner.net,
Remove numeric digits from address before responding via email.
Right you are. All the "work" done in the envelope remains in the
envelope, except for the light going out the window, or the magnetic
fields, all stuff really too small to measure.
But, I suppose that the ratings of the machines would be slightly
higher than the actual power consumption. If you're adding up the heat
production in a space in order to figure the cooling load, those numbers
would give you a little fudge factor.
--
FWIW;BBB
>The subject says it well. How many BTUs does a "typical" computer,
>monitor, laser printer, UPS generate?
>
>Is there a chart of some other reference?
>
>Thanks,
>
>Steven
>
>ste...@main.nc.us - preferred email
>
>ste...@923-tree.com - alternates
>s...@efn.org
Your typical computer uses about 200 watts when everything is running.
That is equivalent to about 600 Btu/hour in very round numbers.
Nearly all the wattage used ends up as heat. Almost nothing ends up
as light and sound. Even light from the screen is absorbed by the
furnishings of the room as heat.
Bruce Birbeck <roc...@tidewater.net> wrote:
Machesis of the North
For the parisites of the net
clas...@psbgm.qc.ca
root@localhost
admin@localhost
ro...@mailloop.com
jo...@mailloop.com
jor...@earthlink.net
Chairman Reed Hundt: rhu...@fcc.gov
Commissioner James Quello: jqu...@fcc.gov
Commissioner Susan Ness: sn...@fcc.gov
Commissioner Rachelle Chong: rch...@fcc.gov
abuse@localhost
postmaster@localhost admin@localhost
hostm...@BEST.COM
ant...@PACIFIC-TECH.COM
>A very simple method(and acurate) is to assume that the computers have
>either 300 watt or 250 watt power supplies, therfore multiply by 3.41
>to convert to btuh.
Wrong and right, in that order. Rated wattage has nothing to
do with actual draw. Put an amp probe on it, and you'll see less than
1 amp peak ( IE, <120 W ). If you size for 250 / 300 W, you'll be
*hugely* over.
>The monitors have amp ratings on the back Watt = amps * voltage.
>If the sytems run 24 hrs/day you can design to this combined value.
No. Do not design to power supply design limits, design to
actual usage.
Paul
>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~>~~
pjm@(remove this part )pobox.com
My WWW site is at http://www.pobox.com/~pjm, featuring free HVAC software.
The Sci.Engr.Heat-Vent-AC and Alt.HVAC FAQ is at http://www.elitesoft.com/sci.hvac/
A good reference for the heat load for equipment is the ASHRAE fundamentals
book. For an office, a good guestimate for computer equipment is 2watts per
square foot.
Bruce Birbeck wrote in message <6e5ugf$71j$0...@207.3.226.21>...
>Aire Kool wrote:
>> Steven , If you can determine how many Watts your equipment uses you can
>> multiply that number by 3.412 and this will tell you the answer in BTU's
per
>> hour.
>But that's not the whole story, Gord. Even running W-95, the computer
>uses at least some of that energy in a usable form. The fan pushes air,
>the fax pushes paper, the monitor excites phosphor,and so on.
> Not sure what the ratio is, but not all the wattage is released as
>heat.
>
>> Steven Lee wrote in message <6e4n0d$11n$1...@haus.efn.org>...
>While it's true that the computer does do some work, it doesn't take 500
>Watts to excite some phosphur. Most of the energy that a computer uses is
>wasted as resistance. The mechanical work is little more than spinning a
>hard drive or a cd-rom around, yet my pc requires a fan running to maintain
>its temp.
>
>A good reference for the heat load for equipment is the ASHRAE fundamentals
>book. For an office, a good guestimate for computer equipment is 2watts per
>square foot.
>
In what way does this answer the original question? And what use is 2
W/sq.ft to the questioner? Please keep to the topic asked.
If the whole system was based on what the computers/accesories actual
heat output was then maybe your method would be viable.
If you look at the what 100 watts is equivalent to in cooling cfm you
will see my point. Good design is not splitting hares because in the
end 25 cfm means nothing.
A client is much more interested in a system that is flexable and will
not require rebalancing every time a small heat source is added, or in
a case of presion design a complete retrofit.
pjm@see_my_sig_for_address.com (paul milligan) wrote:
Machesis of the North
>Paul i will not start a argument but if you look in any reputable
>design manual you will see what you propose is not true.
Presence or absence in a design manual is not equivalent to
truth :-)
>I know what you are saying about acutal current draw but this is not
>the way systems are designed.
>
>If the whole system was based on what the computers/accesories actual
>heat output was then maybe your method would be viable.
Then what _do_ you design on ? Heat load that is not there ?
If the power supply is rated @ 250 W ( typical ), and actual draw is (
at a guess ) 100 W, that's a 150 W difference, or ( in round numbers )
500 BTUH. Put 100 PC's into your load calc, and that's 50,000 BTU, or
~ 4 tons ( of purely sensible) load. This accounts only for the
computer, not the monitor ( which will only increase the numbers )
Work your A/C unit SHR into that, and of course the load number grows.
In these days where hundreds or thousands of PC's in any office
complex are not uncommon, I think that's a signifigant load.
I don't doubt you that many design books fail to account for
this properly, but I would suggest that they are wrong.
{ before I get slammed again - guys, I'm just throwing ideas
around here, so get off my ass :-) }