"Edward Diener No Spam" <> wrote in
message news:AWwAe.1820$ nk.net...
> Vanguard wrote:
>> "Edward Diener No Spam" <> wrote
>> in message news:SluAe.1663$ k.net...
>>
>>> How does one relate the UPS needs for a particular computer/power
>>> supply. I am upgrading to a 530W power supply for my Abit mobo and
>>> system. Does this mean that any UPS I may buy needs to support 530
>>> Watts or higher ?
>>
>>
>>
>> 530 watts is its power output, not its power consumption. You never
>> mentioned the efficiency rating of the power supply. If, for
>> example, the efficiency was 70% then the power consumption would be
>> 530 watts / 0.70 = 757 watts. Are you actually drawing 530 watts
>> from your power supply or actually expect to draw that much with
>> later addition to your computer's components?
>
> Let's say, just as an example, I do. Then what is the minimum UPS
> wattage I need to support my computer and its power supply ?
>
>
> What else are you going to connect to the UPS?
>> Please explain what good is a workstation's system unit on a UPS if
>> the monitor is not also connected to the UPS. A file server can run
>> without a monitor but a workstation is for use by a *user* and that
>> means you need to *see* what is going on.
>
> Of course. I need to add in my monitor's usage also. But I am just
> looking for a good formula to determine what the minimum UPS wattage
> should be that I get. Also I do not understand the relationship
> between the UPS VA and the wattage. So I am looking for hints there
> also.
>
> If you or anyone else wants to point me to somewhere I can read about
> UPS in relationship to the wattage used on my system, and how to
> determine the wattage actaully used on my system, I will be glad to
> read it.
You need to look at the specifications for the UPS in which you are
interested. A 1000VA UPS that has a load of 1000VA wouldn't supply any
time at that full load - except if the battery actually could produce a
sustained output at the maximum load. A UPS with a tiny battery would
go poof in no time at its maximum draw but a huge battery could deliver
that sustained maximum output load for a lot longer. So it depends on
the capacity of the batteries in the UPS. As another example, I can
string ten 12-volt motorcycle batteries together which would then handle
a 120VA load, but a series of ten much larger car batteries could also
sustain the same 120VA load but for much longer. With batteries, size
does matter.
The maximum voltage output of the UPS will be 120 volts (since that is
what the connected equipment is rated for). The maximum current will
depend on how much and how fast it can be drawn from the battery (so
quality of battery and its chemical composition matters) and the size of
the conductors used to connect the equipment. The typical power cords
can only handle 15 amp currents. So theoretically you could draw 1800
VA from the tap except it is likely that the battery cannot supply that
high a current load. So it all depends on the battery in the UPS, not
the VA rating because that won't tell you the time.
For example, the chart at
http://www.apcc.com/products/runtime...dedruntime.cfm tells you
how much running time you have with THEIR products using whatever
batteries they employ in their products based on the load. You say that
you have a 530-watt power supply except you never mention its brand and
model. Lots of PSU makers lie about their specs so figure you'll get
60% to 70% of their rated wattage (before the PSU fries since many don't
provide decent overload protection). At two-thirds of your PSU's
claimed 530-watt capacity, you could suck out from it 353 watts. At 60%
efficiency, and if loaded at 353 watts, it would be sucking in 589 watts
from your UPS. Now add a 19-inch CRT monitor that consumes another 140
watts. With just the system unit and monitor, you are consuming 729
watts. From their chart, APC's BR800 wouldn't give more than a minute
or two of up-time and you probably cannot close all your applications
and properly shutdown the operating system and computer in that amount
of time. If, however, you were loading your PSU much less, say only 200
watts, and it was more efficient at 70% so it sucked in 286 watts at
that load and your 19-inch monitor only consumed 80 watts for a total
load of 366 watts on the BR800 then it would stay up for 12 minutes.
I have a 2500 VA true-sinusoidal output transformer-isolated UPS that
gives me 56 minutes of up-time (the two batteries weigh a total of 70
pounds) with the system unit and a 19-inch monitor (the scanner and
inkjet are also hooked up but they are in standby mode so they consume
very little power for their indicator lights, and I also put the
cordless telephone on the UPS since the telephone line is powered in an
outage but not the cordless function of the base unit to the handheld
unit). The 56-minute up-time was when the batteries were new. After 5
years (and before I replaced them), I only got around 20 minutes of
up-time. As to what up-time you can expect, you'll have to check the
specs for whatever UPS you are looking at and actually know what type of
load you are going to put on it.
You could go to
http://www.powerware.com/UPS/selector/, select your type
of setup (probably a workstation), and walk through their wizard to
figure out what size of UPS you need, but their runtimes are based on
their efficiencies and battery capacity for their products, so the VA
rating they show for their product may not equate to the same VA
capacity for another vendor's product. Because you were vague in
providing any details regarding your particular hardware setup, almost
deliberately so, I choose some typical values and found that their $95
low-end 700VA #3105 UPS (
http://www.powerware.com/UPS/3105_UPS.asp)
weighing 15 pounds might give you about 3 minutes of up-time whereas
their $630 high-end 700VA #9120 UPS
(
http://www.powerware.com/UPS/9120_UPS.asp) weighing 29 pounds would
give you 10 minutes (and a lots more features, too). Same VA rating but
more time because of a larger battery, better efficiency, better shaped
output voltage (sinusoidal or more granular stepped-wave), and other
factors.
Similar Threads
Linear Mode
