Calculate runtime for backup battery

[AK]

This is what I am using for backup for my computer.

Expert Power EXP 12180 12 volts 18 amp hours

My Xfinity modem uses 22 watts.
LG monitor 60 watts
Computer 55 watts
Grand total of power usage is 137 watts.

Is there a calculation that will show how many hours my computer will run using the backup battery?

Thanks.

[R.Wieser]

AK,

> Is there a calculation that will show how many hours my computer will run
> using the backup battery?

Yes.

A "Watt" is roughly equivalent to a "Volt times Ampere". This means your
backup battery contains 12 volt times 18 Amp hours equals 216 watt hours.

By dividing that by the 137 Watt you draw from it you get how many hours it
will last : 216 watt hours divided by 137 watts equals 1.57 hours (one hour,
34 minutes).


Mind you, the above is just a crude calculation, which does not account for
how many amperes you draw from it, the age of the battery (a charge declines
due to age), how many times and how fast it was charged or even temperature.
And all of that is ofcourse dependant on the type of battery (lead-acid
batteries respond differentlly that lithium-ion ones).

Hope that helps.
Rudy Wieser

[Paul]

UPS manufacturers have a prediction curve, which de-rates
battery capacity, for the rate of discharge. Visiting the
web site for the UPS product, can sometimes cough up that
graph.

The capacity is not a "conserved" quantity.

It goes like this.

Load You predict You get

100W 8 minutes 8 minutes
200W 4 minutes 3 minutes
400W 2 minutes 1.2 minutes

As you honk on the battery, more of the energy is lost
as heat inside the battery. You're talking 40 amp flows
from the battery. Notice that the UPS has bloody fat wires
on top to the battery. That means there's a serious DC current
flow at full UPS load. Those wires are kept short, because
among other things, that wire costs money. That wire, at
least the wire on mine, is the "very nice" kind. Not
RadioShack wire :-) (RadioShack refused to sell nickel
plated wire, so I hated them for that...)

*******

https://www.apcguard.com/Smart-UPS-Runtime-Chart.asp

Let's just take the top entry, and Half versus Full load.

Half 24 minutes 240W
Full 7 minutes 480W

You can see the battery doesn't like this. You can go down the
chart and take ratios of half versus full, and see whether the
behavior is a "manifest constant", or whether each unit is
different.

What you'd really want, is check for manifest constant behavior,
for units that use the *same* battery. You would hope in such
a case, the same ratio presents itself, at the same load values.

In any case, that page gives you at least something to plot up
and consider for your maths.

*******

UPS are rated in terms of Watts and VA. The difference between
these two units of measure, is the PF or power factor. This
is a measure of whether the voltage and current waveforms are
out of phase. Modern ATX supplies have Active Power Factor Correction
or PFC, which attempts to make the load PF=1 or resistive, the
best kind of resistive. The power company prefers we use
resistive loads, like an incandescent bulb is PF=1.

If you take all the electrical loads, put them on a power strip,
then plug the power strip into one of these, you can measure
W, VA, and PF, with the one instrument. One of these should
be about $40 or so. I have one monitoring my "pig" computer,
as a reminder of what a pig it is. It costs real money to run
the pig.

http://www.p3international.com/products/p4400.html

There is also a version of this, for other markets
with different plug. The design limit of 15 amps is
a function of the shunt inside, and the shunt won't
change between markets. But one meter expects 120V,
the other model expects 240V max. There's probably
a different scaling resistor used for volts. And a different
firmware for the processor in there (maths are done
in real time).

I don't know if any UPS website has worked examples of
W and VA, and what to do with the numbers. But it would
certainly help if they did.

Paul

[AK]

Thanks Rudy.

My 137 watts came from using a Watt O Meter, so it should be fairly accurate.

It is a sealed lead acid battery.

Andy

[R.Wieser]

AK,

> Thanks Rudy.

You're welcome.

> My 137 watts came from using a Watt O Meter, so it should be fairly
> accurate.

Even better (your message read as if you had taken the wattage values off of
the labels of each device and added them together).

Mind though that when I said "roughly equivalent" I didn't account for any
kind of phase-shifting between the voltage and current cycles. When you
would be measuring the voltage and current independantly with the apropriate
equipment you might well find a higher current than what the drawn wattage
seems to indicate (but gets obsfucated by the phase-shifting).

IOW, the time I've calculated is an *optimum* time, the real one could be
well lower - also because of the number of other factors I named.

> It is a sealed lead acid battery.

I'm sorry, but I've not been enough "in to it" to know how much difference
it makes on the current you're drawing from it. Just don't take the
calculated time as show long it will actually last. Drawing ~11.5 amps (137
watts divided by 12 volts) is not a small ammount.

Also, the voltage conversion (from the batteries 12 to the output 220 volts)
also causes some loss.

Did I already say it was a crude calculation ? :-)

Regards,
Rudy Wieser

[AK]

My old Frontier modem did not draw much current and did not heat up much.

But then I got only 15 Mbps download speed.

With my Xfinity modem, I get 320 Mbps speed.

But it generates a lot of heat. :-(

My power outages tend to last no more than a few minutes.

Andy

[Paul]

With lead acid batteries, even the sealed ones, the idea is
the battery provides time to do a controlled shutdown of the
PC. Then, you flip off the power on the lot.

This reduces the "depth of discharge", and is intended to
make the battery last longer (in years). I got 11 years
from the first battery, only 4 years from the second battery.
The replacement batteries are not very good. The failure
mode of the second battery was different -- it seemed
to be plate failure, destruction of plates. Three of six
cells failed short.

Paul

[AK]

My current battery is 4 yrs old.

Andy