APC question

J. P. Gilliver (John) wrote:
In message , Andy


But the remaining time value kept changing. I think it is only somewhat
accurate.

Or, like time to finish a multi-file move or copy, it might change as
the load changes. Might be worth having Task Manager open and seeing if
there's any (inverse) correlation between CPU load and projected time left.

The metrology on a UPS is only approximate.

*******

http://batteryuniversity.com/learn/a...tate_of_charge

"BU-903: How to Measure State-of-charge

Explore SoC measurements and why they are not accurate.

Voltage Method
...
This makes the *voltage-based* SoC method *impractical*
for a battery in active duty. === UPS discharge activity...
"

My experience with lead acid, over a wide temperature
range (trying to charge my car battery at -20C), is
that the voltage method can be off by 50%. In other
words, when I think the battery is "full", it might
only be "half full". The terminal voltage is
temperature sensitive. Even the density is
temperature sensitive, so if you use a hydrometer
on a wet battery, you must use the temperature
table to correct for the temperature of the battery
when you checked the composition of the liquid in
it. And the temperature is not the air temperature
either. It's the temperature of the positive battery
terminal. In a car, the battery could be heated by
the trip you just took, needing 12-24 hours to track
ambient (air temp) again.

You charge car batteries, to prevent them from "freezing".
A healthy car battery, the liquid might be 12 molar
sulphuric. The freezing temperature might be -55C
in that state.

A bad car battery, one where you've let it discharge,
the liquid is getting closer to tap water. Consequently
the freezing point is only a bit below 0C. If the
car is outside and it's -20C, now the battery can freeze.
The liquid becomes a solid, expansion forces take effect.

So if a person owns a lead acid car battery near the
end of its life, and the temperature at night will
be dropping to -20C, then you try to charge it
back up before the cold weather comes. As you charge
it, the liquid becomes more acidic.

The temperature issue also causes problems for the
charging circuit. Whether charging the battery,
or float charging it to a certain voltage, the
desired terminal voltage is also temperature
sensitive. At -20C or +20C, the correct float
voltage is going to be different than the 13.2V
at room temperature (for a car battery, your
UPS may use 6V batteries).

In summary, voltage and lead acid are loosey goosey.
If you think you know what you're doing... you don't.

The alternative would be a fuel gauge chip.

https://www.ecnmag.com/article/2012/...y-fuel-gauging

"An improved solution over the voltage-based method
is the Coulomb counting technique. In this technique,
the charge flowing into and out of the battery is
integrated to get an accurate estimate of the
remaining capacity. This technique works well,
if the battery]s initial state-of-charge is accurately known."

A problem with that technique, is the battery load may
vary from 10mA to 100A. And this may require a large
dynamic range from the design. That's probably easy
to do with sigma-delta (if someone was starting such
a design today). Older designs would be more
crappy (flash converter or SAR).

The processor in the UPS, may have a 10-bit ADC
for example. Not a lot of anything to work with.
No 24-bit gear in those, although the silicon
cost to do 24-bit, wouldn't be that high. Silicon
is "almost free", and only adding pins to the
microprocessor package costs money. On a UPS,
the electricity is relatively free, so the power
to run a measurement circuit wouldn't be an issue.
In any case, the UPS has to turn off "everything"
at some point, when the battery is too low.

Paul