Re: power consumption batteries with awake/sleep modes...

On Jun 5, 1:14 pm, Yevgen Barsukov <evgen...@xxxxxxxxx> wrote:
On Jun 4, 9:14 pm, Ginu <osheik...@xxxxxxxxx> wrote:

Hello all,

I'm studying power consumption in batteries for various wireless
devices to analyze the efficiency of different wireless technologies.
However, I've come across two separate approaches and am not sure how
these are related. I'm hoping somebody can clarify this for me.

The first approach consists of how batteries typically work (from what
I've seen - I'm a communications engineer so don't know much about
this) - the battery draws one current in awake mode (on the order of
mA) and a different current in sleep mode (on the order of uA) to
extend battery lifetime. This suggests to me that it doesn't matter
what the device is "doing" in awake mode - as long as the device is
awake, the battery draws current on the order of mA for the duration
of the time that the device is awake.

Battery does not do anything by itself, think about it as a giant
charged capacitor in series with resistor. It has tiny self-discharge,
which in case of Li-ion battery is negligible.

It is the system which has different power modes, such as sleep,
active etc. The simplest way to think about the system as variable
resistor connected across battery terminals, which goes
to low resistance in high-power mode, and large resistor in low-power
mode. Current will be determined by the battery voltage
and the resistance.

More meaningful than resistor is a constant power model, as system
actually operates from internal power convertor which maintains
constant voltage to different system nodes, which themselves
consume constant current, so from outside this looks like a constant
power drain.



The second approach is that different wireless technologies have
different energy efficiencies. These are based on time&power to switch
to transmit/receive mode, power related to gates in the transmitter/
receiver electronics (gates), transmit power, time transmitting
(message size divided by transmission rate), etc. This energy
efficiency is typically given in joules/bit for a technology and can
be covered to power consumption (joules/sec) by multiplying the energy
efficiency (joules/bit) by the rate (bits/sec).

This is the only approach that makes sense. However most devices
have so many different functions that it is impossible to determine
efficiency par function. You could try to do it by monitoring
power consumption, while making comment on what the system is doing
at this moment.
It is a very tedious process but that is how it is done properly.

Other approach is to use a standard typical functionality test that
exercises all typical functions,
for example - 10 min talking, 5 hr rest, 10 min playing game.
Same test is applied to different phones and power consumption
is measured.
This is easy to do, and meaningfulness of such test will be
determined by how "typical" such usage is.



In the first approach, power is calculated strictly as awake current x
battery voltage where P = I * V (both constant in awake mode over
time). In the second approach, P = energy efficiency x rate. This
second approach suggests that the power consumed is not constant,
while the first one does.

Can somebody provide some clarification on how battery drain occurs?
Surely these two approaches should be related since both should define
power consumption?

Also, power drain in a battery is not linear. Does anybody have a link
to any material that provides the non-linear equations that map
battery drain (current)?

Here are some typical voltage vs current curves for batterieshttp://industrial.panasonic.com/www-data/pdf2/ACA4000/ACA4000CE215.pdf

More current, more IR drop (voltage loss) across internal
resistance of the battery. R is increasing at low temperatures.

Regards,
Yevgen



Thanks in advance,
Omar

And just to re-iterate, battery capacity operates on my 2nd approach,
as opposed to my first? I've seen studies that calculate daily current
consumption by observing the fraction of a day (in hours) that the
battery is awake, and the fraction of the day it is asleep, and
finding the sum. Remaining battery capacity was then just plugging
this daily current consumption calculated into an exponential function
to get the capacity remaining. So to me, it was seeming like what you
were actually "doing" was irrelevant, and that the current was just
being drawn at a constant rate while in wake/sleep modes from a
consumption perspective.
.

Relevant Pages

  • Re: power consumption batteries with awake/sleep modes...
    ... devices to analyze the efficiency of different wireless technologies. ... extend battery lifetime. ... actually operates from internal power convertor which maintains ... be covered to power consumption by multiplying the energy ...
    (sci.chem.electrochem.battery)
  • Re: power consumption batteries with awake/sleep modes...
    ... I'm studying power consumption in batteries for various wireless ... devices to analyze the efficiency of different wireless technologies. ... extend battery lifetime. ...
    (sci.chem.electrochem.battery)
  • Re: power consumption batteries with awake/sleep modes...
    ... devices to analyze the efficiency of different wireless technologies. ... extend battery lifetime. ... More meaningful than resistor is a constant power model, ... (message size divided by transmission rate), ...
    (sci.chem.electrochem.battery)
  • power consumption batteries with awake/sleep modes...
    ... I'm studying power consumption in batteries for various wireless ... devices to analyze the efficiency of different wireless technologies. ... extend battery lifetime. ... what the device is "doing" in awake mode - as long as the device is ...
    (sci.chem.electrochem.battery)
  • Re: Measuring my applications power consumption
    ... or to get power status information from within .NET? ... To measure actual power consumption the *best* way is with a ammeter inine ... The generic battery driver shell provided by ... Chris Tacke, Embedded MVP ...
    (microsoft.public.dotnet.framework.compactframework)
Loading