AVR-Chat

On Jul 30, 2004, at 1:17 PM, Graham Davies wrote:

> --- In AVR-Chat@yahoogroups.com, "C Sizer" <aussiecol39@y...> wrote:
>
>> ... I want to use 4 x 2300mAh AA
>> batteries in serial/parallel to
>> provide me with 3V.
>
> Sounds like alkaline cells at 1.5 volts each.

He needs a supply voltage of 3V, not batteries that add up to that much.
As he states below, he needs a converter:

>
>> The printer runs on 5V but consumes
>> over 3.5A peak, so I require a DC-
>> DC converter circuit... to provide
>> 3V at 5A peak.
>
> You mean to provide 5V/3.5A, which will suck six amps or more from
> your 3V battery. I don't think this will work. The internal
> resistance of alkaline cells will be too high and your 3 volts will
> sag something terrible.
>
> Like everyone else, I suggest you switch batteries.  Use four NiMH AA
> cells to get 5V and regulate down to 3V. I don't think you'll need to
> buck down, just use a linear regulator, unless you're taking a lot of
> current at 3V. Although the capacity of the NiMH cells won't be 2300
> mA.h (I use 1800 mA.h cells but I think now you can do a bit better)
> I think they will actually last longer. This is because the alkaline
> cells will never give you full capacity at such a high discharge rate
> and you're avoiding power loss in the boost converter. High quality
> cells will be able to keep the voltage up because NiMH cells have a
> low internal resistance. For example, the 7.2 volt 6.5 A.h modules in
> my car have an internal resistance just a bit over 10 milliohms. I
> can get 60 amps or more from the 300 volt battery and the voltage
> stays up at 275 volts.

Unfortunately with a linear regulator, the current required from the
battery is still 3.5A.  With 4 NiMH cells in series and a DC-DC
converter with 90 percent efficiency, you could get the peak current
draw down to about 2.4A.   A linear regulator under the same
conditions is wasting 5 W *more* power than the switcher. That's
a lot of extra power and heat when you're operating off of batteries.

Mike