Battery Theory

[TomM]

Hi,

I have a question about battery capacity and recharging. If I have a battery that has a rated output of 10.8V and 50 WHr, is it possible to recharge it with a power source rated 6V 3W? Or would it need a source rated at 10.8V?

Thanks

[Franklin97355]

You would need a source rated HIGHER than 10.8 in order to push the voltage into the battery at it's rated voltage.

[MrAl]

Hello,

I have to agree, you need a power source that has a voltage that
is higher than the battery voltage.

To charge correctly however, you also need to know the chemistry
of the battery...that is, what type is it?
Here are some possible types:
Lead Acid
NiMH
NiCd
Li-ion

Each type above has a different way of charging. Doing it wrong could
damage the battery permanently and even possibly blow it up and
cause a fire. Li-ions seem to be the most dangerous so much care
has to be used when charging this type.

Typical charge regimes go like this:

Lead Acid:
Charge with limited current until the float voltage is reached,
then use constant voltage. There is however a new technique now
that supposedly increases cycle life but it's not easy to do.

NiMH:
Charge with constant current until minus delta V is reached. Delta
V is typically 10mv.

NiCd:
Same as NiMH, except Delta V is typically 20mv.

Li-ion:
Charge with limited current until 4.200v is reached, then use constant
voltage. The 4.200v spec is very exacting...dont go over 4.250v or
expect permanent cell damage. To make up for meter inaccuracies
when measuring this set point it's a good idea to set a little lower.
I do my cells up to 4.150v just in case the meter is slightly off and it's
really getting charged to 4.200 or so.

[Rodney]

I agree with most of what MrAl said with the following comments. On the nickle based cells (NiMh and NiCad) you can not always depend on the delta V as it will vary with the rate of charge and a few other factors. A better way of detecting when to end the charge on these two types is to monitor the cell temperature; as soon as you detect a significant temperature rise (about 10 degrees F) terminate the charge.

[MrAl]

Hi again,

Rodney:
I like the termperature rise method too, better than any other for
NiMH cells. I was however just listing typical methods and the user
should really read up on all of these methods (other cell types too)
before attempting to charge any cell or battery.

I have not found an NiMH cell that does not obey the minus delta
V law yet, but i still believe it is possible. I also know that these
days commercial chargers also limit the total charge time to some
acceptable level just in case the delta V method fails or something
else goes wrong and the cell continues to charge for a longer
than expected period. The timer is set at the start of charge and if
the timer goes off before the charge shuts off it is then shut off
immediately. Nice idea i think, even if using temperature rise detect.

For what it is worth, there is also a problem associated with detecting
temperature rise also. The problem involves the use of a thermistor
to measure temperature. The first problem occurs if the thermistor
cell contact becomes compromised, so the thermistor no longer can
respond to the temperature of the cell itself but rather only the ambient
air now. This happens if the thermistor is not properly mounted so it
stays in contact with the cell. This problem isnt too hard to avoid though
if the temperature rise does not behave as expected relative to a
second measurement of the ambient air (not hard to do either).
The second problem comes up if the thermistor goes bad. This is a
little harder to deal with and probably the best idea is to include the
timer to time the max charge time allowed. Using two thermistors
isnt a bad idea either i guess.

Just a few more things to think about when charging cells.

I had designed a chip that does charge monitoring of different cell
chemistries so i got pretty deeply involved of these various techniques
and their pitfalls. The chip monitors the charge and sends the
data to the computer for analysis and sounds an alarm at the end
of charge for each cell being monitored. The user has the option
of using cheap thermistors to monitor temperature too, requiring
one thermistor per channel and the user has to enter the thermistor
data parameters into the software window for the thermistors.
I've done charging while monitoring both voltage, current, and
temperature and it's interesting to compare the graphs of these
after the cell is done charging.
My next addition to this chip was going to be another chip that actuates
the charge current for each channel so that it would turn off automatically.

[kheston]

MrAl,

Do you have some more info on the new harder-to-do charging technique for lead-acid batteries? I'm in the process of building a charging system. With batteries getting more and more expensive, prolonging their life would be nice. Got a link or other reference?

[MrAl]

Hi again,

There are quite a few references on the web so your best bet
is to do a search. The 'new' method is called the

"Current Interrupt Charging Method"

so a Google search for that phrase will provide you with plenty of reading.

Doing a search for "Lead Acid Charging" might not turn up anything
too useful except the older algorithms.

[jwax]

Good info (as always) MrAl-
But looking over the search, that method originated in 2001/2002.
I'm curious what's new with it these days. (It's still news to me!)
John

[MrAl]

Hi John,

That's why i put 'new' in quotes he he.

I dont know if anything has changed since then, sorry.

Let me know how you make out with it when you get to it,
as it's always interesting to hear about these things.

[IslandEddie]

Check out the UC2906 lead-acid battery charging IC from TI. I have used it in several applications, works great.

http://focus.ti.com/docs/prod/folders/print/uc2906.html

Eddie