Posted by Ray D. on May 18, 2008, 1:31 am
I'm going to be building a solar powered Li-Ion battery charger for a
class project, and I just wanted to ask a few questions that I hope
some of you may be able to help me out with.
First off, here are the specs for the battery:
Camera: Minolta DiIMAGE G500
Battery: Lithium Ion NP-500, 3.7V, 820 mAh
I think I found a good match for a solar cell and IC - the links are
Unfortunately the only specs given for the solar cell are Voc = 8V,
Isc = 310 mA. I also estimated the power dissipated in the IC (as per
the equation in the datasheet) and it is nearly negligible - about 50
Some assumptions that can be made are the device will be used only
during peak sun hours and will be angled perpendicular to the sun for
maximum insolation on the cell. I don't necessarily need the battery
to fully charge in one shot, but merely keep the battery from dying
over the course of a week or so.
So with that said, is there a way to determine (or estimate) the
operating point of the solar cell without an IV curve? Obviously the
output of the chip will be around 4V (battery voltage), so does that
imply the solar cell will be operating at that voltage? The pulse
charger requires a current limited source (which a solar cell is
inherently), so will that mean the solar cell operating point will be
the lowest Vcc at which the IC will operate, in order to maintain the
highest current? I'm just confused as to how determine the operating
point of a device that uses a pulsed current, where the high current
value is basically the maximum the solar cell can provide.
I'd also like to do a load analysis. What do you think would be the
best way to do this? I know most are simply a static voltage and
current value multiplied to give you the power (and also time the
device will be used to yield an energy figure). Again, though, the IC
is a pulse charger and has two main sections in the charge cycle -
constant current and constant voltage (where the current pulses
decrease until it reaches full charge). Should the load analysis
follow the complexity of the device, or would a simple estimate of
output voltage and charging current suffice?
Please do not reply with comments such as 'this project is pointless,
just use a car battery adapter for power', or 'you can buy a solar
powered charger at this website'. I'm doing this project and want
some helpful input! Any suggestions or help in understanding the
solar cell operation with this chip and battery are greatly
appreciated! Thanks so much.
Posted by Roderick on May 18, 2008, 4:50 am
I've done some work with NiMH chargers, so have some ideas. I've
heard, but do not know, that Li is a more forgiving technology for
Your solar cell will actually have two operating points, one for when
the pulse is on, and another for the time in-between. If you charge
at a 1/8 C rate, you will have an average draw of about 100 mA, but
generally these chargers have a very low duty cycle, so the pulses
could be huge, like on the order of 5 to 8 A or so. You will need a
good external FET to pass the current, and I suggest a good stiffening
capacitor across the solar cell. And don't be skimpy on wiring - it
will make a difference at 5 amps.
If you really wanted to, you could develop your own I-V curve for the
solar cell under a given illumination by varying a resistive load, and
drawing the graph yourself. Or, look up the characteristics of
another photovoltaic cell. All silicon is about the same in shape,
only the magnitude varies. I will say offhand that the max power
current will be only about 10% less than the short circuit current,
and the max power voltage will be the open circuit minus about 20%.
Put another way, if your cell has a short circuit current of 300 mA,
you can expect it to deliver 270 mA all the way up to 6.4 volts (80%
of open circuit). This depends on temperature and illumination, of
course, but that will give you a general idea. Try pulling up the
spec sheet of some commercial solar panel, and you can scale the
values to your cell.
Posted by Ray D. on May 18, 2008, 3:53 pm
I kind of assumed the solar cell would have two operating points when
it is pulsing, but how could that solar cell produce 5-8A when it's
only rated for 310mA? I was thinking the constant current portion of
the charging cycle would operate at around 250mA and the pulsed
current would be that current turned on and off at varying rates until
charging has completed. Also, the IC has an internal pass MOSFET and
does not require an external one.
Posted by Roderick on May 19, 2008, 6:25 pm
The big current pulse comes from the stiffening capacitor, which you
hopefully are charging up between pulses. For best energy transfer,
you want to have the solar cell charge the cap up to a little past 6.4
volts, and when the capacitor dumps during the current pulse, expect
the voltage to drop to 4 volts, or whatever the battery limits you
to. You'll need to choose the minimum capacitor value yourself, based
on what current the solar cell actually puts out, and i = C dv/dt.
In the charger we designed at work, we did not trust the on-chip pass
transistor for the currents needed, but you'll probably be fine for a