77Views8Replies

# design for a battery charging circuit using solar panel? Answered

i am using a 10-w solar panel for charging battery- 4v, 6ah(3 nos 4V, 2ah battery in parallel). This is used for a GPS tracker mounted on a roof top of loco wagon . usage is 24x7x365 . i have found that gps tracker is consuming about 150ma, 4v. But after 1 to 1.5 days of usage the battery discharges and tracker stops working. one more thing the sunshine availability is approx-2-3 hrs /day. the loco wagon is standing in bright sunshine for atleast 2-3 hrs in bright sunshine everyday. where i am wrong? thanks in advace

Tags:

## Discussions

4V from the battery for a device that needs 4V means the device will shut off quite soon.
A battery decreases the voltage while discharging.
And 2-3 hours of sunlight is not even close to getting the batteries charged - do the math with the rated charging current your self ;)

Strangely my calculations show that the solar panel should generate about 20WH-30WH a day based on OP's figures, or about 78WH on a very good day based on my figures. I also figure the GPS is drawing 14.4WH per day, My guess for why OP's solar panel is not doing the business is that is it way overrated, or the classic problem of assuming the product of short circuit current and open circuit voltage gives a nominal power figure.

The real question is how long do the batteries hold up the GPS, if the GPS has a requirement for 4 V and the batteries and panel ALSO give 4V. If the GPS drops out at 3.95V, you are screwed.

This isn't a problem (necessarily) in POWER, but battery volts.

As the batteries charge, the short circuit/open circuit volt problem converges to the open circuit volts.

Downunder35m already mentioned that, and I agree. He also said that 2-3 hours is not long enough, that is what I am not sure about considering what I figured. Some sort of switch mode battery charger circuit designed for use with solar cells would probably be the most efficient way to get that important impedance matching stuff right. IDK what circuit would ideally convert between the optimal voltage/current output of the solar panel to a charging voltage/current for the battery.

That's what MPPT circuits do

MPPT, google tells me Maximum Power Point Tracker.... Umm, sounds like the right tool for the job, Looks like I've got some research to do!

In short, I figure that 10W solar panel should easily supply enough power, at least in ideal conditions. However clearly it is not doing the business for you, so my question is how did you get that 10W figure? How big is the solar panel? Why not do some realistic measurements. Hook up the panel to a load (a light or something) and measure both voltage and current. See if the product is even close to the 10W you are expecting.

The amount of energy the solar panel will produce in 1 day is actually a little tricky to calculate/approximate. It would be best to look at real data to get an a reasonable average of where you live. But my point is, that 10W figure assumes that you get the angle perfectly orthogonal to the sun, that you get a certain luminous intensity from the sun (as in a good sunny day, not cloudy, and that you are close to the equator rather than the poles), that the solar panel remains spotless and clean, and many other factors like temperature. So there is no way you will achieve that figure under practical conditions. So below I try to make a reasonable calculation at what you should expect a solar panel to produce in 1 day.

--------------------------------------

WARNING: basic calculus and trigonometry gets involved below!

Obviously at night the solar panel produces no power, so means we get less than 12H of light from the sun, and less than 10W * 12H = 120WH from the sun.

But how much less? Well clearly you will get the most light at noon (ish) when the intensity of the light hitting the panel is at it's peak, but as the sun moves throughout the sky, the light intensity hitting the panel increase during the morning, hit a peak around noon, and drop off in the evening. So the amount of light hitting the panel can be approximated with a sine curve. So I take that 0-12H period, assuming the peak 10W is generated at noon, and stretch the sin curve to fit over a 12H day. The equation comes out like the watts generated = 10*sin(15x). If you graph that on your graphing calculator with degrees, you will see that the first hump from 0 to 12 (0 to 12 hours) is the wattage the solar panel will produce. this curve makes lots of assumptions and neglects some stuff but is should be close.

So now we have a mathematical model of the sun radiation throughout the day, just integrate that function from 0H to 12H to get the total WH produced in the day! I am lazy and did this on the calculator but it is really easy to do by hand. I figure you will get about 76WH/day IDEAL conditions.

I think you are capable of figuring out if that is enough energy per day to power your devices and charge your battery ;)