# Small solar charging basics

Thread Starter

#### ryzekiel

Joined Dec 17, 2011
8
Hi everyone,

I'm working on a simple solar charger for a bicycle light. Something that I'm stuck on is the relationship between solar panel voltage, battery voltage, and charging.

I have a circuit that consists of a 3V panel, BAT41 diode for reverse current blocking, and two 1.2V NiMH AAAs in series.

If my panel is at 3V and the diode drops it (by ~0.6V, I believe), will the batteries ever be able to fully charge? If not, is it safe to put two 3V panels in series and charge @ 6v?

Thanks,

-ry

#### Kermit2

Joined Feb 5, 2010
4,162
Might I suggest using two diodes and charging the 1.2 volt cells in parallel. This would give you a charging voltage of 1.6 to 1.8 volts, and depending on the current draw of the depleted batteries might go even lower at the beginning of charging due to loading effects on the solar panel. Going to the series arrangement would require the full 3 volts or even a touch more to fully charge the batteries and you would lose .6 to .7 volts with a single diode.

Thread Starter

#### ryzekiel

Joined Dec 17, 2011
8
That's a great idea. The LEDs used would require a higher voltage, but I could put a switch in to change back and forth between series and parallel. Series at night for the LED, parallel at day for charging.

Would you mind explaining how you got to 1.6-1.8 V in the parallel arrangement? Also, Is it not safe to have the two panels in series @ 6V and the batteries in series @ 2.4?

#### Kermit2

Joined Feb 5, 2010
4,162
I got the 1.6 to 1.8 volts by subtracting two diode voltage drops from 3 volts.

Unless the current output of the panels is small, using a 6 volt source to charge the two series cells could result in a serious overcharge and damage to the batteries. Low output current, like 30-50 mA would mean a slow charge that would have little chance of overcharging the batteries unless left on for several days. Anything lower would not charge them in sufficient time for use that night, and higher currents, over 100 mA or so, would be dangerous unless you constantly monitored the process with a voltmeter.

Thread Starter

#### ryzekiel

Joined Dec 17, 2011
8
Oh, I see. If the batteries are wired in parallel and have common nodes at either end, is there a need to have two diodes?

I should have specified the panels I was using. They're rated as follows:

V: 3V
I: 25 mA
Voc: 4.1 V
Isc: 35 mA

The 6V arrangement shouldn't exceed 25 mA. I have four of these to play around with so there are a few different wiring possibilities.

Last edited:

#### Kermit2

Joined Feb 5, 2010
4,162
The six volt arrangement should be fine at that low current level for charging two AA's in series. Perhaps using two or three diodes in series to lower the voltage would be a good idea just for safety of the battery during charge. Three diode drops would lower the voltage to about 4 volts, which is closer to an ideal level for charging than 6 volts. If you have 4 panels, then a series/parallel arrangement with 6 volts and 50 mA, lowered to 4 volts with the diodes would give you about 400 or so milliAmp hours of charge in an 8 hours day light exposure. If you don't discharge them more than that during use you should be happy with the results.

#### fireofenergy

Joined Dec 17, 2011
20
I'm curious about the light... because white, blue and green leds need about 3.2v... which is perfect for 3 AAA's (after resistor), in which case I would wire up two sets of two panels in series, and wire those up in parallel.
If it is a red (also yellow, and orange leds), then two AAA's charged as Kermit2 suggests.
If it is a white light, then there is a boost circuit which will could drain the batteries past their safe discharge level (~.8 - 1v each)

With leds, I try to always make sure that they are the battery protection from over discharge (as white leds will not go below 2.5v which is just about safe for 3 Nimh, and I assume the same ratio with the red leds).

Thread Starter

#### ryzekiel

Joined Dec 17, 2011
8
I'm prototyping the rear light first. It's red with Vf=2.1-2.3V

I didn't realize that different color LEDs have different voltage requirements. I suppose a 3x AAA arrangement would be the best solution for the front (white) light.

I'll also have to do some reading on battery over discharge. I didn't realize that trying to pull all the juice out of a battery could be harmful to it.

Sometime soon I'll wire up two panels in series and do some arduino datalogging to see how much current they're putting out over the course of a day.

Thanks for the tips, everyone. I appreciate it.

--
ry

Thread Starter

#### ryzekiel

Joined Dec 17, 2011
8
A quick note on over-discharging, though. I think for most cyclists this is a non-issue given the ratio of night to day riding. I ride everyday and 90% of the time I don't need my lights because the riding season has more hours of sunlight.

I'm working on this because of an accident I was in a few months ago. Like I said, I don't ride much at night so when I got on my bike to ride home I was surprised to see that it had died over the month or so I had gone without using it. The accident was nothing major, but it could have been avoided if the oncoming vehicle could have seen me.

So! That's where the inspiration came from for this little diddy. If the sun doesn't keep it topped up, then they can be easily charged indoors since they're common AAAs.

#### Audioguru

Joined Dec 20, 2007
11,248
I have seen many bike riders and electric bikes riding at night with very dim lights. They don't know that the charge on a battery runs its voltage down when it is used. When the voltage is down then of course the LEDs are dim.

A solar panel is rated at noon in the middle of summer with the panel facing directly at the sun. Most of the time it is not so its output current and voltage will be less.
Cloudy days in winter and they will produce nearly nothing.
How will you move them so they always point directly at the sun?

Ni-Mh cells are 1.4V to 1.5V when fully charged depending on the charging current. With only 2.3V (3V minus 0.7V for the diode which is only 1.15V each) then your cells will be barely charged and the charge might last for only a few minutes.

Your solar panels are a joke. Solar garden lights costing \$2.00 have double their mAh rating. Would you light a bike with a very dim solar garden light?

The text for your Cheap (CHEEP, CLUCK, CLUCK) Chinese LEDs say 30 degrees "VLewLing Angle" but the graph shows only 8 degrees. Since they are unbelieveably bright then I think they are focused into a very narrow bright beam that cannot be seen a little to one side. Disasterous for a bike at night. You need a bright wide angle beam of light and you need more voltage and current from a solar panel.

Thread Starter

#### ryzekiel

Joined Dec 17, 2011
8
Hi Audioguru, Thanks for the input.

The panels are very low output. I assume that's a necessary compromise for flexible cells. However, I think for this application (trickle charging batteries which are used infrequently) it is possible to use them.

We've discussed using a 6V arrangement to overcome the scenario you've described (3V being to low to charge two 1.2V cells in series.)

I'm not sure what you are getting at with your garden light comment as the lights are never run directly off the panels. Could you elaborate?

You're absolutely right about the narrow beam. I measured it and it's actually >30 degrees, which is still narrow. I have a few differnt LEDs to try out, but I'd like to try this as it is high output. Once I get it on the bike I can swap lights and see what is more appropriate.

#### Audioguru

Joined Dec 20, 2007
11,248
Most solar garden lights use a 2.5V/60mA solar panel to charge one AA Ni-Cad battery cell all day in summer. Then at night the AA cell powers a voltage stepup circuit that delivers only 10mA or less to a dim white LED for about 6 hours.

If your LED is used only infrequently then it might have its battery charged long enough over a few days or a week then the LED can glow dimly for one evening.

A "high output" cheap Chinese LED is an old dim one in a case that focusses and concentrates its light into a bright narrow beam. A bright modern LED is made to be bright then its case shows it as a wide beam.

Thread Starter

#### ryzekiel

Joined Dec 17, 2011
8
If your LED is used only infrequently then it might have its battery charged long enough over a few days or a week then the LED can glow dimly for one evening.
It will be used infrequently. As I understand it, NiMH loses 20% in the first 24 hours and 10% every month afterwards. If the bike is locked up outside and gets a conservative average of 10mA (considering the rated max of 50 mA in the arrangement of four panels discussed before) over an 8 hour period, then that should more than make up for the self discharge and occasional use.

If this battery is topped up and used infrequently then I can see it putting out more than a dim glow. One thing to keep in mind is that this light will never be on for an evening. It will be on for 10-15 minutes at a time, and it will be flashing.

A "high output" cheap Chinese LED is an old dim one in a case that focusses and concentrates its light into a bright narrow beam. A bright modern LED is made to be bright then its case shows it as a wide beam.
Like I said, this is why we make prototypes! I have plenty of LEDs and if this CHEEP, CLUCK, CLUCK LED is not performing then it wont make the cut.

--
ry

#### Audioguru

Joined Dec 20, 2007
11,248
Like I said, this is why we make prototypes! I have plenty of LEDs and if this CHEEP, CLUCK, CLUCK LED is not performing then it wont make the cut.
Everbody who buys CHEEP, CLUCK CLUCK Leds from Ebay are sorry that they didn't buy Name Brand LEDs with a detailed datasheet from a REAL electronics parts distributor.

I helped a guy on another website then he sent me hundreds of name brand Lumileds Luxeon SuperFlux LEDs. Their forward voltages are almost the same and they are extremely bright with a very wide angle.

Thread Starter

#### ryzekiel

Joined Dec 17, 2011
8
Good to know. Thanks!