I'm trying to repair a set of solar powered garden lights. This wasn't working properly, and so I'm in the process of attempting to repair it. Unfortunately the circuit board was covered with some sort of polymer and I ended up destroying it trying to get everything out. Now I've got to rebuild it, and I think I have a solid plan. I'm hoping that I could get some help and just make sure that I'm going in the right direction before I invest my time and money into it. I also understand that I could just pay $20 and buy a new one, but there wouldn't be any fun in that. Additionally, I'm not certain about all the calculations, so if someone could help me out with them it would be appreciated (I don't expect someone to do the calculations, just give me some guidance as to what to calculate )

Details
=====
- The garden lights have 8 LEDs (5 mm, white, I don't know any other details)
- The lights twinkle
- The solar array recharges two Ni-Cd batteries (1.2V, 900 mAh, standard charge 90 mA) during the day, and uses the battery power at night
- Solar panel gives roughly 2V in direct sunlight, not sure about amps (I believe it's the 55x55 mm panel listed on this website: http://www.sunyooo.com/SOLARPANEL1.asp - 2.5v, 80 ma, 0.2 W, but I'm not 100% sure)
- There's a light-dependent resistor in the solar panel to facilitate charging

Basically that's what I had, and that's what I want to make, only better. The lights that I had were be very dim at night, and wouldn't last too long before the lights went out (1-2 hours). I would like to have lights that last longer (i.e. 3-5 hours) and be as bright, or brighter.

I've done some research online, as well as looked through some books, and I've come up with 3 different ways to make this circuit. At the moment I'm leaning towards a 555/556 timer, but there might be an easier way.

#1) 555 timer:

Because I have 8 LEDs to light, I was thinking of using a voltage doubler circuit that I found here - http://www.reuk.co.uk/Voltage-Double...ry-Charger.htm. I'm doing this to provide more charging ability to the batteries, and hopefully get them fully charged during a day of full/semi sunlight. Also, I think, this would allow a small voltage from the batteries to light them up.

I'm also thinking of having the 555 timer be astable, and have the LEDs connected so that they will be "sinked" and "sourced" (I don't know if those are the right terms, but I got the information here - http://www.kpsec.freeuk.com/componen...htm#sinksource. I was thinking this would make the LEDs twinkle ... 4 sourced, 4 sinked; 4 off, 4 on, etc.

Questions
======
- If I were to sink/source the LEDs, would I need a resistor for them? What size should it be? Where should it be placed? I'm assuming the LEDs are 2V each, so that's 16V going to them.
- If the voltage doubler could put out 16V would there be a need for the resistors?
- How could I connect an astable timing circuit with the voltage doubler, and make it compact enough to fit into the light that I have now? Can I combine them into the same 555 timer circuit, or will I need two 555 timers?
- Can I use a 556 timer?

#2.) Reverse-engineered light
This light is shown in this schematic - http://members.shaw.ca/novotill/Sola...t/circuit4.png. I noticed that there is only one AA battery (not my two), and there is an induction coil as well. Would this be a better way to go?

#3.) Solar light
These two schematics (http://www.talkingelectronics.com/pr...olarLight.html) are very similar (I think), but again there's only one batter and an induction coil. Is there a benefit to these? Also, this website discusses a lot about oscillations and positive feedback. I don't know much about this, and it seems that this is more of a complex version. I don't know

Basically I'm looking for a relatively easy, but enjoyable way to fix this lamp. I'm a beginner with electronics. I can solder, and have access to the majority of equipment that I *think* I need to fix this. Any info would be GREATLY appreciated.

Bryan

 

Here's a circuit drawing of the 555 which I was thinking of using. Would it work?

Attached Files

IMG_0001.pdf (1.02 MB, 22 views)

 

Everything should be up now. Any help would be greatly appreciated.

 

If your solar cells put out 0.2W in direct sunlight, then your 2160mWh (900mAh * 2 * 1.2V) of batteries will be about 65% charged after 8 hours of noonday sun. But it isn't noon for 8 hours straight! So the battery will be about 20% charged or less (400mWh) after a very sunny day. At night, if the 8 LEDs draw only 350mW (10mA into each LED at 3.5V) total, you'll get 1 hour, maybe a little more.

There is no 555 circuit that can improve this. You need a more powerful (larger?) solar array.

 

I don't remember such a detailed original post.

 

What if I were to only have one battery? Would that workout better?

I've been doing a *lot* of research lately, and I want to get everything right

 

A single battery wouldn't change the total energy available. You need more efficient or larger solar cells, and/or more efficient LEDs. The battery doesn't figure in the equation unless it's not big enough. In the scenario I see, one cell should be enough. Using two makes some circuits easier to design.

 

So if I were to build this, which option would be the best to use? A timer or non-timer circuit? If non-timer, which one would be more efficient and why?

 

The 555 circuit has some flaws; mostly if Vc is greater than the turn on voltage of 2 LEDs in series they both will be on without current limiting.

I like the talkingelectronics circuit #2 best because it only needs one inductor, but I also like the way circuit #1 doesn't need a CdS cell.

Either circuit might over-discharge the NiCd if it doesn't get any light for a while (such as in storage or the winter). And note that they can use NiMH instead without any changes.

 

Would this be the same if it's with 8 LEDs in series?

Is there any way to prevent the over-discharge? Could I use a diode somewhere? If so, where?

I truly appreciate your help

 

Also, you mentioned that the voltage doubler would not work ... why? Would a voltage multiplier work? http://www.reuk.co.uk/DC-Voltage-Mul...cuit-Plans.htm

 

I missed the part where I said doublers would not work.... I just don't like two LEDs in series without any current limiting.

 

I'm sorry, I took this to mean that a 555 circuit wouldn't effectively double/multiply the voltage.

I guess I was thinking of making the main circuit look like this [solar circuit.pdf]. Depending on how much power I need to give to the LEDs (would it be power, voltage, or current that I'd be worried about?) I would assume that I'd have to have a doubler/multiplier hooked up [doubler.gif]

I guess the main hook-up would be the solar panel to the doubler, and then the Vsource of the doubler lead into the +Vs on the solar circuit.pdf (remove the solar panel).

Would that work?

Attached Images

Doubler.gif (3.3 KB, 14 views)

Attached Files

Solar circuit.pdf (529.1 KB, 16 views)

 

The 555 can double the output voltage, but the resulting current will be 1/2 (no new power created). And, the 55 driving the LEDs would make them flicker like you want, providing the Vs is less than 8*Vf of the LEDs.

But neither circuit will create energy. All of your energy must come from your solar cells.

 

I should correct that: The 555 requires 4.5V input before it can double anything. It cannot double 1.2V from a NiCd or NiMH.