DIY solar powered moonlight

[funkman262]

This project was the one that I had the least amount of hope for because there aren't a bunch of threads detailing it like my other projects. I went to good ol' Home Depot and picked up the cheapest single solar power LED path-light that they had and got started. All I needed was the wiring so I just took it apart and threw out everything else.

What I started with:

What I needed:

After unscrewing:

After just ripping the thing apart (I had a lot of trouble removing the solar panel as it was completely glued to the metal base):

From left to right: solar panel, light sensor, circuit board (with LED sticking out other side), and rechargeable battery.

Now the fun part: wiring it all together to the lengths I need:

And the final product:

*Side note: I replaced the white LED that came with it with a blue one. Unforunately, I didn't realize the bulb only had a 30 degree viewing angle so it's more of a spotlight than a moonlight. It does give the rest of the tank a bit of glow also but I'll most likely eventually replace it with a wider viewing angle or adding multiple LEDs for a better spread.

I just put the solar panel and sensor to the side of the LEDs but hopefully it'll be able to absorb enough light from the LEDs to complete the charge each day. I taped the battery and circuit board together to clean it up and placed it on the other side of the heatsink. I'm a bit worried that I damaged the solar panel because I really had to man-handle it to remove it from the base. The moonlight was on all night and I'm going to continue to let it run until it dies. Then I'll turn my fixture on and hope that when I turn it back off tonight, it still works. Considering I really didn't know what to expect with this and I didn't have a set of guidelines to walk me through the process, I think I'm pleased with it. It didn't take a whole lot of time to put together either.

Feel free to ask questions or leave comments

 

[bassplaya12]

Cool, thanks. I've never been concerned with viewing angles on LED's as I've only used them in guitar effects pedals, how do I determine the viewing angle? Just need to find a way to mount this now

 

[funkman262]

I bought the blue LED from radioshack (the only one they had) and it said the viewing angle on the back but I overlooked it.

 

[bassplaya12]

oh, i've never paid attention to that. I'll have to look into that

 

[eznet2u]

Just sand the head of the LED into a slight V shape. That will spread the light out to both sides.

 

[funkman262]

Just sand the head of the LED into a slight V shape. That will spread the light out to both sides.

Interesting. You've done this before?

 

[bassplaya12]

^^ditto that. that'd be fantastic if that worked. Then I don't have to buy any more LED's. Anyone think that I could chain three LED's together off of one solar pack? Or at least two? I'd like to do blue and UV to try to get some glow as well

 

[Daemonfly]

I got 4 of these at home, but was too busy (and lazy) to deconstruct them yet. Just have em sitting on top of tank for now, lol. But, there was a thread on NR about these, and I was wondering if the charging circuit could handle more than just one battery applied. If so, that would give you longer running time with multiple bulbs. But, then you have to see if the solar panel can output enough to charge more batteries...

 

[funkman262]

Thanks for that info. I'll see what I could find about that. But honestly, if I do end up wiring multiple LEDs together, I'd do it in series so that there's no additional draw from the battery and each LED would have a lower output (I believe that's what would happen, I'm no electrician). That way, the life of the battery won't be affected, I'll have a better spread of light, and the LEDs won't be so bright (I want it to be just enough to see in the tank at night without affecting the ecosystem balance in there).

And bassplaya, I was also thinking about adding UV into there. I found these blue and these UV for cheap and free shipping.

 

[Just2Many]

Awesome thread I'm going to try this on my tank? Think you could use this for cheato growth?

 

[King_Richard]

Glad you posted a detail breakdown, mine was rough since I was anxious to see what they looked like.

I did hook up two pnls and batteries to the one circuit board powering one led to give me a longer run time. Mine went from 4hrs to 7-8hrs roughly, maybe a lil longer.

I'm wondering how many pnls and batteries it would take to run 1 cree xre?

Another thought was to run some solar pnls in my sump since I usually leave the lights on down there 24/7 or atleast opposite of my display tank.

Depending on how many pnls it takes to run one xre, we could use our existing lights to pwr say 4-6 xre's or even better yet 12 of em. My thinking behind this is that if we could pwr up to 12 xre's using the energy from 12 xpg's and the solar pnls then we would have an extremely efficient light fixture.

I also had a wire in the wrong spot one time and whenever the sensor sensed light the led got brighter, wierd.

 

[funkman262]

I did hook up two pnls and batteries to the one circuit board powering one led to give me a longer run time. Mine went from 4hrs to 7-8hrs roughly, maybe a lil longer.

Do you remember what the expected operating time was on the package? The one I bought said 10 hours but I got a longer run time than that last night. Mine hasn't turned off since I set it up (I turned the sensor around so that it won't receive any light and shut off. I'm trying to see how long it will run before shutting off). Maybe the solar panel isn't receiving enough light from the LEDs to charge the battery. If that's the case, maybe consider moving the solar panel to the sump light if you can put it in a better position.

Another thought was to run some solar pnls in my sump since I usually leave the lights on down there 24/7 or atleast opposite of my display tank.

Depending on how many pnls it takes to run one xre, we could use our existing lights to pwr say 4-6 xre's or even better yet 12 of em. My thinking behind this is that if we could pwr up to 12 xre's using the energy from 12 xpg's and the solar pnls then we would have an extremely efficient light fixture.

I pretty much know nothing about electronics but you may have something there. The only problem I can see is the efficiency of the solar panel. Let's say you're running a 36W light on your sump. I don't think you can expect the solar panel to collect enough energy to run a 36W LED fixture in the display tank, no matter how many solar panels you have. Maybe someone else can expand on that a bit.



I do have a question though: With these types of solar lights with the sensor, does the solar panel still charge the battery when the the moonlight is on or does the solar panel only work when the sensor detects light? I ask because like I said above, I turned the sensor around to see how long the moonlight will run so the moonlight is on at the same time as the display lights. Thanks.

 

[funkman262]

Just sand the head of the LED into a slight V shape. That will spread the light out to both sides.

I actually just sanded it flat and it worked like a charm. I barely notice any spotlighting and the entire tank gets a nice glow. Thanks a lot for that :beer:

 

[King_Richard]

funkman262:
I don't remember the time it was supposed to operate, never looked. I'll have to check when I go shopping. The light never went out just too dimm to reach the bottom of the tank (24"). I'm also not experienced in this area just know basics really, I'm learning as I go with this.

I'm pretty sure that any energy the solar pnl recieves will charge the battery. I played around when setting this up. I hid the sensor, and could run the light w/no batteries just using the solar pnl but when I put the battery in, the light dimmed indicating to me that it was charging the battery. I also noticed that we used different lights and the solar pnls are slightly different.

 

[j-dizzle]

sweet i have a couple of these in my garage ! i am trying this for sure. thanks for the write up!

 

[bassplaya12]

Thanks for that info. I'll see what I could find about that. But honestly, if I do end up wiring multiple LEDs together, I'd do it in series so that there's no additional draw from the battery and each LED would have a lower output (I believe that's what would happen, I'm no electrician). That way, the life of the battery won't be affected, I'll have a better spread of light, and the LEDs won't be so bright (I want it to be just enough to see in the tank at night without affecting the ecosystem balance in there).

And bassplaya, I was also thinking about adding UV into there. I found these blue and these UV for cheap and free shipping.

That's actually a really good deal. Don't need that many of them, but it never hurts to have them around.

I actually just sanded it flat and it worked like a charm. I barely notice any spotlighting and the entire tank gets a nice glow. Thanks a lot for that :beer:

Post a pic!!

 

[der_wille_zur_macht]

Nice concept.

I'd ditch the electrical tape before it melts/catches fire/etc. Glue the LED in place with a dab of epoxy, or screw a hole in the heatsink and clamp it down.

Another solution to your too-narrow beam problem: reposition the LED down near water level along one of the edges of the tank, aiming up at the heatsink (which is more or less reflective). The beam will scatter and you'll get a wider spread, though at the cost of some intensity. Or, get a tiny square of diffuser (the thin diamond-pattern plastic stuff used on some commercial lighting fixtures) and glue it over the LED.

When playing with different colors of LEDs swapped onto the fixture, watch the voltage drop differences. If the fixture was designed for an LED that wants 3v to run at it's target current of 60mA and you swap on an LED that wants 2.5v for it's target current of 20 mA you might burn it up. (Numbers hypothetical.)

 

[der_wille_zur_macht]

Thanks for that info. I'll see what I could find about that. But honestly, if I do end up wiring multiple LEDs together, I'd do it in series so that there's no additional draw from the battery and each LED would have a lower output (I believe that's what would happen, I'm no electrician).

It sort of works that way. The problem is, that "driver" likely won't run more than one LED in series - it won't be able to provide enough voltage. You may get a very very dim glow. The response curve for LEDs, in terms of output vs. voltage, is nowhere near linear - so, giving half the voltage will result in vastly less than half the light.

If you want more LEDs, you're really best off just getting a wall wart and doing some math, then using a resistor of the proper power and resistance ratings to limit current. Then you don't have to screw with the solar panel or the limitations of the circuit in that commercial light fixture. This is an interesting recycling effort, but probably more expensive than just DIY'ing with a wall wart and resistor, and isn't really going to create any sort of energy or cost savings (the power to run a few common LEDs is probably a buck or two per year.)

 

[bassplaya12]

^^If I wanted like two LED's, a blue and a UV, then get them to spread via a diffuser, how would I need to wire it?

 

[der_wille_zur_macht]

how would I need to wire it?

It depends on what you're driving them with.

In the case of this solar conversion, it's really hard to say, since we don't know anything about the circuit.

In the case of a (lower voltage) wall wart, you could just put them in parallel with their own current limit resistor. If you had a higher voltage wall wart and LEDs that had similar current ratings, you could stack them in series with a single resistor.