Solar LED Shed Lighting

We recently made some major landscaping changes to our yard, including adding a small shed. The shed doesn’t have an electrical hookup to the house, but I wanted to make sure that it had basic lighting. I decided that this would be a good way to use some of the LEDs that I had lying around from experimenting with building bicycle headlights.

I first looked at commercial products that put all of this together, but I couldn’t find any that seemed to be bright enough. Most were using dozens of dim/low cost white LEDs without defining out bright there were. Plus I already had the LEDs (in hindsight the cheapest part of the system), so I thought I’d build my own.

This is a system diagram showing how everything fits together:

Parts list:

  • Solar Panel — Brunton SolarFlat 15 (about $120)
  • Charge Controller — Brunton SolarController (about $25)
  • Battery — 12V, 7AH lead acid. I had this already from a UPS, but they are about $20 new.
  • LED Controller — TaskLED MaxFlex4 (around $40)
  • 6 LEDs — Cree XR-E Q5 (around $6 each)
  • Aluminum Strip — 1″ wide aluminum channel from the hardware store (around $10)

My goal was to stay under $200, but I think the total ended up being more like $250. It would be nice if there was a single source for all of these components, but I had to use multiple sources. The solar panel and charge controller came from Amazon, the LED controller came from TaskLED and the LEDs came from DealExtreme.

The panel is just placed on our roof and plugs into the charge controller. That in turn plugs into the battery, which is placed on the top of one of the shed walls (inside the shed).

The LED strips are made of 3 LEDs mounted on a 4 foot section of aluminum channel. The LEDs are wired in series and the wires run inside the channel.

The LED controller is mounted in a small box with a push button and a status LED. I used a piece of copper pipe as the heatsink, and that is also bolted to the project box. The box was rescued from another project,which is why it has some extra holes.

The controller itself is tiny (about the size of a quarter) which made it a little trickier to work with. It is designed to be small enough to work in a flashlight. This is a pretty advanced controller and the selling feature for me was having an auto-off function. It is setup to start dimming the lights after they’ve been on for 15 minutes,and they will be turned off completely after about 20 minutes.

That explains howthe system was built, but how long does the battery take to charge, and how long can the LEDs stay on before the battery dies? To answer both of those we need to look at the capacity of the battery. It is a 12 volt 7 amp hour battery, which means it stores 84 watt hours of energy (12volts *7amphours = 84 watthours).

The solar panel puts out 15W in optimum conditions, but Seattle is far from optimum and I didn’t make any attempt to aim the panel. I figure it is probably worth about 7 watts for 8 hours a day, which is 56 watt hours. So in roughly1.5 days the battery will be fully charged.

The LED controller is outputing 350 milliamps. There are 6 LEDs which have a forward voltage drop of 3.7 volts. So the total watts consumed by the LEDs are: (3.7 * 6 * .350) = 7.77 watts. These LEDs are about 10x more efficient than a normal light bulb, so they are producing about as much light as a 70-100W incadescent bulb. The controller is about 80% efficient, which means that about 9 watts are actually consumed from the battery when the lights are on. That means the runtime is about 9 hours. I could run the LEDs at 700ma each and still get a runtime of4or 5hours, but I don’t think the extra light is necessary. In reality the lights will probably only be used a maximum of an hour a day (if Christine was potting a lot of plants).

If this system is reliable I think I will expand it to include some LED landscape lighting. I’ll probably need to add another panel and use a larger battery, but the basic system configuration will be the same.

Photos

6 Comments

  1. Rory says:

    this is a great project. I”m tempted with something like this for my attic, where i don”t have any lights currently, and don”t really want to run a circuit up there for it.
    as far as optimum conditions, in seattle you could actually be running better then optimum during the summer, with the 14-16 hour days of sun. is there some limit to it that you need to watch out for?

  2. AlexWetmore says:

    There isn”t any limit to worry about. The charge controller stops charging the battery when it is full.

  3. Jimmy Livengood says:

    Fun project!

    This reminds me of what the Light Up The World Foundation is doing(and maybe others) -providing solar-powered LED lighting to rural areas.

  4. dan says:

    that is very cool!

  5. Patrick says:

    Have you done any more thinking about applying this to landscape lighting? What kind of LEDs would you use? Were you going to build your own housing, or go with a commercially available unit?

  6. admin says:

    I haven”t given it any more thought Patrick. Our shed lighting is actually not working at the moment, our panel failed and I haven”t had a chance to send it in for a warranty replacement (the manufacturer did say that they”d cover it).

    I was thinking that I”d probably make my own housing. I haven”t found landscape lights that use LEDs that I like the design of. Most of them use very weak LEDs. I haven”t pursued the option too much because building custom housings with a driver and a good LED per housing was adding up to be pretty expensive and time consuming per light.