Hello all,

I'm new here and new to electronics in general. I am an artist and architect very humbly asking for your advice. If this type of inquiry is not welcome on your forum, I apologize.

I want to create a completely sealed block of resin internally lit by a battery powered LED circuit that lasts as long as is currently possible. I'm shooting for years. It should be about the size of a beer can and provide enough light to read by if held very close to the page. Is this possible? I have found examples of this being done in the past but with no stated life expectancy. How could I roughly calculate this? How long do you think an optimally designed circuit might last?

potential problems:
-I understand the heat damages LEDs over time. Encasing the circuit in resin may accelerate this as the plastic will inevitably retain the heat generated by the light. Is this actually a problem? How might I mitigate it?
-From what I have read, LEDs never truly burn out but rather slowly dim. For my purposes, it would be best if the bulb died fairly suddenly. Could I create a circuit that switches off or self destructs when a specified low level of power consumption is reached?

Any comments would be greatly appreciated. Thanks!

 

Huge amount of info here which answers most of your questions. They do every type and flavour.

http://catalog.osram-os.com/catalog...Oid=000000000000010100010023&lang=en&catId=EN


Out of interest , do you want to incorporate the battery in the resin block or are you running 2 wires to a separate battery/power supply?

 

I wouldn't worry about heat or the LED fading over time, if you all you have is a beer can with batteries in, it will have to run at very low power levels.
Just as an example, you need 3 AA cells to give enough voltage for a white LED.
Lithium cells are around 3000mAh so with 3 cells and the LED at 1mA (fairly dim) it will run out after 3000 hours which is about 4 months.
Put 12 cells in there and it will last 9000 hours (about a year) at the same current.

 

You could add an inertial switch that turns the LEDs on when the "can" is shaken, and a timer that keeps them on for a few minutes, and they turn off until the can is shaken again. This would prolong the life of the batteries. In addition, you might embed a small solar cell inside the acrylic to recharge the batteries.

 

Four 19Ah lithium D cells would give you 8mA for more than a year (24 hrs day) and that's enough current to provide ~5 lumens (enough to comfortably read) using the best available LEDs.

Is a non-contact rechargeable system an option?

 

Four 19Ah lithium D cells would give you 8mA for more than a year (24 hrs day) and that's enough current to provide ~5 lumens (enough to comfortably read) using the best available LEDs.

Is a non-contact rechargeable system an option?

I like that, basically a hidden coil that recharges the batteries with no electrical contact.

There is an old concept call throwies where you light up LEDs with no battery included. They couple into a pulsating magnetic field from a coil hidden under a blotter. Just put the LEDs on the surface of the blotter and they glow, with no apparent power source. A common theme is to make them in the shape of bugs.

http://www.instructables.com/id/LED-Throwies/

Another different kind of throwie is a LED with a watch cell battery and a magnet, you can stick them to vehicles and other metal surfaces. It is a form of art unto itself.

 

Four 19Ah lithium D cells would give you 8mA for more than a year (24 hrs day) and that's enough current to provide ~5 lumens (enough to comfortably read) using the best available LEDs.

Pretty expensive if this is a representative price ($20 each):
http://www.futureelectronics.com/en...anical/batteries/Pages/3831500-TL-5930-S.aspx

 

A "joule thief" circuit to drive the LED (I have successfully driven 9 white LED's off of 1 AAA battery with a joule thief circuit) and an inductive charging unit to charge an internal battery/capacitor bank might make this a pretty interesting project......

 

Pretty expensive...$20 each...

That's right, it would be a $100 light. This is the "is it possible?" stage of development, not the "how cheap can we make it?" phase. The OP specified only self-contained longevity, size and luminosity.