[1] How near-sighted? I once went into a place that makes glasses in one hour (while you wait). When I handed my prescription to the optician, she read it, and then screamed (in front of a packed waiting room) “OH MY GOD! We can’t make these!”. Now that’s near-sighted!

Another variation would be to use different colour LEDs (Think of sort of a
binary display clock with the different binary digits represented by different coloured LEDs).

Anyway, just some fun ideas to think about.

Dave

You might also consider some of the super-efficient red LEDs.

As for diffusing the cases, in an extreme case, I’ve even been known to take to the epoxy case with a rasp and turn the dome into a flat surface, which broadens the viewing angle significantly. Don’t go too deep with this, since you don’t want to risk cutting the bond wire to the top of the chip. But, it can be done (I’ve also drilled holes into the epoxy in order to glue a fiber optic into an LED (not that I’d recommend this for the sunrise clock idea).).

As for the clock idea, I’ve long thought about building a clock which, when the alarm goes off, it gradually increases the room lighting, simulating the natural increase of light at sunrise. However, I haven’t yet done anything with the idea yet. My preferred implementation would probably be to use a microcontroller to build the clock with, and have that vary the triggering of a triac which controls an incandescent lamp (Yeah, remember all about safety with regards to isolation from the AC line and such.). I know that incandescents are much less efficient than fluorescents (regular or CFLs), but it might be easier to building the circuit using an incandescent. The other approach might be to use a fluorescent and pulse width modulate the power to it (But, that will get tricky, what with having to factor in ballasts, and the fact that glow discharge lamps are extremely nonlinear devices; Plus fluorescent lamps are even more nonlinear since the glow discharge characteristics depend upon the temperature of the gas and how much of the Mercury has vapourized).

Dave

P.S. Welcome to the wonderful world of optics. I used to work in an opto-electronics lab, and it’s an eye opening experience (if you’ll pardon the pun!).

In the pictures above, I was running about 7V through all the LEDs, which is 50mA order of magnitude. Earlier I had run them up even higher, probably to or beyond 100mA, briefly.

Oh, and then there was the yellow one I burned out because I was using a 100Ω inductor instead of resistor. (Well, it looked like a resistor — even in a tan package — just a little lumpier.) It glowed a strange orangey-red color as it was dying.

http://www.sparkfun.com/commerce/product_info.php?products_id=602

They are expensive, but *very* bright. How many volts are you using to drive the LEDs? I know LEDs like the Luxeon need heatsinks and often a PWM controller is used to regulate power to the LED.