Candle Power?

no

What you appear to propose does not even make for a closed circuit, let alone anything that can drive above the threshold of a diode voltage drop. So I repeat Edignan's succinct answer, no.

Not really a challenge for everyone.. Which candle gets the cathode/anode?

Total bollocks

see http://message.snopes.com/showthread.php?t=48319 and http://hackedgadgets.com/2007/02/02/how-is-the-fake-candle-power-trick-done/

Nuff said.

Well at least we now know where this absurd idea originates. Thanks for the links.

No matter how many times you tell people that they shouldn't believe everything you see on the web, some still will.

It's not even a very good illusion.

Only if you say the magic words.

the LED will not light, but will cause your candles to burn brighter. This effect, called the single malt infarction, has been simultaneously recorded by Lyndoor Industries and Dragon Balls LTD in separate patent applications. We are currently litigating.

You can convert heat of candles burning to mechanical power via sterling engine and run a teeny generator...But your assumptions here with present design is fantasy...

This is a parlor trick. If you want the candle power to light the led I would recommend that you use the trick we used many years ago to run a radio off the kero lamp on the kitchen table. Use dissimilar metal junctions over the flame with like junctions out in the room temp air. Now days this is known as a thermocouple. back then we just knew it made a small amount of electric current (Dc). Combining many into a series - parallel arrangement produces enough current to run the old radio. On kero with no ties to the grid. It worked and will still work. So why use trickery when you could do it the way we did back in the early 50's.

This has the makings for an absolute incredible Middle or High School science fair experiment. Don't answer this, as I send my students to CR4 for research purposes, but think about what student could learn about practical physics, chemistry, metalurgy, and a ton of other applied sciences, considering and researching what exactly makes a thermocouple generate electron flow. This would lead to "Which dissimilar metals, and with what dissimilarities, would create the greatest current flow? Do you generate more power through a chemically bonded, physically bonded, or unbonded junction?" (i.e., soldered, welded, or just twisted). I can think of a raft of other questions my students would be able to pursue. And all of the research would lead toward, if not actually to, practical applications.

Thanks for this input. I never even thought of applying the thermocouple principle in this way.

BTW, and this one I could use a public answer for, does a Peltier junction work in a "thermocouple-ish" fashion? If I apply heat to one side, or cold to the other, does it generate current flow?

I had a Peltier once, but no longer. And I didn't think to test this while I had it. I wonder if you couldn't, with suitable heat-sinking, make a direct-to-dc field generator to lay in your campfire. Thermocouples could do it. Would a Peltier junction device?

How a Peltier-Seebeck device actually works is a quantum mechanics operation that will probably be beyond your high school students. You can use a Peltier-Seebeck device to directly convert a thermal gradient into electric power. It is one of the methods that companies are presently exploring for solar power conversion. Photo voltaic cells directly convert visible light into electric current but a significant part of the solar power we receive on the surface of the Earth is infrared energy.

That's what I thought he was trying to show.