“Her invention, a flashlight that is powered solely from hand heat, took second place at the competition.”
“At roughly 24 lumens, Ann’s flashlight’s brightness falls shy of commercial flashlights, which output dozens if not hundreds of lumens.”
They are being picky! If I were stuck with dead batteries in my torches (hundreds of lumens potentially :Sp ) and needed light, 24 lumens would be fine thank you very much. Good luck to her!
Someone mentioned her project to me recently - they saw it on TV. At the time my response was a total “meh.” I figured that based on my experience with peltiers the device would be saturated quickly. When I played with peltiers it was at higher wattages, so that may simply not apply at such low power.
The Google Science Fair writeup on the subject has been educational to me: Google Science Fair 2013 - The Hollow Flashlight
She’s doing a massive step up of 100:1. I didn’t understand how boost drivers work or the mechanics of why they were so inefficient and I still don’t, but her writeup shed a little light on the subject. I guess the ones we use must feed stuff through an oscillator, a transformer, and then a rectifier… and at some point current regulation must happen as well!
I agree, we got by on stock Mini Maglites for years, they are spec’ed at 14 lumens. On the other hand, 24 lumens is probably 100% BS. Her experiment seems to have used a single 5mm LED per flashlight. The flashlights she’s pictured with may be later designs featuring 2 and 3 LEDs. Even 3 fully driven 5mm LEDs won’t produce 24 lumens as far as I know - maybe 6 lumens?
She’s getting 5.4mW to the LED array based on the big flashlight in her writeup, the peltiers generate about 10x that much and the rest is lost in conversion. That’s 0.00216Amps (2.16mA). I’d say maybe she’s generating… 0.2 lumens?
Lets call them 24 Chinese Lumens :bigsmile:
or 24 worthless reporter lumens. I don’t see where Ann Makosinski ever made that claim.
For her original inspiration (lighting for schoolwork) I wonder why not convert this concept for use as a headlamp? The forehead has double the surface area and you could use commodity extruded heatsinks on the peltiers… assembly should be at least as simple as the tube flashlight I think. I don’t know if we can source the same wattage per surface area there though, and the flesh is awfully thin (won’t conform as well to flat peltier surface) so that may be what prevented her from going that route.
Really it would be nice to see a lot of additional research on this. For me the big question is whether convection actually helps. That has a bearing on future design tweaks.