DIY 100W LED flashlight

Posted on Tuesday, April 15th, 2014 in hacks, LEDs by DP

In this video Julian Ilett demonstrates his DIY 100W LED flashlight:

I discovered that due to a lucky co-incidence of voltage and internal resistance, a 100W LED can be connected directly across the terminals of two 18V Nickel Cadmium power tool batteries. And that means you can build a 100 Watt (7,500 Lumens) flashlight for less than $10 (not including batteries).
Note: Don’t try this with other battery types – you’ll almost certainly fry the LED!

Via Hacked Gadgets.

This entry was posted on Tuesday, April 15th, 2014 at 9:00 pm and is filed under hacks, LEDs. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

10 Responses to “DIY 100W LED flashlight”

  1. KH says:

    Over-hyped and a non-practical toy. Is it going to work for 1 minute with that heat sink? Didn’t even check (or mention) the current draw. This is never going to work as a practical flashlight.

    The LED module is nice, but cooling the thing at 100W is non-trivial. It will die quickly with bad cooling — it’s a few strings of LED chips and one weak chip will kill a string. There will also be a high current pulse before the battery packs settle at a lower voltage. I worry about its reliability. A more practical way is to run the thing at a lower rating. Use 1 battery pack and add a boost converter plus an MCU to control brightness. Add a better heat sink and a small fan.

    For an actual practical flashlight, running the thing at 100W will be rare. Since light is logarithmic to the eye, 10W is much more useful than 100W, and it’s easier to run it at 10W for 5 minutes than 100W for 5 minutes. But I concede that many flashlights these days are marketed for bragging rights rather than as a practical tool.

    I guess the primary reason for the video is for the author to boast that he can make a big ass flashlight for $10. I hope readers here are see these vids with a critical eye rather than accept such nonsense.

    • eLLi says:

      But… but… you can blind people much better with 100W – it would be no fun using a mere 10W candle ;-)

      • KH says:

        Yes, I know it’s fun, but I’ve seen enough failures in the more cheapo white LEDs from China to be cautious about driving them at rated power etc etc. I want my fun flashlight to last for some time…

        IIRC there was only one vid in the original posting. Has this been edited?

  2. hman123467 says:

    To answer KHs question, they only draw 3.5 AH and I have made a much prettier one than Julian’s but it runs on a 4AH battery and is a practical flashlight when used with a collimator and has been able to go up to 1Km (reflector pickup).
    It uses a small fan from a computer power supply and a DC/DC step up converter from 12V to 34V. The diode temperature hovers about 90 degrees Celsius
    I hope that answered a few questions

    • KH says:

      Well, your setup is miles apart from the above showboating. Care to share info about the DC-DC module? I am biased to making reliable flashlights myself but being an intermittent reader of candlepowerforums, I appreciate the delights of fun flashlights. Do let us know how long the LED module lasts with usage at 90degC…
      Unfortunately 3.5AH is not a current unit. Of little use unless you give the voltage + current => power. How many watts is going into the thingy? I am interested to know how many watts translates to 90degC operation and whether there is a heat sink and heat sink material too.

      • magma6 says:

        He means 3.5A of current I guess (this kind of module are generally rated 32V-35V, containing generally strings of 10 LED in series, for what I understood) which fit the 100W advertised.
        Unless he meant 3.5AH per hour :-)

      • KH says:

        3.5A x 34V = 119W, uh, unlikely (why tune to above-ratings, pushing one’s luck?) or unwise (poor LEDs… those thermals…).
        Nah, I don’t want to guess.

    • Shaun says:

      Lol. You’d be the guy who would have stood next to Thomas Edison and tell him how stupid his project was after each of his 10,000 failures.

      See, this is what we call PROTOTYPING. Taking an idea, and over time, testing different variations to improve a design.

      I hardly think this project will be sold in stores in the near future, so you can cease with your nonstop naysaying. You’re something of a nag. And an unbearable know-it-all.

      • KH says:

        Now I will nag on. :-p Edison is one of America’s well-honed myths.
        Actually, we should credit much of his great work as “Edison Research Labs”. I think there is no shame in that, it’s still a remarkable achievement. It’s just he lived in an age alongside companies like Standard Oil, an age when great researchers ceded most credit to the top banana.

      • KH says:

        Or, consider Orville and Wilbur Wright. Sure, being bicycle makers, they epitomized the ideal of the inventor who makes prototypes. But actually, critical to their effort was throwing out old aerodynamic tables and making their own using a wind tunnel they built themselves — by doing so their propeller and power plant were years ahead of others — a critical step in order to lift off the ground, after all, everybody was making gliders. So, their success was based on both theory and prototyping.

Leave a Reply

Notify me of followup comments via e-mail. You can also subscribe without commenting.

Recent Comments