Current fluorescent lamps are great for lighting large areas using very few Watts; however, LEDs are far more efficient at producing light and have less of an impact upon the environment considering there is no mercury within them. [Andrew] sent in his team’s LED florescent bulb. The first revision utilized 87 LEDs, but to increase output the second revision uses 210. The assembly can’t actually be placed in current fluorescent lamp ballasts and must use a 12 volt 1 amp power supply, but perhaps future versions will correct for this. Another problem is the relatively small viewing angle, and while there is a diffuser, we’re wondering if they have any other ideas to spread the light and adjust for the color temperature without reducing output? We wonder how it compares to some of the commercially available LED florescent lamps.
Do what i did, sand the tip of the led to make it flat. it becomes less of a lens and also doubles as a built in diffuser.
[LEDs] have less of an impact upon the environment considering there is no mercury within them.
I’d love to see a citation for this, given that producing solid state devices isn’t the most environmentally friendly thing going. It’s easily true LEDs don’t have the same impact on the immediate environment (though the amount of mercury in a CFL is pretty dang small, about the same as a period on a page (from snopes: http://www.snopes.com/medical/toxins/cfl.asp)), but what about overall costs?
I hate to be nitpicky, but LEDs aren’t “far more” efficient than CFLs.
According to Wikipedia, “In a lighting application, operating at higher temperature and with drive circuit losses, efficiencies are much lower. DOE testing of commercial LED lamps designed to replace incandescent or CFL lamps showed that average efficacy was still about 31 lm/W in 2008 (tested performance ranged from 4 lm/W to 62 lm/W)”, and “The luminous efficacy of CFL sources is typically 60 to 72 lumens per watt”.
LEDs are neat and I *want* them to be better than CFLs because I loathe the light quality of CFLs… buuuuuut… they’re not there just yet.
I remember seeing a how it’s made on ccfls, and i remember seeing a full drop of mercury falling into the large tubes
relatively wide viewing angle LEDs can be had. Combine that with sanding, and you’re good to go.
I think the real issue with light quality for something like this is the color of output. Probably best to select LEDs all from the same factory bin to make sure.
Do not forget the wasted energy if using a 12v walwart for the PS. It consumes some energy while the lamp is used or not. These are also refered to as “phantom loads”. Probably wastes considerably more than the conversion to LED saves.
@Evilspoons
I Agree, Watt for Lumen LEDs are far less efficient than tubes and cannot be because the quality of silicon needed would make them too expencive
oh i forgot to mention the loss of efficiency through the transformer if powered by such as this array is.
Another uninformed ‘save the environment’ project. I know it makes you feel valued, but you really just wanted to make a lamp from LED’s.
Also white LEDs don’t look like they have a very nice spectrum: http://powerelectronics.com/power_management/led_drivers/Fig-2-white-LED-vs-RGB-LED-spectrum.jpg
i thought the point of this was to replace the actual fluorescent lamp. but if its just powered by the 12v wall wart all this is is a bunch of leds and resistors…
@pwrx – nope – we designed this lamp to run on 12 volts (actually 10.5 – 13.5…) so it is operational on a solar or wind energy system – I am aware of the debate over the efficiency of LEDs right now, but lets face it, even if an LED is less efficient than a fluorescent, we have to consider other things like lifespan and the toxins contained.
And to everyone who is commenting on the use of a wall-wart, I totally agree with you. This is a less-than optimal installation. It would be best to use a switching-mode power supply, or even better to use solar panels.
Thanks for all the comments everyone – glad to see your interested!
This is impressive work, if it is the work of grade school children. Otherwise, I am sadly disappointed by it even being considered a “hack”. So they drilled a piece of plastic and wired up a bunch of LEDs in a series parallel array. From the looks, they’re not even modern power LEDs, but instead cheap white indicator grade diodes.
In 1990 this may have been somewhat original. I’m grumpy today!
easy there hackaday. The documentation for this is superb, the site is clean (distinct lack of tags), and the final result is neat.
yes; led’s arent as awesome-tastic as commonly thought, its still a cool idea.
as a note: There are constant current led drivers that are much better than resistors. Most can do pre-set currents (bcd/spi/i2c interfaced), and they wont burn off extra voltage as heat. Otherwise, investigating pwm & higher efficiency dc-dc conversion would be helpful. Best of luck with your endeavours
*distinct lack of blink tags
@justdiy – I think you underestimate this project. We did not use simple resistors. We designed a linear current regulator which requires a resistor to “calibrate” each LED group. You MUST have a resistor on every LED to prevent “unlimited” current to flow through them and to compensate for the minor differences in each LED.
Care to share some insights for our next project?
For those interested in the debate on LED greenness:
http://bits.blogs.nytimes.com/2009/08/04/leds-as-energy-efficient-as-compact-fluorescents/?scp=1&sq=led%20vs%20floursent&st=cse
@peter – nice find, i havent seen that yet!
One aspect of LED lighting is the ability to easily dim the light through pulse width modulation, something I gather most florescent tubes are incapable of doing.
This light could very easily be upgraded with a dimmer, for example it could be done with just 4 components; a Picaze 08M, 78L05 regulator, a FET capable of handling the load of the LEDs, and a variable resistor.
The bottom line here is: If you want to make a 210 LED Desklamp – Awesome! Knock yourself out. We all love LEDs.
Do not however overstate environmental benefits of your projects unless you are on solid ground.
The BS detectors were on high alert today. Good job, everyone. Well played.
One other thing to remember is that the LED lamp may contain far more lead than the fluorescent light does. I’m not sure if the lamp in this post used lead-free solder, but if a lamp like this were mass-produced in Asia somewhere it probably wouldn’t!
I have to agree that LED-based lighting just isn’t there yet, though this is a valiant homebrew effort.
Hello, I am the second developer for this project (alex).
Thank you guys for the comments so far. We will try to use them to improve what we wrote so far and to try to eliminate any misunderstandings.
There is more than one advantage to using LEDs; so please don’t rant about one concept only. I saw that efficiency comparing post (http://bits.blogs.nytimes.com/2009/08/04/leds-as-energy-efficient-as-compact-fluorescents/?scp=1&sq=led%20vs%20floursent&st=cse). I actually agree with it, but we don’t pick LEDs just because one advantage.
Besides green, LEDs are also less toxic than fluorescent(toxic stuff is trapped in plastic, not in a gas as it is in fluorescent). Second, you can run these lamps on low voltage, without the use of balast (our “system” idea was part of the project, just not on the site yet). They’re even recyclable(the only things that make an led end its life is the yellowing of the plastic around it, that can be fixed in the future). And they’re much more flexible, you can use PWM as stated and so on (One guy at the fair was even saying that we could use them for wireless information transmition like seen on here: http://www.sciencedaily.com/releases/2008/10/081006130548.htm)
If you guys want references, we can get you some. But we were under the assumption that the site was not wikipedia.
Most modern LEDs can be dimmed in an analog manner (without PWM) without an annoying amount of change in color.
Sorry, but stupid project is very very stupid.
As someone already mentioned, those are _indicator_ LEDs, not meant for lighting. They’re very inefficient, have a very ugly light, and will not last very long even when running at half the recommended current of 20mA.
This is worthless.
Next time try with modern, high-power LEDs.
Their light quality is infinitely better (>70 CRI), so is their efficiency (>100lm/W if you buy of the good ones), last much longer (70% brightness after 50,000 hours, with a proper heatsink), and you only need four of the 2-watt ones to substitute a 700lm lamp, so wiring is not a pain in the ass.
(Sorry if I’m insulting, but I’m really tired of n00bs giving bad reputation to what should be the lighting of now, not even of the future..)
@visionetiks
I agree, those were crappy LEDs, we brought the cheapest we could find for the project. To anyone that wants to build these I would recommend getting better LEDs.
i am sick of the pseudo-religion of the church of earth-friendliness. equally loathsome are the terms “green” and the nonsense term “carbon footprint.” The only real “green” to have come from this church are the shovelfuls of “green” al gore and his priests have made hawking their dubious wares.
The worst result of the whole earth-first movement is that so many lies have been told, and so much nonsensical hippy feel-good propaganda has been distributed that people are starting to become leery of the whole thing. Yes, good intentions at the expense of fact can undermine the very thing you supposedly believe in.
When somebody has a project that is interesting, I listen. when they have a project that purports to offer greater efficiency, lower cost, or higher reliability, I listen. when they offer a project that is creative, entertaining, thought-provoking, instructive, or inspiring, I listen.
but when they present a project that is supposed to help save the earth, adios. I move on to a different URL.
Having said all that, don’t be too rough on these guys. anyone who dreams, builds, and experiments is worthy of some appreciation and some respect. Modified led fixtures are ideal for Motorhomes and camping trailers, field camping, and emergency lighting in ham shacks.
As an architectural lighting designer I’m excited to see people working on the next big thing in their basement. Nice work y’all.
However, the above quibblers are correct– as the technology currently exists, LEDs are not nearly as efficient as fluorescent lamps, especially in the warm white range.
As such, their environmental benefits are also dubious. Most electrical power comes from coal in this country, which releases mercury into the air by the ton. More power used is more mercury in the atmosphere, where it is a greater threat than the small amounts used in fluorescent lighting (which is reclaimed when properly disposed of, in any case).
LEDs are also plagued by a number of technological problems such as high expense, poor color rendering, inconsistent color temperature, poor lamp life, and rapidly diminishing light output– try running that lamp you’ve made for 5000 hours and I think you’ll see what I mean.
A good reference for the existing state of the technology is the DOE’s CALiPER test, which test and compare real products available on the market today.
Max: Gee, another man who isn’t up with the current technologies…
All those problems you mentioned have been overcome in the latest generations of high-power LEDs (which 99% of the projects and commercial products do not use)
Well, all except the high expense, but I bet this would improve if more people used them.
By the looks of it, you’re still talking about the old 5mm LEDs
If you’ll excuse me, I’ll go cry in a corner..
‘Green’ isses aside, I’ve found that cheap 3mm white LEDs produce a much smoother and nicer beamspread than cheap 5mm white LEDs which produce a pronounced spotlight effect, and they produce proportionally more light for their size than 5mm LEDs, meaning you can pack much more into a smaller area.
I’m close to finishing my new bike light which has 300 3mm LEDs and only uses 10 watts when all 300 are powered. normally only run 60 so I don’t annoy other road users, keeping the full 300 for completely dark roads/paths. The beamspread of the 3mm LEDs mean I get a very nice even light that lights up the entire road/path in front of me.
@haku: 300 leds is not something to brag about. unless you have a very high voltage power supply, you’re wasting as much energy on current limiting as you are on leds. spend your 10 watt power budget on modern 100 lm/watt illumination grade power leds and enjoy some real lighton your dark roads and paths. figure 1-2 watts for smps regulation that will yield you 800-900 lm of light
@Watoosie: Thank you. You nailed it on the head.
I saw the most horrible sight the other day: a hybrid suv decked out with american flags and the word “hybrid” in giant letters across the windshield. I swear, people aren’t being “green” to save the earth, they do it to feel important. If they /really/ wanted to save fuel, they would buy a little vw diesel. those get amazing fuel milage.
How does that relate to this project? Idk, i’m just sick of people trying so hard to be green. I would’ve been fine if it were just an led lamp.
I have a suggestion on how to make the manufacture of this a bit more efficient. Create a jig with a peg the size of the hole to be drilled. Then clamp that to the drill press with the peg offset by whatever distance you want the holes to be apart. Drill your first hole then put that on the peg to get the distance for the next hole. Rinse and repeat. That way you don’t have to measure and score the plastic where you want to place the leds. Just measure the first one and you are done.
I know it is popular in some circles to bash people for trying to be “green”. But can’t we just focus on the awesome concept and how we can make it better? Of course led technology isn’t there yet but when it becomes viable, the rest of the work could be done and waiting. Oh wait, I forgot this is the internet…all it is good for is porn and anonymous shit talking.
@george – Thank you! I can’t tell you how much I agree with you.
UV, This will give you a nice sun tan and cataracts as a bonus. Greenies gone wild – lame.
I want the moon, on a stick please. And I want it YESTERDAY….
It’s just a neat, thought provoking project with good intentions. Give the guy(s) a break. Sure the transformer etc are not ideal but seeing as DC microgrids are an increasingly hot topic of research this sort of work will hold more value further down the line. This sort of item already exists to buy off the shelf, but it’s good to see people thinking about these things and putting them to the test.
The paranoia about Hg in fluorescent bulbs is a bit amusing, especially since no one considers that LEDs aren’t completely benign themselves. (Some of the common semiconductors used in LEDs are compounds of arsenic.)
@justDIY, the power source is the ebike’s 26v lithium battery and using an adjustable switching regulator to step down the voltage. I originally designed the array to run off pure 12v and use 24watts but like you said the resistors dissapated way too much energy and got very hot. But I turned the switching regulator down to pure 10v and the brightness drop-off was minimal but the overall power usage dropped down to 10.2 watts.
I know I could have used Luxeons or Crees for the job like just about everyone else out there but for a few reasons I went with an array of 3mm LEDs;
– the smooth gradient beamspread is extremely nice for cycling along unlit roads/paths http://www.haku.co.uk/pics/BikeLight300-5.jpg
– no need for a collinator to direct the light as the 3mm LEDs have lenses built in
– no real problems with heat dissapation
– very thin design means it’ll sit permanently attached to the bike and to the untrained eye look like a fancy reflector
– I wanted to be different :)
Cool lighting project.
-is it okay to say that?
-with anyone?
?
why dont they use modern high power leds?(google: cree xr-e xp-e mc-e ,luminus sst-90 sst-50) i think it would have been cheeper ( becaus you wold maybe need 2-20 leds ( depending on type) and not over 200
maybe you also want to have a look at candelpower forums(just google it)some people around there know much more about leds then i do and the diy section hase some jaw dropping led projects (im not payed to say this :p)
sorry for my bad english
Hey guys, cool project! Ignore all the internet tough guys above, this is a cool project and as you’ve stated with more expensive LEDs, a different power system it would be amazing.
But, I’m guessing cost is a factor here and as such, everyone should get off their high horses!
It would be great if we all had endless supplies of money for our “hacking” but we don’t, so lets stop the bashing and have more creative comments on how problems can be overcome.
Ta :)
@ alex
“the only things that make an led end its life is the yellowing of the plastic around it”
This isn’t true. Cheap white LEDs become dimmer as the phosphors within them wear out. The plastic package remains water clear. I have looked at a selection of white LEDs from various sources and the effect of dimming with age is consistent. Better ones seem to be composed of several different color chips and no phosphor.
Perhaps you could try the same experiment but interleave red, blue and green or yellow LEDs, adjusting the current in each color to give the best light. I think the spectrum will still be very uneven, as it is with fluroescent and regular white LEDs, but at least you will have some control over the color.
Wow, tough crowd. Good intro project into LED lightning though. Start off cheap, learn the lessons on how to more effectively power a large number of LEDs, and move forward to more advanced (and more expensive) LEDs.
Most of the comments are correct, in that the 3mm (and 5mm and 10mm) LEDs tend to be behind the curve in terms of lumens/watt. They also have a much higher variability in color. The LEDs from Cree and Philips can be ordered by color temp and lumens/watt. There are quite a few companies who sell optics for these LEDs to create spot lights, or diffusers to spread the light out.
So, kudos on your project. Hopefully your future projects will be as well documented and interesting as this one.
Thanks [andrew] for the suggestion in the article for scoring the plastic instead of using a saw. I’m building a case for my NAS out of 1/8″ Lexan & was too impatient last night to wait until I could borrow a table saw. So I got out my X-Acto pen, scored about 20x along the edge of a ruler, and put all my weight into snapping that sucker. A little sanding and some acetone later I had a perfect edge!
@Buzzkill – Our system actually uses batteries to run the lamps, you can read all about it in our upcoming lighting controller section (we even eliminated the “phantom loads” from the charger).
@nes – I’m sorry about my inaccuracy, I will try to research that better before defending. Yes, it would be nice to have RGB leds, but those tend to be a lot more expensive, and they would make the schematics a lot more complex. If I were to change the colour temperature I would just buy the warm LEDs to begin with. But again, for this project we just got the cheapest we could find. This project was mostly a prototype, to see how we can drive so many LEDs the most efficient way.
instead of running a bus wire, when i do these kind of projects i find it easier to space the led’s a little closer and bend the cathode leads over and solder the overlaps together, then clip all but a 1/16th inch of lead off of the led anode and one side of the resistor, solder those together at a right angle, and begin to solder all of the overlapping resistor leads. works very well and saves time and wire
@john – We wanted to do that, but we were afraid of the cumulative resistance. + it looks cooler with a copper wire.
Nice project, alex, even if it doesn’t utilize the LEDs used currently in lighting fixtures.
visionetiks – your slam on Max and the others might mean something if you backed up your statements with any sort of reference. Max’s comment was right on with the state of the art LED lighting fixtures that are currently hitting the market and being tested by the DOE.
I’d love to see some higher efficiency LED products with good color temperature, color rendering, and color consistency, but it’s a challenging engineering problem that the LED, light engine, and fixture manufacturers are still working out. If you can lead us to better LEDs, please post a link.
Great looking final product and nice little write up! Don’t listen to all the naysayers who don’t understand the joy of a nice DIY project like this :)