luxeon project

Rothrandir

Flashaholic
Joined
Aug 17, 2002
Messages
7,795
Location
US
well, i have a few more questions to ask.

i recently gave in and ordered a couple luxeon emmiters. they arrived two days ago, and i have had fun playing with them. they are white and bright with no evidence of either green or blue so far in what i have seen. anyway...

i plan on using as a pr2 base luxeon on 3 d-cell mag light. what resistance value should i use? how many and of what resistors should i use? also, is it possible to run the negative power through the heatsink or is that a bad idea?

how should i solder the wires onto the ls? should i just be careful not to hold the soldeing iron on two long? how long is to long?

oh, and where is the cheapest place to order arctic silver?

how much better does copper conduct heat than arcit silver?

thank you so much for your patience and your help
 
Yup. Use Artic Silver between the Luxeon and the copper--but not much. When it comes to thermal greese...even Artic silver, less is more!

(Cover the whole area to be heatsinked--that's not what I meant by less is more...but use a thin thin layer of it. Thinner is better)
 
thank you very much, i just ordered it at 8.99 on sale.

about the other questions...

p.s. while i am here, i plan on putting a couple of ls' powered by a radio shack 3v, 300mA dc plug thingy, if i pluged this into a dimmer outlet, would it dim? or do luxeons not dim?
 
Yes you can dim them, just add more resistance. I do not recomend trying to run a transformer off a light dimmer. AC light dimmers depend on having a non inductive non capacitive load. they are designed to work with standard light bulbs any thing else may not be safe.
 
Wow! Those are great prices.
icon14.gif


Unfortunately for people in Europe that order from Conrad don't get these bargain prices...
Just wondering why the Arctic Silver adhesive is €32.49 (~US$ 30.-)
shocked.gif
, compared to SVC's $8.99 ...
icon15.gif
icon15.gif
icon15.gif


Running a transformer on a dimmer outlet is a sure way of killing your precious Luxeons.
tongue.gif
Something goofs up and you get magic smoke...
frown.gif
 
hmmm, as much as i love magic smoke, i don't want to pay 15 dollars for it...any thoughts on the resitor issues stated above? they are my main concerns right now.
 
I've been doing a lot of reading around here on using luxeons from other power supplies too. First off, dont trust the resister calculator values for them. (my favorite calculator is at http://linear1.org/ckts/led.php) The forward voltages vary so much that you may be over driving them significantly if you just do that math. You must have a way to measure how much current you are actually sinking through them. You can get a multimeter that can measure that very inexpensively. Quite nice ones under $40 USD. I would categorize that as a must have.

Resisters seem to be a bit problematic for luxeons, or at least frowned upon by some folks around here. The proper way to run one these things is with a constant current regulator. You can find a LOT of info about this kind of thing around here. I've been reading it for weeks and am almost ready to order the parts to start playing with them. Makes working with them a lot more work than just connecting them up like you can do with other LED's though.

In the meantime I'm sure that the proper resister from that link above will work just fine from a wall wort type supply. But I have specifically read that people have noticed MUCH more current going through the luxeons than they expected from doing the math since the luxeons are not always exactly the same internally.

Good Luck,
James
 
It depends on your power source:
If you will use only NiCd or NiMH cells, you will get a flat discharge curve at a voltage of about 3.7 V. With a resistor 1.8 Ohms 0.25 W you will get about 350 mA through your LS.
Taking this resistor while using alkaline cells with 3*1.5 V=4.5 V causes a current of 780 mA through the LS - far too much! (In this case, our resistor, only capable of handling 0.25 W, will smoke up with 1 W).
If you will use Alkaline cells, a resistor 4 Ohms 1 W would first being usefull (you may get only 3.3 or 4.7 Ohm types in our real world...). But while the battery voltage drops while using this flashlight, current through the LS drops, too. With fresh batteries and 4.7 Ohms (3.3 Ohms) you will get about 300 mA (420 mA) through the LS. Using it some hours, the battery voltage may meet the voltage of accus, 3.7 V. With this voltage and 4.7 Ohms (3.3 Ohms) you will get about 130 mA (180 mA) through the LS.
In the real world, there may be a little variation of these currents, depending on the LS etc.
The best way would be taking a current regulator. But most current regulator circuits dislike low voltage drops. And while 3 cells offering only max. 4.5 V, the LS needing 3.1 V, there is a maximum of 1.4 V for the regulator - not enough in most cases.
IMHO for getting first results, using a resistor isn't the worst way, even if you will only use NiCd/NiMH cells with their relatively constant voltage. Even using alkaline cells, to the naked eye the drop of light output isn't much visible. Most commercial flashlights are using this quick and easy (and dirty...) technology. Only very high sophisticated flashlights (like ARC LS, AAA) are using regulation and voltage conversion. You know the difference in price...

Soldering the LS with heatsink seemed not to be very critically.
 
hmmm, i think i might forgo this led pr2 thing. they only way i would consider it now is a value, i have no idea what to do and i don't want to fry the luxeons.

i know there are several dc\dc's out there, what one would be the best for running a 1w white from 3alkalines? i am looking for brigtness and will probably have a custom heatsink made at a local machinist so i could probably afford to run the ls at a little over specs.

how about a boost from 3v? could someone please supply links directly to where i can purchase a fully asembled curcuit? dat2zip has a couple right?

by the way, thanks guys for all of your help, it helps alot
smile.gif
 
Rothrandir,

don't worry. Just give the simple resistor a try. It wouldn't cost much. And with an appropriate heatsink the LS tolerates more current, especially if you wouldn't keep it running for thousands of hours without a break.
With a resistor 3.3 Ohms 1 Watt for less than 1 $ you would stay on the safe side: Fresh batteries will only slightly overdrive the LS with 420 instead of 350 mA, and with still 180 mA at 3.7 V it will produce lots of light. These 180 mA are more than the initial consumption of other commercial designs. And: There is an increase of efficiency of light output at lower current. So you will not see much difference.

As an option you may use a second resistor with a second switch for getting full current with lower voltages.

The other way would be using a voltage converter and current regulating module. dat2zip seems to offer excellent solutions for this task. Indeed, in this case most designs are using only two cells. Three or more cells would need a special voltage converter of SEPIC type, which is more expensive and less efficient. Please bear in mind that most converter modules are designed to fit in a small Mini Maglite etc., so you may need a support for the module and the optics in a bigger housing.

So I think, try the resistor first, and if you are not fully satisfied with it, try a converter module next. You are skilled to manage it.
 
Top