Some joints beg to be silverbrazed

(Snip techy info)

Wow! I see elsewhere that Metal Halides are good for 85 lumens per watt.

formatting link
of page)

Very Nice!

--Winston

Reply to
Winston
Loading thread data ...

My bad. Luxeon claims 500 hours, not 1000.

--Winston

Reply to
Winston

I'd be glad to do that if you'd like. Send me an email, I'll send my address. You could probably do the same thing at a welding store, though and save the shipping cost.

I use mine in presence of four 40-watt fluorescents and a window admitting daylight. I have also used it outdoors in sunlight with a small wirefeed welder. No problems!

(snip)

Yeah, that's pretty huge all right. How do you calculate thermal resistance from geometry? I have some stuff that works with flat plates (adapted from National Semiconductor's audio handbook) but nothing more general than that. I have to use a power resistor and a thermocouple.....

There is a selection of heatsinks for Luxeons shown at

formatting link
aren't quite as huge. There's one that's about 40 mm x 40mm x

18mm that would get you real close to 60 deg C rise above ambient at 5 watts.

For max raw lumens in a small space it's pretty hard to beat xenon incandescants. They're pricey and only last about 10 hours, but they do produce a lot 'o lumens in a small package.

HID lights, as used by SCUBA guys and now in some automotive headlamps, are both bright and fairly efficient -- but they're very pricey, have 500 to 1000 hour lifetime, and don't start immediately.

Reply to
Don Foreman

Thanks! Email enroute.

Not geometry but surface area.

I found a rough 'rule of thumb' equation on the net:

(Degrees C per watt) = (50 / Sqrt(Surface Area))

I mangled this and came up with:

(Surface Area) = (50 / (Degrees C per watt))^2

Where:

Surface Area is the number of square centimeters of heatsink exposed to convection air.

Degrees C per watt is the maximum amount of temperature rise above ambient you can tolerate for each watt of thermal power to be dissipated.

Yup. Things get hairy, quick when you have multiple power sources, too.

I lost my old heatsink handbook that was full of various designs, with thermal resistance labeled *right next to the sketch*.

formatting link
that aren't quite as huge.

Nifty! Looks like a good place to start, but a square heatsink isn't going to give you the packing density you want. Hexagonal, perhaps?

That's 90 C in a 30 C ambient. The 5 watt LED has a die - to - junction thermal resistance of 11 C per watt. So if you held the heatsink to zero degrees C, the die would be up around 53.1 degrees at power. Luxeon says the die must not exceed 135 C in operation.

So it's worse than I thought.

Let's try to hold the worst - case absolute die temperature down to say

120 C. Subtract from that, the (die - to - aluminum substrate) thermal resistance of 53.1 degrees and you get 66.9 degrees. Subtract from that the thermal resistance of a Sil-Pad and you get say, 65 degrees. Subtract from that a 30 C ambient and you have 35 degrees. Divide that by 4.83 watts and you are left with a requirement for a 7.2 degree C / W heatsink.

You could do that with water cooling, perhaps! :)

That is cool!

I read a rumor about a 50 W metal halide. Where do I sign?

--Winston

Reply to
Winston

(snip)

Or a little fan.

That and short lifetime are why I don't think the 5 watt Luxeons are practical beyond intermittent use in a "show off" flashlight.

I've posted my 3-watt Luxeon lantern story here before. A 2" square heatsink seems to work fine, runs warm but not hot. I bonded a

3-watt emitter (not a star) to a short 3/8" dia aluminum stud with Omegabond 101 to reach the focal point of the reflector. The short stud is fastened to the square with a machine screw and thermal grease in the interface. No Sil-Pads.
Reply to
Don Foreman

Don Foreman wrote: (Snip)

I'm convinced.

That makes me wonder what one could do with a cube-shaped aluminum enclosure with one side open for a parabolic reflector. You could heatsink the LED to the enclosure and place the batteries on both sides of the reflector. A shade / dust cover could trickle charge the batteries via photovoltaics. It lives on the windowsill pointed towards the Equator.

Compact, reliable and Very Bright.

The folks who sell batteries would not like that *at all*.

--Winston

Reply to
Winston

I have a small lamp attached to my mini-lathe. Built it myself. It has a 1 watt Luxeon driven at spec (1 watt). The Luxeon was an older unit that only produces about 20 lumens. The small flood reflector gives me about a fairly uniform 250 foot candles (at 1 foot). The luxeon is bonded to a finned aluminum 1/2 inch rod that protrudes 1 inch from the back of the plastic reflector housing. I usually have the light fairly close to my work, maybe 6 inches.

I sometimes think I should add a spot reflector or optic to make it a bit brighter, but for most uses it's just fine. Most people would do fine with a Lux III and a 10 degree optic. That will provide in excess of 7000 fc.

For reference, I get 4100 FC from a 200 watt halogen lamp over my work bench. I get 650 from my desk lamp. Both of them are flood lights, so while they put out more light, the light is not concentrated like a luxeon with an optic or reflector.

A Lux V (5 watt) is not a long life unit. The Lux III is rated for

50,000 hours if run at spec. Both units will have dropped to 70% of their original brightness at that point. You could do the same with halogens, but it's generally cheaper to use a brighter filament and bypass the need for more than a cheap reflector.

The maker of Luxeon LEDs is predicting a doubling of effciency within a year. When Luxeon based task lights come out, give them a try. They don't need the same watts to do the same work as their incandescent brothern.

Daniel

Reply to
dbs__usenet

Can a LPS lamp be used as a monochromatic source? If so, where does one buy a small lps lamp, if it exists?

Max ben-Aaron

Reply to
Max ben-Aaron

No. See page 6, figure 7: chemeducator.org/sbibs/samples/spapers/34jg1897.pdf Four energy peaks, not just one.

Do you need a laser?

18 watts small enough?

formatting link

--Winston

Reply to
Winston

It's monochromatic enough to generate interference fringes, I've used them in interferometers.

Reply to
Nick Hull

Thanks, Nick.

Max, is that sufficiently monochromatic for you?

--Winston

Reply to
Winston

What I did was set up a 12 volt 500 watt transformer (out of the box') - and built a box to hang that holds a Halogen headlamp from my truck.

They say replace both when > Ted Edwards wrote:

Reply to
Martin H. Eastburn

Fog lights are yellow because shorter wavelengths scatter more. Blue/violet is exactly what you don't want in fog.

Haven't seen one close up and personal. How are they getting it blue? A filter? If so, then I would say it's nothing more than an affectation.

Ted

Reply to
Ted Edwards

Reply to
Don Foreman

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.