The truth about Luxeons and Lumens...

...does anyone know it?

It seems that we have come a long way from the days when flashlights we advertised with only candlepower ratings, if even that. More and more lights - especially Luxeon based lights - are being advertised with a "lumens" rating. These ratings are often suspect. Is there any reasonable way for us to acertain the truth? Say if we tested a particular light and knew how much power was going to the Luxeon, could we make a reasonable guess at actual lumen output? Is there a better way to make a reasonable guess? Is there ANY way to make a reasonable guess?

Depends upon many things, not the least of which being the bin code. Lumens can vary very widely between bins for the same drive conditions. You could definitely tell that some claims are false though.

There's also the question about net horsepower at the rear wheels, vs brake horsepower at the engine. /ubbthreads/images/graemlins/wink.gif
Many are advertising lumens from the bulb(s), instead of net lumens through optics, lenses and etc. /ubbthreads/images/graemlins/thumbsdown.gif
I have one where there was a big improvement from switching from optics and plastic lense, to reflectors and ultra clear glass lense. /ubbthreads/images/graemlins/ooo.gif
Did I add more confusion? /ubbthreads/images/graemlins/tongue.gif

Cheers. /ubbthreads/images/graemlins/smile.gif

Well, one thing NOT mentioned is that lumens don't tell high tightly focused the beam is. That's where candlepower comes in.

Folks seem to be advertizing the Lumens off the Luxeon datasheet, which is when the die is held at 25C. It is no where near this. For a Lux III running at 1A, in most flashlights, take 20% off.

Knock off a good 7-35% for the use of a reflector, and in most flashlights, it is towards the higher percentage. If you perfectly polish your aluminum in an intert atomshpere (no oxygen), and then deposit a very thin layer of special oxide, then you can go for the 7% number. Use chrome plating, plan on ~40% loss (not nickle). Or you can use vacuum deposited enhanced silver with a special oxide overcoat and get 0.9-2% loss.

Then chop off another good 10% for protective lens, unless you use a dual sided AR (Anti-Reflective) coating, with a borofloat/borosilicate crown glass.

Example, take an ARC4. Run enough current through the emitter to get to the 60 lumen level. Put it in a reflector and put a lens over it, and you end up at 30-33 lumens.

OK, so them how exactly do the Surefire L4/L5/L6 paly out in terms of Lumens?

As a manufacturer of raw light sources, what is lumileds to do but publish specs on lumens from the light source? It's up to product designers to deal with the (in)efficiency of their collimating solutions to determine actual output.

As far as Lumileds own deceptive data sheets - that's typical. Data sheet info is almost always determined under lab conditions to make the product look better, and to be taken with a grain (or several) of salt.

[ QUOTE ]
idleprocess said:

As far as Lumileds own deceptive data sheets - that's typical. Data sheet info is almost always determined under lab conditions to make the product look better, and to be taken with a grain (or several) of salt.

[/ QUOTE ]

I fail to see any deceptive practice on their datasheets. They provide specifications under well defined conditions, as they should, and they provide the necessary information to compute preformance under conditions different from the test conditions used for parameter specification. What more could a designer want?

Most people here at CPF aren't equipped to measure light in lumens. Only one person that I know of has even intermittent access to the necessary instrument: an "integrating sphere".

The use of 'bins' is a great tool to maximize sales and obscure the facts.

Which is brighter at 3.75 volts? An SWOK or a TWOL? Which one can you sell for more?

The SWOK can run from 51.7 to 67.2 lumens at a voltage between 3.5 and 3.75 volts.

The TWOL will fall between 67.2 and 87.4 at a voltage between 3.75 and 3.99 volts.

If the TWOL has a Vf of 3.99 volts, it will be under driven at 3.75 and fall much lower than it's rated 67 lumens.


Personally, I would not be suprised if the manufacturing process just happened to create lots of LEDs that produce 67.5 lumens. You can always hope for one of the brighter ones, of course.

Daniel