Luxeon die size?

AilSnail

Flashlight Enthusiast
Joined
Jul 11, 2003
Messages
1,299
Location
Bergen, No
How big is the luxeon 1w die?

What size bulb filament can it be compared to with regards to throw?
 

OddOne

Enlightened
Joined
Mar 5, 2003
Messages
495
Not sure if there'd be a direct correlation to a bulb. LED light behaves in its own peculiar way with regard to throw.

The die itself is in the neighborhood of 0.1" square.

oO
 

Orion

Flashlight Enthusiast
Joined
Jun 27, 2002
Messages
1,613
Location
Missouri
Here's another question that could be asked here. Is it possible to make the die itself bigger than it is, and perhaps it would produce more light? /ubbthreads/images/graemlins/confused.gif
 

snake

Enlightened
Joined
Dec 20, 2001
Messages
256
Location
HONG KONG
bigger the die size lower the yield.. so it cheaper to connect several small(or size with reasonable yield) die to make a bigger area as the 5W LS did
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
The 1W die is 1mm square. It appears larger in whites due to the phosphor. The die for a 5W is slightly larger than 2mm square.

A bulb filament may be 3mm long in one axis (between the wires supporting the filament), but may only be a fraction of a mm thick...making it appear as more of a point source.

Making a larger (1.5mm square, 2mm square?) die will allow more current to be run through the luxeon. However, the larger the area of your emitter die, the more difficult it is to provide even current distribution across the junction.

The grille pattern of light that the blue/green luxeons emit is due to this problem. Conductive channels must be deposited across the led (in this case, a ladder pattern of conductors) in order to provide even current distribution across the junction. Otherwise, you get "hotspots" of current flow across the junction near the contact points, and lesser amounts of current flow farther away. This leads to uneven light production, and uneven heating of the die. The grille pattern conductor spreads this out evenly, but blocks some of the light emitted.

Making the die larger may require that the grille pattern obstruct a larger percentage of the die, or may represent an exponential increase in cost/difficulty in manufacturing, or may be beyond the limits of processes when the 1W luxeon was designed (I'm just speculating at this point).

The other thing to consider is that if you double the size of the emitter, you both make it more difficult to collimate the light into a coherent beam - it's a larger light source, and we've seen how poorly some optics work with a 5W luxeon. In addition, if you simply make a single emitter larger, you still only have a single junction, which means your Vf will be the same (may even be lower due to a larger area for current). So, if the area of your LED die is 4 times larger, you are expected to supply 4 times the current.

If we take a Luxeon III as a starting point, at 700mA of current, than you would need to supply 2.8A of current to keep the same current density as with the smaller Luxeon. Getting into the multiple-amp range is now creating other difficulties, as you must construct your bond wires to handle that current, and any device you place the larger luxeon into must hande those kinds of currents, both in distribution wiring, switching, and power supply.

It's kinda easy now to see why Lumileds decided to go with a series/parallel arrangement for the 5W. It's much easier to increase your voltage and current a little, than to crank your current up into the multi-amp range. Supplying power with higher voltage is much easier from a packaging and power delivery standpoint.

Wow, that was way off topic.
 
Top