3 ms·

An interesting article, but it makes me wonder: Do we have to worry about droop at all? If we can't make an LED produce lots of light at high power, then how ab

An interesting article, but it makes me wonder: Do we have to worry about droop at all? If we can't make an LED produce lots of light at high power, then how about using lots of LEDs, each one at low power? If someone points out that we don't know how to do this within certain price & size constraints, then I reply, fine, that's a problem to solve. So is droop. Solving either one of these problems get us the lighting we want. Is there any reason to single out droop as the problem that must be solved?

ars 17y ago
You don't even need multiple LEDs, but rather multiple emitting areas in one package. Maybe the result would be too big to be practical. Or too expensive.
- Huppie 17y ago
  I think it's the latter. I have been looking at proper led-lighting for my house a few months ago but when you want a lightbulb comparable to a 40W incandescence lightbulb it's filled with 80 LEDs (though it consumes only 4W). The cheapest I could find cost about €20,- while a simple incandescence lightbulb costs about €0,50. I could also get 'High-Power' led-bulbs, filled with only 3 - 10 LEDs and producing the same amount of lumen but they all had a higher power consumption than the 80-led bulbs and strangely where more expensive (this probably has to do with heat removal requiring big heat sinks).
loaded 17y ago
Yours is a good idea and is widely used right now as the standard reaction to droop. As physicists we are interested in the droop problem because it is more fundamental to the structure and intrinsic nature of semiconductors and solids in general. It is along the same lines of: if we have type-2 superconductors that work at liquid nitrogen, why do we keep looking for SCs that work at room temperature, or if we have type-1 superconductors, why do we need type-2? From a purely pragmatic point of view we know that more LEDs leads to the use of more money, more time, and more materials. A more efficient LED is a wise idea. I believe you may be missing a link here, that droop, or its physical cause, IS the entire problem that causes strains on price and size constraints. This is the reason to single it out.

ars 17y ago

You don't even need multiple LEDs, but rather multiple emitting areas in one package. Maybe the result would be too big to be practical. Or too expensive.

Huppie 17y ago
I think it's the latter. I have been looking at proper led-lighting for my house a few months ago but when you want a lightbulb comparable to a 40W incandescence lightbulb it's filled with 80 LEDs (though it consumes only 4W). The cheapest I could find cost about €20,- while a simple incandescence lightbulb costs about €0,50. I could also get 'High-Power' led-bulbs, filled with only 3 - 10 LEDs and producing the same amount of lumen but they all had a higher power consumption than the 80-led bulbs and strangely where more expensive (this probably has to do with heat removal requiring big heat sinks).

loaded 17y ago

Yours is a good idea and is widely used right now as the standard reaction to droop. As physicists we are interested in the droop problem because it is more fundamental to the structure and intrinsic nature of semiconductors and solids in general. It is along the same lines of: if we have type-2 superconductors that work at liquid nitrogen, why do we keep looking for SCs that work at room temperature, or if we have type-1 superconductors, why do we need type-2? From a purely pragmatic point of view we know that more LEDs leads to the use of more money, more time, and more materials. A more efficient LED is a wise idea. I believe you may be missing a link here, that droop, or its physical cause, IS the entire problem that causes strains on price and size constraints. This is the reason to single it out.