Is there a limit to LED potential?

[o0o]

I've been very out-of-date on LED technology. My most powerful Cree is only rated at 160 lumens.

Now I find out that Surefire is making an LED rated at 400 lumens.

So, my question is, at this rate of lumen increase, how far will LEDs be in a year? What about 5 years?

Is there any kind of limit as to what LEDs can do?

Will we ever see a 1,000 lumen single LED in 2x123 format, similar in size to a Surefire U2?

 

[Zeige]

I've been very out-of-date on LED technology. My most powerful Cree is only rated at 160 lumens.

Now I find out that Surefire is making an LED rated at 400 lumens.

So, my question is, at this rate of lumen increase, how far will LEDs be in a year? What about 5 years?

Is there any kind of limit as to what LEDs can do?

Will we ever see a 1,000 lumen single LED in 2x123 format, similar in size to a Surefire U2?

Dont think the Surefire is 1 single LED, its 4 bunched toghether?

 

[o0o]

Dont think the Surefire is 1 single LED, its 4 bunched toghether?

So, kind of like the Lux V then? That can't be good? As the Lux V only had a 500 hour rated life?

 

[RyanA]

yes, ssc p7 4 die emitter. It can do 900lm or so, that is if you can provide 2.8amps.

 

[Flashlight Aficionado]

See This Thread

Post #62 and #63 seem to be the most informative.

 

[RyanA]

Wait for the Cree MC-E (also 4 die), should be better/more reliable and more easily packaged. As far as performance, the only limit is $$$

 

[Sir Lightalot]

There was a bunch of threads a while back about led efficiency (search something like: "finally! 300 lumens per watt!") don't know what happened to those leds though...:thinking:

 

[alibaba]

I've been very out-of-date on LED technology. My most powerful Cree is only rated at 160 lumens.
quote]



ONLY 160 lumens, I love it! I'm pissed because I have to turn my flashlight on HIGH just to land my light aircraft! If I want to signal the International Space Station I have to use TWO lights!! Just how long must we wait until we have Photon lights with 1000 lumens and 100 hours of flat, regulated runtime!?

 

[enLIGHTenment]

Is there any kind of limit as to what LEDs can do?

The physical limit for luminous efficiency is 683 lumens per watt. No light source will ever exceed this. LEDs have a long way to go before bumping into this limit. How close we can actually get to 683lm/w is anyone's guess.

There are other physical limits which combine to dictate how much power (total wattage) is possible for an LED. These limits are in terms of how hot emitting materials and phosphors can get without failing and how effectively heat can be extracted from the die. I don't know what these limits are and how close we can get to physical limits is anyone's guess.

Overall, I would say that perhaps 340lm/w (half the theoretical maximum) may be the achievable limit for efficiency and that the power dissipation limit might be around 150W with engineered forced air cooling. Call it 50,000 lumens within 25 years. For a 4W handheld device, call it 1,400lm.

 

[enLIGHTenment]

So, kind of like the Lux V then? That can't be good? As the Lux V only had a 500 hour rated life?

The Lux V only runs for 500 hours because it uses a series-parallel configuration and has poor thermal conductivity. Parallel connections always overdrive and overheat one LED. Combine this with a poor thermal path and a 500 hour lifespan is the result.

Cree isn't repeating Lumileds' series-parallel mistake (all four MC-E dies can be driven independently) and have tended to do a better job with thermal conductivity than Lumileds anyway.

 

[David Gretzmier]

You can buy today, for 40 bucks plus change, an unmodified p7 flashlight that will put out the front roughly 400-600 lumens. run time for about an hour from a single 18650 battery that can be recharged, arguably, 2000 times or so. With very simple mods to make the led run as it should, this same flashlight will do closer to 750 lumens out the front. I think it is really reasonable to expect, in 5 years or so, for 40 bucks, you can buy a flashlight with double the lumens out the front and the same run time. the p7 is not really the most efficient at making lumens out of the amps it uses, and the new 4 die cree will probably put the efficiency where it will sit for the next year or so. battery technology will also creep up year to year.

I think the last 3 years we have seen led's grow in leaps and bounds, and in the next 3-5 we will see solid incremental steps, rather than jumps.

 

[Jarl]

The physical limit for luminous efficiency is 683 lumens per watt. No light source will ever exceed this. LEDs have a long way to go before bumping into this limit. How close we can actually get to 683lm/w is anyone's guess.

Only for green light. For white light from a single emitter it's just under 250 lumens/watt and for multi-emitters (RGB) it's 350 (though I'm not at all sure about this figure). The R2 can do 145 lumens/watt at very low power. The problem is getting the heat away from the die and maintaining this lumen/watt figure at higher powers.

 

[Gunner12]

I think the limits of LED would be close to the limits of efficiency. I'm not sure how fast it would go.

If the LED is close enough to the supposed limits of efficiency, it shouldn't produce too much heat.

As for LEDs now, the Seoul P7(4 Seoul P4s in one package) is probably one of the brighter(and with better efficiency) ones right now. The Cree XR-E R2 bin is pretty good too.

 

[LukeA]

The physical limit for luminous efficiency is 683 lumens per watt. No light source will ever exceed this. LEDs have a long way to go before bumping into this limit. How close we can actually get to 683lm/w is anyone's guess.

There are other physical limits which combine to dictate how much power (total wattage) is possible for an LED. These limits are in terms of how hot emitting materials and phosphors can get without failing and how effectively heat can be extracted from the die. I don't know what these limits are and how close we can get to physical limits is anyone's guess.

Overall, I would say that perhaps 340lm/w (half the theoretical maximum) may be the achievable limit for efficiency and that the power dissipation limit might be around 150W with engineered forced air cooling. Call it 50,000 lumens within 25 years. For a 4W handheld device, call it 1,400lm.

683lm/W is for monochromatic 555nm green light, aka useless. The limit for white light is 241lm/w.

 

[lctorana]

683lm/W is for monochromatic 555nm green light, aka useless. The limit for white light is 241lm/w.

Hypothetical Question:

If by some thermodynamic miracle, I lit up a 241lm/W LED, would that theoretically mean no heatsink required?

 

[Gunner12]

Yes, that should be right.