I'm having fun with this 1W Red-Orange. You might remember this Luxeon from such threads as " 1W Red-Orange luxeon super macro". This is a T-ranked Luxeon, meaning it will pour out 60-80 lumens of red-orange light at 385mA.
The Luxeon datasheets show a brightness curve that seems to increase exponentially as the junction temperature decreases for the Red/Red-orange/Amber luxeons. At 0C, brightness for the red-orange is expected to increase by 50%, and by 80% at -20C!
I wanted to supercool this thing, so I mounted it to a heat sink with some thermal goo, and sumbersed the heat sink in ice water. I also did tests with the same heat sink, but sitting in free air (which never got above room temperature at rated current). I set my camera up with fixed exposure settings, so that i could take photos of the beam on the ceiling, and get a good idea of brightness differences by comparing photos.
A few things about temperatures:
With a heat sink sitting at 20C, and a thermal resistance between the junction and star board of 20C/W, running at 1W, the junction temperature is expected to be around 40C. With a heat sink immersed in ice water, estimated at less than 5C, given the same conditions, the junction temperature is expected to be around 20C. The datasheets normalize brightness levels to 100% at a junction temperature of 25C. Let's just call that a measure of 1.
At 20C, a r-o luxeon is expected to have a brightness of about 1.10. At 40C, about 0.8. At rated current, dunking this setup in ice water theoretically will yield us a 37% increase in brightness.
Cranking up the current to 600mA increases the Vf to 3.4V, meaning that right at 2W is being dissipated. In these conditions, the bare heat sink setup yields 20C + (2 * 20) = 60C junction temperature, at a normalized brightness of 0.55. In ice water, the junction temperature is about 40C, for a normalized brightness of 0.8. In these conditions, dunking in ice water gains about 45% brightness.
At lower power, 100mA, the Vf is 2.26V, for 0.23W dissipated. Here, the bare heat sink setup yields a junction temperature of 20C + (0.23 * 20) ~ 25C, so a normalized brightness of 1. In ice water, the junction temperature is around 5-6C, giving a normalized brightness of around 1.4, yielding an increase in brightness of 40%.
These are all theoretical, and in reality, it would take some more sophisticated equipment than what I have to really measure the total light output difference.
But, what's neat is that i am able to show a brightness difference just with these photos. They've had the bit depth chopped down to emphasize different brightness levels. In all of these, running on the bare heat sink is on the left, while running in ice water is on the right:
100mA:
350mA:
600mA:
And just for fun, in ice water, cranked up to 1A (4.38vF = 4.38W!), next to a 5W royal blue.
Phew! that was long - any questions? /ubbthreads/images/graemlins/grin.gif
The Luxeon datasheets show a brightness curve that seems to increase exponentially as the junction temperature decreases for the Red/Red-orange/Amber luxeons. At 0C, brightness for the red-orange is expected to increase by 50%, and by 80% at -20C!
I wanted to supercool this thing, so I mounted it to a heat sink with some thermal goo, and sumbersed the heat sink in ice water. I also did tests with the same heat sink, but sitting in free air (which never got above room temperature at rated current). I set my camera up with fixed exposure settings, so that i could take photos of the beam on the ceiling, and get a good idea of brightness differences by comparing photos.
A few things about temperatures:
With a heat sink sitting at 20C, and a thermal resistance between the junction and star board of 20C/W, running at 1W, the junction temperature is expected to be around 40C. With a heat sink immersed in ice water, estimated at less than 5C, given the same conditions, the junction temperature is expected to be around 20C. The datasheets normalize brightness levels to 100% at a junction temperature of 25C. Let's just call that a measure of 1.
At 20C, a r-o luxeon is expected to have a brightness of about 1.10. At 40C, about 0.8. At rated current, dunking this setup in ice water theoretically will yield us a 37% increase in brightness.
Cranking up the current to 600mA increases the Vf to 3.4V, meaning that right at 2W is being dissipated. In these conditions, the bare heat sink setup yields 20C + (2 * 20) = 60C junction temperature, at a normalized brightness of 0.55. In ice water, the junction temperature is about 40C, for a normalized brightness of 0.8. In these conditions, dunking in ice water gains about 45% brightness.
At lower power, 100mA, the Vf is 2.26V, for 0.23W dissipated. Here, the bare heat sink setup yields a junction temperature of 20C + (0.23 * 20) ~ 25C, so a normalized brightness of 1. In ice water, the junction temperature is around 5-6C, giving a normalized brightness of around 1.4, yielding an increase in brightness of 40%.
These are all theoretical, and in reality, it would take some more sophisticated equipment than what I have to really measure the total light output difference.
But, what's neat is that i am able to show a brightness difference just with these photos. They've had the bit depth chopped down to emphasize different brightness levels. In all of these, running on the bare heat sink is on the left, while running in ice water is on the right:
100mA:
350mA:
600mA:
And just for fun, in ice water, cranked up to 1A (4.38vF = 4.38W!), next to a 5W royal blue.
Phew! that was long - any questions? /ubbthreads/images/graemlins/grin.gif
