A little while ago, I obtained a white lamina BL4000. For those that don't know, Lamina Ceramics specializes in using metal ceramics for thermal management of LED arrays. Their BL2000 is a 7-cavity array with 6 LED dies per cavity. The BL3000 is a 39 cavity array.
The BL4000 reduces the size of the light source, and uses a circular ceramic base so that phsically, the BL4000 is comparable to a Luxeon star - at the output that lamina rates them for, the BL4000 seems to be competing with the Luxeon V.
Lamina gives a typical output of 120 lumens at 700mA, and a typical Vf at that current of 7.6V. I figured I'd run it through its paces and give some measured values.
First, some pictures:
The BL4000:
The circle is approx 0.8" in diameter. Lamina uses the same ceramic board for an RGB version of the BL4000. The other sets of solder pads are for the RGB version, so there are independent + and - pads for R, G, and B.
Here it is next to a Luxeon III star:
It's about the same size...but the BL4000 is very thin:
It's less than 1/16" thick!
Next to each pair of solder pads is a little oval window. On the white version, only one of those appears to have something underneath. Here is a close-up of that:
There looks to be some kind of semiconductor, and a bond wire. I'm really not sure what this is - maybe some kind of ESD protection? Or perhaps another diode in series with the array, to protect against reverse voltage? There is only one of these - the other two spots don't have the semiconductor or bond wire - I would imagine that on the RGB version, all three are populated.
Given the 700mA, 7.6V rating of this part, and the small emission area, I'd imagine that it's a series-parallel setup of high power LED dies, just like a Luxeon V. Here is the lamina powered up with a fraction of a mA:
You can see the 4 dies lit up - so this looks like it is indeed a series-parallel setup. The dies appear to be spaced rather far apart too, unlike the tightly arranged Luxeon V dies. The phosphor definitely is not a conformal process like on Luxeon LEDs - so there is a significant amount of color shift depending on the viewing angle. This also leads to a large light source compared to the size of the actual dies.
Now for some test data. The datasheet specifies test data to 800mA...but, given the information in the datasheet about thermal resistance and maximum junction temperature, it seems you should be able to drive it to much higher currents. I've done tests up to 1300mA.
First, the Vf vs. If curve:
The BL4000 has a Vf curve that's quite linear, but also quite steep. It's quite a bit higher than a Luxeon V. I'm wondering if that mysterious semiconductor has anything to do with it...
Next, beam distribution.
I've recently set up a system in my garage that allows me to plot out beam distribution of a light source. I have an 8" diameter disk with angle markings, a light source holder, and my light meter, held at a distance of 1m.
Here are the results of the BL4000 beam angle measurements:
It's a nice pattern that's very close to a normal lambertian pattern.
Given brightness measurements and the beam distribution measurements, I can do a lumen calculation based on piecewise integration of the measurements. I don't have an integrating sphere to verify my results, but I've used this technique for measuring quite a few luxeons and the results always have been reasonable.
Here's how the lamina scales with current:
You can see that the efficacy numbers aren't that great - Lamina doesn't actually make their own LED dies - they are simply LED packagers - so they rely on other LED manufactuers for the LED parts themselves. Thus, they are at the mercy of someone else as far as efficacy goes. It looks to me like they're using some older generation power LED dies for the BL4000 - the "typical" values from the datasheet even show that they're only getting just over 22 lm/W. Output is really starting to level off at 1300mA - 1A is probably a good point to be running these at.
It seems that these numbers are pretty low - using the same technique and instruments, I measured one of my Lux V lights to have an output of 103 lumens with a VV0S emitter running at 750mA - so again, the technique seems to produce reasonable numbers.
I'm not sure what I will be doing with this yet...but it was fun to poke at these to see what they're capable of...
The BL4000 reduces the size of the light source, and uses a circular ceramic base so that phsically, the BL4000 is comparable to a Luxeon star - at the output that lamina rates them for, the BL4000 seems to be competing with the Luxeon V.
Lamina gives a typical output of 120 lumens at 700mA, and a typical Vf at that current of 7.6V. I figured I'd run it through its paces and give some measured values.
First, some pictures:
The BL4000:
The circle is approx 0.8" in diameter. Lamina uses the same ceramic board for an RGB version of the BL4000. The other sets of solder pads are for the RGB version, so there are independent + and - pads for R, G, and B.
Here it is next to a Luxeon III star:
It's about the same size...but the BL4000 is very thin:
It's less than 1/16" thick!
Next to each pair of solder pads is a little oval window. On the white version, only one of those appears to have something underneath. Here is a close-up of that:
There looks to be some kind of semiconductor, and a bond wire. I'm really not sure what this is - maybe some kind of ESD protection? Or perhaps another diode in series with the array, to protect against reverse voltage? There is only one of these - the other two spots don't have the semiconductor or bond wire - I would imagine that on the RGB version, all three are populated.
Given the 700mA, 7.6V rating of this part, and the small emission area, I'd imagine that it's a series-parallel setup of high power LED dies, just like a Luxeon V. Here is the lamina powered up with a fraction of a mA:
You can see the 4 dies lit up - so this looks like it is indeed a series-parallel setup. The dies appear to be spaced rather far apart too, unlike the tightly arranged Luxeon V dies. The phosphor definitely is not a conformal process like on Luxeon LEDs - so there is a significant amount of color shift depending on the viewing angle. This also leads to a large light source compared to the size of the actual dies.
Now for some test data. The datasheet specifies test data to 800mA...but, given the information in the datasheet about thermal resistance and maximum junction temperature, it seems you should be able to drive it to much higher currents. I've done tests up to 1300mA.
First, the Vf vs. If curve:
The BL4000 has a Vf curve that's quite linear, but also quite steep. It's quite a bit higher than a Luxeon V. I'm wondering if that mysterious semiconductor has anything to do with it...
Next, beam distribution.
I've recently set up a system in my garage that allows me to plot out beam distribution of a light source. I have an 8" diameter disk with angle markings, a light source holder, and my light meter, held at a distance of 1m.
Here are the results of the BL4000 beam angle measurements:
It's a nice pattern that's very close to a normal lambertian pattern.
Given brightness measurements and the beam distribution measurements, I can do a lumen calculation based on piecewise integration of the measurements. I don't have an integrating sphere to verify my results, but I've used this technique for measuring quite a few luxeons and the results always have been reasonable.
Here's how the lamina scales with current:
You can see that the efficacy numbers aren't that great - Lamina doesn't actually make their own LED dies - they are simply LED packagers - so they rely on other LED manufactuers for the LED parts themselves. Thus, they are at the mercy of someone else as far as efficacy goes. It looks to me like they're using some older generation power LED dies for the BL4000 - the "typical" values from the datasheet even show that they're only getting just over 22 lm/W. Output is really starting to level off at 1300mA - 1A is probably a good point to be running these at.
It seems that these numbers are pretty low - using the same technique and instruments, I measured one of my Lux V lights to have an output of 103 lumens with a VV0S emitter running at 750mA - so again, the technique seems to produce reasonable numbers.
I'm not sure what I will be doing with this yet...but it was fun to poke at these to see what they're capable of...