Interesting LED read...

This is a copy of the article I sent to Jake at reefbuilders...

"While many reefers had their attention pointed to MACNA this weekend, Cree technologies was busy updating their site with some info on their new XP-G line of LED's.
http://www.cree.com/products/xlamp_xpg.asp

These LED's are more efficient, smaller, and higher output than anything on the market so far according to Cree. At only 3.5 x 3.5mm, they take up less than 1/4 the area of the XR-E model, so you can bet on a packaging in the future similar to the MC-E which is 4 of the XR-E's in a 7 x 7mm chip.

Efficiency ratings can be misleading, since they depend on the binning of the LED as well as the temperature and power. LED's tend to be more efficient at lower currents, so the best efficiency numbers companies boast are often lower than what their actual use would allow to be cost effective (350mA is common). In the past, there have been some binning labels on LED's that never saw the open market either. Even with all of this taken into account, the XP-G marks a significant step forward.

According to the Cree Product Characterization Tool, it would take 40 of the R2 bin, cool white XR-E LED's to make 10,000 lm (for reference, Ushio lists their 250 watt 10-14k bulbs to have an output of 10,500 to 11,000lm). At this quantity, each LED would run at 1 amp or 3.7 watts for an efficiency of 67.8lm/W. At 350mA, the efficiency of the XR-E/R2's increases to 98.6lm/W, which also requires 88 LED's to make 10,000lm. So to make about 10,000lm of light, at least 40 LED's at 148 watts would be needed, or 88 of them for just over 101 watts at 350mA.

The M-binned MC-E(4 XR-E's in one) LED is listed to max out at 700mA, and to make 10,000lm would require 14 of these LED's at least, running at 78.9lm/W for just over 133 watts total. At 350mA, 24 of these LED's at 96lm/W for just over 107 watts total would make the same amount of light.

Based on the XP-G's data sheet, there is a S2 bin which makes 148-156lm at 350mA & 3V (1.05 watts), for 141-149 lumens per watt, giving technology like plasma-induction a challenge. Even if the S2 bin never reaches the mass market, the R5 bin still boasts up to 148lm at 350mA. At 1A (3.3W) the LED puts out close to 390lm, for 118 lm/W. This means 26 of these LED's at 85.8 watts total would make 10140lm of light! On the XP-G's main page, Cree states: "The XLamp XP-G LED is capable of 345 lumens at 105 lumens per watt when driven at 1 A, which is 46% brighter and 64% more efficient than the highest-performance XLamp XR-E LED at the same current." Even with those numbers which differ slightly from those derived directly from the data sheets, the increase in efficiency is considerable.

Just by the supplied numbers, these LED's should mean that 13 can do what 20 once did, making due with about 89 watts to replace what once took 148! That is a significant increase in efficiency! If you look at the data sheet, you will also note that the spectral graph indicates a dramatic increase to the proportion of blue in the output compared to the white XR-E/MC-E's. The photometric (lumen) scale doesnt measure increases to the blue spectrums with as much favor as the green and red, so if these calculations were redone with radiometric ("PAR") values, the huge blue spike with the XP-G would likely mean an even bigger performance advantage for the XP-G's. It is very likely that Cree will come out with multi-pack like the MC-E is to the XR-E, further decreasing the cost of high output systems. With a spectral output that almost seems to bypass the need for additional blue LED's all together (which aren't as efficient as the whites and often lowered
the combined efficiency of a fixture to no more than that of a pheonix 14,000K halide), fixtures designed with these LED's for other uses could be direct competition for any fixtures made for the purpose of 'sustaining marine life'. This is awesome news indeed for those who read between the lines...

The only potential downside so far is that the viewing angle on these is getting wider than with previous models. The XR-E's are 90 degrees, negating the need for optics all together for many DIY'ers. This is good since even though lenses may focus the light into a brighter patch (sometimes too much) they also cost some of the output due to geometry and clarity. The MC-E's are listed at 110 degrees. The XP-G's are listed at 125 degrees. This is great for those with frag tanks (at last!), but not so much for the taller/narrower aquariums.

Pricing and availability are still unknown at this point, but I have a few suppliers telling me they will let me know as soon as they do. Other than that, these LED's are the ones that we will likely see taking the place of many other lighting technologies available to reefers... the time for LED's has come (IMO, and as long as the supplied data is accurate).

Now if only Cree would come out with those Royal Blue (450-465nm) MC-E's running at 1A they promised... "

Seriously, those XP-G's look like the bee's knees... higher efficiency than ANY other technology on the market right now at ANY spectrum (3000K HQI halides hit 105 lm/W). The only exception might be the plasma induction Luxim stuff, but the XP-G's CAN beat that at lower currents... just not worth it. Besides, the Luxim unit isnt exactly available on the market...

And did you check out the spectrum? Not much need for supplimental blue with this one. You could use something like 90 watts of these to replace 250 watts of halide (and that is a conservative guess since that huge blue spike does more for PAR than the recorded lm value). For every 250 watt halide you replace with these, thats $40-50 a year you would save in electricity right off the bat, and no more bulb replacement costs.

~This was posted by Hahnmeister, a member of our local club and avid techie and skimmer/T5 tester.

 

yay, the minute I get T5's over my tank, there gonna be old news comparable to PC's with the new stuff thats coming out.

~Will.

 

Oh, T5 is not being phased out yet!!!