der_wille_zur_macht
Team RC
<a href=showthread.php?s=&postid=15266143#post15266143 target=_blank>Originally posted</a> by dtech
Soundwave said before that the blues didn't need to be at max amp since they are more for color than anything, the growth comes from the CW.
FWIW, that is in direct conflict with what other LED-experienced folks have claimed. I'm not trying to start a fight, just pointing this out. IMHO it's pretty early in the game to know details like these, since I don't think anyone has done conclusive testing. Look at the spectral plots of either of these LEDs - the shapes of the curves aren't like anything else we use in reefkeeping, even for lights that end up "looking" the same to our human eyes. To me, this says we shouldn't jump to conclusions yet!
If I remember correctly, the blues have hardly any extra output when driven at 1000mA. I went with 700 to lessen the heat generated by a 1A current.
Looking at the datasheet now, I notice some interesting things.
1) Blue holds output better than white and royal blue vs. junction temp. In other words, if your heatsink sucks, your blues will still perform OK, but your whites and royals will suffer a bit in terms of output. Going from 25 deg. C to 150 deg. C, blue loses only like 5%, while white and royal lose like 25% - 30%.
2) They only provide plots of % rated output vs. current for blue, white, and green. Blue and green have the same curve. Unfortunately, royal isn't plotted. So, at least we know that blue and white will perform the same in terms of current - in other words, you'll get the same % more light jumping from 700 to 1000mAh for both blue and white.
In the real world, I would assume you have to blend these two conclusions, since it's likely that our LEDs will get much hotter as we put more current though them. So, white and royal blue probably lose more performance at 1000mAh than blue does.
In the end, it's all probably somewhat of a wash, since so many people are using dimmable drivers. Personally, I see it as a tradeoff. All of these LEDs will give off less lumens per watt consumed at higher power (in other words, they are less efficient at higher currents), so IMHO you have two choices: 1) drive at lower currents and use a few more LEDs. This is more efficient in the long run, but more cost up front. 2) Drive at higher current and use fewer LEDs. Less up front cost but more long term energy use, and the LEDs will die sooner.
If I was making a large tank, or a deep one, I'd probably use TONS of LEDs with tight optics, driven at low currents. This would be a huge upfront cost, but more efficient over the long life I'd want out of a big tank. If I were building a nano, I'd use fewer LEDs, wider optics, and drive them a bit higher.
I'd be interested to hear if others would buy in to this same strategy?