Jump to content


  • Content Count

  • Joined

  • Last visited

About uglybuckling

  • Rank
    insomniac reefer
  • Birthday 12/08/1982

Contact Methods

  • Website

Profile Information

  • Location
    North Carolina
  • Interests
    HCI/computers, neurosci, violin, fish/reefs
  1. uglybuckling

    ugly's fugly natural tank

    I gotta take some more pictures. Everything is so much bigger now. =P Got a pretty sweet RAW-format P&S camera coming for Christmas, so hopefully the next round of pictures will be easier than these were. Also, the tank is doing great. More than I can say for myself at this point, but December will be over soon. If you don't hear from me sooner, I'll catch you on the flip side.
  2. uglybuckling

    Full Spectrum LED tank pictures

    Interesting. I had a fiddly string awhile back and it turned out to be a grounding issue--one of the solder blobs was touching the aluminum on the edge of a star. But if you've already checked your soldering a zillion times, the issue is more likely what others have suggested--a bad emitter. Checking them with a multimeter is painful but at this point likely necessary.
  3. uglybuckling

    Full spectrum LED layouts

    Sorry for the lack of updates. I'm still alive--work and the holidays have just gotten pretty busy for me. So anyhow, layouts for Steve's LEDs linear heatsinks are probably the most commonly-requested thing in this thread, along with more info about drivers. Since I (theoretically) have a week off coming up, I'll see what I can do on those two issues. Meantime, I'm taking requests. Wanna see a video about what life is like with three reef tanks, five reptiles, a hedgehog, and no furniture? Yeah, I didn't think so. I'll get to work on the driver video.
  4. uglybuckling

    Underwhelmed by NW

    The latter statement is confusing for me. Two cool whites and two royal blues is going to have very little incident red light. Is your point here that intensity of white light (i.e. in the setting of a fixed amount of blue, essentially color temperature) rather than spectral content is of highest importance? I agree with most of what has been said, in particular that there's a lot of green and not much red in that fixture. If that's intentional, and it's an aesthetic preference thing, that's perfectly reasonable, and I can't really say like, "hey, you should build your fixture this other way." But expecting a fixture with that much green and blue (and it really is a lot; that's like fluorescent light levels of cool white, green, and blue) to respond to the admittedly small amount of red in a neutral white LED in a way that is noticeable is....well, probably unrealistic. My guess is that the effect you were looking for was there, but was just completely overwhelmed by what was already present in the fixture, to the point where it was not noticeable. TLDR there's more than one way to skin a cat. One combo of LED colors that works well is the 6:3:3:1 RB:NW:TV:OCW ratio. Another way uses lots of cool white and green LEDs like your fixture. Many commercial manufacturers have gone the cool white route and built some admittedly nice-looking fixtures. Trying to grab an element that has a specific effect in one fixture and get the same effect from it in another fixture may or may not work well. Your testing, however, of this, is really appreciated and is a valuable lesson for anybody who is looking to do something similar with a fixture like yours. The finding may not be generalizable, however--adding a few neutral whites to a fixture that already has a lot of cool whites and greens is in theory and may be in practice very different from basing a fixture solely on neutral whites. In other news, I added four mushrooms to my ice cream the other day and I'm very much underwhelmed by mushrooms. I could hardly taste them, but what little I could taste, just didn't seem like it achieved the effect that people describe when they add mushrooms to things.
  5. uglybuckling

    Full Spectrum LED tank pictures

    It would be more cost effective to use two smaller heatsinks rather than one big one. I have a 48 inch tank and use two 6x9 inch heatsinks over it. Save money on aluminum and use it to buy better LEDs. There's a thread around here that I made awhile back with full-spectrum LED layouts for most common tank sizes. Should be a reasonable starting point for how to proceed using Cree or Rebel LEDs.
  6. uglybuckling

    BioCube LED Build HELP

  7. uglybuckling

    DIY LED for a 20Long

    http://www.nano-reef.com/forums/index.php?showtopic=311998 http://www.youtube.com/watch?v=gdf_6oWsdZ4 Solderless is okay but you have to find sources for all the solderless LEDs. It's a pain.
  8. uglybuckling

    Easiest SPS to keep in a 3 gal

  9. uglybuckling

    Large Fish Only LED Ideas

    You will get FAR more light per driver if you use Crees than if you use Bridgelux. You will also get more light per dollar of initial outlay. And more light per dollar on your power bill over time. The efficiency of the Cree and Luxeon Rebel LEDs is pretty awesome by comparison to most of the other emitters on the market. (another way to think about it is, for the same amount of light, you will need fewer emitters of the Cree or Luxeons, and thus can save money) Furthermore, I don't trust anyone but the above two brands' neutral whites. The offbrand cool whites are okay for cool whites (I still wouldn't use one in one of my fixtures, but more because it's a cool white than because it's a specific brand)...but good neutral white phosphors are very difficult to produce and although I've definitely seen pretty good neutral whites from (companies other than Cree and Bridgelux), I've also seen pretty terrible ones from those same companies. In the same bin. In other words, it's not that they're all bad. It's that they are inconsistent. For same-current, I'd go with 6:3:1 or even 8:4:1 blue:white:ocw. Why? Blue and white should be roughly the same ratios we use for other builds (which is to say, two blues to one white). OCWs usually run at lower amperage than blues and whites, and as such will be over-represented if we run the blues and whites at the same amperage as the OCWs. (In other words, at the same amperage for all, blues/whites will be dimmer and OCWs over-represented). So instead of the usual 6:3:1, I might go as high as 8:4:1. (in other words, a little OCW goes a long way)
  10. uglybuckling

    Phillips vs Cree Royal Blues

    On average I was under the impression that the Phillips royal blues (at least the good-binned ones) were five to ten nanometers LOWER wavelength (more purple) than their good-binned Cree equivalents. Is this more an eyeballing thing or a spectrophotometer thing? Also, if you don't mind my asking (and you can answer in a PM if you like) where'd you get these odd-looking Phillips LEDs from? Tangentially relevant: I read an article yesterday about why the sky ISN'T purple (after all, people answer "why is the sky blue?" with some hand-waving nonsense about short wavelengths cutting through the atmosphere better; it would then stand to reason that since violet light would do better than blue at getting to us, the sky ought to be violet)--as it turns out, the answer is that the sky really is violet, but we just don't see it as that color because our eyes suck at seeing violet. Your experience with these LEDs may be similar--they may actually be violet but you can't see it.
  11. uglybuckling

    DIY LED noob seeks advice!

    Here are some videos that may help: The big one: http://www.youtube.com/watch?v=gdf_6oWsdZ4 Shorter ones that are more focused: To answer your questions directly: Wattage is a measure of power. Power = voltage * amperage. Amperage determines brightness. Voltage is basically "how many LEDs can I have in my string?" because usually the maximum voltage a driver can put out (rather than its maximum wattage) is the limiting factor. By way of example, let's use a 40W Inventronics driver with a maximum voltage of 54 volts and a maximum amperage of 700mA: Assuming our emitters require 3V each at 700mA, If we use voltage to determine how many LEDs we can get on the driver, we get 54V / 3V per LED = 18 LEDs. If we use wattage (= amps * volts) to determine how many LEDs we can get on the driver, we get 40 watts / (3V per LED * 0.7 amps for the string) = 19 LEDs. It is important to note that NEITHER OF THESE NUMBERS IS CORRECT because you have to leave overhead on the drivers (so you should use the fifteen LEDs that Milad recommends!) but you get the point--voltage is limiting. So figure out the voltage your LEDs need (all of them except true violet require 3v or less; true violets are around 3.6 volts), and then just make sure that when you add 'em all up it's like 5 or 10 volts less than the maximum on the driver you're trying to use to run them (to allow for overhead). The remainder of the plan is fine. Going to three clusters on a tank that high is probably unnecessary (you will have plenty of light mixing with just two clusters because there's a lot of vertical space to allow for good light spread) but if you want to do three I'm certainly not going to stop you; I don't think it'll result in more disco or anything as long as you keep the clusters fairly tight. Milad has the rest of the questions covered, I think: +1 to that stuff he said. =)
  12. uglybuckling

    Adhesive to mount optics?

    Hot glue gun to edge (not center!) of bubble worked for me.
  13. uglybuckling


    I picked up this same macro about a week ago from a local grower. I'm not sure if mine is C. racemosa or C. peltata, but it is doing very well and growing...eh, not a fast as some of my macros but still noticeably within under a week, which suggests to me that it may be invasive. Thanks very much for the timely ID. =)
  14. uglybuckling

    Full spectrum LED layouts

    My potentiometer wall warts also run my fans (at 10v DC), so they stay on all the time. So far it's totally fine. If you are using Meanwell drivers, the discharge of the capacitors (when the AC power to the driver goes out but the control circuit is still on) will cause a single bright flash shortly after lights out. As best I can tell this is not harmful; my array has been doing it for awhile and though it annoys me, I can't figure out a good workaround and like I said, I don't think it's hurting anything, so I haven't put a whole lot of effort into "fixing" it.
  15. uglybuckling

    Please help a newbie here :-)

    In short, a Cree XML setup is just brighter than other stuff (with the exception of arrays and the new Luxeon Ms). So you need fewer emitters. Less soldering, fewer (but higher powered) emitters and drivers, etc. Also less spread-out and more of a point light source. Crees are brighter than Bridgelux LEDs when run at the same amperage, too. By a lot. Just for the record. Yes, but it's done in sort of a sneaky way. Potentiometers regulate voltage, NOT amperage. So what is done is to create a separate "control" circuit for the driver, that includes the potentiometer. This circuit runs anywhere from 1-10 volts (hence the 1-10V way of controlling drivers). This control circuit then INSTRUCTS the driver on how much amperage to put out on the main (series-wired LED) circuit. It is, however, important to realize that the control circuit is separate and that even though what the knob changes is the control circuit VOLTAGE, the driver uses this voltage to figure out how much AMPERAGE to put through the main circuit.