Education Energy & Sustainability Furniture & Lighting Technology
Double-whammy lighting/heating energy saving tank hack
LED tankwater cooling hack.jpg

Flickr user fotogra4er replaced the fluorescent tubes lighting his aquarium with LEDs. Which, of course, make way more light and way less heat for the same amount of energy. Then he upped the ante by cooling the LED lighting bank with circulated tank water, exploiting what waste heat the LEDs do generate to warm it, and thus saving even more power that would otherwise go to the tank heater.

[via Hack a Day]

14 thoughts on “Double-whammy lighting/heating energy saving tank hack

  1. The LEDs are brighter, and this is a very clever hack, but I might worry about the spectrum of light that the diodes emit. The plants in the tank are evolved for daylight, and aquarium lighting manufacturers make a variety of lighting elements to match the emission of the sun.

    I don’t know what colors the diodes emit, but there’s a decent chance that his plants will die back. It may also affect his livestock!

    I’m quite certain that LED panels are unsuitable for saltwater reef tanks, as the corals are especially sensitive to the light.

    I’d spend some time researching this before you follow suit.

  2. Many people have made led lights for planted tanks. Yes, there are spectrum issues, but there is research out there showing that plants aren’t nearly as spectrum dependent as people think, and a lot of anecdotal evidence showing leds can be used successfully. As far as aquarium lighting manufacturers matching spectrum, I assume you just mean light manufacturers since all the fluorescent/CF lights that you find in aquarium shops are simply re-branded lights from GE, Phillips, etc. They are specialty lights and so overpriced that everyone that I know would never consider using them because they are a waste of money and the light is UGLY. Instead the most common practice is to buy fluorescents with the highest CRI so that you get the best color rendition. Actinics for reef aquariums are a whole other story, but there are leds that output similar spectrum as those and diy builds are out there to be found. Anyway, if you search for diy planted tank lighting with leds, you will find a plethora of successful projects. They can be damn expensive however.

  3. The plants will adapt, but definitely be prepared for die-back. I went from a 15 watt fluorescent to a 15 watt led setup, the color temp appeared the same to my eye, but the plants can tell the difference. What was once a pale and leggy sword plant is now very bushy and compact. Once it grows more roots I’m sure it’ll explode in size. Of course, the algae appreciated the increase in light as well.

    Concerning cooling the leds with tank water, I don’t like the idea additional cleaning of tubing and pumps, as well as dissolving trace metals into the tank water. Can’t use copper as that’s highly toxic, so you have to use aluminum, which is also going to dissolve oxides into the water. Using SS or titanium would be crazy expensive.

  4. Last I checked, which was admittedly about a year ago, cutting-edge LED lighting technology was just starting to approach the efficiency of run-of-the-mill fluorescent bulbs. A T8 fluorescent driven by a high-efficiency ballast gets upwards of 80 lumens/watt, and when I dug deep into the literature (because LED lightbulb manufacturers had buried this information as thoroughly as possible), it looked like the LED bulbs I could actually buy were getting just a bit over half that.

    Don’t be misled by claims about “equal to a 100-watt incandescent,” as those are easily fudged. For one thing, LEDs tend to put out a much narrower beam than fluorescent or incandescent bulbs, so if you measure just the center of that beam and compare it to a spot measurement underneath a fluorescent bulb (with no reflector), the LED’s number will probably be higher. Put the fluorescent inside a reflector, as you always would for a lighting fixture, and it suddenly looks like a much better deal.

    As I said, I haven’t dug into this recently, so my information could be outdated. However, a quick survey shows that LED manufacturers are still playing some of the same numbers games with their light readings, which suggests that they still haven’t really beaten fluorescents. When I see honest, comparable figures coming from the LED folks, I’ll start looking into them again. In any case, I seriously doubt that it’s appropriate to say that LEDs are producing “way more light and way less heat” than fluorescents, especially if there’s still enough waste heat from the LEDs to warm up a fish tank.

    1. If this person had replaced incandescents with LEDs, it would make sense, however, fluorescent lamps have always been highly efficient, and even modern LEDs have trouble matching fluorescents in efficiency. Heck, 4-5 years ago, LEDs were actually just beginning to match halogens for high-power lighting. (They were superior for flashlights because small halogens are FAR less efficient than large ones, plus they don’t dim well as the battery dies and they’re far more durable.)

      There is the advantage here of the fact that it is probably far easier to extract the waste heat of LEDs, so overall, the total system (lighting + tank heating) is likely more efficient if the tank heater was running reasonably often.

  5. The LED industry is just beginning to grow up and get their crap together with standard measurements. I share the disgust and frustration with others here on how pathetic it has been until now. If you look to reputable manufacturers (Pretty much Cree and one or two others) they are making a stab at being more transparent on specs.

    Water cooling is brilliant in this application as it really helps with the issue of LED heating, which isn’t really much heat, but it’s in a small space and therefore difficult to dissipate.

    To avoid the cleaning/contamination issues, how about a closed liquid circuit? Put a heat sink in the tank ala CPU coolers, but coated to prevent contamination/corrosion?

    1. Yes, dropping a fluorescent bulb releases some mercury, and for that reason alone LEDs will be a welcome change – but only when they actually compete on energy consumption. That’s because one of the biggest sources of mercury pollution in the world right now is coal burning, which is where we in the US get half of our electricity. If you replaced all of the fluorescents with even slightly less-efficient LEDs, you’d likely end up in worse shape for mercury pollution.

      As for the fluorescent bulb disposal “problem,” it’s not really a problem. Just turn them over to your local sanitation department or recycler, and they’ll release zero mercury into the environment. The recyclers put the bulbs under vacuum, break them, recover the mercury, the metal, and even the glass, and turn it all into new products.

      I really am looking forward to switching over to LEDs, but we need to tone down the hype about ten notches and make realistic comparisons. Otherwise, we’ll throw out a solid technology for one that’s still in beta.

Comments are closed.


I am descended from 5,000 generations of tool-using primates. Also, I went to college and stuff. I am a long-time contributor to MAKE magazine and My work has also appeared in ReadyMade, c't – Magazin für Computertechnik, and The Wall Street Journal.

View more articles by Sean Michael Ragan