The myth of LED efficiency

[mcallahan]

Something somewhere in the spectrum is missing.

Ding! Ding! Ding!

I 100% agree. No LED maker is talking about spectrum though, which is smart marketing b/c no one (check the boards) is really interested in spectrum, its all talk about PAR/Shimmer/etc.

As hydro said, the gardening industry has BILLIONS of dollars backing it (including the government), so there has been research done on what spectrum of light is best for plants.

For corals/aquariums, that kind of money doesn't exist so the finding the right spectrum will be long term, much trial/error thing.

 

[mcallahan]

Agreed. However depending on your situation, ROI can be 6-18mos depending on what you do with your existing equipment. So, if they last 3-5yrs, you are still considerably ahead.

The biggest pitfall I see is that LED's are increasing efficiency about every 6 mos. So similar to most electronics, they are outdated quickly.

Based on what fixture and what tank? For my 90G, I figure I'll need to spend AT LEAST $1500 to get a decent LED setup and I could probably sell my MH setup for $400 (best case) for a difference of $1100.

For a power savings of $50/mo (best case, I doubt I'd see that), that's a 22 month ROI.

Now for a small (30-40G), I agree, LEDs are a great investment.

 

[ChrisKirkland]

Its my understanding that this isn't quite an apples to apples comparison. LED's provide a directional light source so most, if not all light is directed into the aquarium. A MH is an omnidirectional lightsource which emits light in every direction. Consequently, significant light is lost to the surrounding room before it ever reaches your tank.



I take particular issue with this. How did you come to this result? I did a side by side comparison with a 250W MH SE bulb next to a 48LED (16 x Cree XR-E Royal Blue and 8 x CRee XP-G white) array (total wattage 133W). Both were mounted at the same height. Although the metal halide emitted 30% higher PAR at a 4" depth, the LEDs showed 20% higher PAR at a depth of 24".

Depth .......MH......LED
4".............744......501
12"...........498.......425
18"...........324.......361
24"...........184.......224

BTW, I do agree that plasma is probably the future of high intensity lighting. However, from this vantage point, you appear to be making some gross assumptions without doing actual testing.

this is a good explanation of the differences between led and MH, I run MH myself but the light emitted from the LED will travel further through water than the light emitted from MH, thus allowing deep tanks to have coral grown on or near the bottom. Someone might have already mentioned this.

 

[serpentman]

Although I agree that more research is necessary, if a deficiency is discovered, its very easy to incorporate reds, yellows, etc. into your array.

 

[James77]

For a power savings of $50/mo (best case, I doubt I'd see that), that's a 22 month ROI.

You won't see a savings on electric that high unless you have a large tank or are in the top tiers of Calis electric rates. At my $.15/kwh I would save about $9/month switching to ~350w of LEDs from my 2x250 HQI halides and VHO actinics.

Even factoring in bulbs expense, the savings are far from amazing with the LEDs. It would take a good 4-5 years for ROI on a DIY LED system for me. I also do not believe every single LED will last their supposed 7-17(or wherever the claims now lie) years, there will be many burnouts and overall repair problems with LEDs, as there is with any lighting.

 

[serpentman]

Based on what fixture and what tank? For my 90G, I figure I'll need to spend AT LEAST $1500 to get a decent LED setup and I could probably sell my MH setup for $400 (best case) for a difference of $1100.

For a power savings of $50/mo (best case, I doubt I'd see that), that's a 22 month ROI.

Now for a small (30-40G), I agree, LEDs are a great investment.

Although I did my analysis on 250W MH's, I think you are way over on your estimated cost to light a 90. However, even using your numbers, assuming you purchase new bulbs for your MH's annually, this knocks the ROI down considerably.

 

[noahm]

I think the other part of the problem is that everyone seems to be comparing their experience with commerically available fixtures which are generally quite a bit behind the technology (or hamstrung by ridiculous patents). A few years back, it was 'You can't grow SPS under T5!'... ...

 

[ganjero]

Like I said before the technology is not there yet, but it looks promising and definitely more efficient. My main problem with the OP is that he keeps referring to lumens (Again lumens have nothing to do with photosynthesis, but we how much light the human eye perceives). Also he is testing this on plants and comparing to 600w MHs which are rarely use in aquaria.

The reason coral farms don’t use LEDs is simply because as a business they cannot take risks on a technology that hasn’t been perfected for this hobby, that is just common business sense. ( BTW Pacific East Aquaculture started using LEDs for some applications)
Your statements are also not very well supported by data but rather seem like personal opinions. It would help you to post what kind of equipment did you use ( for MH as well as LEDs), for how long did you use any specific equip. Take PAR measurements or some type of measurements that has to do with photosynthesis(not lumens). If you are posting in this forum and arguing that LEDs are not efficient to grow coral perform your test on corals not plants. I would like to know what kind and brand of lenses, drivers, LEDs, etc did you use.

If I was to use LEDs right now I would use a DIY fixture that I can tweak, repair, a mix the colors that I really need.

 

[jefathome]

Our corals actually don't all have the same zooxanthellae. The activity spectra of zoox is close enough for our purposes that you could basically ignore the differences, but you still can't come up with an ideal spectrum for corals. The problem is the pigments of the corals themselves. The light actually reaching the zoox is modified by the pigments and fluorescent proteins of the coral in such a way that the light reaching the zoox is not the same spectral composition of the light striking the coral. To make things even more complicated, the pigments change over time even within corals. As a result you could easily tune a light to be ideal for zoox, but the corals will completely undo all of that tuning and each specimen will mess with it in a different way.

It's not a lack of research that's prevented us from coming up with an ideal spectrum for corals, but the fact that there isn't one. We can go out into the field and measure the precise light use of coral across the spectrum, but it's a PITA and the results have limited applicability to other corals and even the same coral over time, so we usually just talk about PAR.

So is this not the same in plants then? Do they indeed all use the same exact spectrum?

 

[jefathome]

You won't see a savings on electric that high unless you have a large tank or are in the top tiers of Calis electric rates. At my $.15/kwh I would save about $9/month switching to ~350w of LEDs from my 2x250 HQI halides and VHO actinics.

Even factoring in bulbs expense, the savings are far from amazing with the LEDs. It would take a good 4-5 years for ROI on a DIY LED system for me. I also do not believe every single LED will last their supposed 7-17(or wherever the claims now lie) years, there will be many burnouts and overall repair problems with LEDs, as there is with any lighting.

:furious:
I hate you... 15 cents per KWH? Bastard!!
I was up in the 35cent range for a while (Teir5 PG&E)


obviously I don't really hate you

 

[jefathome]

jefathome,

I have solar tubes on my main tank. I had some extra LED strips and I was going to use them as supplements, but they just don't have the spread of a MH reflector. The LED's go almost straight down. So, no supplemental light under the solar tube. My corals all browned out over the winter (none died). So, I pulled the PC retrofits and put in 250w MH with T-5 retrofits between the tubes. Only run the MH two hours now. Took the corals about 10 days to color back up. I plan on running the MH about 4 hours in the winter.

Interesting... Maybe I'll use my T5's for supplement then. during the summer (due to heat) I cut back my MH lighting and only used T5's (I have mostly LPS and Zoa's). Nothing showed any ill effects...

So if I run the same T5's WITH the tubes, I should be ok since the tubes will only be in addition to the light that I already had.

In total this will be about 200watts.. 2x55w and 2x39w. Not bad for a 150g system... only about 1.3 watts per gallon.

 

[greenbean36191]

No LED maker is talking about spectrum though, which is smart marketing b/c no one (check the boards) is really interested in spectrum, its all talk about PAR/Shimmer/etc.

There is certainly a lot of interest in lighting spectrum among hobbyists. It's spawned some extremely long and complicated threads here as well as several articles by Dana Riddle. Also, the first question asked about almost every new bulb is "what's the spectrum look like?"

I actually think spectrum gets far more attention than it deserves. It's only of secondary importance for photosynthesis since it only plays a major role at intensities well below saturation. That's why in the lab and in the field we generally only measure intensity (PAR), like hobbiest generally do.

As hydro said, the gardening industry has BILLIONS of dollars backing it (including the government), so there has been research done on what spectrum of light is best for plants.

For corals/aquariums, that kind of money doesn't exist so the finding the right spectrum will be long term, much trial/error thing.

Again, it's not a lack of funding that keeps us from coming up with a "best" spectrum for corals, but the fact that there is no best spectrum for them.

We know quite well what the action spectrum for zooxanthellae is. It only takes a few thousand dollars to measure this, not billions. It's also quite easy to tune lighting to them, and it's done in the lab all the time.

The problem is the corals themselves. The pretty colors we love so much are from the pigments and fluorescent proteins that they use to modify the light hitting them. You can think of them as a series of filters and colored reflectors. The spectrum of the light hitting the coral is not the same spectrum as the light hitting the zoox in most cases because of these pigments and FPs. To give a coral an ideal spectrum you would have to measure how the light is being modified before it gets to the zoox and then adjust the spectrum of your lighting to compensate for the modification. That's where the problem lies though. Each colony uses a different suite of pigments and FPs and it changes even within the same colony over time. You can tune the spectrum to one colony at one time, but you can't tune it to multiple colonies, much less multiple species over long time periods.

We can go out in the field and measure what spectrum a coral as a whole is using, and it has been done many times, but that result only applies to that specific colony at that time and can't be extrapolated to the rest of the species, other corals of the same color, or other species.

 

[greenbean36191]

So is this not the same in plants then? Do they indeed all use the same exact spectrum?

No, zoox don't all use exactly the same spectrum, but they're at least similar. In that regard it's comparable to terrestrial plants.

The problem is that corals don't use similar spectra and the spectra they use isn't tied to their zoox.

 

[boodlefish]

I'm not an engineer or scientist by any means nor have I tested them for anything but lumen output.

I'll repeat what others have said, lumen output is not so relevant for growing corals.

The last fixture that I tested ACTUALLY worked! It was $5,500 and had the same lumen output as a 600 watt lamp. This fixture actually grew the plants quite well....the only problem is that it consumed 800 watts to do it. That is why the topic of this post was "the myth of efficiency" and not "LEDs don't work".

how about-- the myth of LED efficiency when used in: home gardening/ Hydroponics/ or whatever non reef related scenario you tested them in.

There is a reason why none of the major lighting manufacturers, like sunlight supply for example, do not sell them. You only hear about LED only companies producing these...so my point is if they truly worked there would be a Sunlight Supply brand fixture using LED technology. Please note that I'm not talking about some LED actinic strips, I'm refering to fixtures that are meant to be used alone with no supplemental lighting.

Well do Aqua Medic, Gisesmann, and Marineland count as major lighting companies/ manufacturing? They all have full LED fixtures in the pipeline, some are already available for purchase.

Sunlight supply as a big gun manufacturer cant make a profit off a fringe market (LED's) so smaller companies and individuals alike will make the initial R&D investment and once there is a big enough demand companies like Sunlight supply will buy up some designs and produce said fixtures at a fraction of the cost.

Disclaimer....I'm not arguing that LEDs aren't a great lighting choice in many situations, I'm arguing that LEDs can't grow coral or plants without being less efficient than MH lighting. I'm also arguing that b/c there has been little research on a per coral basis to know what color spectrum that each of these corals need to thrive, live long, and have the proper colors that we all are expecting from these high dollar fixtures. MH is proven and is becoming more efficient everyday, using digital ballasts and with more efficient cooling methods can lower the usage of a 400watt light by 25%.

I like that "cant grow corals or plants" dude you only tested it on plants not corals. If anything you should argue LED's cant grow plants (as efficiently) and based on your logic they shouldn't be able to grow corals either.

MH is proven, but even with the improvements in MH tech/ efficiency 25% is not so much. especially when compared to some of the DIY projects published here on RC.
http://www.reefcentral.com/forums/showthread.php?t=1718642&highlight=led+diy

+1 on the t5's although I'm only a fan of the VHO's. You are right about the 24", same with plant lighting. What I really appreciate about T5's is the ability to evenly spread the light over the tank or plant canopy. The VHO's put out their fair share of heat though, Sunlight Supply just came out with an air-cooled version of the VHO which I think was overdue. I keep telling them they need to make a remote ballast for the fixtures so that you can remove that part of the heat from the canopy. I guess I will have to build and sell one myself!

Once again an example of Sunlight Supply lagging on technology/ product improvement until there was enough consumer demand to warrant the expense of changing their manufacturing procedure. Adding active cooling their T5's was long overdue. If you decide to wait for them to implement technological advancements then you have more patients then me.

cheers :beer:

 

[kryppy]

Again, it's not a lack of funding that keeps us from coming up with a "best" spectrum for corals, but the fact that there is no best spectrum for them.

I wonder how much true development in metal halide, HPS and mercury vapor lighting schemes have been driven by the underground hydroponics crowd. Even LED fixtures for consumers come from them...

We know quite well what the action spectrum for zooxanthellae is. It only takes a few thousand dollars to measure this, not billions. It's also quite easy to tune lighting to them, and it's done in the lab all the time.

The problem is the corals themselves. The pretty colors we love so much are from the pigments and fluorescent proteins that they use to modify the light hitting them. You can think of them as a series of filters and colored reflectors. The spectrum of the light hitting the coral is not the same spectrum as the light hitting the zoox in most cases because of these pigments and FPs. To give a coral an ideal spectrum you would have to measure how the light is being modified before it gets to the zoox and then adjust the spectrum of your lighting to compensate for the modification. That's where the problem lies though. Each colony uses a different suite of pigments and FPs and it changes even within the same colony over time. You can tune the spectrum to one colony at one time, but you can't tune it to multiple colonies, much less multiple species over long time periods.

We can go out in the field and measure what spectrum a coral as a whole is using, and it has been done many times, but that result only applies to that specific colony at that time and can't be extrapolated to the rest of the species, other corals of the same color, or other species.

"The problem is the corals themselves.."

I love this quote. It applies to fish as well...lol

Undoubtedly the technology to control the intensity and wavelength of every single photon in the tank at will by computer is only a few years away and people will be including scripts with frags..

I've been running the biggest Solaris LED unit since they hit the shelves and have some very interesting results with all sorts of corals in my 300 gallon display tank. I should endeavor to take some photos that show how the highly directional stream of light cause odd growth on the SPS's along with some up at the lights from the corals perspective shots.

Best technology to hit the hobby since the invention glass if you ask me. Hard to burn your house down with them too!!

 

[blasterman789]

Several months ago I completed a lot of research into LED technology because I've had such good luck building high powered 'artsy' LED lights and wanted to expand. Some issues I ran across have ben touched in this thread, but I'd like to sum up some of the arguements as I concluded them in my own mind. I also build my own DIY lights for my reef tanks and consider 95% of the DIY crowd to be following the same mediocre designs.

- LEDs have not done well in large scale agriculture because of the need for orange / red light for most fruiting crops, and in this respect LEDs *cannot* come close to competing with high pressure sodium. While you can get 670nm with LED, it doesn't matter when you can get 150lpw from a cheap HPS and light a much larger area without needing a heat-sink the size of a small car.

- LEDs are most efficient at producing blue, then green. By adding a smidge of amber/red you get a cool-white LED, and this is what Cree keeps raving about in their press releases. However, when you move to neutral white or warm-white the efficacy curve drops dramatically because LEDs simply don't produce longer wavelength light very efficiently.

- LEDs are much more efficient than MH when it comes to producing 5500-6500k light, but it doesn't matter *see below*.

- Cool-white LEDs don't have any more PAR than a cool-white CFL. Might actually be less. Again, the visual lumens that cool-white LEDs are optimized for don't mean a thing for plants.

- This leaves blue, and LEDs are very efficient at producing 440-470nm light even though current blue LEDs are dated technology. More so than metal halide or any other off the shelf technology. It's light at this wavelength that contributes most of the PAR for marine based corals. I've tested and confirmed that blue light is the cause for bleaching of Acropora.

- It's virtually impossible to quantify the PAR several dozen 3watt LEDs and optics -vs- a much bigger point source 175watt or 250watt metal halide. Or, a highly diffuse source like fluorescent tube. We need bigger LEDs to do a true apples to apples comparison, and I'm still talking to manufacturers in Asia to build bigger blue LEDs and ditch this 3-watt nonsense. Cree is not the standard.

"The problem is the corals themselves.."

The problem with coloration and LEDs are that most LED lights contain cool-white LEDs that not only have mediocre PAR, but kill color rendition. Ditch the cool-whites for neutrals or warms and color improves - dramatically. No NASA scientist or Nobel prize winner declared that cool-whites must be used for reefing.

 

[boodlefish]

Several months ago I completed a lot of research into LED technology because I've had such good luck building high powered 'artsy' LED lights and wanted to expand. Some issues I ran across have ben touched in this thread, but I'd like to sum up some of the arguements as I concluded them in my own mind. I also build my own DIY lights for my reef tanks and consider 95% of the DIY crowd to be following the same mediocre designs.

I would like to see some pics of your not so mediocre DIY build.

- LEDs are most efficient at producing blue, then green. By adding a smidge of amber/red you get a cool-white LED, and this is what Cree keeps raving about in their press releases. However, when you move to neutral white or warm-white the efficacy curve drops dramatically because LEDs simply don't produce longer wavelength light very efficiently.

- LEDs are much more efficient than MH when it comes to producing 5500-6500k light, but it doesn't matter *see below*.

- Cool-white LEDs don't have any more PAR than a cool-white CFL. Might actually be less. Again, the visual lumens that cool-white LEDs are optimized for don't mean a thing for plants.

So you are suggesting that Cool-white (6500k) LED's are more efficient than Metal Halides but less efficient then CFL's or are you saying Cool-white (6500k) LED's are more efficient than Metal Halides (measured in lumens) but less efficient then CFL's (measured in PAR)?

- This leaves blue, and LEDs are very efficient at producing 440-470nm light even though current blue LEDs are dated technology. More so than metal halide or any other off the shelf technology. It's light at this wavelength that contributes most of the PAR for marine based corals. I've tested and confirmed that blue light is the cause for bleaching of Acropora.

Huh? So blue light causes bleaching? Then why can I run all "blue" T5 bulbs over my tank with out a problem but if I add one too many "white" bulbs my corals start to bleach?

The problem with coloration and LEDs are that most LED lights contain cool-white LEDs that not only have mediocre PAR, but kill color rendition. Ditch the cool-whites for neutrals or warms and color improves - dramatically. No NASA scientist or Nobel prize winner declared that cool-whites must be used for reefing.

But you just said that "neutral-white" and "warm-white" LED's are horribly inefficient. :hmm5:

Well with all that being said. suppose you were to recommend a LED configuration, what would you recommend?

Blue LED's for PAR and warm-white or neutral-white LED's to round out the coloration? Or maybe a different mix?

and when you say "kills color rendition" are saying that the actual corals fade or don't produce the correct colors or are you saying that the tank looks too blue?

 

[jcolletteiii]

I would like to see some details of the OP's 60,000lm, 800w fixture.

BTW, this thread attempting to trash LED's (when it appears the OP has no experience with the application of LED's to marine aquaria) does not have anything to do with the water-cooled MH aquarium fixtures that you are trying to develop, does it? You know, RC has a very harsh policy on shilling...

 

[lisafoster]

I would like to hear from people who have leds over their tank for more than 5 years to say how they really stand up over time. They are not cost effective yet in a few years maybe. I just feel there are going to be many improvements. I love my metal halides I have t5s over one tank and look wise it does not compare to my tank with MH. Although I love the way the colors of corals pop with actinic lights.

 

[canesfan44]

I would like to hear from people who have leds over their tank for more than 5 years to say how they really stand up over time. They are not cost effective yet in a few years maybe. I just feel there are going to be many improvements. I love my metal halides I have t5s over one tank and look wise it does not compare to my tank with MH. Although I love the way the colors of corals pop with actinic lights.

The problem with that is that there are very few people that have had LED's for 5 years and if they do, it's not going to be the same LED's that are currently being used in units like the AquaIllumination, MaxSpect, AcanLighting etc, and DIY units that use Cree XP-G's, semi-LED's etc. The LED's commonly used now weren't really around 5 years ago, and the oldest factory-built LED hood I can think of is the PFO Solaris, and that wasn't available five years ago. It first appeared a little less than 4 years ago and it used Luxeon LED's, nobody uses Luxeon LED's anymore. We all would like to know how the LED's that are currently being used hold up over five years, but we will need to wait a few more years until the 5 year mark comes around, and by then there will be even better LED's available.