how efficient are LEDs really?

[sphelps]

My point was that the fact they're coming from China is irrelevant. The prices are dropping because of improvements in technology, not because they're made overseas. I guess I was confused about why you brought it up in the first place.

The point was as new products emerge from R&D they are expensive as production runs are done closer to home where most of the R&D is preformed. This results in a higher cost for many reasons including higher labor costs and R&D investment return. As time goes on and if demand increases they would move production to other locations where labor is cheaper, among other things, in order to reduce the manufacturing cost. This is where the price decrease is likely coming from, not better technology. Better technology would mean a better product and possibly a higher price.

The same will happen with new products they release like the 200 lum/W LED which will be very expensive at first and then slowly come down as it goes through the cycle.

The other point to this was that LED fixture manufacturers are probably already all over this. If they where smart they would already be outsourcing large quantities of LEDs to get a rate much better compared to the rate you DIY'rs get. So you're right if you've already seen a drop in price for fixtures this could have already happened so again you may not see an additional drop unless manufacturers don't invest in R&D to continually increase quality with new LEDs or whatever else.

 

[der_wille_zur_macht]

Perhaps your argument might be valid for some types of products, but I don't see how this tangent is valid in this discussion.

We're currently seeing:

1) Raw LEDs are becoming much cheaper for a given level of efficacy. This is true at hobby levels and mass-production levels. Honestly, I don't care why this happens, but that doesn't change the fact that it is happening.
2) Price/performance is becoming more favorable for finished products at a similar rate. This is directly a result to the improvements in cost/performance of the raw LEDs themselves. There are certainly other factors influencing price, but at the least, the improvement in cost/performance of the raw LEDs will continue to drive prices down for the foreseeable future.

I don't understand your theory that the fixtures will suddenly stop dropping in price. Did the price/performance of personal computers suddenly stop dropping 18 months after they came to market? LED price/performance is essentially tied to the same influences. I know that the price/performance of commercial LED fixtures offered in this hobby will continue to become favorable for the foreseeable future. I'm not speculating - I know, firsthand, what the pricing structures and production costs are for these fixtures, because this information has been shared with me by the manufacturers and suppliers in this market. I'm not speculating based on magic theories about production moving overseas. The price/performance ratio is improving because of improvements in technology. This is a simple fact, and it's not something that's going to go away in the near future. Roughly two years ago when the first LED fixtures hit the reefkeeping market, an LED that gave 60 lumens/watt cost roughly $5 (wholesale). Today, an LED that produces 130 lumens/watt costs about that (again, speaking in wholesale terms - what the manufacturers of finished products are paying when buying 1000 units at a time). This new LED represents a change in technology vs. the old LED. The manufacturers of finished products took advantage of that change in price/performance by adopting the newer 130 lumen/watt LED, which gave their fixtures better performance for essentially the same manufacturing cost. The improvements were passed on to end customers, in terms of better price/performance ratios for the finished product. In a few more months/years, LEDs in the ~200 lumens/watt range will likely be available at this price point, which will represent another shift in price/performance.

Again, I'm really not even sure why we're bothering to argue price, because - again - LEDs offer advantages that some of us appreciate, regardless of price. I'm not trying to change anyone's mind, or tell people that MH is crap (it isn't - I like MH; I just happen to like LED more for certain functions). I'm not claiming that LEDs will create world peace and cure cancer. I'm just trying to speak fairly in their defense, since some people seem bent on tearing the whole concept down.

 

[beeker]

Beeker, I have a question regarding the Par38s. From what I've read, in a quick Google search, Par38 LEDs are only available in 2700K, 3000K and 4200K. Well, that's only from a couple of the Google results.

I'm VERY unfamiliar with LED lighting, but would like to attempt my own DIY, over my 40 breeder. Heck, eventually, maybe I'd do something for my 200DD.

Par38's are made by Evilc66 on nanotuners and nanocustoms

they are a standard screw in light bulb, with 5 Cree leds per bulb for a 12k color they use 3 white Leds and 2 Royal blue leds, these lamps put out 21w's but only consume 15w's of power, they are an all in one lamp, no fan needed and come with optics

the 20k Par38 bulb use's 3 royal blue leds and 2 white leds just like the 12k it is basically a plug and play bulb

track lighting/sockets and plugs for the track are all you need to get them to work, or any standard light fixture you'd like to use will work as long as the heatsink on the PAR38 isn't covered

 

[sphelps]

Like I said before it's simply speculation and I'm not guaranteeing anything, it's simply the way I see it. You may think a fixture for an aquarium can be looked at in a similar way as a computer because it's LED based but to me it's still a fixture for an aquarium not a computer and these can't be compared. The computer industry is completely different than the aquarium industry. Products made for aquariums rarely drop in price once they meet production levels. Yes you will see more competitors and those competing on commodity will have a cheaper product but the quality units aren't exactly going to get much cheaper, IMO. If I look at quality fixtures other than LEDs I see many in the $2000 mark, these fixtures are usually made from aluminum to act as a heat sink and reducing the need for fans. If I look closer that price goes up even further when controller features are added. So do I think a quality LED fixture which required the aluminum heat sink and has controller features and better efficiency will be much cheaper than the $2000 mark, of course not and that has nothing to do with TVs and computers getting cheaper. I'm not looking at commodity but a quality product that will be based on supply and demand. I even wouldn't be surprised to see an increase in price as a reflection of more quality and reliability put forth by the manufacturer.

 

[Flanders]

when it comes to the cost of new technology, the aquarium industry isn't any different from any other niche industry.

 

[der_wille_zur_macht]

Perhaps the disconnect here is due to the development curve of LEDs vs. other technologies adopted recently by the hobby. LEDs are becoming better by the minute, and WILL continue to do so for the foreseeable future. I'm confident that will translate into improvements in price/performance in the hobby.

Meanwhile, many other technologies are essentially "done" by the time our hobby adopts them. The core technology behind a "new" product in the hobby might be making small improvements at best. Hence, finished products implementing the technology in this hobby likely won't see a change in price/performance ratio. Based on the information I have, this is not the case for LEDs.

 

[sphelps]

So are you suggesting the advancement of LEDs is tied directly to this hobby?

 

[der_wille_zur_macht]

So are you suggesting the advancement of LEDs is tied directly to this hobby?

No, quite the opposite. LEDs are getting better regardless of this hobby, we're just riding the wave.

 

[beeker]

Couldn't have said that better

i remember when i got into the hobby 5+ years ago, i looked into PC, T5 and Mh and decided that i wanted to get a fixture that i didn't have to worry about what i'd be able to keep under it with a decent price and something i could use longterm

So i bought a 175w MH retro with a magnetic ballast as my first light ever for a 20g high, and after all this time i still have the retro on a 12 frag tank, though bulb replacement is 60 bucks a year it has lasted and that has been the benefit of hobby and the interest into MH when it became a proven technology to work at keeping corals healthy and vibrant

After all this time i'm happy to see people moving forward into Led technology not only for the aquarium hobby but to replace most lighting around the world for the green impact it has, i'm not a tree hugger but i realise where were at and where we should be headed so hopefully the led technology continues to be driven by us hobbyist as well as the rest of the world looking into cheaper and more effecient lighting

i mean remeber when nintendo came out...and now were at PS3 and Xbox360, technology is always advancing in someway and it hobbyist like us definatly give companies more of a reason to develope products that work

 

[salty joe]

Do the impressive LED builds seen around here yeild more PAR/watt than off the shelf par38 fixtures?

 

[kryptic4l]

to all of you who have owned and used a fixture for 10 year's kudo's.

My day will come when I own a setup longer then a year without wanting something different.

I do like the idea of t5 - led hybrids however. until there becomes more viable led options then royal blue and cool white.

 

[overklok]

Once LED's are really bright enough and widespread, I think the main advantage will be less heat generation. MH's really crank up the heat.

 

[wld1783]

to all of you who have owned and used a fixture for 10 year's kudo's.

My day will come when I own a setup longer then a year without wanting something different.

I do like the idea of t5 - led hybrids however. until there becomes more viable led options then royal blue and cool white.

I'm running 84 Cree Cool Whites, 156 Royal Blue, 4 36" actinic t5 with two 36" 50/50 T5s on a 330 gal SPS and could not be happier with the results.

Great growth and short of 5 amps with very little heat.

Bill

 

[der_wille_zur_macht]

Do the impressive LED builds seen around here yeild more PAR/watt than off the shelf par38 fixtures?

Impossible to answer without more details on the "off the shelf" fixture. We'd need to know the model and bin of LED, plus the drive current.

Sadly, very few off the shelf products provide that, so it's hard to say. But consider that an "average" bin of XR-E or Luxeon I/III/V cool white might only produce 50 - 60 lumens/watt, while the XP-G R5 that's easy to get now is in the 130's. So unless you know EXACTLY what you're getting in an off the shelf product, there can be 100% or more variation in efficacy!

 

[mr.wilson]

The benefit of a light fixture that may last 10 years is moot, if a better quality replacement comes out every two years at half of the cost. I also agree that our marine application will drastically shorten the projected lifetime of these lights.

Sanjay's most recent article http://www.advancedaquarist.com/2010/2/aafeature2#h4 shows that LED's have less PAR and smaller coverage than T5 or MHL. If you factor in the initial cost, the money saved through energy efficiency is diminished somewhat. The heat transfer issue is a plus for LED, but the artificial look and lack of light rays, shimmering & shadows leads me to stick with MHL until LED gets better. I agree with Sanjay's comment " I am excited about the future of this technology for reef aquarium use.", but until I get a couple free lights from the manufacturer, I will stick to a proven method.

I don't have much experience with LED, but I have had overheating issues where MHL has burnt out LED's in combination fixtures. The cheap LED fixtures I have seen lack a decent heat sink and fan, so overheating is possibly a greater problem with LED than with T5 or MHL.

LED technology will certainly improve, but it's also possible that MHL could be engineered to be more energy efficient and last longer before colour shifts and dimming. Even if Osram Powerball lamps were available in 10 or 20,000 K we would be focusing less on new LED technology. http://www.osram.com/osram_com/Prof...tal_halide_with_ceramic_technology/index.html

The lack of any beneficial, long wave UV A spectrum is a major drawback with LED in my opinion. Apparently a UV emitting LED is in the works, but that may take years to develop and as with all LED lights, the spotlight problem will persist. http://************.com/2009/12/14/...-development-for-the-grassy-ledio-spotlights/

 

[der_wille_zur_macht]

until there becomes more viable led options then royal blue and cool white.

FWIW -

There are lots of other viable options. You see those two colors because they match what most people want, and/or people don't understand the whole range that's available. If you're picky, the ability to fine-tune colors is stupefying.

First, you have the "named" colors of cool white (5kk - 10kk), blue (~475nm), royal blue (~455nm), plus greens, blues, reds, and UV if you want.

Then, within each of those colors, there are color bins specifying a very narrow band of the overall color's range. For instance, you can get Royal Blues that peak at 450nm, 455 nm, or 460nm. Then, you can get blues at 465, 460, 470, or 475. So that's 7 different "blues" to pick from.

Among the whites, there's even more variation. Since they're "fully spectrum," instead of binning by a dominant wavelength, the white LEDs are binned by region on the CIE 1931 color space. The bins available for the cool white LEDs range from 5,000k - 10,000k, but since kelvin is a one-dimensional representation and CIE 1931 is essentially 2-d, the bins go "above" and "below" the BBL (which represents the K scale). What that means is that instead of just picking how "white" the LEDs are, you can pick exactly what tint they have. The BBL essentially ranges from yellow to blue - the bins above and below that line can tend towards purple, green, orange, and other colors - again, giving you a very fine degree of control over color. As an example, the cool white XR-E has 13 different color bins available!

Most people only talk about "cool white" and "royal blue" because they buy from retailers that essentially default a color bin near the middle of each color's range. But in most cases, these vendors will also let you specify a certain bin, which gives you the granularity of control described above, if you want it.

 

[der_wille_zur_macht]

Sanjay's most recent article http://www.advancedaquarist.com/2010/2/aafeature2#h4 shows that LED's have less PAR and smaller coverage than T5 or MHL. If you factor in the initial cost, the money saved through energy efficiency is diminished somewhat. The heat transfer issue is a plus for LED, but the artificial look and lack of light rays, shimmering & shadows leads me to stick with MHL until LED gets better. I agree with Sanjay's comment " I am excited about the future of this technology for reef aquarium use.", but until I get a couple free lights from the manufacturer, I will stick to a proven method.

It's worth noting that the commercial fixtures he tested essentially use "old" LED technology - the "best" available today is 30 -50% more efficient, which would have changed the results drastically.

I don't have much experience with LED, but I have had overheating issues where MHL has burnt out LED's in combination fixtures. The cheap LED fixtures I have seen lack a decent heat sink and fan, so overheating is possibly a greater problem with LED than with T5 or MHL.

Again, this is not a problem with LED technology, it's a problem with a specific implementation of the technology. There are PLENTY of cheap, poorly engineered MH or T5 fixtures that fail prematurely, too.

The lack of any beneficial, long wave UV A spectrum is a major drawback with LED in my opinion. Apparently a UV emitting LED is in the works, but that may take years to develop and as with all LED lights, the spotlight problem will persist. http://************.com/2009/12/14/...-development-for-the-grassy-ledio-spotlights/

There are several UV LEDs available, FWIW - and have been for years. Though, I'd be interested in knowing how UV was beneficial, because it's clearly the case that light rigs that don't produce UV can grow beautiful corals.

Oh, and spotlighting being a problem? Again, that's poor fixture design, NOT inherently the fault of LED technology. The beauty of LEDs is that you can slap a $1 optics on the emitters and get the EXACT spread you want. If you have a fixture that's creating spotlights, the designer of the fixture can just use wider optics. Similarly, if a fixture is being designed to mount high above the tank, narrower optics can be used.

FWIW, Sanjay also made the following statements:

From this it should be quite clear that statements directly touting LEDs as being equal to 250W metal halides are correct.

Additionally, it should be noted that since the LEDs have virtually no heat input into the tank the LEDs can very easily be placed closer to the water. This would increase the peak light levels in the tank and also put more light directly into the tank, given the smaller spread of the light distribution. This makes more efficient use of the light as most of the generated light can be directed into the aquarium.

 

[Mustang Boy]

Impossible to answer without more details on the "off the shelf" fixture. We'd need to know the model and bin of LED, plus the drive current.

Sadly, very few off the shelf products provide that, so it's hard to say. But consider that an "average" bin of XR-E or Luxeon I/III/V cool white might only produce 50 - 60 lumens/watt, while the XP-G R5 that's easy to get now is in the 130's. So unless you know EXACTLY what you're getting in an off the shelf product, there can be 100% or more variation in efficacy!

i believe the PAR38 he was talking about is the bulb that evil created that uses 5 cree LEDs. evil has also posted PAR graphs of the bulbs with each optic choice and at varying heights.

 

[Shakes]

I have had my 36" eco-lamps KR-92 for just over 2 months now and I can say I am impressed. I recently converted from a FOWLR tank to a reef tank and decided to go with the KR-92 due to ChingChai stating he had better growth with the KR-92 LED than with his T5s.
I got back from a month long trip to Thailand last week. During that time my wife did not do a water change, the skimmer had stopped working due to the air intake had been partially clogged and my refugium macro algae all died due to (what I suspect) not having enough light. I attempted to use a 6.9 watt Phillips LED that was at the Japanese store (not designed for aquarium). Apparently it did not have enough PAR so I just purchased a nanocustoms 38 PAR 21 watt CREE light. BUT in spite of all these issues I received what I consider very good growth from my SPS. I just purchased the Apex controller and the temperature never gets more than 0.6 degrees above my high setting (Temp set for 74.5-74.8 degrees). I researched many LED's prior to my purchase and found some to be very cheaply built! The KR-92 is built extremely well!! LED's are expensive and I expect the price to come down quite a bit over the next 5 years as better and more LED manufacturers compete but at the current price point I think they will pay for themselves long term due to less energy consumption, no chiller requirements long lifespan etc. I am proof that LED's will grow SPS in spite of poor tank husbandry which I plan to improve upon as long as I can limit my time away from home.
I will post pictures to my album with an update to the SPS growth You can see my first months SPS turn around on it currently.

 

[solos2]

The lack of any beneficial, long wave UV A spectrum is a major drawback with LED in my opinion. Apparently a UV emitting LED is in the works, but that may take years to develop and as with all LED lights, the spotlight problem will persist. http://************.com/2009/12/14/u...io-spotlights/

To quote an article by Dr. Sanjay Joshi:
http://www.personal.psu.edu/sbj4/aquarium/articles/Photosynthesis.htm

The absorption spectrum for zooxanthellae has been shown to have a broad peak in the 400 to 500 nm waveband (blue-green) and a narrow peak in the 650 to 700 (red) waveband.

FWIW, I don't see how UV-A light is beneficial to coral growth, and I've seen UV filters are sold along with MH fixtures to block UV light. I don't think there's any documented evidence suggesting that UV-A light is proven to enhance coral growth either.

Like I said before it's simply speculation and I'm not guaranteeing anything, it's simply the way I see it. You may think a fixture for an aquarium can be looked at in a similar way as a computer because it's LED based but to me it's still a fixture for an aquarium not a computer and these can't be compared. The computer industry is completely different than the aquarium industry. Products made for aquariums rarely drop in price once they meet production levels. Yes you will see more competitors and those competing on commodity will have a cheaper product but the quality units aren't exactly going to get much cheaper, IMO. If I look at quality fixtures other than LEDs I see many in the $2000 mark, these fixtures are usually made from aluminum to act as a heat sink and reducing the need for fans. If I look closer that price goes up even further when controller features are added. So do I think a quality LED fixture which required the aluminum heat sink and has controller features and better efficiency will be much cheaper than the $2000 mark, of course not and that has nothing to do with TVs and computers getting cheaper. I'm not looking at commodity but a quality product that will be based on supply and demand. I even wouldn't be surprised to see an increase in price as a reflection of more quality and reliability put forth by the manufacturer.

I think most of the $2000 in the high end LED fixture you suggested, went to the R/D of the controller (hardware and software), lower end LED fixtures using simple on/off timers these days costs more like $300-400 and cheaper if it's DIY.

So basically you had been comparing the prices of simple on/off MH/T5 fixtures+ballasts, with high-end computer controlled + USB/Bluetooth accessibility as well as fully programmable moon cycle LED fixtures.