This LED lamp vs PC lamp

[HecticDialectics]

First off HECTIC how am I incorrect??? As you shown from your other post

you know nothing about LEDs. So How would you know?

 

neanderthalman outlined it quite nicely Like he said, it's not about being correct or incorrect. It's about learning. I admit when I'm wrong, but I'm pretty sure everything I've said in this thread is more or less correct.

[nynex]

neanderthalman Thats cool....and I am interested in hearing your view on it. If you are schooled

in this, then you of all people would probably be who us people who are interested in further studying

this technology should talk to. I think LEDs are an awesome technology and have quite alot of potential.

I read in forbes magazine that in japan and hong kong they are building hotels that have all LED lighting.

 

The array im building is only for a 12 gallon Nano cube. I think the LEDs I have would work just

fine for that. I thought about not using the resistors and using buckpucks instead. I agree with you

on that guy from Hong Kong's design that being open with copper wiring is not a good idea around

saltwater. But I think at the end of that thread he said that he was building an enclosed hood for it.

 

I think it would be better off to solder the connections instead of the way he did. It looks like he used

all resistors with his setup. He used all 5mm size LEDs too. Which I have 10mm wideangle LEDs. I

also have 5mm...but im only using those for moonlights. I also thought about adding a few rows of

the 10mm UV. Not sure though.

 

Now I agree with you if I was trying to make one for 100 gallon + tank. That unless you use Luxeon,

then it wouldnt be worth the effort..atleast not yet. At that point I would do like Strangelove and buy a

solaris. Given 2 or 3 years I think the technology is going to get even better.

 

Now I went to solaris website http://www.solarisled.com

 

What is your thought on this...

 

"The new H4 series produces PAR light output levels equal to a 400W MH 20k. It uses 40% less

energy than the 400W MH 20k fixtures it replaces. The life of the LEDs is approximately 50,000 hours,

so it almost eliminates metal halide and fluorescent bulb replacements. All heat is radiated up and away

from the tank; therefore, it does not heat the water like Metal Halides or Fluorescents. This eliminates

the need for Chillers. The room air conditioner needs to work ½ as much since the light fixture

produces only ½ the heat of Metal Halides which saves even more energy over traditional lighting

methods.When looking at 4K the LEDs can not compete on a lumens/watt basis. When we look at

10K and 20K the LED's produce more PAR/watt than the metal halide."

[neanderthalman]

Well, for the most part, what they say, or rather, what they intend to say is ok. However, I am a nitpicky bastage. They make false claims.

 

I like the comparison to 20K MH, which is known to have the lowest PAR of all MH lighting. We should all be aware that this comparison is made using a crappy spider reflector, and thats fine as well. What I'd like to know is if this measurement of PAR being made when the solaris fixture is also 20K. It is entirely unfair to compare the two when they are not operating at the same color temperature. Since they do not indicate that this is the case, then I am suspicious of the comparison.

 

I question their measure of lifetime - accelerated lifetime testing is far from accurate for things like spectrum shift, which is the sole reason we replace our lights regularly - not complete failure. The model works, somewhat, for estimating the lifetime before complete failure of a device, but not a shift in parameters. Such accelerated testing is done by running fifty LEDs for 1000 hours, and counting the number of LEDs that fail. So far, nobody has actually run a solaris fixture for 50,000 hours, so we have no way to truly know how stable the output spectrum is. We'll check back in 14 years and find out.

 

The other method of accelerated life testing is to run the LEDs at an elevated temperature. It is also reasonable for estimating the expected lifetime, but the statistical models used are again not designed for things like spectrum shift.

 

I agree with what they claim for reduction of heat production, but the "heat is radiated up and away" is total crap. Any body with an elevated temperature will emit infrared radiation in all directions, thus heating the tank. This is how a MH fixture does it, this is how a fluorescent fixture does it, and this is also how an LED fixture will radiate heat to the tank. There is no fundamental difference.

 

Now, the fact that LEDs operate at a dramatically lower temperature than MH means that the amount of heat radiated into the tank is significantly less. It is not, however, zero. Since it's not zero, don't claim it's zero.

 

The bit about the air conditioner is also crap. They are assuming that all of the heat in the room is from the metal halides. That's obviously not the case - every heater, powerhead, pump, and other equipment produces heat as well. You also have heat penetrating through the walls of the house, and so on and so forth. Will reducing the heat from the lights reduce AC operation? Yes. It just won't be by 1/2.

 

Lastly, the final statement they make is that the LEDs produce more PAR/watt than the metal halide. MH has a significantly higher efficacy than LED. They have a higher lumens/watt (or par/watt if you like) than LED, which is directly contradictory to their statement.

 

What they are trying to get across is that you get more light into the tank with less power consumption - which may be true. What they actually said is untrue, as the MH will produce more lumens/watt than LED. The difference is that nearly every lumen from the LED makes it into the tank, but there is significant light loss in MH setups. If you were to only count the lumens that make it to the tank, then yes, LEDs could have a higher lumens/watt than MH, because of the focused output. I'm certain that they use a crappy spider reflector for this comparison as well.

[coolwaters]

its true that light, no matter when it comes from, will always give off heat about the pigments absorb wavelength's. like if metal halide is shooting off 400 watt of power on something it should give off the same temperature as 400 Watt of LEDs its just that MH has other things types of light (wavelengths) shooting out of it. your probably thinking about cold fusion...where pretty far from that....

 

the thing about metal halides is that its like a 360 degree light angle. (less if its a DE bulb and subtract the side with the base.)

even with reflectors your still losing a good amount of light.

 

dont get me wrong i think MH are great. im just stating the facts.

 

i wonder if its going to help if the base of LEDs are made of silver =P

 

cant wait for cree or luxeon to come out with thos 160+ lm/W.

 

someday just imagine your tank running on like one HID LED that gives off 683.002 lm/W lol

[neanderthalman]

No, sorry, not talking about cold fusion. Cold fusion is a crock, btw. Even if *something* is producing extra energy in those experiments and it's not an experimental error, the amount of energy is too small to be accounted for by a fusion process. I'm talking about Planck's law for blackbody emissions. Really not sure how you made the jump from blackbody emissions to cold fusion.

 

If cree or luxeon come out with an LED that truly produces 160 lm/W, then they have surpassed MH in efficacy. That would be simply awesome and truly revolutionary. I'd buy a solaris fixture if that were the case, no question.

[nynex]

Do you think the technology is going in that direction? I think it will only be a matter

of time till luxeon figure out a way to come up with an LED that will produce 160 lm/W.

Thats why I said before that I truly believe that in 5 years...LED systems like the Solaris

will be standard Reef lighting.

 

Like I said I was reading in Forbes magazine...I also think I seen something in the

New York Times about this too. That in Japan and Hong Kong..they are building hotels

that use nothing but LED lighting. They are basing their designs around an all LED lighting

setup through out the whole hotel. I thought that was interesting.

[neanderthalman]

I totally agree. LED lighting has the potential to become the standard in lighting, not just aquarium lighting. It's just not quite there yet. There are still a few questions to be answered, the biggest one for me is how long before the spectrum shifts. So far, they haven't addressed that, only given a mean time to failure.

 

As far as general lighting is concerned, I think OLEDs stand a better chance of revolutionizing illumination. The technology is a little behind semiconductor LEDs at this point, but has the potential for a greatly reduced cost and increased scalability.

[nynex]

Ya I agree... You know now that you point that out. Ive never heard

any LED manufacturer ever mention anything about spectrum shifts

either. Im wondering if they plainly dont know. I also agree that saying

an LED will last 50,000 hr is a total guess too. But as you said who cares

about how long they will last. Even if they last a quarter of that, thats

still 2 or 3 years..

 

I was reading your post from the LED array you made a in 2005.

I was impressed..Im going to design my array with 3 in a series.

I think thats a great idea. You used all 5mm size LEDs?

 

I have 10mm,8mm and 5mm

 

What happen to your array? Is it still up and running?

[neanderthalman]

The tank got scratched to hell last december, and I was so disgusted and disappointed that I tore it down and transferred the livestock to my 10g. At that point, yes, the array was still working fine. The intensity had diminished some, but it was still miles beyond the PC that I had been using before.

 

I picked up some acrylic scratch remover a few months ago, with the intent to fix the tank. Then I moved. Once I find the scratch remover, I'll be setting it back up.

 

They were all 5mm LEDs, I forget the specs. They're early on in that old thread. I've learned a hell of a lot since starting that project, let me tell you. Keep in mind that with 3 in a series, I was overdriving the LEDs, which is very risky. I didn't know just how risky it was at the time, I got lucky.

 

Funny thing, I had far more trouble getting MH to work properly than I did with experimental LED lighting. Bad ballasts, bad bulbs. I went through three of each before getting good stuff. Always buy quality. ugh.

[nynex]

wow.. whats up with that? You just got a bad run of halide gear..

 

Ok..so scratch the series 3 idea? What about buckpucks?

http://www.leddynamics.com/LuxDrive/conten...ducts_buck.html

 

What do you think of this as an alternative to resistors?

 

Here is where I got my acrylic cleaner..stuff works very well..

 

http://www.rplastics.com/novuspolish.html

[neanderthalman]

That's the exact polish I got - I convinced the guys at my local plastics shop to give me a little foil sample packet. I think I got the step 2, but I need to find the packet to be sure.

 

I don't know much about the buckpucks, I haven't done any work with LEDs for quite a while. If they are a constant current supply, as I think they are, then it is likely superior to resistors. Certainly, they would be less sensitive to voltage fluctuations. However, I don't know how much heat they produce, so you could wind up wasting a fair bit of energy as heat inside the buckpuck itself. If they aren't very efficient, then you could wind up producing more heat in the buckpuck than you would in a simple resistor.

 

You can do 3 LEDs in series, just don't do it without a resistor. A small resistor on each string would bring the LEDs back into normal operating range, and keep the LEDs from being overdriven.

[strangelove]

Strangelove...Cool good to see another Photographer with an Understanding of Light.

 

Actually my Metal Halides for my photoshoots are not Strobes. They are set lighting

with 48" softboxes. For architecture..you dont need strobes too much. Although I use them

when I do Restaurant shoots. The set lighting is easier to incorporate into the natural lighting.

 

 

Oh boy hot lights, hope all your locations have air conditioning. I burned my hands on an Arri light one time, forgot to use some gloves when disassembling it and grabbed a spot on the arm that I thought was cool, but once I got it off a stand that thing started frying, couldn't just drop it so I had to endure the pain until I could set it down on a tile floor. Great lighting for your application though. I've always thought hot lights (fresnels) used tungsten bulbs. You must be using HMI lights and if memory serves me those are a pretty penny.

[strangelove]

I agree with what they claim for reduction of heat production, but the "heat is radiated up and away" is total crap. Any body with an elevated temperature will emit infrared radiation in all directions, thus heating the tank. This is how a MH fixture does it, this is how a fluorescent fixture does it, and this is also how an LED fixture will radiate heat to the tank. There is no fundamental difference.

 

Now, the fact that LEDs operate at a dramatically lower temperature than MH means that the amount of heat radiated into the tank is significantly less. It is not, however, zero. Since it's not zero, don't claim it's zero.

 

 

I think you need to look at how an LED, in particular how Luxeon LEDs are built. There is a semiconductor chip that sits on top of the bulb where the electrical current passes through, this is what gets hot, and if you can draw the heat away from that chip than the heat won't travel down towards the bulb, this is where heatsink and ventilation come in. The other side of the semiconductor has a spacific element which illuminates when a current is passed through it, since it does not use gas like PC or MH than there is very little to warm the bulb enclosure once the heat is "draw up and away" as in the case with the Solaris fixture. You can seriously put your hand under the bulb itself and touch them and they are just warm, just barely warm at that. This is where PFO designed the Solaris right, so their "up and away" claim is totally true. So if your designing a LED array you should take this into account add as much heatsink in direct contact with each bulb to draw the heat up, then design a fan system to blow the hot air away so the heatsinks can work properly, this should also increase the lifetime of the LED bulb, since it isn't subject to heat. PFO claims that the bulbs that they use will shift 70% in 5 years, that's spectacular IMO. By then they should have a retrofit kit for the light hood I have now and the price for the replacement bulbs should have come down dramatically. Either that or I can just buy a new LED light hood. In contrast an MH bulbs color spectrum will begin shifting in 6 months and will need to be changed within a year.

[nynex]

Strangelove.....Look on Ebay....

 

There is a dealer on ebay that sells these metal halide light setups for photography.

He has Buy It Nows for everything and its pretty pricey. But they also do auctions

with alot of it too and if you are lucky you can win an item for a really decent price.

Thats how I got most of my lighting rig. I won one auction that included a 48" Softbox,

Light Head, four 1000 watt metal halide bulbs, four 500 watt metal halide bulbs, a nylon

(good not cheap) carrying case all for $120. You just have to watch the auctions...

 

The softboxes work very well for my shoots.

[neanderthalman]

I'm pretty certain that with my "degree" in "semiconductor fabrication", I am well aware of the structure of an LED, but thanks for the refresher strangelove.

 

I'm referring to blackbody radiation, which does not depend on physical construction, only the temperature and emissivity of the material. You can't "pull" it away, as you claim. Since the solaris fixture runs at an elevated temperature, it will heat the tank. It will, however, be significantly less than a MH fixture - there's no argument there. It's simply not zero, as they have claimed.

 

Check wikipedia for "blackbody radiation" and "planck's law" for more information on this phenomenon if you do not believe me.

 

This is the first I've heard of PFO's claims regarding spectrum shift. I am curious about two different things - what constitutes a "70% shift". How is that quantified? Secondly, how much of a shift is necessary to negatively impact a reef tank? Without that data to compare to the value they have provided, there is no way to know whether or not this is indeed "spectacular".

[strangelove]

Strangelove.....Look on Ebay....

 

There is a dealer on ebay that sells these metal halide light setups for photography.

He has Buy It Nows for everything and its pretty pricey. But they also do auctions

with alot of it too and if you are lucky you can win an item for a really decent price.

Thats how I got most of my lighting rig. I won one auction that included a 48" Softbox,

Light Head, four 1000 watt metal halide bulbs, four 500 watt metal halide bulbs, a nylon

(good not cheap) carrying case all for $120. You just have to watch the auctions...

 

The softboxes work very well for my shoots.

 

 

$120 WHHHHAAAAAAAAAAAAAAAA!?! Do you know a link to these lights? Excuse me for a while I'll be Ebay shopping for the next few hours.

[nynex]

Strangelove

 

Yep...Here is a link to their ebay store...

 

http://stores.ebay.com/amvona-com

 

They have great stuff....although their customer service reps are idiots.

I actually got into a dispute with them on my last purchase. Which made

me mad...cause I really like the products..

 

 

neanderthalman

 

I was wondering that too. I mean how do they know exactly that there is going to be

a 70% shift. Seeing that there has not been a solaris fixture that has operated for that

long yet. Seems to me they should say " Its possible there may be a 70% shift within

5 years" . Or do you think that they are trying to base it on pre existing luxeon technology?

[strangelove]

Sorry for hijacking this thread, but wow those prices for lights are cool. Looks like they have both MH and Tungsten lighting systems. Looks to be a Russian manufacturer like the Kiev cameras. Are you happy with the build quality of these lights? And what's it like working with them, bulb looks kind of awkward.

[nynex]

Honestly I love them. I was worried about the manufacturer quality too.

 

But when I got them I was impressed at how well they are made. The only thing

is...the softbox was a pain in the rear to figure out and it comes with no instructions.

 

I tried for 3 days to figure out how the brackets fit into the material to get the

box to open up...I tried and tried...nothing worked and then one night I woke up at 3am

and I couldnt sleep and I was thinking about it...So I went out there and I got it to work

on the first try. It was so simple after I finally got it to work. But when you first see the

brackets..."you are thinking what the ****??? There is no way that fits in like that!!!"

 

Whats funny is everytime I take them apart and Im in between shoots. I keep thinking

im gonna forget how to put them together again...lol

 

But together they work very well and ive got good results from them. The softboxes are

very well made and have vents that are Velcro shut and can be open super easy. The front

of the Softbox is also on with Velcro and also comes off too...

 

On a reef tank related note....If you wanted to... the light heads have slots cut out of them for the

softbox brackets...if you wanted to...you could use some kind of suspension wire and suspend these

light heads from the ceiling over your tank and get Metal Halide bulbs that are in correct spectrum..

Cause the ones that come with the kit are like 2700k or something...way off for a reef tank...But I

Imagine you could use any other Metal Halide bulbs that are under 1000 watts...

 

Just some stupid thing I thought of one day at a photoshoot..lol

[strangelove]

Yeah these Dynaphos lights do look cool, Arri's are like $1,000 to 3,000 bucks. I'm thinking of getting the rail system for my studio. Lights are still a great deal.

 

You should photo your tank with your 1000w lights, that would be fun to see, if you have time that is. Can you imagine a 1000w 20k light over a Nano tank, that would just be ridiculous

[strangelove]

I'm pretty certain that with my "degree" in "semiconductor fabrication", I am well aware of the structure of an LED, but thanks for the refresher strangelove.

 

I'm referring to blackbody radiation, which does not depend on physical construction, only the temperature and emissivity of the material. You can't "pull" it away, as you claim. Since the solaris fixture runs at an elevated temperature, it will heat the tank. It will, however, be significantly less than a MH fixture - there's no argument there. It's simply not zero, as they have claimed.

 

Check wikipedia for "blackbody radiation" and "planck's law" for more information on this phenomenon if you do not believe me.

 

This is the first I've heard of PFO's claims regarding spectrum shift. I am curious about two different things - what constitutes a "70% shift". How is that quantified? Secondly, how much of a shift is necessary to negatively impact a reef tank? Without that data to compare to the value they have provided, there is no way to know whether or not this is indeed "spectacular".

 

 

 

Here is PFO's info from their Solaris FAQ's

 

 

"The Solaris uses Luxeon LED's. The LED's were originally developed by HP and since have been aquired by Phillips Lighting. They are the highest quality, with the highest light output, available in the world today.

Luxeon rates their Blue LED's for 50,000 hours and the white LED's for 50,000 hours. They state that they will maintain 70% of their light output for the stated life.

These life ratings are stated for running the LED's at 700ma and running less than 120 Degrees C at the junction temperature. The Solaris Junction temperature is less than 50 Degrees C. The lower the temperature the longer the life. Our junction temperature is so low because of our patent pending thermal design. This cool running not only gives longer life, but it also increases the light output compared to running it at the maximum temperature of 120 Degrees C.

One way to increase the light output is to run the LED's at 1000ma or greater. The problem with that is it shortens the life. The gain is, the fixture initially generates more light. We are a certified luxeon manufacturer and so we run our LED's to the manufacturers specifications."

 

 

 

Also here is some info on LED light shift and how thermal dynamics effects the color spectum of LED bulbs.

 

 

LED's produce heat and light. The light is radiated down while the heat is radiated up. If proper thermal design is not developed in the fixture than the light output reduces, the wavelength shifts toward the red spectrum and the operating life of the leds reduce. The following graph shows the light output curve as temperature increases.(The junction temperature is the point where light is emitted from the LED)

The Solaris was built using computer aided heat dissipation modeling to eliminate heat issues. Not only does the solaris pull the heat away from the LED quickly, it also keeps the heat out of the aquarium. The Solaris LED Junction Temperature is about 50 Degrees C. which maximizes light output and lengthens the life of the LEDs.

 

[strangelove]

By the way, I'll need to clean out the acrylic shield, dust is accumulating from the fan intake side. So I will need to service the light, blow out dust etc. To do this I will need to take it apart. So if anyone is interested in seeing the internals of the Solaris LED light let me know and I'll take some photos while I service it and post them.

[neanderthalman]

My big question is spectrum shift, right? You responded that they shift 70% in five years. I asked how this is quantified, and how much shift is necessary to negatively impact a reef tank. You've provided more information from solaris, and now we see where the value of 70% over five years comes from.

 

Turns out it has NOTHING to do with spectrum shift at all. It's measuring the drop in output over five years. Maintaining 70% intensity over five years is fairly impressive, I'll admit, but we do not replace aquarium bulbs because of a drop in intensity.

 

Aquarium bulbs are replaced because the spectrum of the bulb shifts, not because the intensity is reduced. The expectation that the bulbs will run for 50,000 hours, and that they will maintain 70% of their intensity over five years is absolutely meaningless information to reefkeepers. It's simply some impressive numbers that they threw out to make their light sound better than it is. I don't have the data, but it wouldn't surprise me if MH bulbs lasted 50,000 hours and maintained 70% of their intensity after five years. It would also be pointless information, because what is important is SPECTRUM SHIFT.

 

Furthermore, they have blatantly lied when they claim that the solaris fixture does not heat the water. I will agree that heating is minimal, but it is physically impossible for it to be zero. The solaris fixture operates at a temperature higher than the tank, so there will be a net flux of infrared radiation in the direction of the tank. This results in heating of the water, end of story. They cannot claim that the solaris fixture does not heat the water. The can claim that the heating of the water is greatly reduced, compared to MH and fluorescent, but they cannot make the claim that no heating occurs.

 

Their blatant dishonesty regarding the specifications of the light make everything else they have said to sell their lights extremely suspect. We do not have five years of anectdotal evidence to back up any of their claims, or to observe whether these lights start growing nothing but algae after 18, 24, 36 months, or however long it takes. Spectrum shift happens, PFO even acknowledge that it occurs, but they do not quantify how long it takes before their light becomes useless to us. Why? Because they don't want us to know. Why wouldn't they want us to know? Possibly because it's a fairly short time. A short enough time that MH bulb replacement, chillers, AC, and extra energy consumption are more economical than a $1800 light fixture.

[strangelove]

Well that's the best info I could find to answer your question, so I extrapolated light output to answer light spectrum shift for you. To get better info on bulb life I suggest going to Phillips or Luxeon to find more accurate data.

 

As for heat transmission, it's as close to zero as you can get. The Solaris that I have on my tank adds no heat to my system, heat that I can detect. Now about the dark matter or blackbody radiation you are talking about, who knows, I will freely admit that I have zero degrees in physics and have zero desire to do so, but my High School math education does tell me that my Solaris gives off zero heat to my tank.

 

Don't believe me, there is a long thread about this on RC and there is this guy named Sanjay who does testing with another guy called Dana Riddle who did some temp readings with a heat gun. His findings were exactly what PFO claims, the Solaris adds zero heat to the surface of the water in his tank. I'd try to find it and post it, but that thread is like 100 pages long.

 

Heres a better article that Dana Riddle wrote, lots of numbers and data and all that jazz.

 

Heres the link

[neanderthalman]

I've tried to get better info on spectrum shift - and they don't have it. Believe me, I was extremely interested in the solaris fixture last year when they were first announced. I did a lot of research on them at the time, and the information on spectrum shift was one thing I found to be missing.

 

The heat transmission, as you say, is as close to zero as you're going to get. I couldn't agree more, the damn thing barely runs warm. It's simply not zero. It likely was not picked up by the heat gun (I assume you mean infrared thermometer) because those handheld infrared thermometers are the single least accurate and least precise of all temperature measurement devices. I'll read their article, but I'll bet you now that they didn't even account for the emissivity of the water. I'll tell you later if I was right or not.

 

FWIW, those "heat gun" thermometers work on the exact principle that is heating the water - emission of infrared radiation due to an elevated temperature.

 

Now, had they instead used a more reliable temperature measurement device, such as a platinum RTD, then the results of their experiment would be valid. They would have picked up a small increase in water temperature, an increase likely smaller than the error in the infrared thermometer (typically several degrees). As I have already willingly conceded, any heating effects are very small - but they aren't zero.