Flip Chip Opto High Power Build Thread

[alesh]

Also that efficacy curve, I posted to open the thread. stops at 1.8amps. I still am guessing at the 40% efficiency. It could be only 300wP. But, even then the ease of build and the raw concentrated power makes it worthwhile.

So, at first I thought it was a typo in the spec.

But, you do the math, it is not. The only way to get 950w or so @ 50.8vF is 19a.

Now if these COBs are $500, that's a deal killer. If they are only 20% efficient at 300wP, that's a deal killer.

They're about 82.4lm/W @50.8V @19A @Ta 25°C which translates into ~25.5% efficiency. Certainly going down when warmed up. Not really sure what Ta means. What Tj/Tc do they assume?

From my experience, I would hardly call wider angle an advantage.

 

[Doer]

Ts.max = 30c - 25c = 5c/2000w = .0025 C/W woops need a bigger cold plate.

 

[Doer]

Huh? The spec says 79550 lm/m^2 @19a.

 

[Doer]

They're about 82.4lm/W @50.8V @19A @Ta 25°C which translates into ~25.5% efficiency. Certainly going down when warmed up. Not really sure what Ta means. What Tj/Tc do they assume?

From my experience, I would hardly call wider angle an advantage.

We don't know the efficiency curve of this chip. 19a. We know that all bets are off. Why is wider not an advantage if you don't need to concentrate density?

 

[Doer]

They're about 82.4lm/W @50.8V @19A @Ta 25°C which translates into ~25.5% efficiency. Certainly going down when warmed up. Not really sure what Ta means. What Tj/Tc do they assume?

From my experience, I would hardly call wider angle an advantage.

Ta is Air Temp.

 

[Doer]

This is the max rating so I don't know where you got your math?

At Ta = 25c
DC Forward Current(iF) 28500mA @ Tj≤145°C Junction Temperature Tj 145°C
Input Power W 1560W when Tj≤ 145°C

 

[alesh]

Ta is Air Temp.

Ambient temp. But what does it mean. That COB certainly will not have such specs when running @1000W in ambient without HS. Now what does it tell us?

Huh? The spec says 79550 lm/m^2 @19a.

79550 lm @965W. 82.4lm/W. LER estimated at 325lm/W. Efficiency 25.5%. 244.8 PAR W.

This is the max rating so I don't know were you got your math?

At Ta = 25c
DC Forward Current(iF) 28500mA @ Tj≤145°C Junction Temperature Tj 145°C
Input Power W 1560W when Tj≤ 145°C

From their data sheet. They provide specs @19A @50.8V.

We don't know the efficiency curve of this chip. 19a. We know that all bets are off. Why is wider not an advantage if you don't need to concentrate density?

No, we don't. And until we do, there are 3 things for sure. 25.5% efficiency @Test conditions. Higher efficiency when run softer. Worse efficiency when run harder.

 

[Doer]

All true my friend, I see how you got that now. Thanks for clearing me up.

 

[Randomblame]

Whoar!
Heavy light source!
I really wanted to know how they performs driven really soft, maybe @ 3-3,5A! I think efficiency goes to the sky! But this bleeding edge technic is surely more expensive at the moment, but it shows where the way goes! Driven this Cobs soft(150-175w) can allow to cool with normal cooler, because I like no water in my electronics!

Ps. I'm new here on rollitup and for a half year a "read-only user" to learn all over LED's, but this light-bomb earned a first comment!

 

[Doer]

The idea of course, is to not put water, IN the electronics. But, I have a water cooled light. Plumbing is not so bad if we take our time.

And of course, the max spec is traded for cost of KwH to get a happy medium of efficiency. And this is just the largest one.

The point though is these are driven at such high amps, it is hard to say where the happy medium is. So, just a rough swag so far.

I could see running 4 of these at 1/4 power, but it really depends on the COB cost and the curves.

 

[Doer]

Ambient temp. But what does it mean. That COB certainly will not have such specs when running @1000W in ambient without HS. Now what does it tell us?
.

It says you need to run these on a cold plate. I've never heard of trying to run any high power LED without a heatsink.

And it say, that the test Ta = 25c

It says if I run a cold plate at 25c, I am not re-radiating for a room at 25c.

It says if I have a large enough cold plate and heat exchanger, I can put enough gallons per minute of cooling, I can keep 25c on the plate.

And it says with a C/W on this COB Junction of .007 it is much more easy to cool. That's the patent.

How can I harness the break though, (if it is one?) That is the question.

 

[Doer]

Just doing a bit of light reading on radiators.

An automobile radiator typically deals with (at road air speeds) between 50,000 and 100,000 watts per hour.

I won't need to pay $480 for a 2500w heat exchanger. The idea is to put the radiator under the house, and face the exhaust out a side vent.

 

[alesh]

It says you need to run these on a cold plate. I've never heard of trying to run any high power LED without a heatsink.

And it say, that the test Ta = 25c

It says if I run a cold plate at 25c, I am not re-radiating for a room at 25c.

It says if I have a large enough cold plate and heat exchanger, I can put enough gallons per minute of cooling, I can keep 25c on the plate.

And it says with a C/W on this COB Junction of .007 it is much more easy to cool. That's the patent.

How can I harness the break though, (if it is one?) That is the question.

Well, that's classic Tc then.

edit: Watts per hour, Mr. degree in physics? Really?

 

[Doer]

Light reading on LER

Luminous flux is the measure of the perceived brightness or “light power” – it’s different than radiant flux, which measures all of the light emitted. Luminous flux is geared towards what the eye can see and the brain can interpret.

And they also likened to fuel economy. And that makes sense. I like Horsepower better than economy.

But, I also know, I cannot see a reduction in PAR frequencies on the HPS. And "watts of all light emitted" is PAR radiation. But, like lumen themselves, LER seems to have more to do with the watts the plant doesn't care about so much. It is only geared to what humans perceive as brightness?

http://www.digikey.com/en/articles/techzone/2011/feb/radiant-and-luminous-flux
Generally, the radiant flux from a source differs for different wavelengths. The flux at a given wavelength, λ, and in a spectral interval dλ is given by Pλdλ. A plot of Pλ versus λ gives the power spectrum of the radiator. Power spectra may be either continuous or discrete. The usual tungsten filament lamp has a continuous power spectrum like that of Figure 1. The mercury lamp spectrum shown in Figure 2 is a typical line spectrum with the spectral lines corresponding to energy level transitions. Fluorescent lamps have discrete spectra superimposed on a continuous spectra. The total radiant flux P emitted by a light source is the integral of Pλdλ over all wavelengths,

This is the area under the Pλ curve.

Figure 1: A typical continuous power spectrum.

Figure 2: A typical line spectrum, of a mercury vapor lamp.

If you superimpose these energy curves, for PAR you get this.

http://en.wikipedia.org/wiki/Photosynthetically_active_radiation

Luminous flux differs from power (radiant flux) in that luminous flux measurements reflect the varying sensitivity of the human eye to different wavelengths of light, while radiant flux measurements indicate the total power of all electromagnetic waves emitted, independent of the eye's ability to perceive it. Lumens are related to lux in that one lux is one lumen per square meter.

Lumens to lux calculation with area in square meters
The illuminance Ev in lux (lx) is equal to the luminous flux ΦV in lumens (lm) divided by the surface area A in square meters (m2):

Ev(lx) = ΦV(lm)/ A(m2)

or in the case of this COB, the photon energy that makes it to a 1 meter patch is what the spec says 79550 lux/m^2 @19a.

So, if this LED group has been measuring efficiency on lumen that is not correct for plants.

Lumen is the human brightness that leaves the emitter. But, plants don't care bit about lumen efficiency. A plant only can use the photon flux that makes it to the leaves.

Too bad. I've had this discussion before about CFL and HPS.

So, the lux efficiency is what matters, not lumen per eyeball watt.

 

[Doer]

Well, that's classic Tc then.

edit: Watts per hour, Mr. degree in physics? Really?

Watt/hrs. KwH? What is your malfunction? Watts are sold per hour and dissipated per hour.
All of this is time based. Or didn't you realize that? If the car is not running, IS IT dissipating watts of work? What did you really offer but your asshole?

And resorting to mockery? How weak is your mind? Don't you have to be correct to even begin to pull that off?

 

[alesh]

Watt/hrs. KwH? What is your malfunction? Watts are sold per hour and dissipated per hour.
All of this is time based. What did you really offer but your asshole?

And resorting to mockery? How weak is your mind? Don't you have to be correct to even begin to pull that off?

OK. I tried. Troll confirmed.

edit: Just to make it clear. Mr. degree in physics there doesn't know what he's talking about and yet he is very aggressive.
KWh is KW * h. Watts are not sold, energy is (measured in KWh, usually). Power (watts - energy per time [J/s]) is similar in concept to flow rate [m^3/s] - the measurement of how fast something is dissipating energy. Of course no one is billed by how fast his water is running, only for how much water is actually consumed. Same goes for electric energy and power.

 

[Doer]

OK. I tried. Troll confirmed.

Tried to troll? Need to to do better.

 

[Greengenes707]

LER takes the luminosity function out of the equation and allows you to convert lumen output(Human biased photometric measurement) into radiant watts(neutral/unbiased radio metric measurement)

 

[Scotch089]

Can I ask without this just turning completely downhill... why are you trying to build this? Not attacking or any shit like that. Just genuinely wondering. When you're talking 25% efficient...your creating 3 times more heat than light, where an hps may create only 2 times more heat than light brand new... only pro I can see out of it is your dissipating the heat differently? Which could up your efficiency? Is that expectation to counteract the low initial efficiency with cooler temps? Even then how much?? But again, in a more expensive manner than what you would with just a hood and fan.. at this point, I see it being less efficient in up front costs, the performance for the same amount of light (actually less light), possibly the extracurricular watts from parts devoted to making it run without overheating itself and the area it's in, etc.

Maybe I've misunderstood what current your driving the emitters at or something, surely you understand how counterproductive it'd be to run them that hard. Only other pro I could see would be you want an led spectrum, but if that were the case wouldn't you want 90+ CRI or go with cmh.. idk the 25% just reallllly stood out to me which led me to posting.

Maybe some clarifying words could help us understand where this derived from and is really pushing you with the route you're going.


Edit: I must be missing something. Where's the 25% coming from?

 

[Doer]

Can I ask without this just turning completely downhill... why are you trying to build this? Not attacking or any shit like that. Just genuinely wondering. When you're talking 25% efficient...your creating 3 times more heat than light, where an hps may create only 2 times more heat than light brand new... only pro I can see out of it is your dissipating the heat differently? Which could up your efficiency? Is that expectation to counteract the low initial efficiency with cooler temps? Even then how much?? But again, in a more expensive manner than what you would with just a hood and fan.. at this point, I see it being less efficient in up front costs, the performance for the same amount of light (actually less light), possibly the extracurricular watts from parts devoted to making it run without overheating itself and the area it's in, etc.

Maybe I've misunderstood what current your driving the emitters at or something, surely you understand how counterproductive it'd be to run them that hard. Only other pro I could see would be you want an led spectrum, but if that were the case wouldn't you want 90+ CRI or go with cmh.. idk the 25% just reallllly stood out to me which led me to posting.

Maybe some clarifying words could help us understand where this derived from and is really pushing you with the route you're going.


Edit: I must be missing something. Where's the 25% coming from?

View attachment 3372660

Research? You know? Alsesh says 25% not me.