Individual LED wattage Research

[Hurricane Aquatics]

Hey all,

Doing a bit of research I had hoped the forums could help me with. I am looking to see how much wattage that the below fixtures are driving their LEDs at.

For example, a fixture could be using 5w Cree LEDs and only driving them at 1.5w. I would like to know what the following fixtures are driven at.

1) Radion XR30 G5 and G6
2) AI Hydra 64HD
3) Reefbreeders newest models V2 pro+
4) Noopsyche V7 K3 Pro
5) Orphek Icon

Appreciate any information.

Thanks everyone.

 

[Timfish]

No idea but one way to guestimate is divide the wattage of the fixture by the number of LEDs used. Of course they could be mixing LEDs from different manufacturers so best bet would be contact them to see if they'll give you the info.

 

[Hurricane Aquatics]

Thanks Timfish,

That's what I've been doing. I have a theory on certain fixtures and Coral growth I want to test.

 

[oreo54]

Hey all,

Doing a bit of research I had hoped the forums could help me with. I am looking to see how much wattage that the below fixtures are driving their LEDs at.

For example, a fixture could be using 5w Cree LEDs and only driving them at 1.5w. I would like to know what the following fixtures are driven at.

1) Radion XR30 G5 and G6
2) AI Hydra 64HD
3) Reefbreeders newest models V2 pro+
4) Noopsyche V7 K3 Pro
5) Orphek Icon

Appreciate any information.

Thanks everyone.

It gets a bit more complicated than that.
Each channel can have its own current output.
Like a uv or violet driven at 350mA while the cree blue channel can be driven at 1200mA.
Made up numbers for demonstration purpose only.

Unless 1) manuf releases numbers or 2) you measure amperage yourself ( not as easy as it sounds as you generally measure " in- line" or 3) you find the driver ic, find the spec sheet and with the information contained you can identify the current set resistor ( some even use more than one resistor) value and calculate the output current, again from the driver spec sheet.
This is for commonly used driver ic's and may not always be that "easy".

2 and 3 involve disassembly of the light unless the drivers are remote which is uncommon.

Some programs may list watts and in that case you can measure voltage across the led string. Usually much easier or estimate using the led model (IF known) and get a guessimate using w = v x a.

20 rb in the xr-15 g5 blue 37.3 watts
Even this is tricky. Most rb diodes have a v( f) of around 2.9-3.6 v at normal drive currents
20 in series probably exceeds the power supply voltage so it could be broken into 2 strings of 10 each.
1/2 of 37.3 is 18.6w approx.
At say 3v is 30v
18.6 = 30 x XA
620mA....sounds reasonable eh.


Most "5w" diodes are just 2 "3w" diodes in series or parallel btw. They will be driven at the same current as you would a 3 w diode.

Say 3 x .620 = 1.86 ( 3w class) watts or 6 x .620 = 3.72 ( 5w class) watts.

The old days of pushing leds to their class limits is probably over. Lessons learned.

Of course none of that shows efficacy of the diodes.
Watts and a par meter could do that.
Another example using normal white diodes.
With a diode that pumps 150 lumens per watt vs one that pushes 75 l/w
At equal watts one has double the par.

Watts, drive current, string voltage..mostly irrelevant to par measurements unless you are building one.

Chart like this one can guesstimate drive current pretty easy and efficacy.
Each manuf should have one. Getting them to cough it out is another story.
CREE Photo red is approx 1.95 to 1.99v
1 is listed at 5w so I'd assume 2 in series in 1 diode 4V.
1250mA. BUT crees vf is 1.99v?
Assuming 2 in parallel
2v 2.5w each diode, 1250mA.

CREE lists photo red it at 1.99v 1500mA ( split 2 ways internally 750mA each diode) max current sooo
2 x 1.5= 3w.
Pro has 2 of these and lists 7.3w.
That would be 7.3 = 4v x X A
1825mA
2 in parallel
3.65w at 2v 1825 again. Exceeds CREE photo red x lamp rating of 1500mA
Soo what is the issue here?

The forward voltage at driving currents are as follows: 350mA = 2.1, 700mA = 2.3 and 1000mA = 2.5

Wrong v(f).
3.65w @2.5v = 1460mA
1460mA
5v ( 2 in series) and 7.3w = 1460mA
They really push the limit on these.
Redoing the single reduces drive current to 1000mA.
Lime 700mA @ 2.75v recommended specs
1.93watts each
Recommended max drive current of 1000mA

2 @ 6w probably also pushes the limit.

 

[razorskiss]

Also each channel and LEDs from different manufacturers will have different quantum efficiencies, different emission spectra etc. Knowing the rated max current draw of the diode and what's thrown through it is going to be both more challenging and less relevant (from a coral health standpoint) to measure than spectral output and output intensity / spread.

 

[razorskiss]

I'll add that the spec sheet ratings are nominal values and may vary from diode to diode by as much as 10-15% depending on whether the manufacturer paid for binned parts, and even then likely vary by 5% or so

 

[LeftyReefer]

What is more important is the wattage at spectrum. I don't care how many watts it can put out while putting out a terrible spectrum. Setting the light to the correct spectrum should be the first step.

Who cares how many watts a light puts out with all channels being driven at 100% if that isn't how you are going to use the light. if you don't run your reds, or greens, or whites at 100%, then who cares how many watts/PAR it's putting out i in that configuration.

Set the light to the proper spectrum, AB+ for example, then measure the input/output of the light. both current draw and PAR to get an idea on the efficiency of the unit.

 

[Hurricane Aquatics]

It gets a bit more complicated than that.
Each channel can have its own current output.
Like a uv or violet driven at 350mA while the cree blue channel can be driven at 1200mA.
Made up numbers for demonstration purpose only.

Unless 1) manuf releases numbers or 2) you measure amperage yourself ( not as easy as it sounds as you generally measure " in- line" or 3) you find the driver ic, find the spec sheet and with the information contained you can identify the current set resistor ( some even use more than one resistor) value and calculate the output current, again from the driver spec sheet.
This is for commonly used driver ic's and may not always be that "easy".

2 and 3 involve disassembly of the light unless the drivers are remote which is uncommon.

Some programs may list watts and in that case you can measure voltage across the led string. Usually much easier or estimate using the led model (IF known) and get a guessimate using w = v x a.

20 rb in the xr-15 g5 blue 37.3 watts
Even this is tricky. Most rb diodes have a v( f) of around 2.9-3.6 v at normal drive currents
20 in series probably exceeds the power supply voltage so it could be broken into 2 strings of 10 each.
1/2 of 37.3 is 18.6w approx.
At say 3v is 30v
18.6 = 30 x XA
620mA....sounds reasonable eh.


Most "5w" diodes are just 2 "3w" diodes in series or parallel btw. They will be driven at the same current as you would a 3 w diode.

Say 3 x .620 = 1.86 ( 3w class) watts or 6 x .620 = 3.72 ( 5w class) watts.

The old days of pushing leds to their class limits is probably over. Lessons learned.

Of course none of that shows efficacy of the diodes.
Watts and a par meter could do that.
Another example using normal white diodes.
With a diode that pumps 150 lumens per watt vs one that pushes 75 l/w
At equal watts one has double the par.

Watts, drive current, string voltage..mostly irrelevant to par measurements unless you are building one.

Chart like this one can guesstimate drive current pretty easy and efficacy.
Each manuf should have one. Getting them to cough it out is another story.
CREE Photo red is approx 1.95 to 1.99v
1 is listed at 5w so I'd assume 2 in series in 1 diode 4V.
1250mA. BUT crees vf is 1.99v?
Assuming 2 in parallel
2v 2.5w each diode, 1250mA.

CREE lists photo red it at 1.99v 1500mA ( split 2 ways internally 750mA each diode) max current sooo
2 x 1.5= 3w.
Pro has 2 of these and lists 7.3w.
That would be 7.3 = 4v x X A
1825mA
2 in parallel
3.65w at 2v 1825 again. Exceeds CREE photo red x lamp rating of 1500mA
Soo what is the issue here?

Wrong v(f).
3.65w @2.5v = 1460mA
1460mA
5v ( 2 in series) and 7.3w = 1460mA
They really push the limit on these.
Redoing the single reduces drive current to 1000mA.
Lime 700mA @ 2.75v recommended specs
1.93watts each
Recommended max drive current of 1000mA

2 @ 6w probably also pushes the limit.

Excellent write up, thank you.

I've been running LEDs since 2011 time frame. There's always been multiple reasons LED doesn't perform as well as Metal Halide and T5. I have a theory on why certain fixtures of LEDs seem to actually grow coral and not just sustain it.

I can tell you if I'm right, the fixtures in the last 5 years and up to present day, are going the wrong way in their progression. I'm building a new system right now and plan to test this theory once it's stable.

Really appreciate the responses everyone, excellent feedback.

 

[Hurricane Aquatics]

What is more important is the wattage at spectrum. I don't care how many watts it can put out while putting out a terrible spectrum. Setting the light to the correct spectrum should be the first step.

Who cares how many watts a light puts out with all channels being driven at 100% if that isn't how you are going to use the light. if you don't run your reds, or greens, or whites at 100%, then who cares how many watts/PAR it's putting out i in that configuration.

Set the light to the proper spectrum, AB+ for example, then measure the input/output of the light. both current draw and PAR to get an idea on the efficiency of the unit.

Well, yes and no. Spectrum is obviously important, but I believe there's another variable at play.

If it was only spectrum that mattered, any led would worn the same as all the others. Wouldn't matter if the fixture had 100 LEDs or 10. That would eliminate the need for different companies and led models.

 

[LeftyReefer]

Well, yes and no. Spectrum is obviously important, but I believe there's another variable at play.

If it was only spectrum that mattered, any led would worn the same as all the others. Wouldn't matter if the fixture had 100 LEDs or 10. That would eliminate the need for different companies and led models.

Say what?

LED emitters come in all sorts of nm ratings/spectrums.
A light that has all green emitters will not perform the same as light with all blue emitters.
Every manufacturer builds their lights with different arrays of emitters, so each one performs differently. Some have better spectrums than others. Those are the lights that work better.

An LED that provides a terrible spectrum is not going to grow coral as well as a light that provides the proper spectrum.

I don't understand what you mean by saying all LEDs are or would be the same???

 

[Hurricane Aquatics]

Say what?

LED emitters come in all sorts of nm ratings/spectrums.
A light that has all green emitters will not perform the same as light with all blue emitters.
Every manufacturer builds their lights with different arrays of emitters, so each one performs differently. Some have better spectrums than others. Those are the lights that work better.

An LED that provides a terrible spectrum is not going to grow coral as well as a light that provides the proper spectrum.

I don't understand what you mean by saying all LEDs are or would be the same???

No, maybe I didn't state my opinion properly.

I didn't mean that a fixture with all green leds would perform the same as one with all blue. I meant that if spectrum was the only thing that mattered, then there would be no need for all these manufacturers.

If LED fixture A (using 100 LEDs) emits an overall 460nm peak spectrum and LED fixture B (using 10 LEDs) emits an overall 460nm peak spectrum, then why have multiple different companies and fixtures?

Without giving away what I'm thinking until I can test it, I think it matters what you do with those LEDs and how you use them more than just the spectrum.

 

[oreo54]

History.....

Lighting the Reef Aquarium – Spectrum or Intensity?

This month Dana and Miguel discuss the importance of spectrum and intensity of aquarium lighting.

reefs.com

Though as noted this was a static spectrum
Mostly to show spectrum can be less important than photon counts.

Ideas about photosynthesis and specific wavelengths at specific times have been around land plants for awhile
Notably the 15 min ir pre full spectrum and post full spectrum. Not sure if it holds any water to this day.


Oh the days of massive " unused" spectrum has sort of sailed except for things like the 50/50 black boxes
Easy way is add watts of " blue" (everything from about 500nm and smaller) to non- blue (everything else)
Believe red sea is like 80/20.
Cool white can be split since it has a lot of blue.
Of course looks will factor in here as to it's "useable" output.
You can add up the pro and blue to see the ratios.

 

[Hurricane Aquatics]

History.....

Lighting the Reef Aquarium – Spectrum or Intensity?

This month Dana and Miguel discuss the importance of spectrum and intensity of aquarium lighting.

reefs.com

Though as noted this was a static spectrum
Mostly to show spectrum can be less important than photon counts.

Ideas about photosynthesis and specific wavelengths at specific times have been around land plants for awhile
Notably the 15 min ir pre full spectrum and post full spectrum. Not sure if it holds any water to this day.

Very nice, thank you for posting this.

 

[razorskiss]

Spread and photon flux definitely matters but if that was the only thing that mattered then only broad emissive surfaces like the Neptune Sky would be viable LED solutions, but clearly Radions and even Kessil's are able to generate success.

I think spectrum is a bigger issue than you acknowledge, especially with budget vs premium LED manufacturers. The primary absorbance peak for chlorophyll a in corals is fairly narrow about 420 nm. Most "royal blue" on the market emit instead at 450-460 nm which completely misses this absorbance peak, but these are much cheaper and widely available than 420 or even violet 405 nm LEDs. Now clearly just having 420 nm light isnt enough, you need enough photons and the photons need to reach enough of the coral's surface to drive health and growth, but every mW (optical, not electrical) at 420 is going to be an ~ order of magnitude more efficacious than every mW at 450.

I think this talk given at Reefapalooza does a good job of detailing the issue with LED spectra and how it's not as simple as labels like "actinic" vs "daylight"

 

[oreo54]

Spread and photon flux definitely matters but if that was the only thing that mattered then only broad emissive surfaces like the Neptune Sky would be viable LED solutions, but clearly Radions and even Kessil's are able to generate success.

I think spectrum is a bigger issue than you acknowledge, especially with budget vs premium LED manufacturers. The primary absorbance peak for chlorophyll a in corals is fairly narrow about 420 nm. Most "royal blue" on the market emit instead at 450-460 nm which completely misses this absorbance peak, but these are much cheaper and widely available than 420 or even violet 405 nm LEDs. Now clearly just having 420 nm light isnt enough, you need enough photons and the photons need to reach enough of the coral's surface to drive health and growth, but every mW (optical, not electrical) at 420 is going to be an ~ order of magnitude more efficacious than every mW at 450.

I think this talk given at Reefapalooza does a good job of detailing the issue with LED spectra and how it's not as simple as labels like "actinic" vs "daylight"

Chl a is not the be all end all pigment.
Accessory pigments can transfer photon energy at high efficiencies ie 90% plus.
The peridinin / chl complex can't be ignored nor the action spectrum.

Peridinin-chlorophyll-protein complex - Wikipedia

en.wikipedia.org

Think of nature..blue and green go hand in hand as to depth penetration until blue " wins".
Capturing " near blue" should be evolutionarily favored.

Sadly cyan and using green is an orphaned child for the most part.
Funnily in the blue plus t5 it has a rich green spike.

Peridinin absorbs light in the blue-green wavelengths (470 to 550 nm) which are inaccessible to chlorophyll by itself; instead the PCP complex uses the geometry of the relative pigment orientations to effect extremely high-efficiency energy transfer from the peridinin molecules to their neighboring chlorophyll molecule.

 

[Hurricane Aquatics]

Chl a is not the be all end all pigment.
Accessory pigments can transfer photon energy at high efficiencies ie 90% plus.
The peridinin / chl complex can't be ignored nor the action spectrum.

Peridinin-chlorophyll-protein complex - Wikipedia

en.wikipedia.org

Think of nature..blue and green go hand in hand as to depth penetration until blue " wins".
Capturing " near blue" should be evolutionarily favored.

Sadly cyan and using green is an orphaned child for the most part.
Funnily in the blue plus t5 it has a rich green spike.

Exactly and in a T5 or Halide all these colors are blended together before it leaves the bulb per say. A true blended, full spectrum. LED is a separated source for each LED.

That's one of the downfalls of LED is the individual sources of light.

 

[oreo54]

Exactly and in a T5 or Halide all these colors are blended together before it leaves the bulb per say. A true blended, full spectrum. LED is a separated source for each LED.

That's one of the downfalls of LED is the individual sources of light.

Yea nobody ( except Kyocera) has " phosphored" coral centric ind. leds.

Even that wouldn't be easy. Think of the now or in the past range of "tweaked" spectrums.
EVEN a decent 14000k diode would be welcome. This blue plus yellow doesn't really cut it as they don't t at 6500k.

Then there is the lens thing.
Tir lenses ( common w/quality/application differences) are designed to collimate the light beam. This punched par using less watts at the expense of blending with their neighbor diodes.
I know why they did it and for a long time it was good enough or worked around.
A combination of marketing factors has probably pushed this longer than maybe it should.
It still " works"...
The best but generally impractical solution was narrow angle lenses hung high.
Wide and hung low is probably second best.

 

[razorskiss]

Chl a is not the be all end all pigment.
Accessory pigments can transfer photon energy at high efficiencies ie 90% plus.
The peridinin / chl complex can't be ignored nor the action spectrum.

Peridinin-chlorophyll-protein complex - Wikipedia

en.wikipedia.org

Think of nature..blue and green go hand in hand as to depth penetration until blue " wins".
Capturing " near blue" should be evolutionarily favored.

Sadly cyan and using green is an orphaned child for the most part.
Funnily in the blue plus t5 it has a rich green spike.

The 90% efficiency number comes from the spectral match of peridinin fluorescence and chl a's red absorbance peak. The actual quantum yield of that process (fraction of photons in producing a photon out) is only 0.24. Absorbance is always a more efficient process than FRET.

 

[razorskiss]

Exactly and in a T5 or Halide all these colors are blended together before it leaves the bulb per say. A true blended, full spectrum. LED is a separated source for each LED.

That's one of the downfalls of LED is the individual sources of light.

The degree of spectral blending and uniformity from LEDs has been extensively benchmarked by reviewers like BRS and shown to be extremely consistent, at least from tier 1 manufacturers. A few mm of separation from point sources emitting over a 120 angle from 2-3 ft away is not going to generate significant segregation between individual emitter spectra

I'll add that when mixing different T5 bulbs the segregation would likely be even worse than with LEDs, but they're mixed spectra already so it's definitely not apples to apples

 

[oreo54]

The 90% efficiency number comes from the spectral match of peridinin fluorescence and chl a's red absorbance peak. The actual quantum yield of that process (fraction of photons in producing a photon out) is only 0.24. Absorbance is always a more efficient process than FRET.

Seems important. I'll look to verify your numbers

Checking your browser - reCAPTCHA

The degree of spectral blending and uniformity from LEDs has been extensively benchmarked by reviewers like BRS and shown to be extremely consistent, at least from tier 1 manufacturers. A few mm of separation from point sources emitting over a 120 angle from 2-3 ft away is not going to generate significant segregation between individual emitter spectra

I'll add that when mixing different T5 bulbs the segregation would likely be even worse than with LEDs, but they're mixed spectra already so it's definitely not apples to apples

There are other focal points such as dots of condensation on glass and caustic line focusing
A broad sampling point will mask any micro hot spots.

I said they were useable but not anywhere near phosphor blended on chip via t5 " like" nor better blended using reflectors.