DIY LEDs - The write-up

[rpp]

One more thing about controlable Meanwell (0-10V).

Can anyone say how dimming looks like? I heard that from 0 to 50% you can see difference but above 50% there is no visible difference.
Is it true?

 

[Norahsdad]

Thanks for the reply, so in the chart below is the one P4 Q2 Q3 Q4 Q5 the bin numbers?
And the higher the Flux (lm) @ 350 mA the better the LED?
thanks Glen


Min. Luminous Flux (lm) @ 350 mA*
Chromaticity Regions Kit Number Group Flux (lm) Cool White (5000 K – 10,000 K)

P4 Group, 80.6 Flux (lm)
WA, WB, WC, WD, WE, WF, WG, WH, WJ, WK, WM, WN, WP 00901
WC, WD, WF, WG 00902
WC, WD, WF, WG, WH, WJ, WN, WP 00903

Q2 Group, 87.4 Flux (lm)
WA, WB, WC, WD, WE, WF, WG, WH, WJ, WK, WM, WN, WP 00A01
WC, WD, WF, WG 00A02
WC, WD, WF, WG, WH, WJ, WN, WP 00A03

Q3 Group, 93.9 Flux (lm)
WA, WB, WC, WD, WE, WF, WG, WH, WJ, WK, WM, WN, WP 00B01
WC, WD, WF, WG 00B02
WC, WD, WF, WG, WH, WJ, WN, WP 00B03

Q4 Group, 100 Flux (lm)
WA, WB, WC, WD, WE, WF, WG, WH, WJ, WK, WM, WN, WP 00C01
WC, WD, WF, WG 00C02
WC, WD, WF, WG, WH, WJ, WN, WP 00C03

Q5 Group, 107 Flux (lm)
WA, WB, WC, WD, WE, WF, WG, WH, WJ, WK, WM, WN, WP 00D01
WC, WD, WF, WG 00D02
WC, WD, WF, WG, WH, WJ, WN, WP 00D03
For other flux and chromaticity combinations, contact Cree or an authorized distributor.

 

[Norahsdad]

One more question will this dimmer with a 24V wall wort work in replace of a "buckpuck and pot dimmer" to control 6 Cree 3w LEDs and how much light will i lose if it only go's to 700ma and not up too the 1000ma that some buckpucks do?

Don't most people get the 1000ma buckpuck and dial them down to 7ooma with a pot or do most people run there at 1000ma white and 700ma blue?
thanks Glen

http://cgi.ebay.com/12-key-Infrared...ti-3W-LED-/220648782745?pt=LH_DefaultDomain_0

 

[sammy113]

One more thing about controlable Meanwell (0-10V).

Can anyone say how dimming looks like? I heard that from 0 to 50% you can see difference but above 50% there is no visible difference.
Is it true?

I had my hands on meanwell ELN "D" model for the first time and built an easy controller with a 9v constant wall wart and a 10k potentiometer. First thing I noticed is it wont lit until it reaches like 15% maybe more (meanwells have a cut off so it won't dim down to very low rates) after that you WILL see the difference. There is a point maybe close to 90% - 100% were you can't really notice any changes but that's just the human eye. If you check that with a PAR meter you'll see difference. Also the pot is very sensitive to the adjustment so very little rotation makes a lot in light output.

 

[rpp]

So visible changes are between ca. 10 to 80-90% not as someone said up to 50% and then he didn't notice visible change.
If so - it's enough for me.

 

[pheinzig]

Ahhh nice build pheinzig!

Splash guard should not be more than 1/4". Any more is a waste. If it won't be trapped in a moist air location 1/8 would work.

Remember the fastest aging of LEDs is heat around their lenses. Make sure you don't stagnate the air flow around them with the splash guard.

THX. I did find a 36x83 slab of 1/2" acrylic from a local reefer for $50
Which would be pretty comparable to a 1/4 in price. Ill be sure to allow extra room for the splash guard,and Ill probably put in a few rows of 1/4 pinholes between the light's as well to allow breathing.
Since its the same price of a 1/4 sheet, and my Aquarium room is a concrete garage w/lots of moisture n the air, ill get this good deal. If it don't like it, I'm glad to know that there is always 1/4 available, and a swap would be a simple matter.

 

[jusmee]

Thanks for the reply, so in the chart below is the one P4 Q2 Q3 Q4 Q5 the bin numbers?
And the higher the Flux (lm) @ 350 mA the better the LED?
thanks Glen
.

Yes they are the bin numbers, and more recent Cree LEDs go even higher to the R series of bin numbers.

The output in lumens is per watt, and 350ma is about 1 watt (given the typical forward voltage drop). For 3 watts you run them at 1000 ma and they give correspondingly more light. The datasheet gives a graph of current vs lumens, but it's essentially linear.

 

[der_wille_zur_macht]

Just re-found an article on light intensity, duration, and spectrum for reef aquaria:

http://www.advancedaquarist.com/2009/5/aafeature

It's highly relevant to the subject that keeps popping up in here - choosing colors/spectra of LEDs and knowing if we've got the "right" mix and the "right" intensities.

 

[TheFishMan65]

Great article - thanks

SO when are you building the PAR sensor so we can all replicate the sun levels in Hawaii? I wonder how much brighter the sun is in June (maybe april is the brightest month, but I don't think so). Same question for december. Now not only do we need dawn and dusk, but season that maybe more important to corals then storms. Sigh will it never end

 

[der_wille_zur_macht]

I do plan on writing code to somewhat mimic the curves he reported in that article, more for the challenge than any belief that it'll be "better" for my livestock.

Though honestly I think the most important takeaway message from that article is that it's not really critical that we duplicate what happens on natural reefs. IMHO the best judge of "proper" intensity and color is that the reef keeper is happy with the growth and color of his corals.

 

[terahz]

That's a nice article indeed. I'm a little surprised how little light is lost underwater. At 3ft I thought about 65% of the light is lost, yet according to some of the graphs it shows only ~30% loss.

700 par from 3ft distance will be hard... I might have to go to 10 degree optics for that...

On a side note, Cree has posted record revenue (79% up), which can only be good for us (I hope)

 

[der_wille_zur_macht]

700 par from 3ft distance will be hard... I might have to go to 10 degree optics for that...

Yeah, but the good news is:

Is It Necessary to Provide Maximal Illumination?

No. Common sense and a quick look at reef aquaria proves that most photosynthetic invertebrates will thrive under conditions of relatively little light. There is no evidence that I am aware of that suggests corals' zooxanthellae require supersaturating light intensities in order to maintain growth rates and/or provide proper nutriment to the coral animal. Most 'common' corals saturate (that is, photosynthesis is at a maximum rate) at light intensities ranging from 200 to 450 molm²sec.

I've seen Sanjay and other "lighting experts" quote the 200 - 400 range so it seems to be a widely accepted target range for reef lighting.

 

[Jay1982]

I doubt that it is the LEDs. High lighting doesn't normally cause RTN and you don't have a lot of lighting for that size tank. Also, the photoperiod is long, but not extremely so, unless you went from something like 8 hours to 11 hours immediately after switching to the LEDs. I would look for something else. However, to be safe, have you tested the PAR readings?

CJ

Hey CJ, thanks for the input.

I don't have a par meter, and I've been trying to locate one for years in the Montreal area but I have had no luck so I can't give you a straight answer on that. The only other thing I have done differently since is that I replaced my membrane on my RODI filter but my TDS is reading zero. I'll post some pictures up tomorrow and hopefully someone can chime in with some advice.

In terms of photoperiod and light intensity. I was running 7 T5s and 2 250W metal halides before swapping over to LEDs.

 

[jtma508]

We've been talking alot recently about spectra around the various threads. While PAR is useful it doesn't really tell us what we need to know. PUR, specifically for the wavelengths used by zooxanthellae, is what we really need to know. Rather than using a quantum sensor, what we really need is http://www.apogeeinstruments.com/spectroradiometer/ with a waterproof detector. If we can dial-in the light the zooxanthellae need all that is left is to add whatever wavelengths we need to make the tank look good to our eyes.

 

[CJO]

I'm a little surprised how little light is lost underwater. At 3ft I thought about 65% of the light is lost, yet according to some of the graphs it shows only ~30% loss.

I'm seeing 50%+ when looking at figure 2. Is this what you were looking at? The article states that the water level was at 3-3.5 feet at high tide in the afternoon.

CJ

 

[TheFishMan65]

Well I can't link to the picture, but at 11:45 it looks to me like a 25% drop 2100 par to 1600 par. Where do you see 50%?

 

[der_wille_zur_macht]

Rather than comparing points on the graph we'd probably be fine just looking at the DLIs. For figure 2, where the depth ranged from 1 - 3.5 ft (due to the tide) there is a difference of 27%. For figure 3, where we do not know the depth (just that it was at the surface vs. underwater), there is a difference of 47%.

If we look at the two underwater DLIs, 30 and 16 Mol photons/day, there's clearly a very large difference. Consider that the 16 Mol photons/day number can be hit with a typical photoperiod and 400 - 500 molm²sec, and we are actually pretty close to simulating something that approaches natural conditions. Couple that with the fact that the article pretty much states that you don't actually need to hit natural light levels to have thriving corals, and IMHO we don't have too much to be worried about.

As an extension of that thought pattern, as much as I appreciate the science and math aspect of lighting, in the end the thing that matters the most (to me, at least) is observing healthy and colorful corals in a given tank. Though there is still clearly room for refinement, IMHO we are definitely at a place where it's possible to do so without needing a handful of calculations and formulas.

 

[CJO]

Well I can't link to the picture, but at 11:45 it looks to me like a 25% drop 2100 par to 1600 par. Where do you see 50%?

3:23pm and on.

Couple that with the fact that the article pretty much states that you don't actually need to hit natural light levels to have thriving corals, and IMHO we don't have too much to be worried about.

As an extension of that thought pattern, as much as I appreciate the science and math aspect of lighting, in the end the thing that matters the most (to me, at least) is observing healthy and colorful corals in a given tank. Though there is still clearly room for refinement, IMHO we are definitely at a place where it's possible to do so without needing a handful of calculations and formulas.

+1!

CJ

 

[TheFishMan65]

I agree with DWZM too.

I missed the drop. I figured that afternoon would be just the opposite of morning so took all the right side to be clouds (also mentioned), and ignored it. If you look at three the left and right are mostly symmetrical which is what I was expecting. So IMO something else changed between the morning and afternoon. Maybe the waves got bigger?

 

[der_wille_zur_macht]

I agree with DWZM too.

I missed the drop. I figured that afternoon would be just the opposite of morning so took all the right side to be clouds (also mentioned), and ignored it. If you look at three the left and right are mostly symmetrical which is what I was expecting. So IMO something else changed between the morning and afternoon. Maybe the waves got bigger?

Waves and/or it's just an effect of varying depths. My interpretation of the article is that the sensor was at a fixed depth with respect to the ocean floor. Meaning, as the tides came and went, the depth of water that light was filtered through before reaching the sensor (i.e. the depth from the surface to the sensor) changed.