LED vs. PC light comparison - PAR & lumens

[Waterproof]

Has anyone done a watt-for-watt comparrison of high powered, 3W LEDs to PC bulbs? I'm incorporating LEDs into my biocube-14, and I've noticed that 2 of the Cree XR-E royal blue LEDs seem to put out more visible light than my 24w PC bulb. I've been struggling to find a way to fit the LEDs in the stock hood along w/ the 2 PC bulbs. I wonder if it would make more sense to just remove the actinic PC bulb and add a couple more LEDs. Any thoughts?

 

I have no way of measuring the PAR or lumen output. Maybe what I'm interpreting as more light output, is just visible light and not useable light. How does the energy utlitzed by corals compare between LEDs and PC bulbs?

[evilc66]

LEDs will be more comparable to MH performance. For most arrays, LED driven to 1000mA will produce PAR output similar to a 70W MH. This can be increased by the use of optics. Royal blue LEDs produce almost as much PAR per watt as cool white. Not 100% sure on regular blues, but estimate about 75% of what a cool white could do.

[Waterproof]

For most arrays, LED driven to 1000mA will produce PAR output similar to a 70W MH.

 

When you say "most arrays", how many LEDs are you talking about in a given array? How many 3w LEDs (roughly) would you need to get a 70w MH PAR? Are you saying they are roughly equivalent watt for watt? For example, 23 LEDs would match a 70w MH?

[evilc66]

Multiple LEDs have some effect on total PAR that reaches the sandbed, but an array of 100 LEDs doesn't produce more PAR at a given spot than 10 LEDs would, with a given spacing. It's the overlap of the LEDs beam that will increase the PAR over a single LED.

 

For example, if a single LED produces 100 PAR (purely as an example) at a given distance, two LEDs spaced 2" apart may produce 110 PAR. 10 LEDs spaced 2" apart may have higher PAR (maybe 125), but strictly because there are more LEDs overlapping. Take that same single LED and put a 40 degree optic on it, and now you have 500 PAR.

[Waterproof]

Multiple LEDs have some effect on total PAR that reaches the sandbed, but an array of 100 LEDs doesn't produce more PAR at a given spot than 10 LEDs would, with a given spacing. It's the overlap of the LEDs beam that will increase the PAR over a single LED.

 

For example, if a single LED produces 100 PAR (purely as an example) at a given distance, two LEDs spaced 2" apart may produce 110 PAR. 10 LEDs spaced 2" apart may have higher PAR (maybe 125), but strictly because there are more LEDs overlapping. Take that same single LED and put a 40 degree optic on it, and now you have 500 PAR.

 

Alright, evil. Let's beat this dead horse a bit more. I was poking around in my BC-14 hood tonight, and I'm not quite sure how I'll fit my LED assembly in there with the 2 bulbs. So I may have to re-evaluate my design. I'm thinking of only going w/ LEDs. I don't need much more light than the stock BC-14. Based on your familiarity with this hood, how many LEDs would I need to equal the same juice at the bottom of the tank? I'm not using optics b/c the stock hood doesn't allow for the extra height. As another option, I could keep the daylight PC and only use LEDs for the actinic.

 

PS I really appreciate all your help.

[evilc66]

Again, the quantity of LEDs are there for coverage, not for intensity. You will want to set up the array to get light to all areas of the tank. Running the array without optics will get you similar performance to a 70W MH. Try and get a heatsink that will fill the splash sheild as much as possible. The large (8.5" wide) heatsink from HeatsinkUSA should fit the bill. I'd say you should be looking at about 14-20 LEDs, depending on what you want your spacing to be (average is about 2").

[Waterproof]

Again, the quantity of LEDs are there for coverage, not for intensity. You will want to set up the array to get light to all areas of the tank. Running the array without optics will get you similar performance to a 70W MH. Try and get a heatsink that will fill the splash sheild as much as possible. The large (8.5" wide) heatsink from HeatsinkUSA should fit the bill. I'd say you should be looking at about 14-20 LEDs, depending on what you want your spacing to be (average is about 2").

 

Are your statements relative to the 3w high-power LEDs in particular? I don't mean to argue; I appreciate your help. However, it seems I could use 1 watt, or event the low-power 5-10mm leds and get 100% coverage from the light. The intensity must also play an important role. Or am I wrong?

[evilc66]

You don't want to use anything less than 3W LEDs for most reef applications. 1W LEDs can be used in smaller tanks, and especially when used with optics. Anything smaller is an absolute waste. The power (and resulting intensity) is required for penetration. It's the same reason you don't put a 70W MH over a 24" tank and expect to keep sps on the sandbed.

 

As an example, a small array of narrow angle 15000mcd 5mm LEDs can produce PAR in the order of 800-1000 at 1" from the LEDs. Move away to 12" and you drop to around 50. An array of 3W LEDs can produce close to 200 PAR at the same distance, and this is wide angle (120 degree). Strap an equivalent optic on these LEDs (about 15 degrees to make it comparable to the 5mm LED) and you will hit PAR numbers in excess of 1200 at the same distance. 5/8/10mm/superflux LEDs aren't in the same league.

[Ocean Flyer]

You don't want to use anything less than 3W LEDs for most reef applications. 1W LEDs can be used in smaller tanks, and especially when used with optics. Anything smaller is an absolute waste. The power (and resulting intensity) is required for penetration. It's the same reason you don't put a 70W MH over a 24" tank and expect to keep sps on the sandbed.

 

As an example, a small array of narrow angle 15000mcd 5mm LEDs can produce PAR in the order of 800-1000 at 1" from the LEDs. Move away to 12" and you drop to around 50. An array of 3W LEDs can produce close to 200 PAR at the same distance, and this is wide angle (120 degree). Strap an equivalent optic on these LEDs (about 15 degrees to make it comparable to the 5mm LED) and you will hit PAR numbers in excess of 1200 at the same distance. 5/8/10mm/superflux LEDs aren't in the same league.

 

Hi evilc66,

 

Will a array of 22 cool white of 2~2.5" apart with 40 degree len at a distance of 24" depth tank equivalent or high than 150W MH? I meant in term of PAR and lumen?

 

*let say all 3W LEDs drived at a constant current of 1A.

[ajmckay]

Will a array of 22 cool white of 2~2.5" apart with 40 degree len at a distance of 24" depth tank equivalent or high than 150W MH? I meant in term of PAR and lumen?

 

My first thought is that you might need a narrower optic at 24", say 20 degrees. I could be wrong though.

[Ocean Flyer]

My first thought is that you might need a narrower optic at 24", say 20 degrees. I could be wrong though.

 

22 LEDs ( mixed of Cool White and Royal blue) may not enough to cover all area if the len is 20 degree as my tank is L24*W18*H24.

 

Currently, I'm using 150W with 2x24w actinic T5 for my reef tank so would like to have better than current lighting, in term of PAR and Lumen.

[evilc66]

40 degree optics will get you into the 250W range. Exatcly how high will be determined by the amount of overlap. Under my tests, the array was at 2" spacing, and I hit 200 PAR at 24" in water. This is close to what 250W Reeflux 12K bulbs are capable of in Lumenbright reflectors (at least according to Melevs tests)

[Ocean Flyer]

40 degree optics will get you into the 250W range. Exatcly how high will be determined by the amount of overlap. Under my tests, the array was at 2" spacing, and I hit 200 PAR at 24" in water. This is close to what 250W Reeflux 12K bulbs are capable of in Lumenbright reflectors (at least according to Melevs tests)

 

Do you mean that the small degree of the len uses, the higher the PAR?

 

If I would to place my LED as below array with 2" spacing...do you think it can cover all area of the tank of L24"*W18"*H24"? Or need to add more LEDs?

 

CW - RB - CW - RB - CW - RB - CW - RB - CW - RB - CW

CW - RB - CW - RB - CW - RB - CW - RB - CW - RB - CW

[Marteen]

Do you mean that the small degree of the len uses, the higher the PAR?

 

If I would to place my LED as below array with 2" spacing...do you think it can cover all area of the tank of L24"*W18"*H24"? Or need to add more LEDs?

 

CW - RB - CW - RB - CW - RB - CW - RB - CW - RB - CW

CW - RB - CW - RB - CW - RB - CW - RB - CW - RB - CW

 

I would alternate the lights more.

 

CW - RB - CW - RB

RB - CW - RB - CW

 

Something like that. You'll get better light distribution that way.

 

What size heatsink are you using? I'm doing a 17" long heatsink that's 7.5" wide and I'm doing 3 rows of 8 lights.

[Ocean Flyer]

I would alternate the lights more.

 

CW - RB - CW - RB

RB - CW - RB - CW

 

Something like that. You'll get better light distribution that way.

 

What size heatsink are you using? I'm doing a 17" long heatsink that's 7.5" wide and I'm doing 3 rows of 8 lights.

 

L20~22" * W5" ~6" .

 

It's still at initial/design state so not confirm yet and I may switch to 3 rows of 7 instead.

[Marteen]

Well I personally would go with the 3 rows as I think it gives better coverage over the tank. I think you could place them like this:

 

*-*-*-*-*-*-*-*

-*-*-*-*-*-*-*-

*-*-*-*-*-*-*-*

 

Not necessarily that many LEDs but you get the idea. Currently I have my rows even but I left enough space that I could add 2 additional rows of LEDs to my heatsink if I feel I need it later on. I'll upload a picture of my layout for you:

 

 

As you can see I've provided space for an additional 14 LEDs if the need ever arises.

[evilc66]

12x3 would work. You can go with a little wider spacing between the rows to spread things out a little more. What Marteen has been stating is good. You want to alternate the colors to eliminate any banding. Makes the color transitions smoother.

[Ocean Flyer]

Awesome.

 

btw, what is your tank dimension?

 

Well I personally would go with the 3 rows as I think it gives better coverage over the tank. I think you could place them like this:

 

*-*-*-*-*-*-*-*

-*-*-*-*-*-*-*-

*-*-*-*-*-*-*-*

 

Not necessarily that many LEDs but you get the idea. Currently I have my rows even but I left enough space that I could add 2 additional rows of LEDs to my heatsink if I feel I need it later on. I'll upload a picture of my layout for you:

 

 

As you can see I've provided space for an additional 14 LEDs if the need ever arises.

[Marteen]

The actual display tank on my BioCube is around 20"Lx15"Wx15"H. Your tank is a bit longer and much deeper than mine. So I would go with what evil said and try to up your LED numbers. Also you will need optics probably 30-40 degree for that depth.

[Ocean Flyer]

12x3 would work. You can go with a little wider spacing between the rows to spread things out a little more. What Marteen has been stating is good. You want to alternate the colors to eliminate any banding. Makes the color transitions smoother.

 

I'm thinking of 24 or 27 instead of 36 .

[evilc66]

Adjust as necessary. The numbers I gave were just suggestions. I suggested that because 12 LEDs span 22" with 2" spacing. Keep in mind that the optic you use will potentially change the spacing. 2" and 40 degrees works out pretty well.

[Waterproof]

Again, the quantity of LEDs are there for coverage, not for intensity. You will want to set up the array to get light to all areas of the tank. Running the array without optics will get you similar performance to a 70W MH. Try and get a heatsink that will fill the splash sheild as much as possible. The large (8.5" wide) heatsink from HeatsinkUSA should fit the bill. I'd say you should be looking at about 14-20 LEDs, depending on what you want your spacing to be (average is about 2").

 

 

I switched my order at heatsinkusa to the large size. I got a 5" length (8.5" x 5"), which should just fit in the original area designated for the PC lights. When you say 14-20 LEDs, was this to get a light output similar to a 70w HM? I think that will be overkill for my needs. I'm shooting for more of a 72w PC equivalent.

 

I think I mentioned on another post that the output from 2 RB leds appeared much brighter than my 24w actinic PC bulb. I know brightness doesn't always translate into more PAR, and what I interpreted may have simply been the result of a bluer spectral output. However, it would seem that a total of 8-11 LEDs (50% XR-E RB & 50% XR-E CW or XP-E CW) would make the tiny BC-14 bright as hell. I've attached a couple layouts - one with 9 LEDs, the other with 11 LEDs.

 

[Waterproof]

The actual display tank on my BioCube is around 20"Lx15"Wx15"H. Your tank is a bit longer and much deeper than mine. So I would go with what evil said and try to up your LED numbers. Also you will need optics probably 30-40 degree for that depth.

 

 

Marteen,

 

You have the BC-29, right? With 24 LEDs? What PAR are you getting at a depth of ~12"? How would you compare your light to that of a 150w MH?

 

Also, how you do manage the heat? Have you noticed any alarming conditions within the plastic hood? How has the temp of the tank water changed since using the LEDs? Sorry for all the questions.

 

Joshua

[coolwaters]

IMO getting the highest color temp white LED would be nice. i had some pretty outdated blues and decent whites. And i have to tell you the whole thing had a 6500k tint...

after i switched to 4 whites and 7 blues i had a 14k-ish look really nice.

 

im guessing its the organic waste that a SW system generates that makes the water look like that...cuz in my cousin's freshwater a 6500k bulb looks very crisp white while in mine looks yellowish...

[Waterproof]

IMO getting the highest color temp white LED would be nice. i had some pretty outdated blues and decent whites. And i have to tell you the whole thing had a 6500k tint...

after i switched to 4 whites and 7 blues i had a 14k-ish look really nice.

 

im guessing its the organic waste that a SW system generates that makes the water look like that...cuz in my cousin's freshwater a 6500k bulb looks very crisp white while in mine looks yellowish...

 

 

I think you were referring to my layout... I agree with you. The 9-10k XP-Es, when combined with the royal blues, should more easily achieve a color temp between 14k and 20k (depending on how I adjust the potentiometers). I'd need more RBs to offset the yellow light coming out of the 5,500-6,500 XR-E cool whites. I might be able to use a 1:1 ratio between the RBs and the CW XP-Es