3watt LED"regular" vs Cree

[mess7777]

Hi all,

I am curious to know how much of a difference there is from a standard 3 watt LED to a Cree one. I haven't been able to find anything about this on any searches I have done.

I have a home-made kit with 6 heads containing 14 - 3 watt LED each. When tested with an Apogee PAR meter the results were abysmal. (~200 at water level, 40-60 at the bottom). Pic for reference(bought used, prior to cleaning). It also has a DIY control box with dimmer.



I am considering options of swapping out the existing LED for CREE and if that would help to get my lights up to speed without having to dish out $1200 for an all new lighting system.I see many setups on here using less than 82 - 3 watt LED on a 90g tank that seem to be doing quite well.

Thanks for any input.

[xCry0x]

What is a typical 3w led in your mind? Bridgelux & cree are the 2 main sources with Cree being almost too strong. (A lot of people dim them significantly)

[BigKahuna]

I'm not sure how true it is but I've read often that LED's are notoriously difficult to get accurate PAR readings so take them with a grain of salt.

[rwb500]

the data sheets on chinese LED's are often suspect, so its impossible to know exactly how much worse they are. but last time i looked at the numbers, i think the chinese LED's put off about 2/3 as much light as the latest CREE (not the XPE though).

are you sure your dimmer is providing the right about of current? can you test it with a multimeter? also those heatsinks look pretty puny. theres 14 led's on each of them? does it get really hot when on for like an hour? feel the plate of meter right next to the LED, on the same side. they may have been slowly burning up during use by the last owner.

[mess7777]

Originally Posted by rwb500

the data sheets on chinese LED's are often suspect, so its impossible to know exactly how much worse they are. but last time i looked at the numbers, i think the chinese LED's put off about 2/3 as much light as the latest CREE (not the XPE though).

are you sure your dimmer is providing the right about of current? can you test it with a multimeter? also those heatsinks look pretty puny. theres 14 led's on each of them? does it get really hot when on for like an hour? feel the plate of meter right next to the LED, on the same side. they may have been slowly burning up during use by the last owner.

I can live with 2/3. I honestly don't know what they are brand-wise, other than that they are Chinese.

Yes there is 14 LED on one plate. The plate is hot to the touch. On the side of the LED I can't hold my finger there very long, on the other side of the plate it's still quite warm.

As for the dimmer I have heard rumors from other owners that he purposely sets them to not go "full blown" so there are less repairs with blown bulbs or burnouts or what may be. That said, I wouldn't have any idea how to test it and don't have a multimeter. If it's easy to measure, I can probably find someone to borrow a multimeter from.

I am of the opinion that 82 - 3 watt LED should be sufficient to light my 90 g tank for pretty much any coral. Also there are no optics to speak of on the LED, so I just have the light pointing straight down...

I can provide more pictures if that will help to diagnose if I am getting the full potential out of these lights.

[rwb500]

82 3w LED's should definitely be enough. I think anyone would agree that 48 CREE's is enough for that size tank.

Are there optics on the LED's? Usually with a tank that tall, people use optics.

Can we see some more pictures please? I'd like to see pictures of the dimmer/driver and the LED's themselves.

one of these is a good investment, at $16:
http://www.amazon.com/Equus-3320-Aut...0462638&sr=8-2

that will take some guesswork out of this.

to me right now, it really seems like there are too many LED's on each plate. I forget the exact ideal (max) operating temperature, but people usually say that if the plate is too hot to touch, its too hot.

[WingoLED]

Do you have optics on the LED? Most build in this forum are made with 60 lens, intensity of 60 degree lens vs without optics can be upto 4 fold in array arrangement.

Unless you know the model number of your LEDs, there is no easy way to compare the efficiency of LEDs. Even Cree themselves of same model but different bin could mean more than 5% difference in efficiency and a 1000K color difference.

[redfishsc]

Here is part of the answer you need. This is a spectral graph of 4 LEDs (cool white XML, cool white XPG, neutral white XPE, and "pure" white Satistronics--- I did not have a "cool" white XPE or Satis to compare, but the results are still VERY obvious--- using a cool white Satis would give about the same result). This data was taken at the University of North Carolina, Wilmington, on their Ocean Optics spectrometer.

The chinese LED is the Satistronics one--- that looks identical in every way to those "bridgelux" 3-watt LEDs that you see all the time.


Compared to an XML, both at 700mA, the Satistronics is about half the intensity. And I've verified this with a PAR meter. A single XML white at 3,000mA (about 10.5 watts) puts out consistently twice the PAR as 4 Satistronics whites at 650mA (about 9 watts).



Look at the height of the graph and remember, the XML has a lower voltage, which means it's putting out nearly twice the light at LESS total electricity being used.

[WingoLED]

I feel the chart should be broken into two putting the same color temperature range together such as cool white vs cool white and neutral white vs pure white because this is kind of misleading if the reader do not understand the relation between color temperature and PAR output.

[The Grim Reefer]

I am running 96 cree 3 watt LEDs, mostly XR E's over a 120. No optics driven at 700ma and it's strong enough to keep a BTA hiding a little more than half way down the rock at the edge of the tank out from under the LED's. A Orange monti on top of the rocks which are about 2/3 of the way up the tank has very little vertical growth but has spread out across the rock. Monti Cap has gone from a 2 inch pie shaped frag to about a 5 inch disk since late February. I obviously have enough light.

I get PAR readings in the 120's at the bottom of the tank BUT I use about a 2.3 to 1 mix of Royal Blue to white and as mentioned a PAR meter does not read blue light well, especially deeper blue which the Royals peak in.

[rtparty]

I may get flamed for this but so be it.

I did quite a bit of research the other day and contacted a few different companies to figure out why CREE was so "great". Some companies and people actually thought CREE was an aquarium specific LED company. I laughed and informed them to do some research on CREE. Our hobby isn't even a drop in the bucket for them.

What I found though was shocking. CREE doesn't even manufacture some of their LEDs anymore. They have outsourced quite a few of their emitters to companies like Edison and others. Some companies are claiming to use CREE LEDs as well but when they were taken apart they are using cheap Chinese knock offs. So be careful where you buy some of your stuff from.

I got to see some Bridgelux LEDs for the first time just the other day and really like them. I thought some of the colors were better and color rendering was an improvement over CREE setups that I have seen in the past.

In the end, I think CREE is over rated and over priced when compared to other quality LEDs on the market. The problem is that certain individuals pushed CREE so hard at the beginning that they became the norm for the hobby. Everyone believed they must be the best when really there are some really good LEDs out there that are much cheaper.

Let the flames begin....

[heckeng]

LEDs also lose intensity at high temps, so if your heat sink is not sufficient to keep them cool, they will be dim. By the sounds of it, they are cooking along at a pretty high temperature.

[The Grim Reefer]

A friend here has a biocube with an LED kit light on it. The original used Edison LEDs. One of those (strips of 6 LED's) burned out on him and the replacement is Cree LED's. He says the crees are noticeably brighter than the others in his array.

As far as coloring goes the blue and cool white has never given nice colors in my humble opinion.

[The Grim Reefer]

A friend here has a biocube with an LED kit light on it. The original used Edison LEDs. One of those (strips of 6 LED's) burned out on him and the replacement is Cree LED's. He says the crees are noticeably brighter than the others in his array.

As far as coloring goes the blue and cool white has never given nice colors in my humble opinion.

[rwb500]

84 of ANY 3w LED should be enough for that tank. this is not a CREE vs chinese LED thread. were trying to help this guy with his specific problem. but while were on the subject, you can see the chinese LED is barely worse than the XPE, which many people used with great success. comparing it to only the XML is not a useful comparison.

[The Grim Reefer]

That would depend on the "Chinese LED" that you are using. It's the same issue you run into with using any unbranded equipment. There is no standard so you could either get a chip close or even equal to a cree or one that really sucks based on the luck of the draw.

As far as OP's issue I agree about the size of the heat sink. Easy way to check for a heat issue is put the par meter under them first thing in the morning and see if the PAR falls off as they heat up.

[mess7777]

Unfortunately I only had a PAR meter one time that I borrowed, I won't be able to check this out. I'm going to go back a few posts and get some pictures of the controller box and LED plates to get some advice.

I really want to know if I am getting the most out of them. I don't want to invest a bunch of $$$ in them, but if inexpensive solutions are available I will pursue.

[gpx1200]

mabie if you switched to crees you could use a few less led's and get more light with less heat. i'd say if you can't hold a hand on the heatsink theyr way to hot