DIY LEDs - The write-up

[TheFishMan65]

Bill,

There is an internal adjustment to limit the current. It has been recommended/suggested that this be turned all the way counter clockwise (most severly limited) the first time.

Also most of the XR-E that are being used in these builds will allow 1000ma. I think only the greens are limited to 700ma.

 

[wld1783]

Bill,

There is an internal adjustment to limit the current. It has been recommended/suggested that this be turned all the way counter clockwise (most severly limited) the first time.

Also most of the XR-E that are being used in these builds will allow 1000ma. I think only the greens are limited to 700ma.

Your right I was looking at the XR data sheet

The XR-E data sheet shows a max pulse of 1.8A for intermittent on off like a blinking light. 1000mA continuous is recommended to achieve 50,000+ hours.

Its good to know I can drive them a little harder if I have to.

Bill

 

[TheFishMan65]

I just had the opportunity to watch a seminar on "Designing for LED Luminaire Reliablity" by electronic design, Bergquist, and Digi-Key.
The slides referred to Luxean and CREE (maybe more) so I am assuming the information is universal.
Some interesting points:
They had an interesting chart
.........Incan...Fluor...MH...LED
Visible....8%.....21%....27%..15-25%
IR........73%.....37%....17%...~0%
UV.........0%......0%....19%...0%
Total..
Radiant...81%.....58%....63%..15-25%
Energy.
Heat......19%.....42%....37%..75-85%

Another slide showed that lifetime was not limited until the junction temperature got over 120C. We have a goal - if we could just measure it. They also fail in in less than 10,000 hours at 150C. This chart mentions Luxeon.

And one other point I had not considered. I assumed heat would just slowly lower lifetime. However it also stresses the solder connections. This can be affected by material (which we don't have control over) and by over all hottest temperature to coldest temperature.

Session 2 is on the CIE color gamut. I will post anything interesting. They say this webinars will be available in about a week. I will post link, but not sure if they will limit access.

 

[CJO]

Good info. I would be interested in seeing all of it if it becomes available for general access.

Thanks,
CJ

 

[The Grim Reefer]

Not sure who is right but I have read that some white LED's do have UV output because they work like a fluorescent lamp (UV light stimulating phosphors) which contrary to what that chart shows do have UV output.

Good point on the solder connections too. Poor connection will generate heat.

 

[jtma508]

FishMan... can we assume that the measurements were taken from the bare device? That is, MH lists significant UV but is that w/o the shield?

 

[Obi-dad]

Not sure who is right but I have read that some white LED's do have UV output because they work like a fluorescent lamp (UV light stimulating phosphors) which contrary to what that chart shows do have UV output.

Good point on the solder connections too. Poor connection will generate heat.

There may be some LEDs with UV, but I think the Crees we use are blue LEDs with white phosphors.

 

[Obi-dad]

I just had the opportunity to watch a seminar on "Designing for LED Luminaire Reliablity" by electronic design, Bergquist, and Digi-Key.
The slides referred to Luxean and CREE (maybe more) so I am assuming the information is universal.
Some interesting points:
They had an interesting chart
.........Incan...Fluor...MH...LED
Visible....8%.....21%....27%..15-25%
IR........73%.....37%....17%...~0%
UV.........0%......0%....19%...0%
Total..
Radiant...81%.....58%....63%..15-25%
Energy.
Heat......19%.....42%....37%..75-85%

...

That chart doesn't look right - it is saying that LEDs produce 75-85% of their energy as heat? I don't think that can be correct.

 

[TheFishMan65]

That was my assumption it does not say. Looking at the slide there is a note that the MH data came from OSRAM Sylvania, the incan and Fluor data was from IESNA Handbook.

Hope it helps.

 

[TheFishMan65]

There is another note that some LEDs claim over 30% for visible light. And DOE's SSL multi year program plan (March 2006) calls for increasing extraction efficiency to more than 50% by 2012.

I sort of think this might be right. If you look at XR-E. Say they are 25%. Then the XP-G would be 25% * 1.3 = 32.5%. Don't remember where but wasn't the new CREE record for lumen/watt something like twice the CR-E or 50% which would be were things are going.

 

[The Grim Reefer]

Have a link?

 

[akindbro4u]

Have a link?

here you go grim. Just a warning it's a heavy read. I'm sure you can handle it, but some of our "novice" readers should really take it easy trying to digest this stuff.
http://www1.eere.energy.gov/buildings/ssl/techroadmaps.html

 

[The Grim Reefer]

Thanks. I've read some stuff from there before. I am guessing most of it is outdated considering the advances just in the last couple years.

 

[akindbro4u]

the newest round table was just last month

 

[FW fugitive]

That chart doesn't look right - it is saying that LEDs produce 75-85% of their energy as heat? I don't think that can be correct.

I can. If they are 15-25% efficient in converting electrical energy to visible light, and put no energy into UV or IR... there's only one other option. Heat. But it's more manageable than the IR put out by MH or fluorescents, because you can deal with it at the source. The MH IR goes into the tank, where it heats up the water.

One of the major advantages of the LED's is that it's close to a point source. So even if they're a little less efficient than MH, you can get more light into your tank. Add to this that you don't have to replace the LED's every 6 months...

(edit) I just looked at the link - http://apps1.eere.energy.gov/buildings/publications/pdfs/ssl/ssl_mypp2010_web.pdf . See Figure 2.1 on page 23. It's already outdated - it has a great figure which shows white LED efficiencies climbing towards 100 lumens/watt and HID's as most efficient at about 120 lumens/watt. The Cree XP-G R5 is already past that at 130 lumens/watt.

 

[dwl]

OK guys, I need some help.

I have 12 XP-G whites and 12 XR-E RB's. I am planning on using U-channel for the heatsink and have a sample piece of the aluminum trim channel for 3/4" plywood from Lowes.
It roughly measures 15/16" wide and is about 1/16" thick.
I plan to build a canopy box and run fans across the heatsinks and have a plenum box at the side, basically as Kcress has suggested.
I can and will go to 1/8" thick and wider stock if needed.

I have a cube (ish) tank with a 24" square footprint, but it has a false back/overflow, so the front to back inside dimensions are about 18", so I need to light roughly 23" x 18" at 18" depth.


2 questions, what ratio of XP-Gs to XR-E's should be using? Should I just use all 24 leds and dim the whites if/when needed, or use only about 8 or so of the whites and all of the blues?

Also, any suggestions as to how to orient the whites and blues?

Thanks.

 

[The Grim Reefer]

OK guys, I need some help.

I have 12 XP-G whites and 12 XR-E RB's. I am planning on using U-channel for the heatsink and have a sample piece of the aluminum trim channel for 3/4" plywood from Lowes.
It roughly measures 15/16" wide and is about 1/16" thick.
I plan to build a canopy box and run fans across the heatsinks and have a plenum box at the side, basically as Kcress has suggested.
I can and will go to 1/8" thick and wider stock if needed.

I have a cube (ish) tank with a 24" square footprint, but it has a false back/overflow, so the front to back inside dimensions are about 18", so I need to light roughly 23" x 18" at 18" depth.


2 questions, what ratio of XP-Gs to XR-E's should be using? Should I just use all 24 leds and dim the whites if/when needed, or use only about 8 or so of the whites and all of the blues?

Also, any suggestions as to how to orient the whites and blues?

Thanks.

I used 2x1x1/8th channel for mine and they run really cool. That plywood trim stuff is pretty thin but with good airflow they might do the job. I got 6 4" sections from speedy metals online for like 69 bux shipped.

 

[akindbro4u]

OK guys, I need some help.

I have 12 XP-G whites and 12 XR-E RB's. I am planning on using U-channel for the heatsink and have a sample piece of the aluminum trim channel for 3/4" plywood from Lowes.
It roughly measures 15/16" wide and is about 1/16" thick.
I plan to build a canopy box and run fans across the heatsinks and have a plenum box at the side, basically as Kcress has suggested.
I can and will go to 1/8" thick and wider stock if needed.

I have a cube (ish) tank with a 24" square footprint, but it has a false back/overflow, so the front to back inside dimensions are about 18", so I need to light roughly 23" x 18" at 18" depth.


2 questions, what ratio of XP-Gs to XR-E's should be using? Should I just use all 24 leds and dim the whites if/when needed, or use only about 8 or so of the whites and all of the blues?

Also, any suggestions as to how to orient the whites and blues?

Thanks.

if it was me I would do a shoebox sized fixture say 16x10, and mount it about 18" up. I would do all the blues, and eight of the whites. just my 2¢

 

[King_Richard]

OK guys, I need some help.

I have 12 XP-G whites and 12 XR-E RB's. I am planning on using U-channel for the heatsink and have a sample piece of the aluminum trim channel for 3/4" plywood from Lowes.
It roughly measures 15/16" wide and is about 1/16" thick.
I plan to build a canopy box and run fans across the heatsinks and have a plenum box at the side, basically as Kcress has suggested.
I can and will go to 1/8" thick and wider stock if needed.

I have a cube (ish) tank with a 24" square footprint, but it has a false back/overflow, so the front to back inside dimensions are about 18", so I need to light roughly 23" x 18" at 18" depth.


2 questions, what ratio of XP-Gs to XR-E's should be using? Should I just use all 24 leds and dim the whites if/when needed, or use only about 8 or so of the whites and all of the blues?

Also, any suggestions as to how to orient the whites and blues?

Thanks.

I just made a prototype box that's 8"x25"x5.5" W x L x H

I'm only using 12 xpg's currently but plan on adding another 12. I also used the cheap 10$ aluminum u-channel to mount the led's on and added a 120mm fan blowing across the led's. The led's stay cool and would be cooler if I added a splash guard on the box since it would help cool the led's furthest from the fan. I have a build thread that will be getting picture's up of em sometime today or tommorrow if you wanna check it out. It's in the reef discussion section called "60g w/50g sump build." I definitely would put a dimmer on atleast the whites and also use all 12 of your whites since from my understanding is the color that will make corals grow the quickest but at a cost of some coloration (this is just my understanding of it and I could be way off). I haven't put a dimmer on mine yet but I intend to and when you get yours up and running I'd like to see some pic's of the 12 white and 12 blue since this is what I was leaning towards doing also.

Forgot to tell ya, my tanks 24x24x25

 

[TheFishMan65]

Here are the links to the seminars I keep referring to. Now you to can refer to them too . I hope the links work the first link they sent worked, but second one needed a login and they were only available for a year. So these is the alternate site - think they would have used this first since they sponsored it. This set you have to login (free), but hopefully they will last longer. Thank Electronic Design and Digi-Key and a bunch of companies.

• Designing for LED Luminaire Reliability
• LED Color-Temperature Control
• Controlling LEDs with Microcontrollers
• Using Analog Semiconductor Technologies to Power Solid State Lighting Applications
• Characteristics of HB LED