DIY LED 220w Cree XTE + Philips ES

They are only available in the 10mm star size last I checked. They are tricky to hand mount compared to the 20mm. Also there is a typo in the item description which says R2 and the part number says R3. When dealing with cutter typos I always go by the part number over the item description. I think the problem is that when they update the descriptions they leave some old info from previous items. On the other hand maybe it is an R2 and the part number has the typo. I will ask mark if i can track him down on CPF.
 
That pic is of Cali Connection's Buddha Tahoe OG fem (King Louis VIII X Tahoe OG). A direct crossing of two elite clones done under a 600HPS. The idea is to set the bar way up there and see if the LEDs can compete with the 600.
 
Supra- great thread. And what great participation you've garnered. I believe I'm using the same sink as you, and I'm using XMLs. Running at ~2A, for a resulting 300W panel. I have fans.
 
The Cree specs allow for 3A, but recommend 2A, so I'm not anticipating running hot. The panel isn't done yet so I haven't sparked them up yet.
 
Rrog said:
The Cree specs allow for 3A, but recommend 2A, so I'm not anticipating running hot. The panel isn't done yet so I haven't sparked them up yet.
Click to expand...

The LED will get hot, but it will be ok. I run 5 single 42v LED at 1.2A on the same kind of heatsink, with a fan and it's fine. Hot but fine. Total watts on a 5" x 5" heatsink area is about 200w. Problem with these heatsinks is the face of them is thick. Around 1/4" to 5/16".

The thinner the face, the faster the heat will dissipate from the LED, through the heatsink and out into the air when you're using cooling fans. For passive cooling the thick faced and finned heatsinks work well, but for active cooling you want thinner aluminum in the heatsink with more surface area(more fins) and air movement(think computer processor heatsink).

My heatsink doesn't get hot but the LEDs still do because this type of heatsink everyone is using is thicker than they should be for active cooling. Take a look at the LED cooler from heatsinkusa.com. This is the type of heatsink you want, only a lot bigger. If anyone can find a large thin type of heatsink with lots of surface area let us know.

BTW, I like the word "heatsink." lol
 
Rrog are you using warm white XML alone or with colors? At first I considered running at higher amperages but a few problems cropped up. First it got me wondering if it would affect the depth of penetration. If the reds and blues are running at 1A and the whites running at 2A I expect the white would go deeper into the canopy than the other colors resulting in inconsistent spectral mixing. Another downside is that the light would be less evenly spread in the canopy because each emitter would be brighter and there would be less of them.

The next thing that changed my mind is that I calculated warm white XML is only 26.5% efficient at 2A (assuming a Tj of 50c). The XM-L does perform efficiently at 700mA but at that level the XTE outperforms XML as well as being much cheaper.

XML.png

On the other hand, if XTE @ 700mA (2watt) dissipation was just not yielding big enough buds, there is another high power option, Cree MKR. The MKR can achieve high levels of efficiency at higher dissipation wattage. The 3000K warm white MKR is about 39% efficient while dissipating 4watts (350mA - 11.25vf @ 50c). It is 36% efficient while dissipating 8 watts. The downside is that it is harder to achieve low junction temps I like to aim for (50c). That is not a deal breaker though the MKR performs very well at higher Tj and is binned at 85c.
 
Actually I'm running NW and WW. No colors. I would turn one WW off when using as a veg light.

I'm expecting passive 54C, active 42C
 
I never thanked you for the tip on sourcing the XT-E in the bins i wanted, Supra - That particular panel got installed yesterday and is looking tasty. :)

Oh and you weren't kidding about them being a pain to mount, with those 10mm stars! I ended up reflowing the whole circuit together on a clothes iron before bonding it to the heatsink. Wayyyyy easier. :P

The penetration issue (lol say no more) is what made me move away from XTE for my next panel- Hard to find colours with similar power and beam angle. I was considering running a XTE for warm whites / Ledengin LZ-1 for reds and deep reds combo (LZ-1 are also rated for silly currents) but the beam angle on the LZ-1 (90 deg) put me off. For this particular panel i need wide beam angle, as its a big heatsink with a small footprint.

I'm about to pull the trigger on fitting it with entirely oslons - SSL 150's - for both color and white.
 
Ooooosh...pretty steep though. Roughly double what i pay for oslon + star.

Don't see any nice warm whites either, that might just be my fail french though :x

Thanks for the link regardless, i can see it being really helpful for peeps that aren't willing to do the reflowing
 
SnotBoogie said:
Ooooosh...pretty steep though. Roughly double what i pay for oslon + star.

Don't see any nice warm whites either, that might just be my fail french though :x

Thanks for the link regardless, i can see it being really helpful for peeps that aren't willing to do the reflowing
Click to expand...

Hello SnotBoogie, if you don't mind me asking...Where do you source your 'oslon + star'?
 
Technically this is not scientific because there was a 600w HPS on the other side of the room. There were different varieties under each light. Also, this is the first time these varieties have been run so they have no control to compare to. That said, here are a buncha buds that were directly under the LED modules. They are all OG and were underferted just a bit but stayed greener longer than the ones under HPS.

DSC06681b.jpg

DSC06692b.jpg
 
The third pic shows a bud that had one side lit up by the LED and the other side in the "shade". For whatever reason LED light encourages anthocyanins.


OG18

DSC06682a O18c.jpg DSC06686a O18.jpg DSC06693a O18c anthocyanin.jpg




Kosher Kush

DSC06683b Kosher.jpg
 
So far I am very happy with the results. The goal was to see if 200w of top bin LED can replace a bare 600 HPS, a very lofty goal. Estimating at a glance maybe 250-300watts would have been more appropriate. It will be quite awhile before the results are on a scale but it will still be an estimation since they were in the same room, there were different varieties under each light, and there was no previous control for those varieties.

The Sour Kush was under LED and really did not impress me in terms of nugget size but the OG18 and Kosher really did impress me under LED. On the other hand the Kandy Kush was under HPS and yielded nice while the RP OG Kush was also under HPS and did not yield too great. So it will take a few runs to establish the full potential of the LED.

The density of the LED nugs is just as good and possibly better than the HPS nugs. The frosting and dankness on the LED side is flat out better than the HPS. Bud to leaf ratio is excellent, easy to trim and minimal larf, decent penetration (8-10 inches).

The LED modules allowed for better coverage of the canopy. Distance from the LED to canopy was about 10 inches but with more modules I'd probably go with 8 inches. The canopy temp was perfect with 600w HPS and 200w LED <80f. I was able to turn the exhaust fans down saving electricity, cutting down on noise and allowing a longer contact time with the carbon. I could have squeezed 6 LED modules (~300w) in a space that normally has 600w HPS and there is no doubt in my mind the LED yield would be equal or greater than the HPS.

The downside of the LED is the initial cost and the difficulty hanging the modules. The rooms use ceiling fans for circulation (because they are quiet) and height is limited so I had to get creative to get the modules suspended. Since the wiring and hanging is heavier and more complex than HPS it would take longer to put up or tear down.

As far as cost goes I estimate the LED would have to run for at least 18 months before it would pay for itself with electrical savings over the HPS and from air conditioning and HPS bulb changes.
 
You must log in or register to reply here.
Back
Top