DiY LED Grow Lights with CREE CXA3070 COBs and CPU Coolers
- Thread starter Gaius
- Start date
nomofatum
Well-Known Member
Once you start doing that math and realizing that the most efficient chips aren't very efficient if measured in dollars instead of watts, you will find yourself interested in these:that link doesn;t work. i did the math : 1000W lumatek dimmable (+10% super lumens) you need like 700W of CXA3070. That means after 10000 hours you will pay the difference for electric. Than the question is : is this LED really worth the price ? give me the link again please, you made me curious
http://www.aliexpress.com/item/-/2011319079.html
JoeyV
Well-Known Member
OK, what we have here is 'failure to communicate.' It seems that - obviously - has got pounds of primo bud dancing in his head, but has yet to crack a single seed. He's a total noob, which he himself admits to, doing what many of us have done when we were 'total noobs'. He's investing too much of his mental efforts in the minutia of things like 'the difference in efficiency between CREE CX-whatever, and a limp sausage".
Who cares? Yes, for those electrical engineering types and advanced DIYer's who value the chase as much or more than the goal, getting down and dirty into the specs of this COB vs. that COB vs. HID vs. the sun, is all part of the fun. (Though I fail to see how 10-15 % increase in available light power correlates 1 to 1 to a 10-15% increase in final product. But that's another issue...)
But - obviously - has stated that he's NOT particularly handy, doesn't know or have any experience in electrical wiring, and has never grown weed before.
Like so many noobs, he's looking for a boiler plate instruction set on 'easy to build' instant success. And that's fine. Who hasn't colored between the lines before? But he's also getting bogged down by bad reviews, not realizing that these products, like Lumatek ballasts for example, are made and sold by the thousands, and most people DON'T post reviews on them until they have a bad experience.
obviously - Start small. Start simple. Spend more time learning how to grow first, and less time worrying about the little details. Choose products that are generally rated high by a large number of users and don't let a few bad reviews spin you off track.
A DIY LED light fixture is NOT for you, at this point!
You will have plenty of work to do setting up a system that works for you in your situation. You can improve on it as you go. And yes, you WILL be throwing money away, from time to time, on 'off the self' products that don't live up to your expectations. It's all part of learning.
If you want to throw your time and money away (and possibly your freedom or your life), tackle an advanced project like a DIY LED light fixture. It's guaranteed you'll rue the day you set eyes on this project.
Yes , you are right. It's not ok for a noob like me to spend money on expensive lights , BUT if I do the math is the same price for 300W LED like for 600W HPS because for HPS I really need proper ventilation. And a reflector. And the yield is much better for LED than HPS/Watt. Some people told me : get the LED , you won;t regret even if you are begginer. Other people said to me : go to HPS cause you are a completly noob at the minut, so is no point to invest in LED. And because I have no experience , that make me so confuse ... With CXA I read that average yield is 1.2 gram/w and hps is 0.8 g/w. I have no other information , only what I read on forums.
nomofatum
Well-Known Member
If you want to up your g/w get the most efficient lights you can afford (to lower the watts part) and hit something close to the following (to increase the grams):
Veg 60% 6500k (or MH) 30% 3000k (or HPS), 10% UVA/UVB (UVB light)
Flower 60% 3000k (or HPS), 30% 6500k (or MH), 10% UVA/UVB (UVB light)
It doesn't really matter what type of light puts it out. Use those spectrums or similar and it will support a thick plant that can pump more nutes/water and support bigger buds.
Then think about if you care more about watts really or $? If $, maybe get the same grams and use a few more watts but spend a lot less on lights.
Last edited:
I thought is not required a good ventilation for LED. nomofatum : about the lights 6xCree add everything else I need is around 350 pounds. And a 600W Lumatek dimmable hps dual spectrum Lumatek bulb and cooltube is 169 pounds. Add 100 carbon filter and fan. But the lumatek electricity cost is 660W + 100W fan. Led , much less. I start to think that you are right , it;s an easier method for me at the minut to get the hps , it;s more simple to install. I was scared about Lumatek review. The problem is that I will get it from ebay UK , but I will go home soon ( I'm west european ) , so the warranty won;t help me alot. I hope the Lumatek balast will be allright.
yes , I found cheaper in Holland for example d-papillon , but not Lumatek. I get from ebay from mr. hydro. And at the minut I am in UK and I can get free shipping . So what do you think about the cooltube ? Is important which brand it is ? Is powerplant better than a normal one ?
. So what do you think about the cooltube ? Is important which brand it is ? Is powerplant better than a normal one ?
r8rfanizme
Member
Nice setup Gaius, I want to make something like that.
2 Questions:
1. How big is the room being lit?
2. If you knew then what you know now, what would you have done different?
Thanks in advance.
2 Questions:
1. How big is the room being lit?
2. If you knew then what you know now, what would you have done different?
Thanks in advance.
stemkey
Member
Thats what im realizing^ Issue- what rate the lm output starts to decrease? The crees lose little over their lifespan, but how quick do the cheap cobs start to fizzle>?
+What about a good uvb 280nm cob?
nomofatum
Well-Known Member
Where do you see a 280nm for sale? Can you actually buy it? Do they have anything in the 290-305 range?Thats what im realizing^ Issue- what rate the lm output starts to decrease? The crees lose little over their lifespan, but how quick do the cheap cobs start to fizzle>?
+What about a good uvb 280nm cob?
stemkey
Member
Idk but reading this I want closer to 280...(c)Cannabinoid pathway: Anywhere in this pathway UVB does a better job than UVA in energizing a phytochemical reaction that will produce more fully realized THC because "all cannabinolic compounds show an absorption maximum between 270 and 280 nm in the ultraviolet region." Pulled from a 420 thread- 'An elaboration on the phytochemical process that makes THC'.
http://www.alibaba.com/product-detail/China-wholesale-20w-280nm-uv-led_60124739709.html
http://www.ledwv.com/uv/dvc-280nm-cob-16-chip-p-771.html
I just hate waiting for stuff from china, anybody know usa seller?
nomofatum
Well-Known Member
Be careful with those wavelengths. 270nm-280nm is actually UVC. Make sure you know the risks of UVC, don't expose yourself to UVC light, at all if possible.
From what I can find no natural UVC reaches the earth except maybe below the ozone holes.
Last edited:
stemkey
Member
I know what you're saying. The easy route would be the 10.0 reptile uvbs for 5 hours in the middle of lights on in flower for guaranteed increased potency but thats not a diy led lol. Maybe adding a few low w 280nm chips to run with the other cobs or a couple 20-30w 280nm that run a few hours in the middle of lights on? Any one done this or know how, safely for plants? The leds here need uvb -
uvb thread--
http://www.420magazine.com/forums/c...-setup/79209-uvb-light-what-s-real-story.html
uv increases thc thread--
http://www.420magazine.com/forums/b...boration-phytochemical-process-makes-thc.html
nomofatum
Well-Known Member
I'm running a reptile bulb in my temp setup right now. I have UVB tanning bulbs for this very reason in my new grow room. From what I saw when I read up on this the target was more like 290-305nm. 270-280 would make me setup a switch on the door so I would never be exposed even by mistake.
Just FYI 270-280nm should not make it through the ozone, so naturally your plants wouldn't be exposed unless you live near a ozone thin spot or hole.
Gadgetboy
Member
I have just started to build one of these for a friend who has no soldering skills, he has asked me to add 6 deep red LUMILEDS LXZ1-PA01-0350 ( http://www.ebay.co.uk/itm/271444899513?_trksid=p2060778.m1438.l2649&ssPageName=STRK:MEBIDX:IT ) and 4 Royal blue LXZ1-PR01-0600 ( http://www.ebay.co.uk/itm/LUMILEDS-...al_Components_Supplies_ET&hash=item3f29c3307f ) My first question is will these make any difference to the light performance? secondly what would be the best drivers to use (preferably one driver for red one for blue)
