LED grow lights... input please.

[Opie Yutts]

You can use anything for the veg stage, including LED's.

I have yet to see anything worthwile come out of flowering with anything other than HPS. The possible exception might be flouros, but a shitload of them instead of one.

Because I have not seen it does not mean that it can't happen (are triple negatives OK?). I have a feeling that you could equal the yield produced by one 600 watt bulb with several thousand LED's. I haven't done the math but there is a formula to figure this out. How much would this cost? Much more I am sure than 1 HPS bulb and the additional ventalation necessary. How long until the LED's would pay for themselves? The cost to power thousands of LED's - would that be similar to the cost to power one 600 watt bulb? And if it is less, how long until it pays for itself in electricity.

And do you have to replace them? HPS bulbs need replacing every year or so due to decrease in intensity. My guess is that LED's last much longer. I Think the real cost we are talking about is the electricity to power HPS and it's ventalation.

I am pulling for LED's. I hope some day they will be economically viable. Right now they are way too expensive for flowering.

Good luck, and keep us posted.

[ledgrower]

440s and 460s...lookin good monkey the best thing is , is that thats only burns 36 wats of power.... keep it up. u could also ad a small uv light in there if u have the space..that would give it even more of an edge to pick up some of the differnt spectrums that the leds dont have....even though they are shining the best ones plant uses...theres still some smaller ones that u could shine. a full spectrum bulb would cover all that..

[Opie Yutts]

The goal here is to use only LED's through flowering. Let's not add other lights. If we do, there's no way to tell for sure how well LED's work.

At least that's what I would like to see, but then it's your experiment. There has been many flowerings using LED's plus other lights. I have yet to see a flowering using only LED's. I think this is probably because you would have to buy thousands of them to get any kind of yield. Please prove me wrong.

[Moose101]

I was looking for a way to build one of these suckers. Found a nice website. http://www.superbrightleds.com/led_info.htm. Would be even better if there was a complete guide on here. One that would take the latest in LEDs into concideration. It's still something I don't understand. Think I might have to read up on electrical engineering just to understand this stuff. Worth a shot. Here are two other websites, http://ledmuseum.home.att.net/ledleft.htm, http://members.misty.com/don/ledx.html.

It seems easier to go rob the red traffic lights. But I don't wanna attract any unwanted attention. Me wonders if there would be any complications in removing them. Aside from the car crash every now and then. Calgary drivers are not so bright.

I would guess it's best to aim for the highest lumen count. Or does it not matter in LEDs? And whats with the 1 watt leds, would they be a better buy?

I think I might build myself red and white flashlights just to have around for camping an stuff. Red light might be good to have in gorilla growing.

[knna]

Opie Yutts: I agree with u, today isnt profitable substitute 600w HPS lamp for leds. Probably in 2-3 years.

But leds setups need far less ventilation than HPS ones, if u design the leds arrays in a way wich convection cooling will be enough. This mean not compacted arrays.

Where leds can be now a good choice is for microgrowers. CFLs and small HID lamps are a lot less efficient than 600w HPSs. Using 30% efficients leds at a reasonable price, wich can be found, in about 3-4 years they paid themselves in energy savings against a 250w or smaller HPSs (i did this calculation for my country, Spain, where we are thinking about a leds group buy).Good leds mounted correctly will have a useful life of 30000h, up to 60000h (not white ones, or led bulbs, or cheap 5mm leds)

Today the weak part of a led grow system is the blue part of the spectrum. Blue leds are still too inefficients and expensive. Valid for experimental purposes, but not for energy savings. Now, for small and microgrowers, the profitable way is combining fluo light (for blue and a bit of the others, ensuring no lackings of certain wavelenghts) and red leds.

I would guess it's best to aim for the highest lumen count. Or does it not matter in LEDs? And whats with the 1 watt leds, would they be a better buy?

I think I might build myself red and white flashlights just to have around for camping an stuff. Red light might be good to have in gorilla growing.]

When talking about leds, is a trade off between initial cost and energy savings.

High power leds are cheaper in terms of output mW/$ and in mounting cost, but they are less efficient than medium and low power ones. Lumileds Luxeon III (or the new K2) have a very good initial price, but they are too inefficients at specs currents, and you may need to use more led watts, with less energy savings. A good compromise is run them below specs current, increasing efficiency and ensuring no heat problems (for leds themselves, not for plants), and avoiding active cooling.

Probably the best relationship between initial cost/efficiency (energy savings) can be found in superflux (piranha) leds, runned at 50mA. And they can be founded in all wavelenghts, while high power leds are usually in the range of 630-635nm peak wavelenghts, valid but less effective than longer wl. In the near future, probably the high power leds achieve enough efficiency and good price to be the best option, but not today.

[Moose101]

Today the weak part of a led grow system is the blue part of the spectrum. Blue leds are still too inefficients and expensive.

Could have sworn I read somewhere that you need 7% of your total grow lights to be blue, anymore wouldn't do anything.

BTW, I haven't seen any experiments that match the wattage yet. Would be cool to see what 600 watts of led can do to a plant. Look at this experiment, http://www.ledgrowlights.com/univers...nn%20study.PDF

Another question: would it make sense to put a larger viewing angle in the middle and then decrease it on the outside. Lets say I create a circular bulb. Use 45° in the center, 30° halfway between the center and outside, then switch to 15° on the outside.

[knna]

Could have sworn I read somewhere that you need 7% of your total grow lights to be blue, anymore wouldn't do anything.

Yes, ive seen some studies (none specifically about cannabis) wich concludes that a minimun of 6% of micromols of photons blues is enough for some plants species. And, in general, between 6% and 12%.

Other studies concludes that the blue requeriments are absolute, rather than a percentage of total. Usually, a minimun of 30uE of blue light per sq meter for healthy grow, and about 40uE for healthy growth with short internodal spaces (the difference between a generic 400wHPS lamp in one sq meter against an enhanced blue 400w HPS bulb; with the diference between leds and HPSs consisting in a HPS blue emision near 500nm, very less effective than a 450nm blue led).

The problem is only see the 6% figure. It refers to percentage of total micromols of photons per second (uE), not a emitted watt percentage. Blue photons carry a lot more energy than reds, so a 6% of uE means near 20% of emitted watts, and like blue leds are more inefficients, problably means near 35 or 40% of the total installed led watts.

Probably the way to go is having blue arrays wich emitts 40uE/m2, and go adding more and more red light as plants grow, anf forgot the % of blue light.

[knna]

Another question: would it make sense to put a larger viewing angle in the middle and then decrease it on the outside. Lets say I create a circular bulb. Use 45° in the center, 30° halfway between the center and outside, then switch to 15° on the outside.

I think is a good idea when using circular bulbs. But i prefer a different aproach, consisting in slim linear strips distributed along all the grow space, thinking in 3D, and seeking for a even flux distribution over the whole plant.

[Moose101]

That sounds like a better idea. But isn't there some kind of ineffiency in the wire and current if it gets too long? I'm no expert, heck I don't even know if I'm saying this right.

[Curiosity]

Resistive losses in the wire will be negligible unless you're stringing your arrays rediculously. If you manage to make it noticeable, all you'd have to do is slightly lower that chain's resistor's value.