Author Options:

LED Grow Lights Answered

I've been an avid rosarian for many years, and am finally getting around to trying to grow miniatures indoors. I've been looking at some of the commercially available LED grow-lights, and they're insanely expensive, not to mention bulky and unattractive. It occurred to me that a lot of the info here would be highly relevant to such a DIY project. What I'm wondering is how many LEDs would be needed to produce around 2000 lumens in an approximately 1-foot square for one rose? I'd gladly listen attentively to anyone who can comment on this with some degree of authority.



5 years ago

greetings all. First, the small 5mm leds don't really work for a grow light. Yes, one can grow with them but not as well as the higher powered leds. Right now, the sweet spot for flux density/ watt consumed is best at the 3 w lvl. Well, 1 watt leds are better but it is hard to get enough of them in a given area. Here be the problem, I have tomato plants that are 3f to 4f tall. I want my light to be "bright" enough at bottom of the plant as well as at the top of the plant to support robust growth. 5mm leds would work for a pancake shaped plant with all its leaves in a flat disk sort of like the shape of a dinner plate. But that isn't what I have with my hypothetical tomato plants. So, I need to start with more flux density to reach the bottom of the plant to support robust growth down there too. So, 3 w leds work good. Even better is adding little reflectors to each led to turn each one into a spotlight. Now my lights will work from the top of the plant to the bottom. Ah but using 3w leds means that I have to remove heat. I mean a 3 watt blue led that has a 500mW radient power heats up the environment the other 2.5 watts. Anyway, I am making a light for a friend. So far, I have 16 leds: 4 660 red and 4 2700 k white leds for the red end of the spectrum; 4 445 blue and 4 12000k white for the blue spectrum. That should work out for most any plant. This is the second light that I am building. The first light was just blue and red leds and didn't really work all that well. This one is full spectrum just enhanced in the red and blue light frequencies. Cost is 72.00 for the led's and drivers. I have scrounged the floodlight housing. Pulled the 100w halide ballast from it. I have to find an aluminum plate for the heatsink to mount the leds on. Find aluminum rod to make heat pillar to xfer the heat in the center of the aluminum plate to the case as there will be no fans or holes for ventilation. I plan to mount the led drivers on the outside of the case as they are already waterproof. So, as long as I don't have to buy anything else the light will cost in the neighborhood of 83 bucks. (wire and thermal heat compound) that is for 48 watt light. If lens are required then that will add about 22 bucks more. That is a one off home made light. If I were doing this light commercialy, I would order the heatsink and light pcb as one unit. That is the pcb for the leds is printed right on the aluminum heatsink. Can get higher "bulb" density that way and no wires to solder. Still, the cost is quite high. Untill the cost of the leds drops (trust me they have already dropped a lot) the lights will just cost a lot. Notice, there is no labor cost in assembling the light. Home made. Can't tell you the effectiveness of the light till we hang it over the seedlings. I can tell you that just red and blue didn't work as well as halide light did. This light is designed to replace 1 100 w metal halide grow light. 1/2 the power and 1/3 the heat.

Well, for example, look at the prices on this page;


Likewise, I have some friends whose new kitchen has an optional little bank of red and blue LEDs configured as a grow light for which they paid $300.

Compared to the hobbyist projects I'm seeing here, these prices seem crazy to me. Am I mistaken?

Yes those prices are crazy but if the quality of those lights are really good then I think that should justify the price. Like these lights from http://growblu.com/shop-leds I know they are costly but the fact that they did a great job in my garden, I'm sure its worth it.

Becuase LEDs will be much simpler to manage space and heat-wise, it seems. I'm envisioning some kind of semi-enclosed, quasi-terrarium thing. Besides, LEDs would be a heck of a lot more fun. Also, any degree of success with this would probably benefit a lot of techno-forward gardening types besides myself.

I have a ton of UV Led's. Could I use just UV instead of a red and blue mix?

If you use UVB, you can boost antioxidant levels in red lettuce

no, you need both, red and blue/uv here is a good article about it http://www.sunmastergrowlamps.com/SunmLightandPlants.html


7 years ago

I have been looking at buy some LED Grow lights like this: http://satandcable.com/left-categories/hydroponics/led-grow-lights

What is the best ratio of Red to Blue lights? I'm a bit confused of the options? Do I actually need any white LED lights?

If you do a 90 W<a href="http://www.ledgrowlightsoutlet.com"> led grow light </a> yourself, one 1w 660nm Red led chip bulb will cost $1-1.5 with wholesell price, bulbs only for 90w cost you about between $100-140, plus led drive and your time, that is not worthy of DIY in economic sense. You can get a decent commercial 90w UFO at $180-200 with 3 year warranty , even cheaper at ebay (no idea about quality and warranty of cheap stuff, you need to balance your invest and risk), if LED chip price goes down, led grow light and other high powerful led products have a lot of room to reduce the price, although super bright led is insanely expensive currently, it will be lower in short future. <br />


9 years ago

You would need 8-10 CREE or Luxeon LED emitters, along with a current controller. A good place to get LEDs like this is at www.dealextreme.com. The current controller could be as simple as a few resistors. Do you really need 2000 lumens for a single rose???

um, incandescent (filament) would work better As you know, you can't heat anything to save your life using LEDs Incandescent lights are less efficient, but they produce a wider spectrum of light If you want, you can use CFLs (compact florescent) i'm not a biochemist, and this is only my opinion (oh, and incandescent light are WAAAY cheaper to buy) don't hate my if I'm wrong : )

5mm LEDs are used in over 95% of LED applications, making them the 'state' of the art. 1watt+ LED have made claims for over 10 years to be state-of-the-art, but are RARELY used, except in flashlights mostly, due to un-manageable thermal problems..short effective life and color shifting. Every major advance in LEDs in the last 10 years has been incorporated into 5mm tech...still LOOKS the same, but 5mms have changed the most, and are the most efficiently used, with the longest effective lifetime generally speaking for the vast majority of the lighting market...horticulture included...

Could you drop a link to the commercially available LED grow-lights, which are insanely expensive? It may help to give you a better answer. L

I saw some of the cheapest on Ebay in one of the stores I think it was ledgrowlights and www.solutionsgrowlights.com both places were about 30 to 35 a bulb. They seemed like high output bulb compared to the others I've seen


11 years ago

State of the art LEDs are about 100lm/W, so 2000lm is about 20W worth, or about 10 3W LEDs. Lumens are probably a bad measurement to use, since they're based on eye sensitivity to green, and plants would rather not have green (they just reflect it anyway, eh?) All Electronics currently has some good deals on clusters of 1W Leds...

Wow, that is a hell of a deal. And it's almost exactly what I need. My plan was to do mostly red, a few white (for aesthetics) a couple blue, and maybe one small UV - so I'd like to be able to mix and match colors. But that's so inexpensive I may have to try it regardless. Thanks for the info.

Oh - and you think it's more useful for me to think in Watts? I was fixated on lumens simply because that seems to be how these things are rated.

Thinking in Lumens is fine (although many of the smaller LEDs are rated in Candela, which is different.) Starting with SotA efficiency in lm/W give you a quick look at just how many LEDs you'll need.

Cree and Seoul are getting a better rep than K2 (but I don't know that I've seen colors other than white in small quantities.) Don't overlook lumiled's official distribution channels via http://www.lumileds.com (Future Electronics); prices there track the reductions of the manufacturer quicker than the "indie" shops; better selection of colors, worse ability to select particular bins...

I did a research project on the effects of LED lights on plant growth, and instead of using lumens, i used einsteins, which is basically a mole of photons.

Also, proceeding on the idea of making my own array of LEDs - I've found quite an array of seemingly smallish, indie shops selling these parts to the hobbyist, each offering a variety of products. Do I want to go straight for the Luxeon K2s, or can I make do with one of the Luxeon knockoffs I think I'm seeing? Which vendor do people seem to like best? How well do you think the super-simple current regulator plans here could be made to scale up to a bank of 10 K2s (or similar)?

LED's .... LED's are the up and coming efficient way of lighting! .... LED's have a long life, very efficient, and when powered correctly can produce very nice lighting. The LED is very sensitive to heat produced by electrical amperage. Most LED's use only 20 milliamps. LED grow lights or lighting panels can be found at ... www.UltraLEDLights.com ....... They custom build LED lighting and at an affordable price.

Hi, Here is a picture of our first grow test of our new Light, the LumiGrow 125. Test plant is Basil. 125 watts of power. Custom designed driver and emitter arrays. The main challenge has been thermal management. Good growth overall, but messed up the test. Turns out I am not a very good gardener. Next prototype is in construction and currently searching for a manufacturer to take this puppy retail!


10 years ago

We have been using LEDs with great success for the veg cycle. We are getting good growth and tight internodes using lights from WWW.solutionsgrowlights.com, at 1 1/2 blue bulb (409lumens 12watts) per sq ft. They are 110volt and screw into a regular light base. ($1.50ea base) We built a movable rack out of one x two's (wood) and mounted the bases to the rack. The bases are the type that mount on a the power cord. Home Depot has the bases. We are still experimenting with the grow cycle to get the proper amount of light and color mixture for optimum growth. But it looks promising at 25 lights on a 4x4 mixed 1/2 red and 1/2 blue. By the way you can get red bulbs in the 640 to 650 spectrum which greatly increase the usability of the light. In the mean time we are saving 80% on our electric bills. We tried building our own also and found that for $30 a bulb it was not worth the time and energy and expense. aloha

For some reason it didn't post right but that ° nonsense was supposed to be a degrees symbol.

I've got a few hours of research in to these LED grow lights and was also trying to figure out if it's more cost effective to build my own... ( I know it's the norm but I just can't justify $300 on a light :-/ )

Thanks to fellow growers (of slightly different motivation) I found an experiment here that references a few different wavelengths and ratio's of LEDs needed. Does anyone else out there have any experience with different wavelengths or can anyone recommend locations/LEDs to buy? So far the most useful search that I've found is here but the information that I'm pulling from seems kind of scarce and I hate to base my experiments off only one source's information. Thanks!

The LED’s recommended (I understand everything up the comma):
1 part 400: 25°, 40mW/sr
1 part 465: 18°, 11cd typ
4 part 630: 20°, 14cd typ
4 part 660: 30°, 4cd typ

i have some really good ideas on where to get LEDs and colours to use... i just need help putting together the most efficient driver... i'm aiming at 25w-50w per lamp with possibly different modules for vegetative and flower/fruiting stages. Maybe just one main module and an auxiliary module of reds for flower fruit... These will be at least 5 times more efficient than HID in my opinion.


11 years ago

i suggest finding the Grow Master patent as it will contain full details of how their device works inculding the color mix. as long as you are not selling it you can just copy the patent (and post your plans here for others to make too)


Reply 11 years ago

try this one:
6,921,182 it looks really trivial, they have a ratio of red+orange+blue leds.

First i want to say that, based on my research, lumens are useless when rating a light ability to stimulate plant growth. lumen is a measure of how bright a light source appears to the human eye. Our eyes are most sensitive to the middle (green) part of the visible spectrum and sensitivity decreases toward the blue and red. Plants on the other hand respond most to red and blue and little use for the middle part of the spectrum and is why chlorophyll appears green (green light is reflected not absorbed.
Unfortunately most artificial light sources are measured in the massaganistic lumen. Keep this in mind when thinkin about light source. This is one reason LED can be a great boon to agricultural/horticultural lighting. [properly driven]LEDs don't waste much energy on heat and have very narrow color spectrum thus you can use more energy in the colors plants use and omit those they tend to just reflect. That's my two cent for now.
Here's some links: (please also note that though these are red and blue grow lights, most of them do not contain optimized wavelengths for plant's production of chlorophylls)
LED Grow Lights
Groovy Grow
Grow With
Grow Master
This last link is the only LED lamp that is optimized but the LED mixture is patented and thus is most expensive. This is what i have been trying to emulate in my personal LED 'grow light' design but of course i could only make it for myself and could only market it if i made enough change to the colour mix as to not infringe on the patent (which i believe is possible while making the light more effective).