LED Lab 2009

[knna]

"If i wrote my last post is because i feel most people want to follow your way. There is nothing wrong on it, but i think is impossible to improve and develop better LED systems without answering "how much" and "how many" questions."

"There's that word again! I disagree.
Difficult, yes. But, Impossible?
Dis li'l pig stumbles blindly along and find acorns everywhere!"

You are absolutelly right, Weezard. Its not impossible, but it depends on pure trial/error and for sure thats a very slow process.

Use 8 of these, n 12 of those,
Cross yer fingers an' yer toes
Addem to da red an' blue,
Makes yer buds swell-up times 2!

If you use known brand LED models, and you know what bins are they, there is no problem calculating how much light they are emitting.

An average uE/m2 (or uE/sq ft) is very similar that currently stating watts/m2 (or watts/sq ft) of HID lamps. We can work with relatively wide error margins, as 15%, and still get good references.

Im thinking of doing a table with the most used LEDs on the market, with how much uE emits each bin at the typical currents used (350 and 700mA). So any LED grower using them may know how many uE he is using. Im currently too busy to doing it, but i hope to have time for it soon or later.

Aditionally, this info would allow people interested on doing DIY LED arrays to calculate the cost of each uE for the different options avalaible, aswell to decide at what current to drive them.

I read here that it's impossible to grow good buds with just red and blue.

I disagree that too. It is perfectly possible to grow good buds using just red and blue.

What i say is they produce less per uE than a more complete spectrum, so you need more uE of light for each gram harvested. So its less efficient than other combinations.

But maybe very little of other wl are required to increase spectral efficacy. How much of any other wl is what im trying to determine.

So let me to explain better my purpose with the post #45:

Split the talk about each other LED setups from the "develop the perfect LED grow light". There is nothing wrong on using LEDs to grow pot and share it with other growers. But researching for improvements needs to a more obliguing conditions.

 

[knna]

Just out of interest, what currently available LEDs do you think offer the best performance potential for the cannabis grower?

Red leds: all top bin of Cree, Lumileds and Seoul SemiConductor (SSC) P4 are very similar. Its the S for Cree XR-C, S for Luxeon K2 and R for the SSC P4. The SSC seems abit better than the other, but for little margin. Larger differences are made by the Vf of the lot you get.

For LEDs, light output is directly linked to current. But Vf only affects watts consuption, and not light emission. So given two LEDs of the same bin, they are going to emit aprox the same light at the same current, but that with the lower Vf is going to be more efficient, as its going to burn less watts for the same light output (remember that Power (watts)=Vf (forward voltage)*If (forward current)).

So those bins of those manufacturers are very similar on efficiency, although the K2 consumes more watts (higher average Vf) but emits more light too.

For any of them, max advisable running current is 700mA, although all supports higher currents. Over that point, lasting of LEDs is compromised, and its very probable a fast output degradation with time (at the point of lasting less than a good fluorescent).

Other option is the Osram models, either the Golden Dragon, the Platinun Dragon or the Argus (its a lensed Platinun Dragon, with 160, 40 and 20º lens). Problem with Osram is they not serve individual bins, but group of bins. For example, the Argus 160º is avalaible on the group of HZ to JZ bins (including the JX and JY bins between them). Of those 4 bins, the lower (HZ) is below the top bins cited earlier. JX is very similar to them, JY is an excelent LED, better that any other, and getting a JZ would be a dream, with radiant efficiency near 50%. But you dont know what bin you are going to get, its a kind of lottery (and you may suspect that larger clients gets the better bins).

Ive been told by SSC that their S bin is going to be avalaible before this years ends. Im waiting its avalaibily to order a large batch and do LED grow lamps with it, as i think that with its efficiency we could strongly outperform HPSs (it would emit way more uE/w than them at medium power, letting appart spectral efficacy). I hope to have found very good working spectral distributions at that time.

CoolWhite and blues: currently, the EZBright chip from Cree is the clear winner. Luxeons are very far for reaching its efficiency. Not only Cree uses it, but have it licensed and sell it to SSC, Osram (recent agreement, i believe Osram still arnt manufacturing using it, or just starting to do) and some others, as Edison Opto (Edixeons) or Kingbright.

Cree XR-E bin R2 is the top im aware of currently. Still produced in small amounts, it has a premium price. While the Q5 is currently avalaible on large volumes, and its efficient enough to avoid paying much more for the R2.

SSC P4 U bins, especially U2 bin are very similar.

The Royal Blue group 15 is the top of the Cree XR-E currently, although recently they have listed the bin Group 16 (so probably its going to be released soon).

 

[SouthernGuerila]

I've been checking the prices of LEDs on mouser.com

LEDs aren't cheap!

How many would you need to equal a 400watt hps/mh 600watt, and 1000watt grow light?

 

[indifferent]

Very interesting info knna, I'm hoping my LEDs aren't too bad in terms of binning! I hope within a couple of years we will see some big strides made in the manufacturing of LEDs. I'm, not much interested in the multi-die LEDs that manufacturers seem to be focusing on, let's hope they start developing LEDs specifically for horticultural applications. LEDengin seem to be the only ones ho are doing this, the availablility of Edixeon 660n and 740nm LED may suggest they are also targetting the horticultural sector with those products.

What do you think of the new Avago Moonstone coolwhites? They are perhaps a cheaper alternative to using CREE XR-Es.

The Golden Dragons I;ve seen for sale have rather low quoted outputs compared to the CREEs, I reckon, like you say, it's down to binning issues.

 

[knna]

I've been checking the prices of LEDs on mouser.com

LEDs aren't cheap!

How many would you need to equal a 400watt hps/mh 600watt, and 1000watt grow light?

Yep, LEDs arnt cheap, and less at mouser and similar. But its due the abbusive profit margins they charge to the product. Still without being cheap, prices on bulk quantities are way less expensive.

Im aware of this problem for many people wanting to go to LED growing, and thats why im thinking on selling LEDs and other components to the people wanting to build their own LED lamps. Its possible to sell them way cheaper still having a profit margin (but not an abussive one). There is a problem for many business to buy LEDs to resell them, due how fast they get obsolete (similar to computers): if you cant sell them fast, you have large losses (stocked product loss value quickly).

I believe is still too soon to replace high wattage HID lamps with LEDs. Its due the non linear price of HID lamps: a 400w HPS cost just a little more than a 250w (not 4/2.5 more) and so on. While cost per watt of LEDs is linear.

This does replacing low wattage HID lamps and CFLs way more profitable than replacing big lamps currently. As smaller the grow space, more profitable the LED alternative. Probably, on microgrows currently is cheaper to install good LEDs than other types of lighting, with the added benefit of low heat (with aluminium walls, half of the heat goes directly out the cab).

But for larger grows, i think currently the most wise application of LEDs is to add side lighting of mostly red leds (or red plus some blue), to improve overall spectrum, enhance production of bottom areas or allow to grow taller plants.

Aswell, for long and narrow spaces, combination of fluorescent tubes and red leds is an excelent alternative

 

[knna]

Very interesting info knna, I'm hoping my LEDs aren't too bad in terms of binning! I hope within a couple of years we will see some big strides made in the manufacturing of LEDs. I'm, not much interested in the multi-die LEDs that manufacturers seem to be focusing on, let's hope they start developing LEDs specifically for horticultural applications. LEDengin seem to be the only ones ho are doing this, the availablility of Edixeon 660n and 740nm LED may suggest they are also targetting the horticultural sector with those products.

Main manufacturers are absolutelly into the white LEDs war. And they mostly lead technology advancements. So i wouldnt expect large improvements on the LED we want to use. At least, until they have reached the 150 lm/w gate and can concentrate on other targets. Their priority is to introduce LED lighting in the residential and comercial sector.

Multidie LEDs or larger chip leds (as the Osram Diamond) are the key to get enough light to replace incans and CFLs. For us, those LEDs are a PITA, both about drive and thermal management.

But 1-3watt LEDs are going to continue.

Other interesting trend is that of AC driven LEDs. The SSC Ariche works directly from the main AC and its improving its perfomance very fast. Currently, its only avalaible on whites, but its expected the release of monocolor versions soon. Red ones may be very interesting for people wanting to avoid messing with drivers. We will see how efficient they are.

What do you think of the new Avago Moonstone coolwhites? They are perhaps a cheaper alternative to using CREE XR-Es.

Avago (formerly Agilent and HP) is a great brand. They develop the Luxeons, but after sell it to Phillips, they have started again from 0. They are doing very well, but they are still a step behind Crees. Interesting for the next future, i think.

But of course, all depends of at what price you can get each model of each brand. On large amounts, price differences between brands are way smaller than on retail.

The Golden Dragons I;ve seen for sale have rather low quoted outputs compared to the CREEs, I reckon, like you say, it's down to binning issues.

Yep, thats the trend. Check at Digikey, and note carefully the group bin. Price differences between the low bins on offer and top bins is not so large:

High power LEDs avalaible today: Golden Dragon group bin GZ-HZ (very low) is from 2.93$, while the HY-JY group (higher avalaible) is priced from 3$ (that drops to 2.5$ buying 10 and 2$ buying 100, which is not an excessive amount).

The Platinum Dragon (700mA) bin JY-KY is priced from 3.3$ (1 unit) to 2.2$ (100), up to 1.46$(2400). The lower JY is a little low (similar to a K bin Cree), but all the rest are good.

 

[indifferent]

Yes, I agree with all those thoughts. The economics are most certainly different when buying in quantity. For over a year I tried to find a stockist of the edison 660nm and 740nm LEDs and could only but them in a tray of 000, and then they were under 2 dollars a piece.

I totally agree that we should be using the best binnings for the peak efficiency, the only reason I am not currently is problems with small-scale economics.

 

[jaysplifferton]

I'm surprised noone has made a real nice LED prefab system yet. Even though it's expensive for consumers like us to experiment around, 100,000 of R&D is nothing for the BIG commercial firms considering how much they would save in power costs from HID and AC combined. Plus, if they could mass-produce and market such a product they could mark it up a fair amount and it would still be way cheaper than most HID products. My advice would be to find some enterpreneurs (sp) in Cali or Canada and get backed by VC from somebody in the know (IE: a rich hobbyist, a dispensary owner, etc).

KNNA, the prospect of everybody collaborating on such a project is a noble goal but the coordination of such an effort seems that it would just be ridiculously hard if it wasn't centralized. Nevertheless, this is a great thread, even though I don't get all of the technical stuff, I think it's great that there are so many scientific minds on these boards.

Another suggestion: You should conduct all the trials on the same genes of plant; skunk or afghan, something "universal" that's a part of lots of other genetics. That way, it'll be easier to fine-tune it to other plants later. (or so is my thinking... idk)

 

[MrNebulized]

KNNA if you are serious about building these kits I'd like to be your first customer. I would be more then happy to lend my garden to experimental growth using LED lights.

Indifferent while I understand your obsession with learning and researching these lights some people are looking for a plug in system. For you to rob a grower of info such as him choose lights and etc. You are holding the community back from more growers using these LED systems. My question is did you know exactly how a HPS or MH HID light worked when you first used them? Did you know why they worked better then Fluoros or incandescent bulbs?

You have a chance to push forward LEDs by letting some of us pick your brain, don't hold back. While I think they should read the thread before they ask you any questions, we are all but humble farmers who should be happy to inform the grander population. I have been researching these lights for about 4 months. I still feel clueless, every time I think I have enough info to build a light it seems more information pops up and a different angle is used. For instance, I was measuring for the most part using watts as a measurement, which I realize now seems more then a little foolish. My research will continue.

 

[knna]

jaysplifferton, im already running that project on a spanish forum. 25 growers participating. The key is to have standarized and know LED's output.

Beyond that, what each grower tries is at his own. We all use same LEDs, but there are several differents color mixes on trial. Most of then are Royal Blue (450-455nm), Coolwhite (WG Cree bin, 5700-6500K) and red (635-645nm) mixes. I did tables of RGB distribution of each possible (reasonable) combination. I built too aproximation of the average SPD of each.

Some are going to try more reddish spectrums, some more bluesish, some with more white. Most are going to use similar light densities, as we choosed to start with 300 uE/m2 for the top 30cm (1ft) of plants, while some are going to try light densities over that a HPS setup produces.

There are some setups that are identical, but most vary. For sure it would be better if we all use same strain, but i think that multistrain tests may provide very good insights aswell. At least for the starting test.

With this test, i hope in just one or as max, two runs, we are going to know what a LED balanced spectrum are able to yield, on a uE basis. What differences does the differents spectrums, and select those that works better to further improvement and more detailed testing.

Some test are going to be performed just by adding supplementary LED modules to already working HPS and foros setups, in order to check whats the yield improvement.

One advantage of the modular systems we are going to use is any grower may change the position of them, and easily change how light is distributed along the grow. One of the main goal of this first trial is to determine how much light deliver from top and how much from sides and intracanopy, and if giving placing side leds in angle instead of perpendicular to plants makes a difference.

Once these basic questions have good enough answers, we will talk again about next testing priorities. But always each grower decides what he want to try next. We meet at chat and talk about it, and often is not difficult to agree on how to share the task. Sometimes two growers want to test same thing, and really its not a problem, but a plus, as far as there are enough growers trying other things.

Its all about how many growers participate on the research.

Aswell, several styles of growing are going to be used. Organic, soil, hydro, free grow, LST, scrog, sog (some of them, vertical sog). As more info from different styles of growing, the better. As we are trying to get answer valid for all growers and styles.

Of course, fine tuning of the systems will require to use same strain on same setup and change one thing at a time. But those test are going to answer the generic questions most growers, not interested on research, are wanting to know: how much light i need to use per sq meter (or sq ft)? what color distribution? how i distribute the light along the grow space? how much yield may i expect from a given setup?

The systems works with adjustable constant current drivers. Most of the growers are using a different channel for red leds and blue and white ones, so just with a potenciometer, we can change both light intensity and spectrum.

On the same forum, there has been a group buy of UFO and similar lamps. Although they dont know the light output of them, at least they know the spectrum. They are starting journals with them right now, too, so we are going to be able to compare those adjustable systems with commercially avalaible ones. There is a new LED growing subforum created to hold all these journals.

Im going to wait until this trial is in progress (i hope at the start of March) before offering similar kits. I want to see what problems have normal growers mounting them (if any), and think if it would be better to offer them already mounted.

Attending to LED industry expectations, for the next year is going to be possible to build LED lamps of top efficiency for 5$/watt (I mean selling price of plug and play lamps). My purpose is we know enough to many people jumping into LED growing next year, with confidence and realistic expectations of what they may yield, without the need to know anything about lighting. I believe the 5$/watt of top leds is the treshold from LED growing becoming profitable for most growers.

 

[sonnyblanko]

Hey guys,
Sorry been away. Wholesale or bulk will depend on how many people are interested as I buy the materials wholesale in bulk. I really want more people to try these out because the only way to see if these work if it's affordable and we will also be able to figure out where changes will need to be made. The results are promising thus far but the goal is to have the most efficient to help growers reduce their energy consumption foot print. If anyone is interested PM me an email address (preferably hushmail)as I will send photos and so on.

If anyone has any other wavelength grow experience I would be interested in your findings or results.

Thanks,
Mike

 

[eenymeenygreeny]

great thread guys keep up the good work. mahalo

 

[N707]

Here they come!!!

Here they come!!!

These just hit the market and are well on their way to being affordable.
These my friends will be the ones that helpbring LED's into the grow game.
They are still rediculously expensive but should be are plummeting in price due to new production methods. The 100 watter was over $400 less tahn 6 months ago.
Remember guys/gals, you can stick 100 1 watt led's in a fixture, and it still will not produce anywhere near the intensity that a single 100 wat one will. In essence, I hopw that you realise that you can pack as many leds as you want into a fixture, but you cannot increase the viability of photons by using bulk rather than qaulity. The true high power leds are unrivaled in grow capabilities.
http://stores.ebay.com/Shop4leds_10...0QQcolZ4QQdirZ1QQfsubZ342523015QQftidZ2QQtZkm

Imagine a micro grow with a couple or even a few of the 50 watters!!! I just got the pleasure to see a reef fixture using the 35 watt ones, HOLY CRAP!!!! 5 35 watters put out far more visible light than a t-5 fixture with about 250 watts and very little heat. I would wager to bet (based on the growth of the stony corals uncer the leds) that the 35's put out far more par and 1/2 the heat of wat for wat t-5 fixtures. I say this based on seeing and playing with the fixture for a couple hours.

Has anyone else here had the pleasure of working with or seeing the new high powers yet? Even the little 10 watters are rediculous in way or percieved brightness, they are so tiny.... the 50 wat one I saw had the footprint of a postage stamp and was about 5 mm in height.

 

[Weezard]

Good stuff!

Good stuff!

Thanks for the link, N707.

Now I wait for the 100W. Reds and blues.

Have had no luck trying to grow with 5W white leds.

Any chance that some of you folks are willing to blow your tax return on a few of these pups for a test grow?

Aloha,
Weeze

 

[N707]

Thanks for the link, N707.

Now I wait for the 100W. Reds and blues.

Have had no luck trying to grow with 5W white leds.

Any chance that some of you folks are willing to blow your tax return on a few of these pups for a test grow?

Aloha,
Weeze

The 5 watters didin't work for ya?
They may have been bunk led's.
Try cre or the p4's (both 5 watt star leds), a cluster of like 10 should be bangin for a little micro grow.

 

[Thundurkel]

Wow!! Interesting, so 100w of LED is 7000 lumens huh? thats nice!! I can't wait till it's in my price range cuz that is the way for micro growing holly shit!!!

 

[Weezard]

The key word here is "should"

The key word here is "should"

The 5 watters didin't work for ya?
They may have been bunk led's.
Try cre or the p4's (both 5 watt star leds), a cluster of like 10 should be bangin for a little micro grow.

5 Watt reds n blues work a treat.
But seedlings under 5W. white leds stretched and died.

Have you had any luck growing with the white leds alone?

Weeze

 

[penis]

im building a light as well, made from 15w + 10w, but it wont be ready for a while :\

 

[indifferent]

CREE and P4 LEDs aren't 5W, they are generally around 2.2 to 2.7W, you can work it out for yourself, say a CREE LED running at 700mA has a v drop of 3.7, 3.7V x 0.7A = 2.59W.

Those high wattage LEDs are not what we want for growing at all, they are just a load of LED dies mounted together in an array and as knna has touched on before, this is the worst way to arrange the dies if they are to be used for growing, you want individual die LEDs so you can space them out to give a good light distribution. Only 7000 lumens from 100W is shockingly inefficient, only 70 lumens per watt, this is about half what the best LEDs, fluorescents and discharge lighting is achieving these days so all in all, those 100W LEDs are about as bad a choice for growing as you could make.

 

[Hydrobandits]

jaysplifferton, im already running that project on a spanish forum. 25 growers participating. The key is to have standarized and know LED's output.

Beyond that, what each grower tries is at his own. We all use same LEDs, but there are several differents color mixes on trial. Most of then are Royal Blue (450-455nm), Coolwhite (WG Cree bin, 5700-6500K) and red (635-645nm) mixes. I did tables of RGB distribution of each possible (reasonable) combination. I built too aproximation of the average SPD of each.

Some are going to try more reddish spectrums, some more bluesish, some with more white. Most are going to use similar light densities, as we choosed to start with 300 uE/m2 for the top 30cm (1ft) of plants, while some are going to try light densities over that a HPS setup produces.

There are some setups that are identical, but most vary. For sure it would be better if we all use same strain, but i think that multistrain tests may provide very good insights aswell. At least for the starting test.

With this test, i hope in just one or as max, two runs, we are going to know what a LED balanced spectrum are able to yield, on a uE basis. What differences does the differents spectrums, and select those that works better to further improvement and more detailed testing.

Some test are going to be performed just by adding supplementary LED modules to already working HPS and foros setups, in order to check whats the yield improvement.

One advantage of the modular systems we are going to use is any grower may change the position of them, and easily change how light is distributed along the grow. One of the main goal of this first trial is to determine how much light deliver from top and how much from sides and intracanopy, and if giving placing side leds in angle instead of perpendicular to plants makes a difference.

Once these basic questions have good enough answers, we will talk again about next testing priorities. But always each grower decides what he want to try next. We meet at chat and talk about it, and often is not difficult to agree on how to share the task. Sometimes two growers want to test same thing, and really its not a problem, but a plus, as far as there are enough growers trying other things.

Its all about how many growers participate on the research.

Aswell, several styles of growing are going to be used. Organic, soil, hydro, free grow, LST, scrog, sog (some of them, vertical sog). As more info from different styles of growing, the better. As we are trying to get answer valid for all growers and styles.

Of course, fine tuning of the systems will require to use same strain on same setup and change one thing at a time. But those test are going to answer the generic questions most growers, not interested on research, are wanting to know: how much light i need to use per sq meter (or sq ft)? what color distribution? how i distribute the light along the grow space? how much yield may i expect from a given setup?

The systems works with adjustable constant current drivers. Most of the growers are using a different channel for red leds and blue and white ones, so just with a potenciometer, we can change both light intensity and spectrum.

On the same forum, there has been a group buy of UFO and similar lamps. Although they dont know the light output of them, at least they know the spectrum. They are starting journals with them right now, too, so we are going to be able to compare those adjustable systems with commercially avalaible ones. There is a new LED growing subforum created to hold all these journals.

Im going to wait until this trial is in progress (i hope at the start of March) before offering similar kits. I want to see what problems have normal growers mounting them (if any), and think if it would be better to offer them already mounted.

Attending to LED industry expectations, for the next year is going to be possible to build LED lamps of top efficiency for 5$/watt (I mean selling price of plug and play lamps). My purpose is we know enough to many people jumping into LED growing next year, with confidence and realistic expectations of what they may yield, without the need to know anything about lighting. I believe the 5$/watt of top leds is the treshold from LED growing becoming profitable for most growers.

Where is this trial taking place? on ic mag?

and I would be very willing to buy a kits or even near whole sale price leds for an experiment.

Very nice thread to all contributors