Advanced LED Development Thread.

[Guest]

This is interesting for sure. I recently saw some new stage lighting using LED'S that replace 250 watt stage lighting and use much less power. Of course my first thought upon seeing them was to wonder if they'd be any good for growing pot. They weren't very expensive either so maybe I'll have to pick one up and do an experiment.

 

[LED_experiments]

greg, there is only 1/2 of your 17W setup in this picture (for this plant). the other 1/2 of the setup is in your clone box?
really nice bushy plant. can't wait to see it in flowering with all 17W.

 

[Neptune]

Can we please see a grow where someone yeilded over 5 pounds of dried marijuana with LEDs only? ....So he yeilded 1g/w... good... but I am guessing he was probably using 20watts?

I will buy you a case of wine if you can show me the money.

and by the way, 2g/w+ is not unheardof with HPS lights in verticle/horizontal coliseum/rotary setups.

Really, I think it comes down to growroom design in the end.
Using LEDs will not be like using HPS/MH in a "flat garden" setup.
People who are trying to build high output flat boards may be working in the wrong direction...

What I envision is a totally different kind of growroom, where the lights are inside the canopy, on the sides of the canopy, and above. something like an electric acid LED tomatoe cage for each plant. LED light poles going every which way, and instead of reflective mylar on the sides of the growroom it will be walls of LEDs.

This is why working with tomatoes is a moot point, it just doesn't grow the same way physically. There is no way to prove you got something good when you yeild 5 lbs of tomatoes, that may translate into 6oz of bud... which is not interesting.

We already know that LEDs CAN grow plants.

 

[LED_experiments]

i think there is still so many undiscovered things about all that stuff, that we can't just say "this ratio is the best" and start producing huge setups. this is why we are doing experiments with smaller setups. the price also is a bit to high, to make huge test arrays.
btw, this lamp that greg uses is 2 year old model (knna, i think you know which one), there are only 628 and 470. we are moving forward, but at this time i have some issues, so i can't start my new project. it will be much more powerfull (better leds), changed spectrum (not sure if better, but hey, we said we are experimenting),... maybe we will make comparation between this 2 year old model and the new one (at the same power consumption), or maybe we'll put all of that together, using new model at the top and this one arround the plant (like knna is planing).
i also had an idea putting leds into the canopy, but i don't grow and i can't give someone setup like that if i'm not sure he can handle it. i also wanted to reduce the costs of the setup using dirrected 'wall' voltage, but again, i don't want to kill the grower with 300V+. if i was growing, i would run much more experiments.

 

[Guest]

i would very much like to see one of those 'lights into canopy' grow with extensive documentation.


p.s.: 5 ponds is 2.26 kg, so if we still had a 1g/1w ratio (i belive we can do better) we would need 2260 watts of leds...c'mon man, thats over 15k€, roughly calculated

 

[Hal]

Im going to use in my experiment a pure sativa (not comercial) strain, flowered in 45x45cm (1,5ft x 1,5 ft) and 2m tall space, surrounded by high power LEDs in all directions, wich will provide an average 400uE/m2 along all the plant (as reference, a 400w HPS in 1 sq meter provide an average 550uE/m2 at canopy).

Im thinking in the last days we are overcomplicating things when designing LED light for growing.

What do you think, guys?

Hey knna...

I am real interested in LED lighting for the cannabis garden, as are most who are familiar with it. I can't wait for it to become affordable...it will really make our gardening easier.

While I can't comment on the technical aspects of growing with LED, I have spotted something that I feel needs pointing out. Why would you choose to do your experiment with a sativa? In my opinion, this is a bad choice for two reasons: 1) they take so much longer to flower...you should be choosing a quick flowering indica dominant strain which will allow you to get your results in a quicker fashion, and 2) most people who are watching your experiments will be looking at one piece of data more than any other...yield. I don't think that indoor grown sativas would provide nearly the yield that most indicas show.

So, I guess I'm just confused why you're going with a sativa. Most growers cultivate the shorter flowering varieties, and it seems like choosing to do your experiment with a sativa is almost setting yourself up for what will seem like a less-than-successful experiment. I think that might be a sign of "overcomplicating" that you mentioned above.

That being offered...much thanks to you and all the others who are doing the hard, and expensive, work on getting this technology up and running.

 

[Ono Nadagin]

Before LED growing of MJ on any scale other than experimental becomes feasible, its inherent lack of canopy penetration must be addressed.

In my opinion the only way to overcome the light penetration issues associated with LED growing of MJ is to manipulate the shape of the plant canopy as well as the lights...

Having said that I have no idea if this would work but it is the only way I have invisioned you might be able to grow MJ with LEDs and have results that make LED MJ a legitimate option for MJ growers

Here is what I am thinking, a colosseum or a cage with vertically hung LEDs in the middle of the colosseum/cage or any system that suspended the clones around a vertically hung LED array... the use of a clone to flower grow style as well as a columar/low stretch strain imo would be necessary to keep the LEDs penetration as high as possible

http://www.icmag.com/ic/showthread.php?t=30081&highlight=coloseum

Here is a thread with pics of a cage grow to give you an idea of what I am talking

 

[LED_experiments]

i think coloseum is not such a good suggestion. most of leds have very narrow ligh, so it's not 360° source like hps.
scrog or sog is still the best technicque which can be used with leds (IMHO). you make a flat canopy using net and then the light doesn't have to penetrate deep in the canopy because it is all at the top, just a few inches from the led array.

i agree with hal about choosing a strain. i would also go for indica.

but we'll se. i'm almost 100% sure that the guy i'm selling my 70W system to will do a scrog or sog (or a combination of that). i think it is hard to cover a 1.5m plant with leds. if you want all the buds to get enough light, you maybe should make some 'condom shaped helmets' with leds inside and cover every single bud with them. i'm sure scrog is much less complicated.

but as they say in my country: why making something simple if it can be complicated?

 

[WhoDAT]

I think LEDs have a lot of potential and I have been enjoying the thread so far. It seems as the PWM and overdrive are the key to it. After reading i did some research about photosystems. I found, from wikipedia (http://en.wikipedia.org/wiki/Reaction_center ,under mechanism), that the photosynthetic reaction center processes photons at 10 picoseconds at a time. The way I understand it a perfectly efficient light system would emit light sustained for only as long as it takes for every receptor to receive one photon and then cease for 10 picoseconds until light is emitted again, correct? So wouldn't cycling the led on and off as fast as possible and overdriving it in accordance with the time it takes to dissipate energy be the most efficient way of lighting with LEDs? The reason I say to cycle the LEDs as fast as possible is because it was mentioned earlier in the thread that LEDs could cycle as fast as 100 nanoseconds. As fast as this may be 100 nanoseconds is substantially longer than 10 picoseconds, so pulsing the LEDs as fast as possible would deliver the photons in as short a time as possible and come as close to 10 picoseconds as LEDs possibly could.

 

[LED_experiments]

whodat, do you know what you are talking about? theory is something, practice is something totally different. it all sounds good in theory, but...
and how do you suggest to make a cheap pwm which will run at couple of GHz? i have been listening 'leds are too expensive' over and over again. do you think pwm would be cheaper? now there are maybe 100 growers who use leds (i said maybe). do you think there will be more of them if using couple times more expensive pwm hi-tech?
that thing with powering leds with a couple of GHz is for now IMO impossible or maybe possible with hudge budget. have you ever heared of 'diode capacity'? if not, ask yourself again if you will really get a pulsed light or will there be continuous few-times-owerdriven powered lamp if you consider diode capacity. they will die in maximum 1 week.

 

[WhoDAT]

whodat, do you know what you are talking about? theory is something, practice is something totally different. it all sounds good in theory, but...
and how do you suggest to make a cheap pwm which will run at couple of GHz? i have been listening 'leds are too expensive' over and over again. do you think pwm would be cheaper? now there are maybe 100 growers who use leds (i said maybe). do you think there will be more of them if using couple times more expensive pwm hi-tech?
that thing with powering leds with a couple of GHz is for now IMO impossible or maybe possible with hudge budget. have you ever heared of 'diode capacity'? if not, ask yourself again if you will really get a pulsed light or will there be continuous few-times-owerdriven powered lamp if you consider diode capacity. they will die in maximum 1 week.

Nothing I said in there was stated as a fact. It was for clarification. And as a matter of fact, no I don't know what I am talking about when it comes to this. Thats why im ASKING.

 

[WhoDAT]

-Myself, although i want to perform some experioments with PWM, im very skeptical about obtaining large photosynthetic gains by PWM respect to the energy usage, due to PWM beeing far less efficient than dimming in continous.

--Lumileds just has announced that their new high power device are capable of running at 1A without large efficiency losses against running at 250mA. If practical use confirm that, maybe PWM would be a reasonable way of increasing the light wattage use efficiency.
:

Can you explain what dimming in continous is and what gains it has over PWM.

What are the advantages of running LEDs at 1 amp as opposed to 250ma? Is it more intense light?

What is light wattage use efficiency?

Thanks for the response man, sorry about all the questions its an interesting topic.

 

[Ono Nadagin]

I wasnt talking of a single row of leds in a colosseum... you would mount them around a colum mount in the center

ehh... anyway I will leave it to you guys as Ill be sticking with Electronic HID untill I see others pull some fat colas... to this point I have seen nothing that makes me think I can get the same reults with LEDs that I get with HIDs.

So come on guys lets see what you can do... someone lets see some LED nugs

 

[daddyweez2000]

check this out what do you think?

 

[daddyweez2000]

http://cgi.ebay.com/48-LED-GROW-LIG...089217777QQcategoryZ43555QQrdZ1QQcmdZViewItem

 

[marx2k]

Heh.. I'd be a little careful of that seller.. 10 feedback, only one from someone who purchased something from this person (and it was some sunglasses).

Other things the user has purchased recently has been "EBOOK HOW TO WRITE GREAT CLASSIFIED ADS" and "Make Money from Misspelled Ebay Items!"

Other than that its been mostly books on hydroponics that the user has purchased. Proceed with caution.

 

[marx2k]

Check this out...
http://www.batmink.co.uk/led_wash.php
http://www.made-in-china.com/showro...China-LED-Fluorescent-Light-HFL-T8-W-H1-.html
http://www.atgelectronics.com/Products_detail.asp?ID=97

 

[sy9942]

I think LEDs have a lot of potential and I have been enjoying the thread so far. It seems as the PWM and overdrive are the key to it. After reading i did some research about photosystems. I found, from wikipedia (http://en.wikipedia.org/wiki/Reaction_center ,under mechanism), that the photosynthetic reaction center processes photons at 10 picoseconds at a time. The way I understand it a perfectly efficient light system would emit light sustained for only as long as it takes for every receptor to receive one photon and then cease for 10 picoseconds until light is emitted again, correct? So wouldn't cycling the led on and off as fast as possible and overdriving it in accordance with the time it takes to dissipate energy be the most efficient way of lighting with LEDs? The reason I say to cycle the LEDs as fast as possible is because it was mentioned earlier in the thread that LEDs could cycle as fast as 100 nanoseconds. As fast as this may be 100 nanoseconds is substantially longer than 10 picoseconds, so pulsing the LEDs as fast as possible would deliver the photons in as short a time as possible and come as close to 10 picoseconds as LEDs possibly could.

Disregarding that a picosecond is 1000 times smaller than a nanosecond (and the implicit problems with sub-nanosecond timing), the cited article states reaction center processing timescales for bacteria as being 10 picoseconds. Nothing is disclosed for green plants. And even if it takes 10 picoseconds for the reaction center to do its job, the electron acceptor may not be ready for another period of time, and it may also take a period of time for the acceptor to pass energy along to the next step in the production of NADP+. I'm still of the opinion that a 1.5 microseconds on, 148.5 microseconds off cycle is the best, but even sub-microsecond timing is difficult to achieve with any significant accuracy.

 

[daddyweez2000]

I was just remarking on the light and what is claimed

 

[flux]

These LEDs are not going to do well during flowering, there's just not enough intensity there (not yet at least), no way your going to get anything near the 500 microeinsteins using the LED, they will however suffice for vegetation.