Designing LED grow panel

[WeayLay]

Anyone that could be any help in my LED project, feel free to chime in.

This light panel will be put in a single Rubbermaid grow box. I plan to run Luxeon Rebel LED series, in a Saber Tri-Star chip. Essentially, there's three Rebel LED's on one chip, having a 10V forward voltage, and the ability to run 700 mA. I want to run 3 reds, and 3 blues. The reason why I want to use equal colour amounts is because it makes it easier to wire for drivers, and the red colour LED's are more efficient than the blue LED's.

I've been doing a lot of research lately, and I honestly just wanted the straight up opinions of people who already have been through the DIY route. I found an amazing website for LED growing: ledgrow.eu
I plan on using active cooling; a fan blowing on the heat sink. I would use two of those drivers, and have a AC to 24V DC power supply with at LEAST 5 amps to power both drivers. The LED's will be wired in three parallel circuits, giving each LED unit 700mA. The final setup should be around 35 watts for the lights only, and more in total due to inefficiencies and fans. What do you guys think?

[Weezard]

Efficacy aside, let me suggest that you add another red module.

Royal blue is a high energy frequency.
Red is quite low-energy, relatively.
Blue triggers photo-tropism, etc. but red light is the "workhorse" here.
If you add a red module, and you intend to flower with this light, let me suggest using 660nm. deep red.
They made little difference in veg., so I use the 627nm lights there.
In flower, however, the difference was notable.
Fatter, harder colas.

Only other advice that comes to mind is, do not cheap out on the heat-sink.
You need to move heat away from the emitters as rapidly as possible.
As the junction heat rises, efficiency fall off a cliff, and life expectancy takes a dive as well.
Protect your investment and get the most bang for your buck, brah.
Honkin' big heat-sink, with a muffin fan on a thermostat is a great investment at this point.

Check my albums for ideas.
http://boards.cannabis.com/members/weezard-albums-let-there-light-picture269761-under-one-roof.html (http://boards.cannabis.com/members/weezard-albums-let-there-light-picture269761-under-one-roof.html)
Learn from my mistakes :)

When you get it built, I'll have a link to a thread called.

Discussion of the pragmatic application of LED's

(hard to pass links here, so ya might wanna google dat)

Aloha,

Wee'zard

[WeayLay]

Hey Weezard, primarily wanted your advice first. The heat sink that I have chosen out is very good quality, and has a 1.7 degree Celsius/Watt thermal conductivity. It is further decreased with active fan cooling. I've been looking at the Cree XT-E Royal blue in a tri-star setup, but I can't find any that have been reflowed onto a Tri Star MCPCB (Metal Core PC Board - better for thermal conductivity). So you suggest Deep red instead of red, I guess that would be the smart idea to do, because this panel is going to be my one panel for flowering and veg growth. But honestly, I'm going to veg for a very short time, so I think putting 4 reds and 2 blue would be more beneficial.
You think the drivers would be happy with having one red and two blues to power?
You know of any place that will reflow LED's?

[Weezard]

Hey Weezard, primarily wanted your advice first. The heat sink that I have chosen out is very good quality, and has a 1.7 degree Celsius/Watt thermal conductivity. It is further decreased with active fan cooling. I've been looking at the Cree XT-E Royal blue in a tri-star setup, but I can't find any that have been reflowed onto a Tri Star MCPCB (Metal Core PC Board - better for thermal conductivity). So you suggest Deep red instead of red, I guess that would be the smart idea to do, because this panel is going to be my one panel for flowering and veg growth. But honestly, I'm going to veg for a very short time, so I think putting 4 reds and 2 blue would be more beneficial.
You think the drivers would be happy with having one red and two blues to power?
You know of any place that will reflow LED's?

The beauty of constant current supplies is, you can mix leds, wire them in series, and set the current that you want.
The forward voltage drops will sort themselves. :)

With skill, you can reflow at home with a mini-torch.

A good soldering iron is safer though.
You just need lots of magnification, and a very steady hand.
Bead the pads and, set a clamp, (wooden clothespins with shaved tips work well), but a large hemostat will also sink heat.
(I used small aluminum clamps that are sold as transistor leads, heat-sink clips, 'cause I had them on-hand.)
Then clean/tin a decent iron and dance around the chip.

I know that it sounds easier than it actually is. :)
I can't do it myself, anymore.
Give it a try, maybe you can.

If not, hmmm.:detective1:

If you know a jeweler, try bringing the pads, chips, and solder with you and ask what he'd charge.

I'd give you a link to a tutorial on home re-flow soldering, but it's on a different board, and they don't allow links to better boards.:(
Sorry, I no pay da bills here, so I no make da rules.

Aloha,
Weeze

[WeayLay]

I'm definitely sticking with a Tri-star mounting system, but I've been looking at reflowing LED's with heatguns, oven tops, electronic skillets, and even soldering irons. I have the heatsink, drivers, and power supply all figured out, but I'm definitely sticking with 2 Tri-star XT-E Royal Blue's for my blue's, and then I'll get 4 tri-star Lumileds Deep Red's for my red range. I've been checking out your albums, but I think I need more people with experience to give me advice with the LED's.

[ChrisGGG]

I'm doing a very similar setup. I found a very cheap constant current driver 75v @ 680nm. I plan on reflowing my own tristar chips w/ Epistar 460nm and 625nm. I know 625nm is not ideal but the price was right. How do you plan on mounting the tristar? My plan is dap of thermal paste with a tap/dye.

Weezer do you have a link for the comparison between 660nm and 627nm?

[WeayLay]

I know a place where I can get them already reflowed on a tri-star MCPCB. Google Luxeon Tri stars.:thumbsup:
A good reference grow is the first link when you google "growing cannabis with LED".
I decided to half the power of LED's and add a 30W+ CFL. The lights+ballast for three tri-stars and a 2.1A driver is a hundred bucks, so I thought I'd keep it simple. Or... I've been thinking lately, in order to boost up my efficiency should I run six parallel Tri-stars with the 2.1A driver? I'm planning on getting a 24V 5A+ power supply for the driver.
Chris, if money is a problem, just get less far reds. I've been looking into grows and far red seems to help flowering especially.
What I posted on 'Calling out to Weezard for LED advice':
in order to lower costs, I'm only going with half the LED's for the single rubbermaid grow box. Should be around 22w, but I'll probably put in a decent powered CFL to make up for the rest of the spectrum. I'm pretty sure 22W will do fine, Pinstripe over at ICmag got 38g from 22W in a PC box (Anyone that respects any grow should check that out, crazy micro grow, probably the best documented one), and I'm going to be using his technique (LST) and further improving it with a SCROG.
The LED's and the drivers are going to be 100 shipped. I've been looking at Phillips Lumileds Rebel LED's, specifically three of them, mounted on a MCPCB (called Tri-star). I'm going to have 1 royal blue and two far reds on each tri-star, what do you guys think?

[sideshowmel]

The beauty of constant current supplies is, you can mix leds, wire them in series, and set the current that you want.
The forward voltage drops will sort themselves. :)

That's a really good point. I don't know why I didn't realize that. They have to divide the potential based on their resistance at that current.

I think you just saved me one of the four power packs on my next experiment.

[Weezard]

Happy to help.

You can save a tad more if you wish.:)

The UV is not necessary.

Cannabis will grow, flower, and become extremely potent with 0 UV.
It does not create resin glands to "protect it's self" from UV.
Quite the contrary!

Outside, the UVc degrades the resin glands almost as fast as they are produced.
Fries them down to tiny black dots that blow away in the wind.
I have watched it happen with my Eyeclops.
283664

And the far red, 730nm., is counterproductive unless it is applied at the appropriate times.
It's function is phytochrome conversion.
Far red can be used to hasten flowering when applied for an hour at "light's out".

Just an aside on the CFL addition.
Nothin' wrong with that, it won't hurt.
Did it myself early-on when I thought there might be "missing wavelengths".
Turned out it was the strain.
She was a photon hog.
Was not about spectrum, was about raw power.

I am breeding her with 2 strains that love the leds hoping to change her greedy mind.
Working?
Too soon to tell, I'll keep y'all posted on any progress.

Aloha,
Weezard

[sideshowmel]

The UV is not necessary.

Cannabis will grow, flower, and become extremely potent with 0 UV.


Necessity aside for a moment, do you think that it's even helpful?

I thought some might be, as it was supposed to stimulate repair, help regulate morphology, and maybe boost disease resistance. One thing I read (plant photobiology
E. P. SPALDING & K. M. FOLTA) said;

[UV-b] induces DNA damage repair responses as well as a suite of
unrelated outcomes. These range from increases in cyto-
plasmic Ca2+ (Frohnmeyer et al. 1999), induction or patho-
genesis-related transcripts (Green & Fluhr 1995),
hypocotyl growth inhibition (Shinkle et al. 2004), and sto-
matal opening (Eisinger et al. 2000; Eisinger, Bogomolni &
Taiz 2003).



And the far red, 730nm., is counterproductive unless it is applied at the appropriate times.
It's function is phytochrome conversion.
Far red can be used to hasten flowering when applied for an hour at "light's out".


As I understood it some far red helps to maintain the balance of the two conformations of one of the phytochromes (can't remember which, will look), which makes them healthier. Of course healthier is a biased word, and the authors of those papers were talking about general food crops.

From what you're saying the plant would be more vigorous if this balance was not maintained for most of the time. Frankly it seems unusual to me that an organism evolved in a particular environment would thrive just as well in an environment missing one of it's constituents, but then I had that reaction to the idea of a 24/0 light cycle which did turn out to work.

I'd really like to hear if anyone has done more specific experiments on these questions. Should do them myself and stop complaining I suppose.

[sideshowmel]

when I thought there might be "missing wavelengths".

How many wavelengths would you say were absolutely necessary, and how many for the ideal?

What do you think about choosing emitter chemistry based on the highest efficiency available in a roughly similar wavelength versus a less efficient type which is bang-on the photochemicals' peaks ? Does 10-15nm of margin seem sensible to you?

[Weezard]

I get by with 2 now.
Red n blue will do.
Used to try a lot of things
Until I got a clue.

(Sorry, sometimes I jus' gotta.)

But it's a bit less simple in practice.
For instance, I've found a strain that does poorly in veg. with 660nm red, but seems fine with 625nm?!

It also appears that R:B ratio is a player, strain-wise.
Can't tell you much about flowering stage though.
Doing all my flowering outdoors now.
Not that the leds didn't cut it.
They flowered impressively, but the sun is free here, and is almost 12:12 year round. :D
I jus' filter out the UV and watch 'em frost up with trichs.
Now I don't need to keep two rooms and chase light leaks etc.
(Way to make a lazy guy happy) :)

And, yes, 10 - 15nm margin is no problem.
Go for efficacy/wavelength.
(The plants grade on a curve) :)

"I'd really like to hear if anyone has done more specific experiments on these questions. Should do them myself and stop complaining I suppose."

Gobs of it, ad nauseum.
Google "Martian method" It was a great ride with no real destination. IMO
Aloha,
Weeze

[sideshowmel]

The martian thing looks interesting, but not really what I had in mind.

The sort of experiment I was thinking of would involve a set of boxes, each big enough for a small plant, and designed to consume the same electrical watts but produce different light spectra. Maybe have some with 1 red 1 blue, some with 2 red 1 blue, some with added far red and far blue, some with broad spectrum white and so on. Control for temperature, atmosphere, nutrients and growth medium, and then calculate yield per watt hour for each. Of course the comparative efficiency of different LED wavelengths will probably continue to develop over time, so maybe calculate yield per photon flux aswell.

Maybe test two spectra at a time with several under each spectrum. Wouldn't take many seasons to get results with a reasonable likelihood of statistical significance. Could probably do it in a few years if I took a bit of overtime to increase the tinkering budget.

I find the whole subject fascinating, but I haven't had much luck finding data collected under controlled conditions, so it's a good excuse to solder things.

[Weezard]

Pleased to meet you Ssm,

Yer my kine guy.
An' you speak solder?

Box 'em, light 'em, grow 'em, post 'em.
I'd be tickled to assist in such a worthwhile experiment.
My brainstorms may lack some donder n blitzen in my dotage, but I still get the occasional widely scattered showers.

Did you dig for the NASA research?
It's a valid, if linear, pursuit and will save some "legwork".

Gonna be difficult to collaborate on this forum though.
No visitor messages, or PMs.
Might do to move it to a less restrictive board.

The children here tried to commit various felonies when they had PMs,
so management punished everyone by shutting down most of the useful features.
Cdot is kind of a kiddie pool now.
Most of the serious growers have bailed, and moved on to less restrictive places.

I hang on to see if any serious talent stumbles in the door.
This was once a first class board, so we still find intelligent folks that google in to the archives.
(IMO C-dot has the most comprehensive cannabis archives on the web.)

But, the "no freedoms" and the number juvenile postings are starting to wear me down lately.
Aloha,
Weeze