Discussion of the pragmatic application of LED's

[DraGonReBorn]

Temp Issues

Temp Issues

I'm a lurker, but I love the thread and hope I can contribute some small input! : )

Some of my background is in operations of large power plants and practical thermodynamics. When you are talking about 500 Megawatt plants, even a .5% increase in effienciency (sp, buzzed) is huge over time! (It makes me want to use a heat exchanger on my bathroom vents/fans. Yeah, I'm a nerd ) But is that increase in effienciency cost effective?

Has anyone thought about an "air to air" heat exchanger for a LED grow? While using a bottom exhaust works, it is fighting the natural air circulation. One is already paying X for a light trap, X * Y for an air to air heat exchanger "could" be an option. As it would also serve the function of a light trap.

While it would only make sense on large grows, an air to air heat exchanger would allow for higher air flows and still use the heat produced by the lights (otherwise wasted energy) to help keep the temps within the range needed. More cost effective the further one is from the equator.

Or, this same heat could be used in keeping needed temps for seedlings/cloning. Seems strange to worry about KEEPING heat instead of losing it. Only pretains to indoor growing, not power plant operations. <G>

Well, power plants need to lose heat too. Depends on where and why you need to move heat.

Feel free to delete this post if it is off topic, or not helpful! Sometimes my Asberger issues make me tunnel vision and not see the big picture. I'm sorry if this is not helpful!

Again, I love this thread!

Peace

 

[knna]

DraGonReBorn, actually some people did exactly what you mean (at least, what I think you are meaning by a air to air heat exchanger).

Some people using sealed grows chambers with CO2 independent supplementation, so not requiring to vent air out the cab, used a heatsink for LEDs which fins is fully out the cab, thus releasing most of heat out. Others used metallic walls directly.

But actually, with LEDs usually the problem is not high temps, but low ones. There is many way of heating an space, but the question is doing it at as low cost and maintenance free as possible.

If you have a heat source near the LED cab, sure its possible to use that heat to raise temp of grow area. But we try to use general solutions that most people can use, and a heat source next to grow area cant be considered as the usual situation. Anyway, sharing any actual solution you use in your case and how it works in the practice is always helpful for other people on similar situation.

 

[Terramoto]

Knaa u still making that wholesale thing? I saw it on cannabiscafe and the leds seem be quite cheap.

 

[knna]

Yes, I keep serving growers components required to build DIY LED lamps, for those willing to do it.

Of course, doing a DIY LED lamp is cheaper than buying it list to plug. But I not put too much emphasis on this point about DIY LED lamps, although I know many people is interested on it just due it. I think the main advantage of building your own lamp is you decide exactly what components you use, the size and the footprint you want it has, thus you build it exactly for your cab, with the features you want or need (exact spectrum, dimming function, adjusting spectrum for multi stage lighting or lighting treatments...)

Most commercial LED lamps out there has the problem of the very cheap LEDs they use. Although design can be improved a lot, it works decently on the short term. But LEDs used range from pure garbage to barely decent ones, with efficiencies always way lower than LEDs available today. A good LED usually emits as minimum 30% more light for same power burned, and often difference is over 100% (meaning a good LED gives off twice the light for same power). And that makes a world of difference on light delivered at a given power: you need to use lot more watts with commercial LED panels (specially the ubiquitous chinese ones, UFO style) than with a well designed and build DIY using good LEDs.

But those chinese LEDs are very cheap, while good LEDs are costly. So the price difference when building your LED lamp is not huge, when comparing watts burned (which most people takes as reference). Its when comparing light emitted, difference becomes huge.

I carry in general top LEDs of top bins. I think, at least until now, that it was a must to use top end LEDs if wanting to compete with HIDs and beat them on a watt basis. Top end LEDs are relatively expensive. Some people ask me to carry cheaper ones, still with lower efficiency (but known one). I feel this is going to be possible soon. LEDs improvement is so fast that currently is thinkable to not use top bins of each model and still be able to beat easily a HID using fewer watts.

When this possibility becomes real, then probably you could built a DIY LED lamp very cheap.

As many people ask me components for their LED lamps, Im going to open an online shop. Im working on it currently, and hope to have it running before this year ends (but on worst case, it wont last later than January). Im going to upload there very good mounting tutorial so anybody with a solder iron can mount his DIY lamp when willing to do it.

For the moment, a friend is showing some of my LEDs on his page, US based.

 

[Terramoto]

Hmmm, the thing is im growing with 210W CFL (i know, i know it kinda sucks) on a 3.3sq ft closet and my buds are very small with 3 weeks old. I was thinking that maybe replacing 91W for 35W of LEDs would give my plants a "boost", or should i buy a 150W HPS instead?

The wavelenght of the leds is something i dont know what would be the best mix choice for my closet.

 

[DocLeaf]

One positive of LEDs, unlike conventional lighting (florescent and HPS), is the bulb / element does not need replacing every few months to produce the same about of lumens/light output. The unit we were testing was tested in the factory for 1000s of hours,, and the first thing to break was one of the computer fans in the unit.

Also individual LEDS are cheap and simple to replace.

peace

 

[knna]

Hmmm, the thing is im growing with 210W CFL (i know, i know it kinda sucks) on a 3.3sq ft closet and my buds are very small with 3 weeks old. I was thinking that maybe replacing 91W for 35W of LEDs would give my plants a "boost", or should i buy a 150W HPS instead?

The wavelenght of the leds is something i dont know what would be the best mix choice for my closet.

On small cabs, like an 3.3 sq ft one, you can use LED lighting not costing much more than the HPS lamp and get better results using less watts. With the LEDs I carry, you can lit such cab with 85W of LEDs and get very good results. With other LEDs less efficient, you may need to use more watts, but anyway less than a 150W of the HPS.

210W of CFLs is a good amount of light for a cab that size. They should yield fine. You cant see big buds at just 3 weeks into bloom. And anyway, its a good amount of light, if buds dont fatten fine, dont blame the lighting, but mostly the genetic used and your ability to grow or read plants.

People often thinks they need to choose between HPS, CFL or LEDs. When all them can work together perfectly. In your case, replacing a good portion of those CFLs with just red LEDs, that are the cheaper ones, may achieve excellent results at low cost. Likely 50-60W of red LEDs and half of those CFLs can beat your actual setup with much less watts. Making some bars with that wattage may cost you less than 100€ including all (LEDs, driver, heatsink,mounting materials). CFLs plus red LEDs is a cheap and easy way to light small grow areas, and ideal on low temp areas where using just LEDs is often problematic.

 

[Terramoto]

Knna, thank you very much i think im going to buy some of the LEDs to mix with my 125W CFL cause the 90W i have is from a chinese store, im guessing replacing those would be a big improve to my closet.

Also found very interesting what Weezard said about the trichomes "color" and UVb and IR.

About the dwarfs made under LEDs an awkward solution might be electrical stimulation.

 

[Weezard]

Interesting, yes. but not shockingly so.

Interesting, yes. but not shockingly so.

Knna, thank you very much i think im going to buy some of the LEDs to mix with my 125W CFL cause the 90W i have is from a chinese store, im guessing replacing those would be a big improve to my closet.

Also found very interesting what Weezard said about the trichomes "color" and UVb and IR.


Thanks brah!
But, keep in mind, it's just a weak hypothesis of mine.

I observed something, then did a few sloppy experiments.
Got half baked, and jumped to some half-baked conclusions, yah.
Not sayin' it's bunk, jus' sayin I have a lot of testing left to do before I offer those ideas with any confidence.

Please do help us devise tests to disprove those "best guesses".
Throw some cold water on da conjecture and you do us all a great favor.
I welcome dat!
Cold water is da usual happenstance when ya get a lot of "hot air" out over "thin ice".

About the dwarfs made under LEDs an awkward solution might be electrical stimulation.

I just read dat thread.
Thanks!

Then thought about it for a while.

I feel that all the "energy substitution" talk is nonsense.


JMO, folks!
Based on physics.

So, did another quick n sloppy 'speriment.

PV stimulus.
Lotta talk about electrical stimulation enhancing plant growth.
I have serious doubts about that.
I have seen good result, but do not think that direct stimulus is the cause.
So.


Photovoltaic cell

I cannibalized a "hat fan".

Under 2 volts and very low current.
"First, do no harm."
Should be enough for electrolysis though.



Neg electrode

Will produce Hydrogen gas near the surface, which should outgas rapidly and hopefully not acidify the soil moisture too much.
I'm more worried about the muck at the bottom of the container than the top 1/2".
Also hoping the Hydrogen will confuse would-be surface egg-layers like gnats n root aphids.
Stranger things have happened.



Pos electrode

Mounted low to produce oxygen at the bottom of the container.
I'm guessing it will be of benefit to the roots by keeping anaerobic McNasties at bay.
If so, root growth should be impressive.


Now, I wait.

Thanks for the link, it made me think.
Aloha

Weeze

 

[sm0k4]

I want to take the thread in a new direction. The electrical side of this new technology we are adapting to our growing needs is just as important as the luminance side of it.

For a basis to my reasoning I want to list why I am switching to DIY LED.

1. Better at spreading light evenly across a canopy.
2. Less heat to worry about(Less heat equals more efficient use of light).
3. Less expensive long term solution
4. LEDs are just too damn cool and I am a huge electronics nerd

There are probably more but these reasons offer plenty justification.

When we use the standard Buck drivers that you just plug in and dim with a potentiometer, you aren't running the most efficient setup and are losing 5-10% efficiency due to heat losses because you are overworking the components if you don't have your strings set up properly.

In my situation I designed 13 light bars that are Aluminum U channel with dimensions of 1.5" wide and 20" long to cover the top of a 3' x 2' cabinet. I want each of my light bars controlled by its own driver. Then I can control each of the drivers with a separate PWM circuit to limit the string current how I see fit for that particular color / string.

You are limited by the limits of the manufactured driver and must base your light design around it. Providing separate voltages to separate driver circuits allows most versatility of light output as well as color ratios. If I only want three White and one blue on a string, I will run the string with a 12-15V supply tuned to the proper voltage that allows max. efficiency through the driver since it will only be using .5V of that input by the end of the LED string. Minimal heat loss should occur unless you are PWM dimming. PWM dimming changes the frequency and as that goes up, so do inefficiencies.

I have never built a light so maybe my interpretation of it is a tad overkill, but I don't mind. I want to design my light around my room without linking two light bars together in a daisy chain fashion. Each light bar having its own driving circuit allows for best rearranging of the lights. Unplug, move the light, plug back in. Or unplug the light bar and leave removed without having to jumper to the next string and recalculate your Vf and efficiency again. So in summary if you build your own seperate driving circuits, you will have maximum flexibility in every aspect of the LEDs. Sure it may cost a bit more than the manufactured drivers, but you can source your own components of better selection and it sure is fun building it.

My design may not suit someone else's needs though, so I'm sorry to digress. Just wanted to point out how efficiency can be improved by building your own driving circuits.

Of course I don't have any hands on knowledge(YET), this is from the last couple week's worth of researching the matter. Boost configurations still confuse me about the voltage output increasing, but I will learn that later and how to best apply this topology. I have a few NCP3066 chips on the way to play with Boost configs. It seems in these configs you are limited by the inductor you choose.

If it is the end goal to achieve the most amount of herbs from the least amount of kWh (the only unit that matters in the end) then maximum efficiency must take place. Thank you for reading. I hope we can prove all the nay-sayers wrong by proving LED yields more per kWh than HID. I believe its already proven, but who says we can't get even higher?

 

[knna]

Once you are an electronic nerd, there is many valid alternatives to driver LEDs, each with its own advantages.

But be aware you solution is not more efficient than a good driver working off directly from mains rectified. Your solution is very efficient (easily over 96%) working from a DC source. But the DC PS you use likely will loss more than 10% of total power. Typical DC PS are 80-90% efficient. So getting an 85% efficiency with your system is the best you can attempt generally.

 

[sm0k4]

Yes, I see what you mean. The only way to get absolute max efficiency is to run large strings at rectified AC. However, I am not that committed to designing a well built circuit or design the board to put it on right now. I want to get my operation up and running asap and designing such a circuit to rectify AC and drive LEDs will get involved. I have little experience in this type of circuit so I have a slight learning curve to overcome.

As long as I hit over 90% eff. with the driver I will be happy. The losses on the PS I can live with for now. I will look into a mains AC circuit after I am growing and I become bored again. I need all this new info to soak in some more so I get a better understanding of how each driver topology works inside and out. Buck topology seems easy and plenty efficient so I started there first.

 

[shlepp]

I am new to this site and am currently looking at designing my own LED lighting system
From what I see there seems to be a lot of bs on the part of the off the shelf LED systems currently available. The biggest ones being wavelength and output. I am looking at the Ultra Bright AlGaInP & InGaN manufacturing process of SDK a japan based company for the red 660 nm and blue/green at 500 nm, I have worked with this company for 20 years in the semi industry mainly on the equipment side. Does anyone have experience with these LED's and if so, any comments on quality.

 

[sm0k4]

This thread is dedicated to the pragmatic use and improvement of LED arrays for growing. Without converting to a how-to, I cannot find datasheets on SDK products, so couldn't say how they are. If you want to know how to go about this, Google DIY LED array and the first suggestion should get you started. Lets not make this thread full of cluttered How-to questions when the info is already available.

 

[shlepp]

Not interested in help designing a LED array, I design equip for thin film apps. Iterested in quality of LED's based on application.
spec sheet for reliability??? guess Im in the wrong place.

 

[Weezard]

Wish I could help ya.

Wish I could help ya.

Not interested in help designing a LED array, I design equip for thin film apps. Iterested in quality of LED's based on application.
spec sheet for reliability??? guess Im in the wrong place.

But I know zip about the top shelf leds.
Your go-to guy on that will be KNNA.

When I built my light there was only one company offering 15W. 660nm. emitters and they were not the most efficient.
Nor was the data sheet much help.
I ended up measuring the junction voltage for each emitter at 80% of Imax.
They were all over the ballpark.

The data sheet wasn't even needed for thermal, I just over-killed on that.
Been running it for over a year, and according to my meter it is as bright as it was on day one.

I figure it will be obsolete in less time than it takes to lose output.
So, spec. sheets were not needed, and bin choices were not offered.
(It was a small run by Ledengin and it seems they are out of stock now.)

Once you get what you are looking for, and start growing with leds, this thread will be the "right place", yah?

Thanks for the bumping platform.

BUMP!

Aloha,
Weezard

 

[Fin]

Hi,

What do you all think about growing seedlings under led lights? How close to lights a seedling can be, and can it be too close?

I had problems with seedlings in coco under leds, and I think maybe too much light was one part of the problem (surely not the only reason).

It might be easier to give them too much as the temperature is not so much a warning sign as it would be in a hotter setup.

What are your guidelines for growing seedlings in led light?

 

[SOTF420]

This is what I have to say about LED @ almost 4 wks flower

 

[Weezard]

I like my cake with icing.

I like my cake with icing.

Hi,

What do you all think about growing seedlings under led lights? How close to lights a seedling can be, and can it be too close?

I had problems with seedlings in coco under leds, and I think maybe too much light was one part of the problem (surely not the only reason).

It might be easier to give them too much as the temperature is not so much a warning sign as it would be in a hotter setup.

What are your guidelines for growing seedlings in led light?

Howzit Fin?

You are right about stressing seedling with too much light.
But you also gave them too much water, too early food, etc.
In short, loved 'em to death.

Not to worry, I killed my first sprout under leds with too much water and too much light.
It's light intensity that drives transpiration, so when you cut back on da photons, cut way back on the water.
Wait for the cotyledons to shrivel and then start with a 1/8th nute concentration.

That said, it's not all about proximity.

It's also about the blue to red ratio.
B:R
Give them 1:3 and 12" of headroom, they will be squat and have fatter leaves.
Give then 1:4 or more and they will stretch for wanting more blue until they saturate from the red.

Once they "see" the equivalent of about 91k Lux sunlight though, they stop growing.

A major advantage of leds is the spectral control.
It allows us, (within strain restraints), to control the morphology of our wee, green, friends.

The low heat and power savings are jus' icing on da cake, yah?

Aloha,
Weezard

 

[alkalien]

I build different Panels for different jobs:

Seedlings and cuts until they are big enough for flowering:

And there is one without blue and way less white for flowering:

Reds on both of them are mixed 630nm and 660nm.

Still I got a few problems with seedlings on dirt. Won't work out for me. Looks like I'm either loving or hating them to death. Works great in DWC though:

Still wondering how to germinate seeds for the DWC. I will have to learn how to germinate those seeds. Looks like the less heat from the LEDs doesnt dry the soil fast enough,I'm extremly curious what I'm doing wrong, will keep you posted when I find out!