The Truth About LED Lights

[Buckshot]

And, no, mr picture guy, I'm not saying one can't grow plants under LEDs. The fraud is about "3x better". Hosebomber has at least offered a quantification that it's 50% better (400WLED==600WHPS), but it took a while to pull that truth out of him.

I offer images because I am only a "picture guy" and not a tech. I am a grower. So I finally get what pissed you off "3x better" OK

This is the quote

"If you get about 300-400watts of LED lighting, you can replace a 1000watt HPS or MH system. You will not need a ballast and you will not need to vent the heat from the light as LEDs emit a near zero heat signature"

The text did not define grow space. If it did I could agree or disagree based on space. As I said before in a small space the statement above is true. In a larger space I would agree with you ...

"picture guy"

 

[lenngray]

I don't know why nobody gets the point. It's a well-acknowledged FACT that 70% of the energy put into an HPS is lost as heat. This is partially in the balast, and partially in the bulb, which gets HOT. This allows the potential for 3.3x as many IDENTICAL photons to the HPS to be generated by LEDs assuming the ability to choose LED wavelengths. Where's that result?!!! Don't give me "oh, space would count". I'm saying there should be the ability to get 3.3x the photon flux density of the SAME photons per unit power. No appeals to "penetration" necessary. No appeals to "inverse square". No appeals to PAR. We should be able to see a 233% improvement in the same situation as the HPS. If not, something is being ignored. Now, I acknowledge that the LEDs do have their own heat loss, but that seems to be on the order of 10%. So, let's see the explanation of where the bulk of that 233% improvement is going !!!!!!!!

 

[lenngray]

Greetings savolainen,
Just in the interest of having you avoid unnecessary glitches, except for the Haight lights, I don't think any current lights count on passive cooling of the LED chips. There's a radiative curve that goes down significantly as junction temperature rises. Presumably that means that even more heat is generated by that power going somewhere other than into radiated light. Early on, there were water-cooled lights, and, if we can cool computer CPUs for gaming that way these days, it can't be such a bad idea for LEDs. Keep that factor in mind in your design ruminations.

 

[WillyP]

The light energy emitted by LED's at 25C running with 350 mA (1 watt) ranges from 20 to 40% of the power input. Running at 700 mA (3W) or 1000 mA (5W) these percentages drop making them less efficient. Also as the diode temperaure increases they become less efficient as well. What is is not emitted as light is emitted as heat. Bridggelux recommends sizing the heat sink for 80% of the power input

 

[Buckshot]

I don't know why nobody gets the point. It's a well-acknowledged FACT that 70% of the energy put into an HPS is lost as heat. This is partially in the balast, and partially in the bulb, which gets HOT. This allows the potential for 3.3x as many IDENTICAL photons to the HPS to be generated by LEDs assuming the ability to choose LED wavelengths. Where's that result?!!! Don't give me "oh, space would count". I'm saying there should be the ability to get 3.3x the photon flux density of the SAME photons per unit power. No appeals to "penetration" necessary. No appeals to "inverse square". No appeals to PAR. We should be able to see a 233% improvement in the same situation as the HPS. If not, something is being ignored. Now, I acknowledge that the LEDs do have their own heat loss, but that seems to be on the order of 10%. So, let's see the explanation of where the bulk of that 233% improvement is going !!!!!!!!

I don't give a shit about photon flux density or % of anything about my lights and I never will ... I am a grower

Space does matter, ask ANY grower

this is what I know

I am using far less than 1000 watts and I am growing beautiful plants. It costs less to run and I can flower all my plants in the space and light they provide. I have a 1000 watt system I can use and won't

I will continue to grow and post images

 

[WillyP]

What exactly is your reasoning for using 410 and 660 other than that is what the name brand companies are doing? You are only hitting 1 photo receptor with each of those (yes I know there is reason for doing so). Likewise, there is a very good reason for using 450 470 and 630. Each of those 3 wavelengths hit more photo receptors than all of yours combined. I would love to see pictures of your great buds with very little red wavelength for the flowering process.

As for 410nm, the phototropins PHOT! and PHOT2 that adsorb at 410/450/470 and my design covers the 450/470 portion and the 630 nm with the warm white

with 100 watts for 7 weeks of 12hr days
dried yield 95 grams of good smoking ICE

 

[savolainen]

Greetings savolainen,
Just in the interest of having you avoid unnecessary glitches, except for the Haight lights, I don't think any current lights count on passive cooling of the LED chips. There's a radiative curve that goes down significantly as junction temperature rises. Presumably that means that even more heat is generated by that power going somewhere other than into radiated light. Early on, there were water-cooled lights, and, if we can cool computer CPUs for gaming that way these days, it can't be such a bad idea for LEDs. Keep that factor in mind in your design ruminations.

I got the feeling that you have not made lights by yourself and are just speculating..? I just counted mine and there were 14pcs. of them. Most differ from each others and are having many wavelength components and sizes like 1W, 3W, 10W, 20W and 50W.

The fact that commercial lights use fans does not mean they are needed. If using 1W or 3W LED components with star plates they don't actually need any additional cooling, unless very much of those are placed side by side.

It was just because there were lots of opinions out there that stressed out how important cooling is, I made too good (big) coolers to most of my lights at the beginning. I thought that coolers must be really big because everyone were fussing about it.
So I made too big coolers and noticed later that those LED's don't get even warm. It was waste of space. I now consider that I should put those parts and rebuild them to get some more focused and useful lights.

I have also noticed that if LED's are 10W or bigger components (those which have internal serial+parallel connections) they get much hotter and coolers need to be lot bigger. 50W LED needs bigger cooler than 5x10W and so on.

So far only problems that I have had has been with 50W component which made approximately 1m long and 1cm thick aluminum rail too hot. At first I used PC case cooler with it but as I like passive systems I removed it later and put it behind timer which switches power on/off every 15min.
Now when it starts to be too hot, timer shuts it off and that 15min off time is enough to let it cool again. It's workable solution but I think am gonna make that cooler even bigger at some phase so that I can use it constantly.

But components smaller than that I have not had problems with cooling. Especially those 3W star LED's are easy. I used thin aluminum plate for those as extra cooler and like I said, they didn't get even warm. After constant using several hours, I can hardly sense any warmth in those.

Maybe the key is spacing. Commercial lights have LED's side by side, my models have lots of space between every LED.

---edit---

Actually I have thought water cooling as an option for that troublesome 50W component. As well as same kind of passive cooler that I have in my computers CPU.

BTW. I also have passive PSU in that computer which was expensive at that time. I use MB integrated GPU so there is no any fans in my computer either. It's in rack case which has top cover dismounted. I have used it years that way. Only occasionally upgraded some parts.

So it's not exaggeration to say that I am passive cooling enthusiast.

 

[savolainen]

As for 410nm, the phototropins PHOT! and PHOT2 that adsorb at 410/450/470 and my design covers the 450/470 portion and the 630 nm with the warm white

Why you don't use 660nm? Fits well to chlorophyll-a peak, and it is higher than 620-630nm peak. Of course they are in different chlorophyll curves, is it it?

Does there need to be 620-630nm RED's or are those used just because they (are?) were cheaper?

 

[WillyP]

I have been building my own LED lights for more than a year using different wave lengths. I have several set ups. The light I use fpr my mothers has only 430/660/warm white LEDs and the vegitative growth allows me to get a dozen clones per week per plant. For raising the plants and flowering, I add 410 nm to the mix \to get more resinous plants. I do not use 630. 450.,or 470nm they are inexpensive and do make plants grow, but are not as effective for plant growth (more watts to get the same growth rate.).
For my lights I use 50 and 100w led arrays which I purchase from Chinese manufactures for $1 to $1.25 per watt made to my specs and I use fan cooled heat sinks for compactness. The Chinese use theses arrays sto make high bay lights with only passive cooling and wattages up to 200W. I have used EBay fan/heat sinks and they have worked well and I have used COOLIANCE heat sins which are more compact and the fans are powered by the LED supply and do nor require a separate 12V for the fan.

 

[savolainen]

I have been building my own LED lights for more than a year using different wave lengths. I have several set ups. The light I use fpr my mothers has only 430/660/warm white LEDs and the vegitative growth allows me to get a dozen clones per week per plant. For raising the plants and flowering, I add 410 nm to the mix \to get more resinous plants. I do not use 630. 450.,or 470nm they are inexpensive and do make plants grow, but are not as effective for plant growth (more watts to get the same growth rate.).
For my lights I use 50 and 100w led arrays which I purchase from Chinese manufactures for $1 to $1.25 per watt made to my specs and I use fan cooled heat sinks for compactness. The Chinese use theses arrays sto make high bay lights with only passive cooling and wattages up to 200W. I have used EBay fan/heat sinks and they have worked well and I have used COOLIANCE heat sins which are more compact and the fans are powered by the LED supply and do nor require a separate 12V for the fan.

Hi, thanks for sharing this. BTW. you seem to have made that 430/660/warm white light purposely so that also blue fits in chlorophyll-a, instead b.
But because also 660nm red hits the same curve, doesn't it bring any problems because there is no LED's at all that would be dedicated specifically to the chlorophyll-b?

Somehow, it would feel more right to have both curves represented, like having 470/660/warm white instead. And maybe having both cool and warm whites mixed so that they would cover wider spectrum. Though that might have only minor effect.

BTW. I also bought those UV LED's from EBAY right after I found article that UV rays increase resin production. Just little later I found out that they were UV-a LED's while cannabis plants would increase resin/THC amount to protect itself from UV-b, not UV-a.
Someone at some forum then said that UV-a has only same kind of benefits that deep blue wavelengths have. So, are you sure that also UV-a has had increased amount of resin?

I bought Repti Glo 10.0 CFL's so that I would be sure my plants get that UV-b. I put 13W to smaller space and 23W to bigger, both were 10.0%. Someone recommended to schedule those to be lit only approx. 4h/day in the middle of the light cycle so that they don't get over exposure. I have recently tried longer periods and I have not seen that plants would suffer from it, yet.
Maybe it would be wise to increase UV-b amount little by little while plants get closer to the end of their lifecycle.

 

[WillyP]

Just to let you know 410nm is not considered UV, Also a warm white LED has wave lengths that cover Chlorophyll b. A warm white is a 460nm LED with red phosphors that provide 600 to 690nm wavelenth coverage
I've also found that LEDs on Ebay are expensive and the wave lengths I needed were no available

 

[savolainen]

Yeah, you're right. I can't remember all the specs very long time... UV-a is from 315nm to 400nm.

If all warm whites are made from those wavelengths that you said, it would be good light even with just those LED's used. AFAIU there should be small amounts of other wavelengths too because blue+red should look more pink than warm white. Or..? Ok, I can google.

Yes, this one was one of the first search results:XP-G warm white LED wavelengths

Based to this picture i'd say that cool white looks very well too, thinner red but higher blue peak..

 

[Swami]

You will not grow anything under 13.8watts.

This is just plain wrong. I vegged a very healthy plant to about 18" h X 14" w under 13.4 watts of LED.

I am new here and could use a little help uploading some photos as evidence.


I am just finishing my fourth LED design which I believe will better anything currently out there in its class. I will start two new threads before the end of the month.

1. Will show my light design. It is almost finished. Getting the last of my drivers in on Tuesday. It is about 150w actual draw made for a 2' X 2' grow space.

2. Will do a Grow Journal showing what it is capable of.

 

[savolainen]

This is just plain wrong. I vegged a very healthy plant to about 18" h X 14" w under 13.4 watts of LED.

I am new here and could use a little help uploading some photos as evidence.


I am just finishing my fourth LED design which I believe will better anything currently out there in its class. I will start two new threads beofre the end of the month.

1. Will show my light design. It is almost finished. Getting in the last of my drivers in on Tuesday. It is about 150w actual draw made for a 2' X 2' grow space.

2. Will do a Grow Journal showing what it is capable of.

I have vegged one bubblegum recently with 7x1W E27 LED lamp. Worked fine. I guess criticism is about fractional LED's, like those with has 168 small LED's in their lamp.

Later, when that BG was bigger, I moved it to grow space where is much more LED power. Self made lights that use approx. 200-250W.
Flowering is not finished yet so I can't tell how big yield it gives... It look's decent but nowhere near that kind of yields that more professional growers have had with LED's.

I have not monitored ph-value and even temps might have slightly off. As well as fertilizers. I think my lights are the best part of the process at this moment.

BTW. Could you specify what this means? "2' X 2' grow space"
What it is in cm's or meters?

I don't know those foot's, feets and elbows which many countries use as measuring units.

 

[dubbrollin]

I have vegged one bubblegum recently with 7x1W E27 LED lamp. Worked fine. I guess criticism is about fractional LED's, like those with has 168 small LED's in their lamp.

Later, when that BG was bigger, I moved it to grow space where is much more LED power. Self made lights that use approx. 200-250W.
Flowering is not finished yet so I can't tell how big yield it gives... It look's decent but nowhere near that kind of yields that more professional growers have had with LED's.

I have not monitored ph-value and even temps might have slightly off. As well as fertilizers. I think my lights are the best part of the process at this moment.

BTW. Could you specify what this means? "2' X 2' grow space"
What it is in cm's or meters?

I don't know those foot's, feets and elbows which many countries use as measuring units.

.6096 x .6096 meters= 2'x2'

 

[evilled]

cree led is top ??

 

[Hosebomber]

Yes, Cree currently produces the highest lumen per watt white diode. There are other options out there for colored diodes that are just as good if not better, those being Osram and Phillips.

 

[svenms]

Where can I buy those LED Chips if I want to crate my own LED Panel ? I mean the good ones like said Hosembomber, Cree, philips.

Thanks !

 

[Hosebomber]

Arrow, Mouser, Digikey, Future Lighting Solutions, Avnet, Steve's LED, Rapid LED, MakersLED... that should get ya started.

 

[evilled]

helios panel is good ??
have loking panel helios creed led 420w
for you is good or mars 2 900w is batter ?