LED lumens, again

lenngray

New Member
BEWARE, all ye potential buyers of LED Light systems:

In another post, I attempted to begin discusssing a
phenomenon, that numbers for radiant flux (which I
called, rightly enough, even acknowledging PAR, as
lumens) are not being given to us for evaluation.

Professional obfuscators, for marketing purposes(?),
are literally colluding to keep this information from
being presented to the public, so they can sell us
whatever garbage they happen to find easy to build,
or happen to have on hand, to maximize profits,
regardless of availability of more efficient products.

100 LUMENS of Red (the main component of grow-lights)
PER WATT _is_ available, and I just exchanged posts on
this board with one of those obfuscators (I'll just
call him Skater), ending with his attempted hijack of
my presentation by talk about the fact that, so far,
nobody gets 100 lumens out of a single 1 Watt diode.
He ignores the engineering issue that nobody should
give a damn about that. Using as many diodes as it
takes to get the job done is just how we'd currently
get that efficiency number.

My first post about a company selling an admittedly
cheap light, but one that only got 30 lumens per watt
was exactly what I wanted to say, and that that was
something to beware of.

Skater closed his argument in that thread with "hey,
people, disregard this ... lumens should not be used
for LEDs"

The "we don't need no measurability", part of the
anti-science marketing campaign Skater brought to my
claim, is what needs to be fought.

The actual term that should be used and cannot be
obfuscated away, is radiant watts, but that can be
confused with watts of power coming from the wall,
so, IN A SPECIFIC BAND, lumens is fine, and is what
the manufacturers use in talking to the engineers
(of which I'm one) they want to use their stuff.

One probably-good $1500 345W light that a top company
advertises, compares chips with its "patent pending"
_LENS_ system to something else, saying it gives a
369% boost in delivered light -- but that something
else is their own chips without _any_ lens. Cute!

Another article, guiding potential buyers on what to
look for, recommended that we only buy lights built
of chips _above_ 1W. But, apparently, 3W chips do
_not_ yet give 100 lumens per watt. A likely reason
they may be saying to go to the higher wattages (even
though they're less efficient) is to make construction
easier or to make reliability of a board with fewer parts,
better.

We need to be able to get the numbers and talk about
them without paid liars doing their marketing hype.

One reason these light manufacturers would want to
keep such comparability from being utilized is that, given
"Haitz's Law" (the expectation that LED efficiency
will likely double every 2 years), they'll have to
keep chasing state-of-the-art and dump (and not sell)
their obsolete crap. And that's bad for profit.

That's my heads up.

The useful question is what _does_ each light give
(and lumens is fine for red though it'd be truly
meaningless for infra-red) but instead, we're getting
claims of "I get so much yield". But we can't measure
what "so much" is.

People, stop tolerating this kind of bullshit in the
so-called "information" we get.
 
I'm gonna be very nice about this and then we can just delete this thread. YOUR comment about 30 lumens per watt is over the whole of a 90 watt panel. LEDs below 475nm are not measured in lumens the same goes for any LED over 630nm. They are measured in mW. As most users of LEDs know, the majority of commercial panels are made up of 455 and 660nm red and blue LEDs. Lumens are a bad measurement for LEDs because it does not record these wavelengths as it pertains to the growing of plants and photo receptors. Yes, when speaking with an engineer and in the designing of LED light panels we use the luminous flux numbers to decide what to use where and in what ratio. The end users' are not engineers and not designing the product so they do not need this information. It doesn't help them in any way. PAR is the only measurement value that matters. How much of the light produced can their plant use. Your attempts at saying random things and quoting them as something I have said is a poor attempt at best as they can simple scroll up and read that I never said "we don't need no measurability" or that I am in any way unscientific.

We recommend using LEDs ABOVE 1 watt because the .025mA 1/4" LEDs simply do not put out enough power (luminous flux). 1 watt leds may be strong enough for your grow. It depends on how you grow and how much intensity you need. MOST grow panels now are 3 watt leds for penetration reasons. (do a search of the forums, there are studies and grow journals on it) Yes 3 watt chips are less efficient than 1 watt LEDs (in most cases) but they do also produce a higher luminous flux and give deeper penetration due to the inverse square law. You're an engineer tho so I'm sure you knew that.

You comment that "(and lumens is fine for red though it'd be truly meaningless for infra-red)" is simply false. Only a narrow band at the top of blue and bottom of red wavelengths can be measured in lumens. 660nm red is measured only in mW and is the largest portion (rightly so) of the red LEDs in most panels. It is also a requirement of the Emerson Effect. If you want to see exact yield numbers in a grow, look at SteveHman's grows 2-4. He has some very nice pictures, took detailed notes and gave a nice spreadsheet at the end.

Fact is, while you have post in 2 topics, I've been on these forums for over a year as a poster. I have called out nearly every manufacture to change parts or all of their marketing when it is a outright lie or misleading. I have also given a huge number of peer reviewed articles on LEDs, photo-receptors, plant growth, lighting effects, and plant chemical reactions and inter-actions. Anyone can say they are an engineer and there are many forms of engineers (I played on for a while and it's part of my resume). A structural, civil, mechanical, aeronautical and most electrical engineers will have no working knowledge of LEDs and grow lights.

I'm not sure if you are confused about LED grow lighting, or you are attempting to make a sub-par 1 watt led grow light and sell it here, or you just simply took 2-3 hours to look over some very basic information and thought you knew it all instantly, but please take a little more time and do some more research.

Hosebomber (aka Skater)... I never sold a LED light panel or worked for a company that has in my life and I approve this message.
 
Ahhhh. I now see how we got so oppositionally oriented without either of us being evil.

Until your mention, in this post, that the 620-630 red, itself, was not fine-and-dandy, it seemed, to me, that you were defending that $99 light, saying that I was just too unsophisticated to understand. So my searching for chips, and a solution to making a high-powered light just kept getting more of your same answer. What I was saying about "lumens are just fine" was about comparing a source of a band with another source of that same band. That would apply even to a 660nm red, as it's luminosity, way down at 5%, would still allow two actual sources of the same band, to be compared.

If you go back, you'll note that you didn't even hint that that light was unsuitable for anything because of its red wavelengths, even when I ragged on its low light-output. It seemed to me you were saying that I didn't know enough to criticize that light, also implying that any shit-ass manufacturer deserves respect, and has a right to con the people of the list, while hiding behind the arcane science.

Okay, so I do appreciate the heads-up that the reds need to be 640-685nm.

This is not to say I have no other disagreements with your expert opinions.

I'm an old-school Electrical Engineer, from before chips, and the equations we learned to work with were not limited to what we had worked with before. And this includes the whole electromagnetic spectrum, all the way into light and quantum-mechanics.

Spouting the phrase "inverse square law" doesn't say you know where it applies. It's about the spherical wavefront of a source of uncollimated and unreflected light. Using reflectors, even silvered mylar (!!!), those photons are far from lost over distance. In fact, mylar is quite good -- proponents of white-painted walls notwithstanding.

Also, in reasoning about 1 watt LEDs, the "it depends what intensity you need" you and others ignore the fact that photons are photons. Once emitted, whether from 525 x 1W LEDs, or from 200 x 3W LEDs, it's the same photons. Penetration depends on those photons, not on the emitters they came from.

Ultimately what we're talking about is turning our electricity dollars into light, and then into plant matter. And the radiant energy per dollar is what counts for us. Although my FoxFarm feeding schedule alludes to 4 weeks of 18hrs/day of vegging light plus 8+ weeks of 12hrs/day of flowering light, for some 1176+ hrs -- in my own experience, it's really more like 1500 hrs of running the lights. In California, it's like 21 cents/kwh, though I've seen numbers that say some people are only paying 11 cents/kwh. If I could really run 200 watts instead of 600, that'd be a savings of $120 per harvest. This would be the meaning of such an actual improvement, and is what it'd be worth to me -- but only that much. That's the only reason I was looking at a $99 grow-light at all.

I might be willing to build a light if I couldn't just go out and buy what I wanted, but that'd just be for myself. I wouldn't be interested in setting up a business just to find myself having to keep chasing "state of the art". I especially wouldn't waste my time setting up a scam.
 
First the cost of electric use. I live in CA and we are on a sliding scale depending on how much you use from 12 cent per KW to 26.5 cent depending on usage. I believe all of the state is set up this way in some degree but don't quote me on that.

The key to the 3 watt over 1 watt is that you have a greater luminous flux from a single point source. That is how they provide more penetration. That and the fact that you get more output from a smaller number of diodes in a smaller space. Smaller core coverage with more light = more penetration.

Don't get me wrong, I'm not saying not to use 630nm red or 480nm blue lights. There are just proportionally smaller in ratio to the 660 and 430 or 450. This will give you a lower "lumen" count while still burning the watts. I'm one of the few LED guys that believes in the whole light theory, a plant needs light from all spectra to properly transfer the chemicals for growth and budding to it's fullest potential.

As for building your own panel. If you have the knowledge and ability, it's a great idea. You can custom build the size and shape and coverage area to fit your needs. It will however be a bit more expensive than buying an off the shelf version. From experience, once you start testing it gets very expensive in the "what if I try this" category.
 
To date no other company has shown the proven results as well as Advance IMHO. However there are a few good companies on the market. If you see a company claiming 1/3rd the power use of a HPS or claiming the light is a 900 and actually draws 387 (true statement about 1 company) close that browser tab and move on to the next. If they are lying about their product there is a reason. It is not possible with current mass produced models to get the same results as an HPS with anything less than 2/3rd the power. ie with a very good 400 watt led you could get about the same yield as a 600 watt HPS. To get better efficiency than that it needs to be a custom panel built for your grow area and then 40% power reduction is close to max. I've gotten as high as 47% power reduction with almost exactly the same yield (within 1 gram of tomatoes) in a hydro setup with light barriers and running the same nuits to both sides and light being the only variable.
 
I read your guys post and just have a quick question....

Why are LED companies not using warm and cool white LED emitters in there products. I mean yes I understand PAR but plants have what are called antennae pigments that absorb more than just the blue and red ends of light... So why are companies not including "some" white led's of both warm and cool sides of the spectrum to fill in the gaps?

I understand that the purpose of LED lighting is 2 things... 1 to create a more energy efficient light, and 2 to produce a light that specifically uses the PAR frequencies and does away with the others that are un needed or used in extremely low proportions... but I figure since CFL growers have such good results, wouldn't using warm and cool white led's produce near the same results? for much less power being used?

This is just a question that I have wondered about and since both of you seem very educated on led, I wanted to hear your thoughts..

Also.... I saw one of you mentioned that your running your lights 18 hours.... try using the Gas Lamp Routine...it works and saves 5 hours of electricity per day in Veg...
 
But those single points are really, as I gave in my example, 200 single points rather than 525. It's the difference between an array of, say 14x14 versus one that's 23x22.

As for the DIY, since starting this conversation, I'm considering a transplant of some Alibaba 660nm chips into an old Chinese UFO that has 2/3 of its chips dark from random chip-overheat burnouts, so I have appropriate current supplies in hand.

When I finally went to the Wikipedia charts for PAR to see the locations of the red peaks, I noticed the difference between Chlorophyl-A, -B, and Carotenoid Absorption, and what they called Photosynthesis Rate. Now, clearly, if HPS's are working at all, that Photosynthesis Rate must have meaning with a far-from-zero rate in the mid-green range, but what does it mean to get photosynthesis without absorption by, for example, the chlorophyl? I ask because you say you're a whole-light kind of guy.
 
Those are nice reads Ice, I believe I have linked that site before on these forums. If you go to the base site there are a large number of links to other resources and articles as well.

Those links are exactly what I was referring to when I said I was a whole light kind of guy. Even though we know that certain types of light wavelengths do certain things, other spectra can do the same job, contribute to, or assist in the process. Using only blue and red LEDs will grow a plant, however, it will not grow to anything near it's full potential.

Using 50 or 100 watt white leds will grow plants. There are a couple of issues with them tho. First the cooling requirements of those diodes are very high. There is a large amount of energy burnt off as heat and you HAVE to keep the chip set cool. Secondly the output of white LEDs are not ideal. I agree with Ice that warm whites should be used to some extent. Cool white LEDs on the other hand have proven not to be worth the replacement of a colored diode. Cool whites have a initial peak around 445nm and then the relative intensity for each bandwidth above 490 stays below the 70% mark. Warm and neutral both have a much better dispersion wave in my opinion. Advance LED's new light is going to use 10watt Cree LED's so they say. Not sure on the color or bin they are using just seen some info. I have no idea what testing they have done with it but I have a feeling that SteveHman has already used it for at least 1 grow.

If you look at the light distribution of HPS lights you will see that they do put out a fair amount of blues and reds. Even though the peak is in the green and yellow range, they over compensate by the amount of light they put out. The plant will simply stop using the greens and yellows when it has what it needs. There are also pigments that can transfer the yellow light into red for photosynthesis purposes.

As for the GLR, I'm more of a natural guy. I have not tested the GLR and prolly won't until I get my light test 100% complete (is that even possible?). Things work better in nature than we can do in a lab/grow room for a reason. In nature the light levels ramp up and the days get longer then the nights slowly get longer till you hit 12/12 and flowering begins. I have never liked the idea of 24/0 or 20/4 for that matter. There just are not a lot of plants that grow well in that natural light schedule. If you are taking clones and they are in a 16 or 18 hour light schedule, start them in the cloner on that same schedule. You will like the results.
 
Those are nice reads Ice, I believe I have linked that site before on these forums. If you go to the base site there are a large number of links to other resources and articles as well.

Those links are exactly what I was referring to when I said I was a whole light kind of guy. Even though we know that certain types of light wavelengths do certain things, other spectra can do the same job, contribute to, or assist in the process. Using only blue and red LEDs will grow a plant, however, it will not grow to anything near it's full potential.

Using 50 or 100 watt white leds will grow plants. There are a couple of issues with them tho. First the cooling requirements of those diodes are very high. There is a large amount of energy burnt off as heat and you HAVE to keep the chip set cool. Secondly the output of white LEDs are not ideal. I agree with Ice that warm whites should be used to some extent. Cool white LEDs on the other hand have proven not to be worth the replacement of a colored diode. Cool whites have a initial peak around 445nm and then the relative intensity for each bandwidth above 490 stays below the 70% mark. Warm and neutral both have a much better dispersion wave in my opinion. Advance LED's new light is going to use 10watt Cree LED's so they say. Not sure on the color or bin they are using just seen some info. I have no idea what testing they have done with it but I have a feeling that SteveHman has already used it for at least 1 grow.

If you look at the light distribution of HPS lights you will see that they do put out a fair amount of blues and reds. Even though the peak is in the green and yellow range, they over compensate by the amount of light they put out. The plant will simply stop using the greens and yellows when it has what it needs. There are also pigments that can transfer the yellow light into red for photosynthesis purposes.

As for the GLR, I'm more of a natural guy. I have not tested the GLR and prolly won't until I get my light test 100% complete (is that even possible?). Things work better in nature than we can do in a lab/grow room for a reason. In nature the light levels ramp up and the days get longer then the nights slowly get longer till you hit 12/12 and flowering begins. I have never liked the idea of 24/0 or 20/4 for that matter. There just are not a lot of plants that grow well in that natural light schedule. If you are taking clones and they are in a 16 or 18 hour light schedule, start them in the cloner on that same schedule. You will like the results.

Nice post hosebomber...definitely I recommend the GLR, it works, and from what I've read, us MMJ growers are stressing our plants with un-natural amounts of daylight hours, most areas that landrace marijuanan comes from only see's 15 or less hours of light per day in the summer so anything more than that actually will stress the plants too... If I can find the link I will post it for you. Since plants use dark to determine flowering/veg, and not the amount of light, the GLR is relatively harmelss because all it does is revert the flowering hormones back to veg.

I definitely see potential in the white LED's and that's why I'm glad to see your input in them :) I definitely think LED's could be the new future for growing.

I made a link that might help those interested in seeing what spectrums your bulbs are putting out...helped me in determining what my HID bulbs were showing...

DIY Spectrometer - Check what wavelengths your lights give out

And here are the results of HPS, Ceramic Metal Halide, and Ushio MH

Ushio MH
Ushio_Optiblue.JPG


Ceramic Metal Halide (BTW I love these bulbs)
Ceramic_MH.JPG


HPS Ushio
Ushio_Optired.JPG
 
Back
Top Bottom