DIY Grow Lights with high-efficiency LED strips

Growroom Design
, , ,
81

True, I guess you get what you pay for. With most COB lights you get all the specs and know the exact parts being used, that is good. Just trying to make the point that with COB you are burning some watts on extra yellow & green the plants don’t need. Bottom line for me is COB is definitely nicer than HPS for many reasons, but the professional horticultural blurples are unbelievably efficient.

The P.L. Light full-specturm Hortiled I have even combines white diodes with blue & red. So does Fluence. That way you get a whitish-looking light that is “pumped up” a bit at the critical photosynthesis wavelengths. there’s no reason to discriminate against “blurple” when it’s done well!

No question these COB strips are very efficient and a great value for smaller wattage. also the coverage area will be uniform, without the “hot spots” some COB lights have. These are similar to the Fluence “Ray” light bars which are pretty much the same thing with a few red diodes added here & there.

1 Like
82

The plants will grow fine without the whole visible spectrum… but they will use yellow and green light too. Blue and red are more photosyntheticly active. So your not wasting any photons, you’re just not getting the absolute maximum umols per Joule.

The rest of the spectrum might have other benefits (or detriments) compared to just blue and red light.

1 Like
83

HLG claims over 3 umol/j when under driving and 2.45 umol/j at the board (40C) when running full current using the white light LM561C diodes, same as solstrips

1 Like
84

Only the shitty blurples are 100% blue/red though - the PL Hortiled cannabis lights are 10% blue, 10% green, 80% red, so technically they’re full-spectrum. Their actual full spectrum (veg) lights are 30% blue, 30% green, 40% red.

Also most companies use the McCree curve to calculate PAR, which counts green and yellow nearly the same as blue and red. I don’t trust the McCree curve that much, I believe it was done with a single piece of leaf tissue under unrealistic conditions and is only an approximation. I think lights that blast the plants at the most sensitive blue & red wavelenghts are going to yield more. this starts to be like religion - you choose your beliefs! :smile: But PAR of COB lights and PAR of individual red & blue diodes is not the same thing.

85

PAR is measurement of light between 400-700nm so i don’t understand what you mean par of cobs and par of red/blue diodes is not the same. while looking further in to the mccree curve i found the quote below, is that the information you don’t trust? still not sure what a mcree curve is or what it has to do with PAR. also found this pic from someone who was calculating these numbers using a spectrum or SPD curve. does that differ from a mccree curve?

Absorptance of green light (550 nm) is only 15% less than the average absorptance across the PAR region. The relative action of green light (550 nm) is only 9% less than the average relative action across the PAR region. So let’s dispel the myth that green light is reflected wasted energy?

2 Likes
86

OK, this is what I’m talking about - here is the McCree curve from his 1972 study:

The PAR number indicates the sum of all the light inside this region of the spectrum. However, it doesn’t match more recent analyses of the wavelengths where photosynthesis occurs - first graph is from Wikipedia, with a couple other ones:

spectrum.jpeg500×651 69.8 KB

These 3 graphs all have big drops in absorption and photosynthetic activity in the green/yellow/orange areas compared to the McCree curve. The professional horticultural LED’s I referenced all have less electricity going into producing those regions, and a greater share going into the red and blue peak areas. So they may have the same PAR number, but they are delivering more red and blue to the plants, and theoretically driving more photosynthesis. This is the concept of a weighted average, yield photon flux (YPF) - versus PPF, which is dumping all the PAR wavelengths into one bucket.

Realistically I don’t think it’s that important to obsess over this stuff compared to “culture” issues like medium, ventilation, nutes, etc. Just trying to explain what the pro hort. companies are chasing in their LED’s, which at this point are all blends of blue, red, and white diodes. Clearly Philips, PL, and Fluence think there is benefit in providing additional “blurple” diodes along with the white ones. hopefully in the near future we’ll start to see studies exclusively focused on sole-source cannabis lighting.

3 Likes
87

Crap! What are we arguing about again? All this photosynthesis talk makes my brain hurt.

I’ll summarize what I think, and I think Muleskinner are actually in agreement;

Most blurple LEDs suck…but a couple are actually good (actually most LEDs suck, but some are really, really good)
Plants use all light on the McCree spectrum from the peaks on the graph you can see that they use more red and blue. That’s the theory behind Blurple.

If you can DIY an LED fixture you can make a pretty good one for maybe 1/2 or 2/3 what it would cost to buy a good one.

If the expensive LEDs seem too expensive get an HID not a cheap LED.

@Baudelaire sells really good, high quality LED strips, if you want to DIY a light his SolStrips are a good deal.

Sorry for highjacking your thread @Baudelaire

4 Likes
88

This is really my focus with the SolStrips. Efficient, high quality LED lights for the non-institutional grower, especially personal and small scale growers, for about $1/watt. They aren’t buying Fluence or Hortiled because they don’t have $2,500 for a lamp, and it’s hard to squeeze a Spydr into a 3x3 Gorilla tent.

I use “blurple” derisively to refer to cheap red/blue lights mass produced with generic, unnamed or tested diodes to supposedly mimic the “improved McCree curve” Muleskinner refers to. I don’t consider the two companies named as blurple sellers. More like “tuned lighting” producers, using real research and real electronics to produce highly specified spectral ratios.

It’s not that the research is wrong, its that 95% of non-white LED grow lights fail miserably at delivering on the promise of more good wavelengths and less wasted wavelengths at much greater efficiencies than HID - and now by extension, white LED. Their diodes, when they are actually disclosed by their manufacturers, are often less efficient than CFLs. Their drivers run at 50% or less of their advertised power. Their circuits are poorly engineered, the PCBs are cheaply designed and fabricated.

Mono color LEDs are inherently less efficient that white diodes due to the nature of the technology. And red spectrum diodes are some of the least efficient of all mono LEDs. One way to look at the red/blue vs white LED efficiency debate is to look at the “wasted” green-yellow spectrum produced by white LEDs as “free photons” because they are so much more efficient compared to mono LEDs.

Except they aren’t “wasted” photons. Put aside the recent studies that highlight green-yellow wavelength’s contribution to plant growth deep in the canopy, which has all the top mono makers scrambling to introduce “lime” colored diodes. Humans use green-yellow light to see cannabis, and the various bugs, molds and plant health cues growers need to observe to successfully grow their crops. Anyone who has used blurple lights knows this is not a small issue.

I’m not against using colored mono LEDs. I’m testing a red/blue SolStrip right now:

I want folks to be able to supplement their white SolStrips with additional red/blue spectrum with a mono diode strip that can be easily mixed into SolStrip arrays.

It’s a great strip. Forty 2w 665nm red and four 3w 450nm blue Cree XPE chips, giving a 7:1 ratio on a 36 watt strip. Only one problem: it costs 400% more to produce than a white Samsung LED strip. And it’s far less efficient than the Samsung strips. If anybody wants a $60 red/blue SolStrip, PM me.

LED lighting discussions have a tendency to run deep on the theoretical and the technical, and give short shrift to the practical and the economical. I’m trying to balance the quest for the best with the desire to deliver a quality light at an economical price. I don’t expect to steal any of Fluence’s business. I do expect to steal a lot of Chinese blurple box makers’ business.

-b420

10 Likes
89

yes i use blurple to refer to cheap low quality garbage lights, not necessarily blue/red monos. i think every manufacturer is experimenting with adding in some reds blues or UVs etc. but if i had to choose between only blue/red or only white, i already chose only white.

1 Like
90

I’m hope we’re not arguing :smile: I just have all these graphs & stuff on my hard drive and like to dump them somewhere from time to time!

It’s not that the research is wrong, its that 95% of non-white LED grow lights fail miserably at delivering on the promise of more good wavelengths and less wasted wavelengths at much greater efficiencies than HID - and now by extension, white LED. Their diodes, when they are actually disclosed by their manufacturers, are often less efficient than CFLs. Their drivers run at 50% or less of their advertised power. Their circuits are poorly engineered, the PCBs are cheaply designed and fabricated.

Yes!!! it’s painful to watch people spend twice as much as a good HPS on these crappy blurples that probably don’t even have the right spectrum. And horticultural lighting companies aren’t really interesting in making hobby-sized lights - most home growers don’t need much horsepower for vegging cannabis, or for a micro-grow flowering area, there aren’t many high-quality options in that niche.

4 Likes
91

I am very happy with mine. 10 strips running at 300w and not even hot to the touch. :smiley:

IMG_2298.JPG4032×3024 1.67 MB

15 Likes
92

Very nice and clean design. What driver did you go with?

93

@Jellypowered What was the final cost and how big is it? :slight_smile: looks really great! I would like to want one as well for my small grow box :slight_smile:

94

Salutations MuleSkinner,

Based on my limited understanding of this complex subject i’d expect a mix of specific and generic, for example:

[ Chlorophyll "2-Level" Ionic Pump - Hledat Googlem ]
Google Images: Chlorophyll “2-Level” Ionic Pump

812×293 54.7 KB

…

This schematic below is about Chlorophyll more specifically, not in cannabis only. It appears to illustrate why the double peaks at 680 and 700 nm based on an analogy to resonant “antennas”. Eventually additional hints are gained, like the 559nm peak and the names of dedicated researchers, for example:

[ https://www.google.ca/searchq=Photosynthesis+Antenna+Calvin+Benson+680+700+550+559&tbm=isch ]
Google Images: Photosynthesis Antenna Calvin Benson 680 700 550 559

811×419 124 KB

So i suppose using similar Googling tricks one might find more useful information to help in narrowing the searching parameters somewhat, which wouldn’t hurt a bit trying to find our way through such jungle!..

Good day, have fun!! :peace:

3 Likes
95

I’m currently using a 24v 20a adjustable switching power supply, Saving up for a meanwell right now, the HLG-480H-24A will be my driver once I can afford it.

@Fisch If you add the driver I haven’t purchased yet, the total will come up to $506.00 I have a cost breakdown somewhere above in this thread.

3 Likes
96

Salutations,

Another question: Who many “bands” can there be?

For example, from 380 to 880 nanometers if each band is 25 nm wide then that’s 500 divided by 25, which totals to 20 bands 25 nm wide each.

Good day, have fun!! :peace:

1 Like
97

Ok nobody mentioned it here :smiley: or atleast I didnt see it but if I would like to have 10 strips for my 3x3 space. what kind should I use of I want one panel from seed to flower?? 2700, 3000,3500 or 5000K??

Thanks :slight_smile:

2 Likes
98

Personally, I would go with the 3500k for a light that can be used to veg and flower. The higher Kelvin are also usually more efficient due to the coating on the diodes. A mix of 3000K and 3500k could be nice too.

4 Likes
99

And what wave lenghts does the solstrips got? How much Red and so…?:slight_smile: Thanks

100

Hi there @Fisch , I just post the intensity vs. wavelength graphs for the different color temperatures for the LM561c diode which the SolStrip uses. This should be what your looking for:

https://overgrow.com/t/diy-led-150w-solstrip-build-incomplete/6828

3 Likes