What are the best COB leds nowadays? onboard drivers good?

Discussion in 'Lighting' started by ClydeWalters, Nov 22, 2018.

  1. Cool. You should check out the solo grow. Good bit of fun. Lots of different grow styles. Doesn't take up much space.

    THE INFAMOUS #GrassCitygrowmies SOLO CUP CHALLENGE 8?

    So. Do you have any good knowledge to lay on me about these full spectrum cobs? I understand it's not the safest prototype in the world, but they have been working. I'm most curious about the film that is used to turn blue light into what everybody uses. Most on here are fully convinced that the warm whites are best. I'm not so sure. I understand why blurples suck, not enough spectrum. I'm fairly convinced these newer full spectrum films don't have the same issue. That said, I do mix in white. Opinion?

    Sent from my XT1093 using Tapatalk
     
  2. The white light LEDs are more full spectrum than those blurple colored LEDs. Just look at the spectrum chart for HLG quantum boards
     
  3. They look a little lacking in red and high in green and yellow. These cobs seemed pretty good, but lacking the other direction. That's why I mix them with white. I had always heard the reds and blues were more important. Nobody ever grew under yellow halogen. Trying to suss out what's real and what's just been passed along. I see lots of commercial grows using the full spectrum, not blurple. There is a difference. The full spectrum uses the same type of film as the whites just with different phosphorescent spectrums. The underlying chip is the same efficiency. I might just have to change my grow to do a side by side.

    Sent from my XT1093 using Tapatalk
     
  4. The hlg quantum and the cheapo China full spectrum.[​IMG][​IMG]

    Sent from my XT1093 using Tapatalk
     
  5. You aren’t looking at the right spectrums. White light leds are very high in reds and can have blue spikes in 4000k. A much more full spectrum than any of the blurples
     
  6. Much better example. The Samsung diodes are top notch for a reason. Wouldn’t waste time with anything else currently except the qb96elite v2’s
     
  7. I see... The qb96 v2 does look like a pretty good spectrum. I don't doubt that they work. I've been following your grow. I do like the output for the footprint too. What if the Samsung diodes were the spectrum of the China light? Would you consider it, or are you against the off color lights in general?

    Sent from my XT1093 using Tapatalk
     
  8. I run a mars Pro II 160 that I’m putting one of each of their new SP-150 on either side for autos. They were free so I’ll put them to work. The blurples just don’t grow as well as full spectrum white light LEDs. They hurt my eyes and hard to diagnose issues as well. It’s just old technology and I don’t like wasting time with it. Mars Pro II has better diodes than most blurples but the qbs are still king
     
  9. Yeah. I don't see any Mars light with the same tech. These aren't blue and red. They're not like epileds with multiple spectrums on one chip. It's a film just like the yellow that is on a white LED, except with a different spectrum. It IS hard to look at and hard to diagnose anything under, but it's not the old tech.

    Sent from my XT1093 using Tapatalk
     
  10. Phosphor

    Check this out.

    Sent from my XT1093 using Tapatalk
     
  11. I just want the best lights for my investment. And I have them now so until something a lot better comes along, this is where I’ll be. No other diodes compare in efficiency, efficacy, spectrum, output, and versatility.
     
    • Like Like x 1
  12. It definitely seems to work good for you.

    Sent from my XT1093 using Tapatalk
     
  13. For anyone really. It’s an easy way to become a better grower. Buy the best lights and set them right with even spread and 1000 ppfd across the canopy. The light is the engine after all
     
    • Agree Agree x 1
  14. Look at chlorophyll absorption spectrum. It was used as the basis of a NASA hydroponics study. This lead to blurple lights to maximize chlorophyll A and B. Also single wavelength LEDs are more efficient in umol/J because there is no phosphor conversion step so burple lights claim crazy efficiencies especially back when white light was inefficieny. Efficiency of white LED has more than doubled in the last 5 years.

    Ive seen different graphs for cannabis but the principal is the same: green, yellow, orange range common in white light drive other metabolic processes besides chlorophyll beneficial to the plant. (And efficiently at that). This is why balanced spectrum gives healthy plants. chloroabs.gif par_light_mccree_curve.jpg

    Sent from my SM-G950U using Grasscity Forum mobile app
     
    • Like Like x 1
  15. As for cobs. Id do min CRI of 80. I havent done much study if 90 CRI is better or worse. Lumens are less on 90CRI but was you should know lumens is white light not PAR. I'm just against the driverless COBs. Dangerous, especially if it isn't a name brand, and not efficient. Samsung has a 10w that looks like this but is LM302B instead of cob.

    COBs themselves are just blue leds on a substrate with silicone phosphor poured on it like 1 big LED. They are less efficient than individual chips because their heat is dense. They were designed for inexpensive designs like can lights, br30, par20 type bulbs. They caught on because distributors like digikey were selling and people figured out how to put together and because they were legit LEDs they performes way better than cheap blurple lights. A QB array style is more efficient than cob every time. Their upsides are cheap, easy to find heatsinks, and modular. They also have optics more readily available that can help with canopy penetration. lux2.gif

    Sent from my SM-G950U using Grasscity Forum mobile app
     
    • Like Like x 1
  16. It that phoshor conversion that has me spinning. White LEDs are blue underneath. How is that any more efficient? Same type of tech. No single spectrum beside the deep red on the qb.
    [​IMG]

    Sent from my XT1093 using Tapatalk
     
  17. It's the most cost effective way to make white light. You could use individual spectrum LEDs and try to fill in evenly but there would be gaps because making different colors is a chemistry issue (think fireworks - doesn't work the same but same concept of different base materials). So it is hard to make every color under the rainbow.

    Then each die material technology has different efficiencies for par/watt. The pickups we've seen in horticulture lighting are from the benefits of having full spectrum (not gaps in the spectrum). Only way to get this is to add white light. You'll still see blurples with white thrown in there and this is why. So even though the white light doesn't have the same umol/J efficiency of some of the blurple lights they do better because of the absorption spectrum vs the action spectrum. So while PAR (PPF) is very handy unit of measurement understanding efficiency, understanding spectrum is beneficial too. And different plants respond differently to spectrum too. Basil and romaine respond opposite to each other with cool vs warm light even though they are both small green leafy plants.

    Shuji Nakamura at UCSB had a breakthrough in the the early 90s with GaN as the die material and revolutionized the efficiency of blue LEDs. So even with light loss effect of the refraction step it is still very efficient.

    Seoul Semiconductor, which always has cool products, has a purple diode they phosphor coat and makes an really good spectrum out of white. . Problem with them is they are crazy expensive, impossible to get parts through distribution, and a warning for any mfg out there look at them is as soon as you design your part in and commit to the 100K pieces MOQ from their factory they discontinue it. I've given up on using them ever for anything but always had cool technology.
     
  18. Just want to make sure we're all talking about the same thing. These are the lamps I'm referring to. Not blurple with white added.[​IMG]

    Sent from my XT1093 using Tapatalk
     
  19. I know, that started a conversation that got me rambling away on how light is made. I wouldn't use those purely because you hook them to AC from a safety standpoint. Secondarily your efficiency will be sub par both from its die quality and circuit design. If you know those things go for it and get one. I wouldn't.

    Better off assuming a Cree or Samsung COB is 2x as efficient and designing for half the power as these although its still going to cost you more. If you do that though you can get meanwell apc-25 pretty cheap and make 2 25w modules.

    I'm not condoning this but I like this driverless link from amazon because they show you whats going on and it as a NTC resistor in case you get thermal runaway. The only thing your version has going for you is that it is 110 vs 220 and it looks like it might, emphasis on might, have a fuse. The one I linked doesn't appear to which is a crucial component. I'd try to get a cord from xmas lights that have that fuse in the plug and then use these in case something happens.
     
  20. I like the Christmas cord idea. I've added a thermal switch to shut down if it gets too hot.

    Sent from my XT1093 using Tapatalk
     
    • Like Like x 1

Share This Page