Is this a Vero 29 killer?

Discussion in 'LED and other Lighting' started by JorgeGonzales, Apr 19, 2016.


    DrMisunderstood Member

    You're a funny guy. I don't know how to power CoBs???? I'm not the one powering them in parallel. Tell me this, when connected in parallel how do you balance the current to each CoB? Are you familiar with the pitfalls of connecting them in parallel, such as thermal runaway and uniformity?

    Did I say something that offended you? Do you think I have something against CoBs? CoBs have inherent limitations when compared to discrete LEDs. That's fact not opinion.

    There is no way a CoB's performance can be superior to discrete LEDs.

    You do realize CoBs are made with the same die as a discrete LED? It's only packaging. With discrete LEDs a design can tune performance to the task at hand with many LEDs to choose from. One place where a discrete design is going to kick a CoBs ass is when the discrete design can tune the uniformity and drop the fixture closer to the canopy than a CoB fixture. The CoB's 120° view angle with the bell curve is not your friend. A discrete design is going to get a bump in performance when it cools the single die down to a temperature lower than a CoB die can go. A heatsink can be positioned closer to a discrete die giving less thermal resistance. When the dies are spread apart, thermal management is simplified.

    And then there is the one undeniable fact where there can be no dispute. The Deep Blue discrete LED that is NOT covered with a layer of phosphor. That being the same Deep Blue LED used in your CoB that is covered with a layer of phosphor.

    If a CoB's Blue LED with a Red Phosphor Converter were more efficient at creating Red, why do Red phosphor converters not dominate the Red LED market?

    So what is it I am missing?
    MrTwist1 and Randomblame like this.

    DrMisunderstood Member

    I cannot disagree with you on that. Maybe we need to work on that. Putting cost and efficiency aside, can anyone here match or beat an HPS?

    I think the solution is in a Red White and Blue fixture. Have you seen the Cree Horticulture Reference Design? It's a 553 Watt White (144 XPG3) Red (48 XPE Deep Red) fixture. It's a very expensive design with lots of room for cost reduction. I estimate it would cost $800 to build one using their BoM. I think I may be able to build one for under $300 plus $100 for the power supply.

    The Cree HPS test results were done with the Cree fixture at a height of 4.9 feet and the Gavita 1000W HPS at 3.2 feet. I believe there is tremendous opportunity to lower the fixture increasing canopy radiant flux 5x or more and still match, if not improve, the uniformity.

    An inexpensive fixture that costs less and exceeds the performance of an HPS. I think that would be a good place to start. Want to try?

    I need to get some 0.0625" copper bars drilled and tapped for 4/40 screws. I also need a break press to crunch the ends of the copper bar to have an 0.3125" radius. If someone here can do this or knows someone that can for a reasonable cost I would appreciate the help. Also I would like to purchase a vertical mill (to drill and tap the copper bars) if anyone has one they want to get rid of.
    Silver or lead

    Silver or lead Well-Known Member

    Lol, I'm not trying to be an asshole here.

    There are no pitfalls wiring in parallel. Thousands and thousands of cobs have been wired in parallel and I've not seen one single example of thermal runaway or lack of uniformity.

    There really is no upside to wiring in series and there is actually a couple of disadvantages. First off high voltage DC can be some scary shit. Second if an led in series goes down then the whole array goes down, this is not the case when wiring in parallel. Thirdly wiring in series is much less forgiving, you have to have exactly the right driver and number of leds to make it work. Parallel wiring does not have this problem I can hook up 3 or 30 cobs to my drivers because they don't care.

    Next up the picture you posted actually shows how shitty monos are at growing, the plant under the full spectrum look better. I don't care how efficient the red and blue monos are, they lack a healthy full spectrum which plants need to maximize their growth so they cannot compete with full spectrum leds. Your picture clearly shows this.
    Silver or lead

    Silver or lead Well-Known Member

    Pretty much everyone here using cobs already does. Tons of examples here of people ditching their hids for cobs because they perform better at less wattage.
    Heil Tweetler and sixstring2112 like this.

    CobKits Well-Known Member Rollitup Advertiser

    all day long on yield and quality and operating cost over its lifetime

    blue is kind of useless we get all we need from warm white. its very efficient and you dont need much. warm white augmented with red is where people are moving to now that red LEDs are much higher flux then just a year ago.

    yes its been discussed here ad infinitum. Its a joke of a study they did there discarding half the light from the gavita which fell outside their test area. The entire test was biased towards their focused fixture

    were already there with cobs. for the price of a gavita and a bulb and reflector change you can build a a 700W COB rig that lasts 20 years and returns $150-$200 a year in electricity savings

    nogod_ Well-Known Member

    What's offensive is your pompous attitude.

    I hate to break it to you but no one cares that you grow cucumbers for some secret university.

    This is a ganja growing site. Hundreds of members grow ganja here with myriad spectra and intensities of both discrete diodes and cobs and submit their results here every single day.

    Believe me when I say, you have a lot more to learn from all the data being presented here than we have to learn from anything you are bringing to the table, all your grants and studies be damned.

    Your conventional wisdom does not apply here. Everyone here has been warned countless times about thermal runaway and current hogging from wiring in parralel and people still do it anyway. Do you know what the result is? Sticky icky and lots of it!

    Now hit the books junior!

    t-bot, nfhiggs, Greengenes707 and 2 others like this.

    DrMisunderstood Member

    Very nice PDF. I am very familiar with the Luxeon Color C and Sun Plus. I did not know about the XtraLight fixture.

    RE Strips: There are 3 strings of 16 LEDs per strip. The current new layout has Every string has a Buck LED driver with analog and PWM dimming and thermal fold\back.
    I had been looking at Zigbee to control the PWM dimming of each string. Now I thinking a 3 channel grey scale PWM shift register (e.g. WS2812) driver to drive the Buck Driver's PWM.

    My new layout has the Luxeon Color C foot print (same as SunPlus 20). My only concern is the footprint size of the Color C and SunPlus 20 is so small there may be manufacturing issues.

    Below is the PCB I have been using the past year. Just ran out. This is two strips of two strings of 16 LEDs. One string of Luxeon Rebel and one Cree X-Lamp.


    The theory behind the PCB layout is to have the thermal pad that is run on a 2mm, 2oz copper pad directly to a cooled copper bar.

    Where I have the thermal vias that are recommended to connect to a heatsink on the opposite side of the board, I now only use them to measure the case temperature.
    I figured it was better the connect the heatsink directly to the copper pad on the top side of the PCB rather than dealing with the thermal resistance of the board and saving the cost of a metal PCB. The thermal path from the LED's thermal pad to the copper heatsink is 100% copper.
    Randomblame and SaltyNuts like this.

    CobKits Well-Known Member Rollitup Advertiser


    DrMisunderstood Member

    Thanks, you have been so kind. So you are not a jerk to other newbies on this site, just being a jerk to me because you mistakenly think I'm a pompous asshole? Not that I give a fuck, but you have a few other things wrong also.

    I am not an employee of the University, my son in law is a professor there that heads all the Horticulture LED research.
    My son-in-law hired me to design some LED fixtures for their research use. I am a consultant paid by the University.
    I've been doing this shit for 40 fuckin' years and I don't need some jerkoff that misunderstands me, calling me Junior.
    I started growing weed in 1975.
    And it is very presumptuous of you to assume I have nothing to bring to the table. Whether I do or not is irrelevant, but what is known you are being presumptuous. A presumptuous fuckin' jerk.
    Please never reply or comment on my posts. I don't need your bullshit attitude.
    A guy and MrTwist1 like this.

    sixstring2112 Well-Known Member

    Hey did anyone say happy 420 in here yet?
    :weed: :hump: :eyesmoke: bongsmilie:bigjoint:

    SaltyNuts Well-Known Member

    @DrMisunderstood I don't think your comments are misplaced although a lot has already been addressed here and on other forums in the past, but I see no harm in revisiting stuff for review or for newcomers, or simply for the role of devil's advocate. I think people are prone to confirmation bias and there is somewhat of that trait here but in truth cobs seem to work legitimately well enough to please many people. The plants seem to do well. Pure science is great and I welcome the insights into that and recognize that's the only way to conclusively prove anything, through exhaustive testing and research. Research is ongoing because the field is still full of mysteries. That is good. Then there is what people feel to be true based on their own observations. In practice it's clear that white light is generally used by plants to good effect. One source of passion for people here is that the use of white light cobs is getting results for people and is in fact helping to bring a new perception of credibility to horticultural LED's in popular use. The poor history of dim LED's sold on the promise of efficient spectra has been ameliorated by the power of the white light. The concern about cobs having potential heat issues is well founded but in reality most are run at lower amperage with excellent heatsinking to run significantly cooler than rated. Although some people run them hard. They do present an increasingly economical source of white light. I will go so far as to argue that most plants will tend to benefit the most from a balanced white light that mimics sunlight, in terms of overall growth and vigor through the life-cycle. But, obviously, the "imperfect" light from an HPS bulb will produce desirable traits in many cultivars. So go figure. I look forward to the day we have a more perfect understanding of light and plant light metabolism and can tune light spectra with more predictability. Stony segue: My guess is that buried in the plants genetic memory is the ancestral history from early algal species that lived deeper under the water surface to avoid the ultraviolet rays, and so became more efficient at using red light because there is less of it deeper down. Plus the relative energetics of the red spectra is favorable in some ways, to them. Blue light is used less efficiently partly because it is easier to get. Sic green light as well. And that is an example of philosophical meandering thought that has enough to do with observable reality that it sounds plausible. Good enough for me. Don't let people here get you down, I like your commentary. This should be a place to present facts as well as opinions. If anyone thinks I too sound pompous or "talk like a fag" (Idiocracy reference) then go fuck yourself, smoke some more weed.
    Last edited: Apr 20, 2017

    sixstring2112 Well-Known Member

    cobz rock :eyesmoke:

    big difference from tomatoes to cucumbers to kush to panama red to wheat lol :wall:


    SaltyNuts Well-Known Member

    you forgot algae.

    I love your grow journal Six, you are one sick fuck. How do you do it? Do you put gnome bones and Elf blood in your soil? With your fucking 12 plants or whatever in constant rotation. Two pounds a week. I hate you. I want to be you.
    sixstring2112 likes this.

    DrMisunderstood Member

    Thanks for the feedback. I was beginning to think I stepped into a huge pile of shit. Good to know everyone here is not a cob snob.

    It must of got lost but I have said I have nothing against CoBs, but they do not work in research and that is what I do. I have nothing against whites. Probably half of the boards I have built in the past years have been white. I'm using white to do landscape supplemental lighting and inter-canopy lighting experiments.

    I assume there have been some ugly debates here in the past and I have been mistakenly singled out as one of those guys. I'm not one of those guys. I have no bias. When I say something I believe it to be true and I do the research and or fact checking to make sure what I say is factual.

    Lumens is a pet peeve of mine. Lumens are used way too much here to the point where the facts get distorted. I posted here to vent about Lumens and the way they were being used in this thread. I really do not give two shits about CoBs. I understand the economics. The economical solution is not always the best technical solution. Sometimes I am backing a very small trivial point. But what I have learned over the many years is the small stuff counts. It adds up. What I have been hearing a lot of here is: hey it works, that good enough for me. That's not good enough for me. I want perfection. That's always worked best for me. At least I try.

    SaltyNuts Well-Known Member


    DrMisunderstood Member

    That's not a bad product. Good for someone that does not have a PCB, I guess that's their market. I think the 390 W/mK is bullshit. The layer of solder and the dielectric are going to reduce thermal conductivity more than 10 W/mK below copper. That's why I attach the heatsink to the LED side of the board and not the backside, to avoid the dielectric of the MCPCB. And FR4 is much cheaper even if I go 4oz copper. I do like that their PCB Metal Core is copper rather than aluminum. I love copper.
    Silver or lead

    Silver or lead Well-Known Member

    You should go into hybridways and a couple others threads and see some of the lights these guys run.

    With cobs as a base and running red and blue monos for spectral supplementation, try to capture the Emerson effect, promoting flowering with with 730nm after dark and using Agromax Pure UV to promote turpine and trichome formation. There are people that do all that at the same time and still come here to read because it is the cutting edge of lighting, because we are always learning.

    So when you come here with these claims that the red blue burple mono people claimed and called cob guys whatever is nothing new around here. They all have heard it all before.

    But the cob guys won with results so if you want to impress these guys you are going to have to bring results.

    You should stick around and read for a while. Men in your position would be fools to not utilize this resource.

    DrMisunderstood Member

    That is the plan. I'm thinking about making it an open source project. I'm not quite there yet. I need to make some more PCBs. I am trying to find the perfect distance between LEDs on the strip.

    My avatar is a screen shot from an app I wrote. The 3 pyramids are the view angle with the percent flux from the radiation pattern for 3 CoBs. I calculate the radiation from the CoB for every angle. Then I set a 48" floor under the the CoBs and start at one end of the floor and measure the distance to each CoB. Add them up, save the sum and move along the floor 0.10 inch. And repeat until I have the relative radiant flux factor for all 480 increments along the floor from all the CoBs. I then change the height of the floor from 6 inches to 50 inches and record all 480 points for each height. I then change the distance between the CoBs from 1 to 48 inches and Repeat.

    That gives me 48 (X) positions times 44 heights (Y) or 2,112 sets of 480 points. I then calculate the standard deviation for all 2112 sets. I then can find the ideal height and distance to place the CoBs above the canopy. I believe the one in my avatar is the 2200K Luxeon Fresh Meat with a 120° view angle. The ideal height was 12 inch with a distance of 17 inches between. The standard deviations were too high and will not be able to use CoBs in the intended project due to the close proximity to the plant and I have to keep the PPFD below 150 µmole for inter-canopy supplemental lighting for a Manhattan high rise condo that does not get enough sunlight.

    I am using the same app to find the ideal distance between the LEDs on the PCB for the Gavita killer. Except I will start at a height of 1 inch and try various floor lengths to find the ideal length of the PCB. I only have to keep the length under 22" to keep the cost of the PCB under $4. I use a more expensive PCB laminate. It's a high temperature version of FR4.

    Just an example of how I sweat the small details. Unfortunately it takes time.
    SaltyNuts likes this.

    DrMisunderstood Member

    I am being misquoted. I have never promoted mono wavelengths.

    I have said I prefer discrete LEDs over CoBs but never specified any color. The purple LEDs were brought up by someone else and I only gave my opinion. And actually I said I do not like the purple.

    There is almost no published studies on cannabis. Lots on tomatoes, lettuce, and basil. There are rumors that the feds may give in and the Universities will be able to do cannabis studies a couple of years down the road. So guess what? You guys are the one of the few places that have data on cannabis. But you already know that.

    I'm not that guy that came here to make trouble. I have 40 years of electrical engineering and had a manufacturing businesses for 20 years.

    I think you guys are doing fine with the CoBs. I think you you can do better. Mainly to improve uniformity to get the LEDs closer to the canopy. Much closer. I do not care what color, just what ever works the best.

    If we can agree that 1000 µmole PPFD is the same whether it comes from a CoB or a PCB with discrete LEDs then we don't have much to disagree on.

    I certainly have not said anyone here is doing any thing wrong other than using Lumens to make bad decisions. I have expressed one opinion on CCT and CRI and that is within a the same series of LEDs (or CoBs), the Deeper the Red, the better. The higher the wavelength on the Red Peak is better and more important than any other characteristic. That's my opinion, not a fact. Do I care to debate that, NO. Live and let live.
    MrTwist1 and Randomblame like this.

    Randomblame Well-Known Member

    Interesting app! I would like to know more about these it, is it available open source or is there a beta version? Is it iOS or Android based, are other tools needed, Windows, sensors,... I'm waiting since a long time for an app with the abillity to use the full potential of the build in camera sensor but unfortunately there are only millions of lux-meter apps, Lol!

Share This Page