Driverless AC COB LED build

[ZxcStaz]

I just purchased (70) - 50watt 110V AC COB LED chips, and I am going to create a water-cooled aluminum matrix to power my 10' x 10' flower tent. This 3.5KW lighting system should deliver optimal PAR while keeping the grow chamber below 85 degrees Fahrenheit. If I am successful, this should be a breakthrough for small growers, and data for future lighting systems.
This system is created from scratch, and the design is my own. It is created from a compilation of research, thoughts, and visions. If I impinge upon prior designs, I would claim serendipity, as I have viewed no schematics detailing my present idea. All in all, I doubt that anyone would endeavor to create such an infernal device as my own, because mixing mains voltage with water is pure insanity. Alas, chaos is where I dwell, as my mind is asunder. Therefore, I present you with this, my Daedalus invention:
The LED's are (50) 380nm - 840nm full spectrum 50 watt chips, (10) 4000K white COBs, and (10) 7000K white COBs. The spectrum should mimic solar radiation with an emphasis on the blues and reds. These chips will be wired with break-away connectors onto the cold plates.
The cold-plate/heat-sinks will be self-contained on a standalone recirculating system powered by a 17L/min recirculating pump. The heat will be dissipated to waste in the summer and redistribute to habitation in the colder months. Utilization of a non-electrolytic or polar medium should circumvent corrosion at the expense of higher heat capacitance, yet afford infinite system recharges from cold city water in the summer months. Conversion to an ethylene-glycol or suitable alternative carrier in the winter will be ascertained based on the summer performance.
As the project commences, I will post updates and pictures. Comments, suggestions, experiences, ideas, and criticisms are implored. Please refrain from negative, inflammatory, or hurtful posts; they are not conducive nor warranted. This thread may deviate from initial conception, as long as it contributes to our knowledge of cannabis cultivation. Here is my plan...

- ZXC

 

[1212ham]

I have a little experience with water cooling and look forward to your project. I'm interested in the cold plates/sinks you will use?

 

[CobKits]

unless your power is DC, in all cases "driverless" means "i am replacing drivers with these inefficient on-board AC-DC circuits"

spend less now... or spend less later

 

[nfhiggs]

I just purchased (70) - 50watt 110V AC COB LED chips, and I am going to create a water-cooled aluminum matrix to power my 10' x 10' flower tent. This 3.5KW lighting system should deliver optimal PAR while keeping the grow chamber below 85 degrees Fahrenheit. If I am successful, this should be a breakthrough for small growers, and data for future lighting systems.
This system is created from scratch, and the design is my own. It is created from a compilation of research, thoughts, and visions. If I impinge upon prior designs, I would claim serendipity, as I have viewed no schematics detailing my present idea. All in all, I doubt that anyone would endeavor to create such an infernal device as my own, because mixing mains voltage with water is pure insanity. Alas, chaos is where I dwell, as my mind is asunder. Therefore, I present you with this, my Daedalus invention:
The LED's are (50) 380nm - 840nm full spectrum 50 watt chips, (10) 4000K white COBs, and (10) 7000K white COBs. The spectrum should mimic solar radiation with an emphasis on the blues and reds. These chips will be wired with break-away connectors onto the cold plates.
The cold-plate/heat-sinks will be self-contained on a standalone recirculating system powered by a 17L/min recirculating pump. The heat will be dissipated to waste in the summer and redistribute to habitation in the colder months. Utilization of a non-electrolytic or polar medium should circumvent corrosion at the expense of higher heat capacitance, yet afford infinite system recharges from cold city water in the summer months. Conversion to an ethylene-glycol or suitable alternative carrier in the winter will be ascertained based on the summer performance.
As the project commences, I will post updates and pictures. Comments, suggestions, experiences, ideas, and criticisms are implored. Please refrain from negative, inflammatory, or hurtful posts; they are not conducive nor warranted. This thread may deviate from initial conception, as long as it contributes to our knowledge of cannabis cultivation. Here is my plan...

- ZXC

Curious about your cob choices... Links? why those particular cobs?

 

[wietefras]

Please refrain from negative, inflammatory, or hurtful posts; they are not conducive nor warranted.

Sorry to ignore your notice, but all the "driverless" COBs we have seen here before, perform very poorly. You are not the first to try these.

You can indeed course grow something under them like you can with CFL even, but the quicker you understand not to use them the sooner you reap the rewards. Especially since you are willing to go through the extra effort and cost of water cooling.

Those AC COBs might be cheap to purchase,but they burn a lot more electricity which means your costs will be higher in the end.

 

[Rocket Soul]

Subbed, lets see how you go.

 

[ZxcStaz]

Whew, let me try to address these questions and comments, but before I do, let me just say thank you guys for your interest and contributions. Your input will help me with this project. I'm pleased to have multiple perspectives and inquiries, as it expands my thought process and increases the likelihood of success.

I became interested in the AC COB's when I came across an article claiming increased electrical to light efficiency. The article detailed the direct conversion of alternating current to light utilizing an array of LED's. Previously I thought that the electron movement had to converted from the back and forth motion to circular direct current before being converted to light. It seems that now we have developed technologies to circumvent this step. Here is a link to one of the articles I came across that more eloquently details the process.

https://www.ledsmagazine.com/articles/print/volume-11/issue-7/features/developer-forum/driverless-led-light-engines-deliver-up-to-93-efficiency-and-no-perceptible-flicker.html

The COB's I purchased were acquired on eBay. They are inexpensive 50 watt chips. I searched for the least expensive chips considering price and shipping. I cannot yet provide more information about their performance. I will post these data as the project develops.

The cooling aspect of the project is simplistic in design. I intend to make a heat sink with four 3/4" square aluminum pipes side by side. This will make a 3"x3/4" rectangle that I will mount the IC's upon. The pipes will be eight feet long and have the COB's positioned down the length approximately every six inches. Thermal adhesive and screws will retain the chips, and the inner two pipes not breached by screws will contain cooling fluid. The fluid will circulate from a reservoir, through the pipe array, and back to the reservoir. In the summer months the hot water will drain to waste, and cold tap water will replenish the reservoir. In the cooler months it will dissipate the heat to the house and be a closed loop.

 

[CobKits]

the cobs you bought are prob in the 80s in efficiency in their onboard driver circuit, and the diodes themselves are low quality, efficiency and longevity

at the end of the day they are likely no more efficient than HID

quality $10 chips (or mid power strips, also cheap) air-cooled would almost certainly outperform those cobs water-cooled

i dont want to discourage your water cooling project but a the end of the day you will be limited by the chips you are using

 

[Aolelon]

I watched a video on the 50w smd 100w cobs off ebay and they are pretty much trash, with half the diodes not even working in them. I would be weary, maybe test them if you have the ability.

 

[ZxcStaz]

The chips are scheduled to arrive on Wednesday. I'll solder them up and test them before I adhere them to the heat sink. I still have to obtain the aluminum squares, but that company is only open during the week. Considering that they were $160 in total for 3500 watts of COB LED chips delivered, if only half of them work it will still be score. I'll post the pictures once the testing commences. I will do more extensive testing once they are on the heat sink. I'll take some PAR readings and I'll read the actual wattage draw per chip and average it out. Those data should inform us as to their performance.

 

[1212ham]

I became interested in the AC COB's when I came across an article claiming increased electrical to light efficiency.

https://www.ledsmagazine.com/articles/print/volume-11/issue-7/features/developer-forum/driverless-led-light-engines-deliver-up-to-93-efficiency-and-no-perceptible-flicker.html

The COB's I purchased were acquired on eBay. They are inexpensive 50 watt chips. I searched for the least expensive chips considering price and shipping. I cannot yet provide more information about their performance.

The thing is, the cobs in that artificial are almost certainly far more advanced and expensive than the cheapest parts on ebay. You may want to reconsider, cheap parts have a way of getting expensive. On the other hand, you already have the cobs and they can be upgraded later.

I guess it's just personal preference, but I wouldn't double the number of tubes to provide a place for screws. I used zip ties on this water cooled strip light.

 

[CobKits]

I'll take some PAR readings and I'll read the actual wattage draw per chip and average it out. Those data should inform us as to their performance.

spot par readings will tell us nothing about their performance unfortunately

 

[ZxcStaz]

I guess it's just personal preference, but I wouldn't double the number of tubes to provide a place for screws. I used zip ties on this water cooled strip light.

Very nice! Beautiful array. Did you weld the square tubes to make the closed loop? Additionally, did you put thermal glue between the metal plates and the tubes while securing them with the zip ties? I like that configuration, it is similar to what I envisioned. The connecting tubes act as a manifold for the distribution of the cooling fluid, right? I considered using aluminum strips to increase the width of the plate, but I am thinking that the conjoined tubes will increase surface area and fluid turbulence allotting additional thermal conveyance. It will be an added expense, but the total heat capacitance will be increased too. How hot does your system run vs. flow rate? Thanks 1212ham, much appreciation!

 

[ZxcStaz]

spot par readings will tell us nothing about their performance unfortunately

Without including an alternative testing method in your statement, your statements are making me question their altruism and motives...

 

[CobKits]

no motives other than to educate

spot par readings tell you nothing about a lights efficiency other than the fact that it is producing light

at a very minimum you need a chip of nearly identical CRI and CCT and known efficacy on the same cooling setup, to compare it against and even then its just an estimation. if you can find a chip that is close on spectrum you can measure both individually on axis at a fixed distance and plot ppfd/W vs wattage for both chips like this. Those were actually measured in a sphere but you can get close with on-axis measurements in a dark room

 

[ZxcStaz]

spot par readings tell you nothing about a lights efficiency other than the fact that it is producing light

I might add that they are producing light in the photosynthetically active range, but the wavelengths of light vs. biological anabolism of biomass is another discussion altogether.

Sure, if we are to compare apples to apples, we would need to test all of the available LED chips within the same parameters, with the same, well, everything. Alas, I cannot replicate lab tested samples. I do glean information from the graph you linked.

From the graph regarding the 50W chip efficiency range there was a variation of approximately 39% - 43%. That deviation of 4% should influence decisions regarding purchase price vs. returns. So, If I do a quick cost to benefit analysis, the $2 eBay chips are 1/5th the cost of the $10 chips you mentioned, yet only lose 4% of the performance. This analysis is based on the posted graph, assessed as the least performing chip, and may not be indicative of actual performance. That being said, I still do not have a viable testing method that can ascertain the efficiency that would satisfy your objections to the utilization of the aforementioned chips.

I do find it interesting that as the power was increased the efficiency decreased in the tested chips upon the graph, and I make the assumption that it is correlated to the heat induced resistance within the chips. This may be the efficiency increase with effective cooling that you alluded to. Thus, I am pondering the optimal temperature of the cold plates with consideration of the grow room chamber. Of course an IC would function best at 0K, but the plants might object.

 

[CobKits]

Sure, if we are to compare apples to apples, we would need to test all of the available LED chips within the same parameters, with the same, well, everything.

well you really only need one known chip, not all of them

From the graph regarding the 50W chip efficiency range there was a variation of approximately 39% - 43%

those arent "percent efficiencies", its unitless. more like the best chip is about 6% better than the worst (but those are all HQ chips)

So, If I do a quick cost to benefit analysis, the $2 eBay chips are 1/5th the cost of the $10 chips you mentioned, yet only lose 4% of the performance.

Thats a baseless assumption. You have no data on your chips so you cant draw any conclusions. the $10 chips i mention would be near the bottom of that set of curves.

1000W of chips that is 10% less efficient than another will cost you about $70 a year in extra electricity. Any you have no lumen maintenance data on the cheap chips. they could be 10% less efficent to start, then 20% less efficient after a year ($140/year per 1000W), then they might die and youre rebuying chips again. TBH if they were 10% less efficient at 50W id be impressed, id guess 20-40% less efficient out of the box

That being said, I still do not have a viable testing method that can ascertain the efficiency that would satisfy your objections to the utilization of the aforementioned chips.

im not objecting, just trying to save you some headaches by letting you know that dozens of people have come before you looking for the silver bullet of high performance for pennies on the dollar, and wasted lots of time and money in the process. same for all the people who post weekly regretting their mars and other cheap options.

I do find it interesting that as the power was increased the efficiency decreased in the tested chips upon the graph, and I make the assumption that it is correlated to the heat induced resistance within the chips.

thats fundamental to every LED. Heat is a factor and you can use any of the simulators from cree, luminus, citizen, bridgelux, etc and change temps and see the effects

Thus, I am pondering the optimal temperature of the cold plates with consideration of the grow room chamber.

it takes energy to move and dissipate heat with pumps and fans and such. removing heat works but there are dininishing returns as you are using energy in the process. most people dont have a need for water cooling as then the plants are out of the optimal 80-85F temps which work well with LEDs

 

[ZxcStaz]

those arent "percent efficiencies", its unitless. more like the best chip is about 6% better than the worst (but those are all HQ chips)

Are we looking at the same graph? What do the abbreviations on the side and bottom mean?

You are correct, I don't have data yet. I have been asking you how to obtain data to gauge the chip performance that will correlate to the tested chips. Is there a test that I can perform that will quantitatively assess the performance, as to compare it with "HQ" chips, without utilizing a research lab?

that dozens of people have come before you looking for the silver bullet of high performance for pennies on the dollar, and wasted lots of time and money in the process.

Please reinforce these statements with citations or links, no offense, but there should be justification to these statistics.

Thats a baseless assumption. You have no data on your chips so you cant draw any conclusions.

I based my assumptions upon the lowest performing chip in your graph, as stated in the response. Without data, nor an adequate method of testing, we can only base observations on the information available.

Regardless, I'll post the development of the array as it progresses. I'll provide pictures of the components and the creation process ideologies, unbiased. Any data and tests I perform will be provided, and an analysis of the data per my perspective will be performed. The raw data will be available for individual analysis and consideration. As this project unfolds updates will follow.

Thank you CobKit's for your contribution to my endeavor. Your interest and ideas are appreciated. I look forward future debates that fuel creative innovations.

- ZXC

 

[nfhiggs]

how to obtain data to gauge the chip performance that will correlate to the tested chips. Is there a test that I can perform that will quantitatively assess the performance, as to compare it with "HQ" chips, without utilizing a research lab?

Just get a high quality chip with well known efficacy parameters. Do a side by side comparison of the intensity (PAR) values measured at 24" at the same power levels. Compare the difference.

 

[ttystikk]

Are we looking at the same graph? What do the abbreviations on the side and bottom mean?

You are correct, I don't have data yet. I have been asking you how to obtain data to gauge the chip performance that will correlate to the tested chips. Is there a test that I can perform that will quantitatively assess the performance, as to compare it with "HQ" chips, without utilizing a research lab?



Please reinforce these statements with citations or links, no offense, but there should be justification to these statistics.



I based my assumptions upon the lowest performing chip in your graph, as stated in the response. Without data, nor an adequate method of testing, we can only base observations on the information available.

Regardless, I'll post the development of the array as it progresses. I'll provide pictures of the components and the creation process ideologies, unbiased. Any data and tests I perform will be provided, and an analysis of the data per my perspective will be performed. The raw data will be available for individual analysis and consideration. As this project unfolds updates will follow.

Thank you CobKit's for your contribution to my endeavor. Your interest and ideas are appreciated. I look forward future debates that fuel creative innovations.

- ZXC

@CobKits is doing you a big solid by trying to help you understand the low efficiency of the solution you've chosen.

I'm interested in the water cooling aspects of your setup, having no small amount of personal experience with the tech myself.

If you ask for help, you would do well to listen carefully to those who give you specific feedback.