Fanleaf's Huge 42 COB Array Build Plus Other Builds

[Islandgrow]

No problem.
In the next day or two I'll get the list/s together for where I bought all the stuff. I shop and shop and shop around until I feel I've got about the best deal I'll get. For several items I have a few places I'll buy from. It just depends on who's got stuff in stock at that point.

What about the citizen cobs are they comparable to the cree cxb's .

"carpe diem"

 

[fanleaf]

Wow fanleaf epic build. I always like to DIY if possible. I assume you can build a cob cheaper then buying the pre made equivalent. Great job man.

Yes, much cheaper. And you can build it to your needs.

 

[fanleaf]

What about the citizen cobs are they comparable to the cree cxb's .

"carpe diem"

I won't down any product here. There is no doubt that you can build a fine grow light with the citizens. Many folks have done so. Again, what you are building the light to do exactly would weigh on the decision of which cobs to use, Vero, citizen ect. The goal with my light is to be as efficient as I can possibly make it at lower drive power. In this case the Cree CXB3590 is unmatched. The right bin Cree 3590's can provide upwards of 200 lumens per watt at right drive currents which handily beats out all competition. Most others are around the 140's to 150's. That being said there are several cob makers that when the cobs are driven VERY HARD start to overtake the CXB3590 in efficiency. I realize plants don't use lumens however that's the information given in the spec sheets for all of these cobs and the fact is that lumens is what humans see. Also a fact that the light in the spectrum that humans see, plants use! So, I guess I would say if you plan to really push your cobs hard with fewer of them at high drive currents then yes, maybe a different choice can be made. If you are trying to build a good grow light and don't think you can afford the 3590's then maybe the vero's or citizens fit the bill. But if you are trying to build the best light you can build, something proven most efficient with 2gpw proven results the 3590 is the king of the hill at this time.


Now, I don't really care to hit 2 grams per watt, I really don't. But I do want the best built light I can get. I want it to be a proven, reliable light that is the best in efficiency. If not I would just go with a new double ended hps. Those are very efficient actually. That being said, when I do get this light finished the first grow I do will be a grams per watt grow. I will do 1 grow that way to show just how efficient these lights can be. I don't know that I will hit 2 grams per watt but even 1.5 would be great right? They can be just over 60% efficient actually. That's amazing.

 

[fanleaf]

Since we will have 14 Cobs per panel I wanted to share the Driver layout with you guys.
We will have 11 of the 3000k's per panel and 3 of the 5000k's per panel. I want all of the 3000k's to be adjustable from 0 - 50W each and the 5000k's to be adjustable from 0-80W each so here's how we do it.

Per Panel
6X 3590's in 3000k on one HLG-320H-C1400B. Those will be 0-50+ Watts each
5X 3590's in 3000k on one HLG-240H-C1400B. Those will be 0-50 Watts each
3X 5590's in 5000k on one HLG-240H-C2100B. Those will be 0-80Watts each

That will be the driver layout per each panel.
So with all three panels I will be using;

3pcs. HLG-320H-C1400B
3pcs. HLG-240H-C1400B
3pcs. HLG-240H-C2100B

 

[fanleaf]

You gave us a parts list, do we get supplier lists where you ordered most of the parts? It would take longer for me to hunt down the parts than it would take me to build a 6 cob light... The way you're laying it out, a light will be an easy build..... One thing i can do is follow directions, and I've got the monkey see, monkey do, down pat..

Best deal on Reflectors and drivers. https://led.cdiweb.com/Home/2
Chip holders Electronic Components Distributor Newark element14
Cree chips I buy from 2 places. Cutter electronics in Australia ships quick to the USA. Their prices are good. Just remember their prices show in au money so don't be scared because it looks way too expensive lol. So Google conversion "au to USD" to figure things out. I do like these guys. They have the heatsinks too. Cutter Electronics Home Page
2nd place for the Cree chips is https://www.arrow.com/
The reflector adapters are at newark element shown above also.

 

[fanleaf]

After waiting for more thermal paste to come in I was able to get some work done today. Panel #1 of 3 is ready for wiring and drivers. Wiring and drivers will be gone over with you guys thoroughly when I get started on that.
If you look at the Led chips closely you can tell where the three 5000k's are located. The center of those are a lighter shade orange or yellow than the more orange 3000k's.
Before mounting the LEDs on the panel today I did weigh the complete frame alone. The complete frame with no LEDs are heat sinks weighed 7 pounds 1 ounce. It is built that I can also hang from the center for added support if needed butt preliminary testing tells me it will be unnecessary. I will probably use the center as a stress relief anyways. Just to be sure I'm spreading the weight evenly on the ceiling of the 8 by 8 tent.

Sent from my android from outer space!

 

[kelticBlue]

Absolutely Monster.


I just put together a timbergrow light kit myself. I used 4 3500k. I was wondering if you could tell the difference in cob between 3000k and 5000k. Thanks.
I would think more people might do this.

 

[fanleaf]

Absolutely Monster.


I just put together a timbergrow light kit myself. I used 4 3500k. I was wondering if you could tell the difference in cob between 3000k and 5000k. Thanks.
I would think more people might do this.

Sure thing. Ill hop on the laptop quick.

Sent from my android from outer space!

 

[VladimirPutin]

WOW that's crazy !

but in a good way

Absolutely Monster.


I just put together a timbergrow light kit myself. I used 4 3500k. I was wondering if you could tell the difference in cob between 3000k and 5000k. Thanks.
I would think more people might do this.

5000K is "natural white" while 3500K is "warm white", 3000K is simply even "warmer"; I'd notice the 500K difference

 

[fanleaf]

Absolutely Monster.


I just put together a timbergrow light kit myself. I used 4 3500k. I was wondering if you could tell the difference in cob between 3000k and 5000k. Thanks.
I would think more people might do this.

The 3500k 3590's are a great cob for start to finish as are the 4000k's!

There are different "bins" out there for the 3590's. The bin is based on the light output of the cob. When they are tested at Cree they get labeled a specific bin with each one.

The some or all of the top bins are
CB lower of the top bins
CD higher
DB Highest
And the rare DD bin is even higher yet

There are many lower bins than the CB bin but I'm not mentioning those.

The more red (3000k) you get the harder it is for them to be made in the higher bins. In 3000k the top bin made is the CB bin. They have a MINIMUM Luminous Flux of 12,134 lumens @25 degrees C.

The 3000k's have fantasic flowering spectrum coverage. As you can see, the 80CRI version that I use even has some 730nm spectrum. It heavily favors the red spectrum and has it's secondary peak at about 460nm in the blue. The blue peak is a little light for my veg taste though and the reason I add in a few of the 5000k 70CRI chips. They cover some of the deep blues and heavily favor the whole blue spectrum for veg with their secondary peak very wide along most of the color spectrum.

The best bin in the 5000k's are the DB bin which is what I use also. They offer a MINIMUM Luminous Flux of 14,340 lumens!

While looking at the cobs while running the 5000k cobs are a much whiter blue color while the 3000k is a soft white more like a hps because of all of the red.

So looking at the chart it makes perfect sense that for those who want to build a smaller light that you would probably want to run all 3500k or even all 4000k. You get good spectrum for veg through flower with those and they are offered in higher bins than the 3000k so you will get slightly more lumens per watt from them as well.
If I were to build myself a small CXB light like a 5 to 8 cob it would still probably have all 3000k's and 1 either 5000k or 6500k for the blue.

Now, since I'm not building a smaller light that gives me far more options to make the light do just what I want it to do.
So even though the CB bin is the highest bin offered in the 3000k color just understand that the CB bin still rocks! They put out tons of light per watt.
That being said the 5000k (more blue) is the DB bin and can put out over 210 lumens per watt at the right drive current!

So, what I like is the option to run just a few of the 5000k's at a higher power for veg with the 3000k's running very low power. In flower I will tune just the opposite and run the 5000k blues soft and run the red's higher.

When you make your own light the advantage you have is that you can customize your light spectrum.

You see, the 3000k CB bin @12,134 Luminous Flux will slightly lack in light output compared to a 3500k in CD bin @ 13,237 Luminous Flux but the 3000k does cover more red spectrum for flower than the 3500k so it's just a trade off. And while my 3000k's put out very slightly less light than the 3500k adding in a few of the 5000k's in DB bin @ 14,340 Luminous Flux put out more light than the 3500k's do and favor veg (blue).

So again, you can make some trade offs to customize your light to do exactly what you want.

Heres a chart on the bins

 

[kelticBlue]

cool beans. I am looking at the pictures here. I guess i never figured you could tell the difference when they are off. sort of orange COB and yellow COB.

I really like that the frame is taller than the door in the first picture. It is more than 4 times the size of mine, that is so cool.
An epic build you got going.

 

[fanleaf]

cool beans. I am looking at the pictures here. I guess i never figured you could tell the difference when they are off. sort of orange COB and yellow COB.

I really like that the frame is taller then the door in the first picture. It is more than 4 times the size of mine, that is so cool.
An epic build you got going.

Yeah, my camera just don't show the difference like you can see with your eyes. The 5000k is really just like a yellow Crayon and the 3000k looks just like an egg yolk. Very different color.

yeah man, That panel is 86" tall. I got 2 more to put together too lol.

Thank you for the props man! I hope you stick around for the whole thing. This will be a truly epic 3590 build. One for the books. I haven't even seen anyone do a bigger one on Youtube lol

 

[kelticBlue]

i will not want to miss this.
Thanks so much for the LED info. The way you describe the working and ins and outs, with decision making and conclusions, even I can understand.

 

[fanleaf]

i will not want to miss this.
Thanks so much for the LED info. The way you describe the working and ins and outs, with decision making and conclusions, even I can understand.

Maybe it can be seen better in this one. Top is 3000k bottom is 5000k

Sent from my android from outer space!

 

[kelticBlue]

I totally see egg yolk on top and yellow on bottom, you nailed the description.
I am not good with colors, but i totally see egg yolk and yellow.

 

[fanleaf]

For those who want to know how to wire such a light, if you can follow my kindergarten drawing this is how it will be wired.

I will be taking pictures of the process of adding the dimming function to the drivers and detailing that whole process. It's fairly easy.

For those who want to add a dimming potentiometer to a compatible driver I will stress the importance of buying Linear potentiometers.

Standard potentiometers are very uneven in the resistance from 0-100%

Turning the dial will look something like this below. It makes things hard to pinpoint an exact power level you want because the standard ones get so touchy.

0-------10%-------20%-------30%------40%----50%---60%--70%--80%90%100

See how touchy they get at the upper end?
A linear potentiometer stays completely even from 0-100% making easy adjustments.



Anyways, heres my drawing of how the panel will be wired.

 

[Islandgrow]

Why are the 5000k only on one side of the fixture. Why not 2 on 1 side and 1 on the other?

"carpe diem"

 

[fanleaf]

Why are the 5000k only on one side of the fixture. Why not 2 on 1 side and 1 on the other?

"carpe diem"

Imagine the 3 panels just like that one put together. That way the 5000k's will be spread very very even. Edge to edge, front to back. Since I'm only adding 9 high powered 5000k chips I dont want to waste any off to the edges of the tent. This way every single plant in the tent will get some coverage from the 5000k's. Remember, each panel is only 19 inches wide and center to center of the cob chips is only 13 inches. If I put 2 on 1 row and 1 at the tent edge row it would leave a hole in the middle with less coverage. I know this because of the 25 cob rig I'm already running. I think once you see all 3 complete panels in the tent you will fully agree.


This way I believe I could veg this whole tent with only the 9 pcs of 5000k chips if I wanted to. Once I get all 3 panels in I will test that theory with my Apogee MQ500 meter to test par/ppfd levels with just running those 9.

 

[Islandgrow]

Imagine the 3 panels just like that one put together. That way the 5000k's will be spread very very even. Edge to edge, front to back. Since I'm only adding 9 high powered 5000k chips I dont want to waste any off to the edges of the tent. This way every single plant in the tent will get some coverage from the 5000k's. Remember, each panel is only 19 inches wide and center to center of the cob chips is only 13 inches. If I put 2 on 1 row and 1 at the tent edge row it would leave a hole in the middle with less coverage. I know this because of the 25 cob rig I'm already running. I think once you see all 3 complete panels in the tent you will fully agree.


This way I believe I could veg this whole tent with only the 9 pcs of 5000k chips if I wanted to. Once I get all 3 panels in I will test that theory with my Apogee MQ500 meter to test par/ppfd levels with just running those 9.

Oh yes I forgot this was one of 3 . Would still spread better if it were staggered. But then again they are running at a higher wattage so it will be intense.

"carpe diem"

 

[fanleaf]

Oh yes I forgot this was one of 3 . Would still spread better if it were staggered. But then again they are running at a higher wattage so it will be intense.

"carpe diem"

I circled each 5000k twice. Just examples of 2 different coverage areas depending on height.

Sent from my android from outer space!