Fanleaf's Huge 42 COB Array Build Plus Other Builds

Fanleaf's Huge 42 Cree CXB3590 COB Array Build Plus Other Builds

Rider509 said:
Another system uses the 185H-C1400. Lots of different drivers will operate four 3590 COBs and I'm curious as to what the deciding factor is in making the best choice of driver.
Click to expand...

I'd go 1750mA. I just did a built with 4 cxb3590 6500k 36v DB bin. The 1750 maxed the chips. At 9-10 they didn't get any brighter. Almost no heat on those 6500k chips either.
The client uses the light at about 70% power. Any more and the plants curl. His effacy is about 180lm/w.
The reasoning is that plants can only handle so much light. You can keep the lights at max but higher up. This will give you poor effacy. Or you can drive them low (35/45w) and get closer to the plants. Better effacy.

That's the reasoning I guess.


Sent from my iPhone using 420 Magazine Mobile App
 
Let me restate.

Four 36v 3590s.

Possible HLG drivers:
Driver V_Min V_Max Current_(mA) Max_Power_(W)
120H-C1050 74 148 1050 155.4
185H-C1050 95 190 1050 199.5
240H-C1050 119 238 1050 249.9
185H-C1400 71 143 1400 200.2
240H-C1400 89 179 1400 250.6
320H-C1400 114 229 1400 320.6
320H-C2100 72 152 2100 319.2

What is the design criteria that would determine the best driver for an application?
 
Voltage. :cheesygrinsmiley: In this case, they're all constant current drivers at either 1050, 1400 or 2100 ma. So first you decide how hard you want to run your COBs which is the amperage. If they're 36v COBs, you'll get 36, 50 or 72 watts each (roughly).

Then it's a matter of voltage - how many COBs can you put on a driver. Divide the driver's max voltage by 36 and you get maximum COBs for that driver. For instance, the 185H-C1400 is perfect for 4 COBs at 50 watts each. 143/36 is almost exactly 4, and 200/50 is 4 - ideal. Also, when the driver is maxed out, it's the most efficient.


:Namaste:
 
JimmyJames905 said:
You forgot hlg240h-c1750. Fits 4 perfectly with 0.6 v to spare.
Click to expand...

Hahahaha, thanks JJ. I wasn't trying to be exhaustive but that helps make the point. So many choices and I don't know enough to understand how a device manufacturer or knowledgeable DIYer can select the best given their design parameters.
 
Rider509 said:
Let me restate.

Four 36v 3590s.

Possible HLG drivers:
Driver V_Min V_MaxCurrent_(mA)Max_Power_(W)
120H-C105074 148 1050 155.4
185H-C105095 190 1050 199.5
240H-C1050119 238 1050 249.9
185H-C140071 143 1400 200.2
240H-C140089 179 1400 250.6
320H-C1400114 229 1400 320.6
320H-C2100 72 152 2100 319.2

What is the design criteria that would determine the best driver for an application?
Click to expand...

There is no "best" driver. If you were powering a street light you would use 2100mA. If you were lighting documents in a museum you would use a 700mA driver. But since you're using it for horticultural purposes you have to decide if you want high luminosity at a distance (high watts, high heat) or low luminosity up close.
Choosing the 1750 or 2100 driver will allow you to turn it up if you want to but you will burn your plants if always at driving at max current. what chips are you using specifically? There's a big difference from bin to bin.


Sent from my iPhone using 420 Magazine Mobile App
 
Graytail said:
Voltage. :cheesygrinsmiley:
Then it's a matter of voltage - how many COBs can you put on a driver. Divide the driver's max voltage by 36 and you get maximum COBs for that driver. For instance, the 185H-C1400 is perfect for 4 COBs at 50 watts each. 143/36 is almost exactly 4, and 200/50 is 4 - ideal. Also, when the driver is maxed out, it's the most efficient.

:Namaste:
Click to expand...

I understand that, Graytail. In this case there are four COBS and many driver choices. And the driver may be most efficient at max but the COBs are more efficient (to a point) when under-driven by dimming.
 
Rider509 said:
I understand that, Graytail. In this case there are four COBS and many driver choices. And the driver may be most efficient at max but the COBs are more efficient (to a point) when under-driven by dimming.
Click to expand...

Are you building an array or just inquiring?
If you have chips already you should tell us what they are (full serial number). Like I mentioned, Crees chips vary widely by bin codes. A CD bin can't touch the luminosity of a DB bin.


Sent from my iPhone using 420 Magazine Mobile App
 
Right, but first you have to decide how hard you want to run them.

Do you want no more than 50 watts each, or the option to run up to 60-70 watts? Then you go with a 1750 ma or 2100 ma driver with enough voltage to run 4 COBs -144 volts. The 320H-C2100 maxes out at 152 volts, so that would give you as much as 70+ watts. But the driver will run less efficiently when you dim it - not a big deal, but a few points.

I've been all over the numbers for months now, and it always seems to work out best at one COB per sqft at 50 watts. :cheesygrinsmiley: I don't think most of us have an environment that can take advantage of higher PAR than that.
 
JimmyJames905 said:
There is no "best" driver. If you were powering a street light you would use...
Click to expand...

In other words there is a "best" driver given certain design parameters.

The drivers in question are all being used to power grow lights by DIYers or manufacturers. They all get the job done but there must be a metric by which a knowledgeable manufacturer or DIYer makes an informed choice. My choice of the 320H-C2100b was by no means a knowledgeable choice based on an understanding of all the variable options.
 
Fanleaf's Huge 42 Cree CXB3590 COB Array Build Plus Other Builds

1 cob per sq ft at 35-50w dimmable will give you the best results in terms of even coverage and high luminosity during flower.

I know there is a lot of confusing information out there. I have been building and selling these COB arrays for the last month pretty steadily. And one thing I will tell you is that customers are mostly stupid. They can't seem to wrap the idea around their head that LED is not Hid. Most hid lights are only 9-12% efficient. That means the other 90some percent is being used to create heat. COBs are up to 44% efficient. Huge difference. I get many customers who refuse to buy from me simply because I cannot provide them with a "1000 W COB". In their mind the reasoning is...they used a 1000 W metal halide so why wouldn't they use a 1000 W LED?
COBs have a fuller spectrum and twice the efficiency of even the best HIDs on the market.
About 2/3rd's of my customers all say the same thing. "This thing is way brighter than expected".
Most of my systems are 1400mA.


Sent from my iPhone using 420 Magazine Mobile App
 
Rider509 said:
In other words there is a "best" driver given certain design parameters.

The drivers in question are all being used to power grow lights by DIYers or manufacturers. They all get the job done but there must be a metric by which a knowledgeable manufacturer or DIYer makes an informed choice. My choice of the 320H-C2100b was by no means a knowledgeable choice based on an understanding of all the variable options.
Click to expand...

Is it starting to make sense? :cheesygrinsmiley:

It's all so simple once you "get it". I've worked out every possible configuration for a 4x4, from 36 COBs at 20 watts each to 9 COBs at 72 watts. I could go 36 with cheap 1212s with $5 individual drivers or power Veros at 72 watts with Mean Wells. The math always leans towards 16 COBs at 50 watts each. And if you're using reflectors, the 90 degrees projects a great overlap pattern at one COB per sqft.

My "expert"advice is to go with the 185H-C1400 with four 3590s @ one per sqft. You won't really want more wattage per sqft than that. That's a LOT of PAR. :slide:
 
Rider509 said:
In other words there is a "best" driver given certain design parameters.

The drivers in question are all being used to power grow lights by DIYers or manufacturers. They all get the job done but there must be a metric by which a knowledgeable manufacturer or DIYer makes an informed choice. My choice of the 320H-C2100b was by no means a knowledgeable choice based on an understanding of all the variable options.
Click to expand...

Look at Timber. They sell a Vero C series 69v system paired to two 1400mA meanwells. That's 100w per cob. There is no way you would be able to drive this system at max output. It's massive overkill. There's another company that uses the hlg 48 driver (parallel wiring) this is dangerous due to runaway thermal effect. Then there's a 3rd company that sells 5 COBs on a single hlg 1050 (37w per).
So which is the best? None of them. Without information on what your personal needs are the consumer is left in the dark. This is a relatively new technology and lighting companies like Mars are never completely forthright with their ratings. ie. they're 400w system that only uses 212w.

My best advice is to first decide how much you want your hydro bill to be vs how much you want to spend on a system.
That would be the first step.

Most of my clients are ACMPR patients on a tight budget. They don't want 1000w systems that require mortgaging their house. The Vero 29 D series at 1400mA is my biggest seller. It uses 170w and puts out 37000 lumens. Less than half the consumption of a HPS and 7000 lumens brighter. A great system for a small tent.

The next step up is the same array driven by the 1750mA driver. 225w with 42000 lumen output. Only a $50 price difference.




Sent from my iPhone using 420 Magazine Mobile App
 
JimmyJames905 said:
Look at Timber. They sell a Vero C series 69v system paired to two 1400mA meanwells. That's 100w per cob. There is no way you would be able to drive this system at max output. It's massive overkill.
Click to expand...

Timber recommends the 4 COB Vero setup for 4x4 veg at like 3 feet away. I can;t speak for them, but I don't think they designed it to be a close up light like their CXBs.
 
Back
Top Bottom