CXB3590 1500W

bassman999

Well-Known Member
with some extra driver headroom you dont need extra cobs. turn that shit up like freedom rock, man

looking at the HLG-240H-C datasheets, they are about as efficient at 70% as they are at 100% (91% eff@120V), and still 88-89% eff. at 30-60%

the HLG-185H-C can be dimmed down to below 50% with no drop in efficiency (93%), and as low as 30% and still be above 90% eff

thus if your design was based on an ultra efficient 0.7-1.05 A, if youre cool with 4 cobs per driver you can go with a 1.4A driver and crank up 30-90% more par watts when you want/need them.

think of a system that condenses down to use more cobs per square foot in a smaller canopy at a higher efficiency for the summer, then spread em out and crank em up in the winter.....
I have several flavors of cobs now, and am working on getting more to experiment with.
I am flowering in a 24"Dx48"Wx60"H tent
(6) CXA3070 AB 3000K
(2) CXB3070 BB 3500K
(4) CXB3590 CD 3500K
(1) CXM-32 3000K
(1) CXM-32 3500K
(2) HLG-240H-42B drivers
I am working on a design to use them all or most of them while keeping them 50% efficient, and even higher in the case of the cxb cobs
This means th CXAs need to be run at 25W each or less

A lil comparison between the CXM-32 Luminus, and the CXB3590 Cree"


CX B3590: at 85C, 2.4A 12,500 lumens, at 50C 13137 lumens

CX B3070: at 85C, 1.925A 9750 lumens, at 50C 10486 lumens


The Luminus CXM32 per the data sheet is:

Min flux of 14545 at 85C, 2.640A

This would translate to approx:

13,381 at 2.4A and 11054 at 1.925A

At 50C should be:

14,451 at 2.4A and 11938 at 1.925A


Efficacy would be:

Cree CXB3590: 13137/(36*2.4)=152

Cree CXB3070: 10486/36*1.925)=151

Luminus CXM32: 14451/(52*2.4)=115
Luminus CXM32: 11938/(52*1.925)=119

Cree CXA3070: 8750/(39*1.925)=123
 
Last edited:

bassman999

Well-Known Member
The Luminus are not very efficient it seems, and actually will most likely fall short of even the CXA

Luminus has the Gen 3 releasing next month, and hopefully they will have some advancements.

The Cree has ceramic backing, whereas the Luminus is approx half as thick and feels cheap in comparison with just aluminum backing.
 

PurpleBuz

Well-Known Member
The Luminus are not very efficient it seems, and actually will most likely fall short of even the CXA

Luminus has the Gen 3 releasing next month, and hopefully they will have some advancements.

The Cree has ceramic backing, whereas the Luminus is approx half as thick and feels cheap in comparison with just aluminum backing.
aluminum conducts heat better
 

bassman999

Well-Known Member
aluminum conducts heat better
Everyone says they are better.
Watching tv inevitably an auto insurance commercial comes on and says, "Save $500 against all others". Later their competitor says save $450 over all others.
Cob Mfg are no different.
I read their specs and they seem to all say they are 2 or 3x better with thermal management that the others.
Aluminum would seem the better on the 2, but the cob efficacy is lackluster, I wonder if Cree
has tried the aluminum surface?
 

PurpleBuz

Well-Known Member
Everyone says they are better.
Watching tv inevitably an auto insurance commercial comes on and says, "Save $500 against all others". Later their competitor says save $450 over all others.
Cob Mfg are no different.
I read their specs and they seem to all say they are 2 or 3x better with thermal management that the others.
Aluminum would seem the better on the 2, but the cob efficacy is lackluster, I wonder if Cree
has tried the aluminum surface?
dunno, flip chips have lower thermal resistance, but the crees are still more efficient at lower currents. the flip chips may be better at higher currents where the better thermal is a bigger factor.
 

bassman999

Well-Known Member
dunno, flip chips have lower thermal resistance, but the crees are still more efficient at lower currents. the flip chips may be better at higher currents where the better thermal is a bigger factor.
That makes sense!

Is there any difference with thermal resistance in favor of the Vero29 compared to the Cree CXB that makes the Vero lose less efficiency slower at higher Currents.
Obviously it wasnt enough to win in efficacy, but made up some ground all the same.
 

welight

Well-Known Member
just wondering are Vero 29 USD28? if so would you buy CXB3070 if they were $28, thinking of running a special on these for a week
Cheers
Mark
 

kmog33

Well-Known Member
just wondering are Vero 29 USD28? if so would you buy CXB3070 if they were $28, thinking of running a special on these for a week
Cheers
Mark
Vero 29s are 28 at digikey. I think Kb has them for $23 ATM.

28 is good for the 3070s though.


Sent from my iPhone using Tapatalk
 
Last edited:

bassman999

Well-Known Member
With 5$ shipping
36$ x 2= 72...+5$ shipping = 77$ to your door...plus free ideal holders if I remember.
28$(week long sale) x 2= 56$ + 25$ shipping = 81$ to your door on sale.
Think of it like chip efficiency(just the led) vs total system efficiency(led+driver+optics+fans).
Just saying, you did fine and came out ahead.
This is true, shipping is crazy especially for small orders, when you add PP fees from a certain seller it get worse.
 
Here are some units you could scale up to fit your space. To some extent, you can adjust the coverage by adjusting the distance to canopy.

@45.7%
(2) CXB3590 3500K CD 36V @ 2.5A (91.65W ea) $95
(2) HLN-80H-36A $90
183 dissipation W -> 99.37W heat
(2) Arctic 64 Plus CPU coolers $26
(2) lenses $8
83.63 PAR W covering 5 ft² = 749 PPFD
$219 = 2.62/PAR W

@49.7%
(2) CXB3590 3500K CD 72V @ 1.05A (75.5W ea) $95
(1) HLG-185H-C1400 (running at ~1050mA) $65
151 dissipation W -> 76W heat ->
passive cooled heatsink 9120cm² -> (2) 10"X10" heatsink $75
or
active cooled heatsink 3040 cm² min -> 3.5" X 30" heatsink (for spread) $40 + 80mm fan+psu $10
or
(2) Arctic 64 Plus CPU coolers $26
(2) lenses $8
75 PAR W covering 4ft² = 840 PPFD
75 PAR W covering 5ft² = 672 PPFD
$243 = $3.24/PAR W passive cooled
$218 = $2.91/PAR W active cooled on bar heatsink
$194 = $2.59/PAR W active cooled on CPU coolers (uniform spread, best value?)

@ 56.3%
(4) CXB3590 3500K CD 36V @ 1.4A (49W ea) $190
(1) HLG-185H-C1400 $65
196 dissipation W -> 85.7W heat ->
passive cooled heatsink 10284cm² -> 5.88" X 38" heatsink $72 (could split in 2 to improve uniformity/spread)
or
active cooled heatsink 3428cm² min -> 3.5"X36" heatsink (for spread) $45 +80mm fan +psu $10
(4) lenses $16
110.3 PAR W covering 6ft² = 824 PPFD
$311 = $3.07/PAR W passive cooled (great value point IMO, very efficient use of driver)
$326 = 2.96/PAR W active cooled

@61.2%
(5) CXB3590 3500K CD 36V @ 1.05A (35.5W ea) $237.50
(1) HLG-185H-C1050 $65
177.5 dissipation W -> 68.9W heat ->
passive cooled heatsink 8268cm² -> 4.9"X36" $70
or
active cooled heatsink 2756cm² -> 3.5"X36" heatsink (for spread) $45 + 80mm fan + psu $10
(5) lenses $20
108.63 PAR W covering 6ft² =
$392.50 = $3.61/PAR W passive cooled
$377.50 = $3.48/PAR W active cooled

@64%
(8 ) CXB3590 3500K CD @ 700mA (23W ea) $380
(1) HLG-185H-C700 $65
184 dissipation W ->66W heat ->
passive cooled heatsink 7920cm² -> 4.23"X44" $68 (could split these into 2 or 4 units)
or
active cooled heatsink 2640cm² -> 3.5"X36" heatsink (for spread) $45 + 80mm fan + psu $10
active cooled heatsink 2640cm² -> 1.8"X44" heatsink $34 + 80mm fan + psu $10
(8 ) lenses $32
117.8 PAR W covering 6ft² = 879 PPFD
117.8 PAR w covering 7ft² = 754 PPFD
$545 = $4.63/PAR W passive cooled (great option if your area is high electrical cost, Hawaii, NYC, SoCal etc or if your grow is already heat limited and you want to increase yield without running more AC, ducting and fans)
$532 = $4.52/PAR W active cooled 3.5"
$521 = $4.43/PAR W active cooled 1.8"


So in summary, the higher the efficiency goes, the more appealing passive cooling is. Less wires, less moving parts, less points of failure for about the same cost.
That's some great info! Thank you for sharing this...
 
Top