-
Happy Birthday ICMag! Been 20 years since Gypsy Nirvana created the forum! We are celebrating with a 4/20 Giveaway and by launching a new Patreon tier called "420club". You can read more here.
-
Important notice: ICMag's T.O.U. has been updated. Please review it here. For your convenience, it is also available in the main forum menu, under 'Quick Links"!
Building a Home Made LED
- Thread starter tenthirty
- Start date
Why on earth would he want to run 64 3W chips?
Dion
Active member
earthwyrms
Active member
how so? alot of builds seem to be using white, not specifically targeted wavelengths of light, found to be effective in plant growth. any "7 band" or bridgelux "full spectrum" LED should target the sweet spots in plant growth far more effectively than a standard white LED.
please explain (no bad vibes meant), i don't understand.
earthwyrms
Active member
it may be a pain to wire; however, every COB chip is essentially, tiny chips of the same type in a smaller package, all connected in series and parallel. wiring individual chips, with their own PCBs and heatsink packaging, together, seems to be the same thing, only taking up more space.
again, these have an altered spectral output, which is not white, but a spetral distribution which is meant for plants. most cree COBs seem to be for lighting, house lighting. even if the W to mols of photon is higher in a cree, more of the light is wasted on non vital wavelength bands. unfortunately, bridgelux sold to a foreign company, they were US based. what makes them less efficient?
https://www.icmag.com/ic/showthread.php?p=5574369
post #66, #69
cree seems to have the right nm types in chips, if they are put together, but the white arrays seem to be inefficient (still more efficient than HID/MH/florescent though)
Attachments
Last edited:
Dion
Active member
but surely u understand the reason those peak nm have a higher conversion it is because they are less provolant in the wild, seen as the sun puts out white light around 5000K
hence the plant becomes more sensitive to those (specifically red) wavelengths in order to harvest enough from them to do it's thing
when u blast it with a buffet shit doesn't get 'better'
Dion
Active member
find jerry at kingbrite
tell him you want 2 HLG-240H-C1750
and 8 vero 29 gen 7 3500K 36v model
and all the holders and reflectors and shit
he also has the heatsinks but it may cost too much to ship to you
jikko77
Active member
may i suggest you to check beta test team work?
is a user on icmag.
what you've linked are peak, but other wavelenght have a use as well.
earthwyrms
Active member
ok. also, i have a feeling that the "full spectrum, used with the "7band" will take care of a grow good. the full spectrum seems like it fills in the gaps well.
if yellow or green are needed, a small light of those bands could supplement the plants.
i think, a mixture of "7 band" and "full spectrum" with a low watt yellow/amber might do a good job.
maybe, around ~150W 7 band, ~100W full spectrum, ~3-9W yellow/amber
(www.ebay.com/itm/3W-E27-Amber-LED-12V-Home-Studio-Spot-light-EY12-/291317657003)
where can things like http://plantsinaction.science.uq.ed....au/files/imagecache/figure-thumb/Fig_1.1.png be produce? how is it graphed, what programs are opensource to do it?
it looks like the quantum yield an be related with the absorbtion, for each component in some sort of equation.
Last edited:
I was looking at running 3x BXRC-30E10K0-B-73 on a HLG320H-C2100. Would this be inefficient running these chips so hard? Planned on 2-3 rows of this configuration over a 5x5.
Is the layout you posted a more ideal one from an efficiency standpoint or is it about the spread?
Dion
Active member
both efficiency and spread
i really like the idea of running the new vero hard so ur plan seems good BUT:
you have a 5x5 right? so you will need around 750w in there, or 30w per sq ft
AND jerry only has the D version
AND try 3500/4000K u will do better than 3000k
AND 3 on that driver will still only be 240w u need 4 to fill it
so instead try 4x 3500/4000k(personally id go with 4000K) on that driver= around 320w
and you need 2 of them
so 8 cobs
2 drivers
ya dig?
After many hours of reading I purchased the mausIV kit and have just powered it up.
Damm the COB's are good and as a bonus adjustable 40w-210w.
Can't wait till I get into the next place where I can build a couple of IX units.
Once it is fully complete I will post a pic. At the moment it looks like drone/ ufo.
Damm the COB's are good and as a bonus adjustable 40w-210w.
Can't wait till I get into the next place where I can build a couple of IX units.
Once it is fully complete I will post a pic. At the moment it looks like drone/ ufo.
earthwyrms
Active member
what is the mausIV ? what spectrum is it?
also, sin, i think i was wrong about the led efficiency, 500mA should run the 700mA chip fine because i read voltage is what makes the light output.
http://www.ebay.com/itm/10W-20W-30W...r-Led-Power-Supply-AC-100V-240V-/391134587736
INPUT:
AC 100-240V 50~60Hz
OUTPUT:
constant current
DC 27-36V 1500mA 50W
$9.98
that should run 30 of them
36V 7.5A would run 15 * .500mA chains of 36V, 150 chips 270W
i don't know how to wire resistors for parallel.
Last edited:
Oh I can dig it. Thanks for all your help man. I'll be sure to post an update once this beast is up and running. Hopefully get everything ordered today.
Dion
Active member
if u have the budget get 3 of these rigs for a total of 960w(that you can dim down to increase efficiency
2 of them will be 640w at full blast....just a tad too little
Hey man unless I'm goofing here I'm seeing that the Vero 29s typically have a forward Voltage of about 50V. The driver I picked has an output of 154V; would this not mean the driver is nearly maxed out at 3 chips rather than 4 (4 would be over 200V and would not be sufficient to light as I understand it.)
Is there something I'm missing here? Am I looking at the wrong chips/driver?
What about the amps of the driver? Are those in the good range to drive the leds? If you aren't sure about the technical details of it better post the specifications of both the driver and the leds. Would be much easier to judge it that way.
explosiv; I posted the current specs in a separate post above.
I'm realizing now that the Vero 29s are available in a 36.8V configuration (BXRC-40E10K0-L-23) which have a max current of 4.2A. This would add up to fit the aforementioned HLG-320h-c-2100 in a four cob per driver configuration (driver specs: 76-152V output,2100mA.)
The only issue I see with this set up is these chips are listed as outputting 126lm/W; whereas the chip I mentioned prior in the 50V configuration outputs 149lm/W (BXRC-40E10K0-B-73.)
As I stated above the issue I'm finding with this chip is that to power four one would require an HLG-320H-C1400 which puts out a less than nominal drive current for these chips (1.8A for the COB, 1.4A output from the driver.)
Is the concept of running COBS at 50V out of reach due to heat issues? Is the difference between the chips a circumstance of binning and that is why the 50V chip is putting out more lm/w? Would that benefit of more lm/w be negated by having to run it at a less than nominal current?
I'm realizing now that the Vero 29s are available in a 36.8V configuration (BXRC-40E10K0-L-23) which have a max current of 4.2A. This would add up to fit the aforementioned HLG-320h-c-2100 in a four cob per driver configuration (driver specs: 76-152V output,2100mA.)
The only issue I see with this set up is these chips are listed as outputting 126lm/W; whereas the chip I mentioned prior in the 50V configuration outputs 149lm/W (BXRC-40E10K0-B-73.)
As I stated above the issue I'm finding with this chip is that to power four one would require an HLG-320H-C1400 which puts out a less than nominal drive current for these chips (1.8A for the COB, 1.4A output from the driver.)
Is the concept of running COBS at 50V out of reach due to heat issues? Is the difference between the chips a circumstance of binning and that is why the 50V chip is putting out more lm/w? Would that benefit of more lm/w be negated by having to run it at a less than nominal current?
