Have any of you DIY COB Growers finished a crop under 1000W DE HPS? - POLL

Astro Aquanaut said:

Need to know the efficacy lumens per a watt... need a spec sheet... However, that was the question I was going to ask Stephen are we going to see higher wattage at similar efficiency or higher efficiency at existing wattage?

Click to expand...

EDIT: @Stephenj37826 just covered this, above.

Your other guess would amount to a reduction in performance, would it not?

And lumens are for humans. The difference is green; our eyes respond to it much more than plants do. Folks here are using PAR watts, because it's measuring the right thing and thus easier to calculate waste heat... which is now simply total watts dissipated minus PAR watts.

Sativied said:

LED is the future then.


Exactly. Don't you just love the irony in that...


For completeness it's a veg light with a spectrum less efficient for photosynthesis. It's quite funny actually, on paper anyway, but the dude is not right in the head so people just let him... Don't blame shittystikk for not mentioning that little detail though, he doesn't really understand what he parrots.


I thought someone dug up an old thread... Supra still going on about hot gavitas, a non-issue for most people who are not in the desert and know how to build a proper hps setup, skewing comparissons by downplaying the difference with DE... Pretending gavitas need plasma supplement to grow hi quality mj...(that's just desperate).

Good times but I thought you guys got over all that trying to beat hps nonsense and evolved.

Click to expand...

i think people got tired of talking about it after every single cob grow yielded more than hps watt for watt

BOBBY_G said:

i think people got tired of talking about it after every single cob grow yielded more than hps watt for watt

Click to expand...

This is the shit I'm talking about guys. Probably a great dude, but wants to hang on to measuring dicks. I just don't get it anymore. I did at first, but not 2 years latter.

ttystikk said:

EDIT: @Stephenj37826 just covered this, above.

Your other guess would amount to a reduction in performance, would it not?

And lumens are for humans. The difference is green; our eyes respond to it much more than plants do. Folks here are using PAR watts, because it's measuring the right thing and thus easier to calculate waste heat... which is now simply total watts dissipated minus PAR watts.

Click to expand...

look at the datasheet, pg 4

BD= 10000 minimum lumens at 85C
CB=11000 min
CD=12000 min
DB=13000 min
DD=14000 min

on average, these bins run 10500,11500, 12500, etc.... as you see from supras tables

so what happens is they build these cob arrays, out of tiny led elements with variability, and statistically some are brighter than others, as their tech improves, their yield of 'top bin' chips increases/ they literally manufacture a lot and test each one and separate them into the bins at the last step. so a 3500K CB bin might have come out the same batch as a CD bin, but luck of the draw one was brighter

realstyles has had some DD chips he got from jerry mos ago. but they were the rarest of the rare. presumably cree got more efficient and now have enough to move into retail channels

the good news is the new top bin will prob be the same price and you can expect some cheaper selloff of any old stock


so its just a category. and lumens are the yardstick. and we want par but who cares, its the same (model minus bin code) cob , same spectra, measured brighter, so for your purposes more lumens = more par.

Stephenj37826 said:

About 10% more output not 10% more efficient unfortunately. So basically the 56% efficient cxb3590 cd will jump to over 60%. It will be DB bin now. Still a notable gain though. I'm thinking you can run your cxb3590 3500k at 1.7a and still be more efficient than we currently are at 1.4. It's really late here but I'll do up some numbers tomorrow. Also if you can't wait it a good learning tool is call the cree product characterization tool. Google it and do the drop down menus. There numbers are the bottom of the bin so those numbers are the least you will get. To be more accurate here on the forums we go by middle of the bin because some will go over the middle some under. Compare the two bins to see the increase cxb3590 CD vs DB I think. Next one down from CD.

Click to expand...

thats the very definition of efficiency. lumens/watt of dissipation

brighter for the same input= more efficient.

again look at supras tables or a 5000K DB vs a 3000K BD

efficiency higher across the board

or alternatively:
https://www.rollitup.org/t/top-bin-cob-comparison.891010/

ttystikk said:

I just watched the video, and sure as shit, Growmau5 is killin' it, 1.9gpw.

@Stephenj37826 or anyone, if I have my numbers right, he's running the very same driver I am and he's running twice the number of chips I am per driver. I'm running CXV3590 72V chips. Does that mean he's running them at only 25W apiece?

Click to expand...

slightly higher than 25

cut amperage in half and wattage is actually slightly higher than half as the voltage/current line is a curve not a straight line

but yeah, *roughly* the 36V ones pull half of the wattage of a 72V at the same current

look at the driver matching table, you'll see where you can run 4.1 36V chips but only 1.9@ 72V due to the reduction in efficiency when driven harder

AND FURTHERMORE

people getting 1.6 GPW will be getting 1.76
people getting 2, will be 2.2

so you can anticipate by sometime oh prob next year 2GPW is the new '1GPW' benchmark from years past

guess what? HPS is using the same bulbs it was 5 years ago.........
COB is getting 10-20+% more efficient year over year over year over year (for now)

BOBBY_G said:

thats the very definition of efficiency. lumens/watt of dissipation

brighter for the same input= more efficient.

again look at supras tables or a 5000K DB vs a 3000K BD

efficiency higher across the board

or alternatively:
https://www.rollitup.org/t/top-bin-cob-comparison.891010/

Click to expand...

That's the sound of HID being loaded onto the trailer and put out to pasture...

BOBBY_G said:

slightly higher than 25

cut amperage in half and wattage is actually slightly higher than half as the voltage/current line is a curve not a straight line

but yeah, *roughly* the 36V ones pull half of the wattage of a 72V at the same current

look at the driver matching table, you'll see where you can run 4.1 36V chips but only 1.9@ 72V due to the reduction in efficiency when driven harder

Click to expand...

I thought his were the C700B drivers? I'm running 4 CXB3590 at 54W apiece, somewhere I'm not getting the driver math.

Do you have the link to supra tables? Sorry, about the hijack this...

Can we start with the LER


Am no math whizz and new to LED...

sorry i assumed you were running same number of cobs. and you were at 49V and he was at 25

go play with the pct and put in 72 vs 36V for the same bin. it doesnt go down to 700 ma (well it prob does for 3070... but no matter

at a given current, youll see wattage isnt exactly half

i gotta bounce but will check in later

you gotta dig thru here to find the actual xls file

http://rollitup.org/t/cob-efficiency-spreadsheets.865238/

[QUOTE"a mongo frog, post: 12347138, member: 311103"]Thank you. I see the savings in these lamps i really do. Big time savings on electric bill alone. The cooling thing i don't know about. Too much math for my pee brain to do. I do know you save in cooling but i just don't know how much.[/QUOTE]
easy peasy

1000W gavita at say 38% efficient = 380 watts of light, 620 watts of heat

1000W cob at 60% (easily and affordably attainable, unlike 70% which is bleeding edge expensive for now): 600W of light, 400W of heat

or alternatively you could replace that gavita with 633W of cob at 60% and get the same 380W of light, and only 253W of heat

ie same amount of light as the gavita, with only 40% of the heat.

AND you cut your electric bill by 37%.....

or alternatively, if you ran the 1000W cob vs 1000W gavita:
same electric bill
58% more yield
and
36% reduction in heat

BOBBY_G said:

sorry i assumed you were running same number of cobs. and you were at 49V and he was at 25

go play with the pct and put in 72 vs 36V for the same bin. it doesnt go down to 700 ma (well it prob does for 3070... but no matter

at a given current, youll see wattage isnt exactly half

Click to expand...

I think I'm asking a different question. I'm using this driver to run 4 CXB3590 72V chips at 54W. I get 54W each because the B model gives 8% more current when the dimming leads are capped off. Add 8W for the 96% driver efficiency factor and the resulting 224W is exactly what my Kill-a-watt says.

Those 8 COBs would get slightly more than 27W each, according to the curve of the graph?

ttystikk said:

I think I'm asking a different question. I'm using this driver to run 4 CXB3590 72V chips at 54W. I get 54W each because the B model gives 8% more current when the dimming leads are capped off. Add 8W for the 96% driver efficiency factor and the resulting 224W is exactly what my Kill-a-watt says.

Those 8 COBs would get slightly more than 27W each, according to the curve of the graph?

Click to expand...

you dont just add that to the efficiency, youre just driving the cobs harder

your driver is ~94.5% efficient at 240V, about 212 W makes it to the cobs

those cobs run at 53W each are about 55% efficient

BOBBY_G said:

you dont just add that to the efficiency, youre just driving the cobs harder

your driver is ~94.5% efficient at 240V, about 212 W makes it to the cobs

those cobs run at 53W each are about 55% efficient

Click to expand...

Okay, I think I know what I'm trying to ask; can I run 8 CXB3590 36V chips on this driver and replicate his setup?

OK everybody slow down, lets look at some fundamentals:

a 72V cob run at a given current looks identical to a driver as a 36V driven at double that voltage. the arrays real limitation is WATTAGE.

from CREE PCT for CD bin 3590, 72V on right, 36V on left:



look at 3.6A on the left vs 1.8A on the right. see how the wattage and efficiency (lumens per watt) are basically identical, and the voltage is exactly doubled?

thus is the nature of a 36V vs 72V cob- how its wired internally, its really the same array jsut wired differently.

advantages to the higher voltage cob are primarily they can be run hotter on a given driver. meanwell doesnt make drivers (in the popular constant current versions) over 1.4 or 2.1A depending on model. were leaving a lot on the table by only runing them so soft. so in the example above, a 1400ma driver cannot run a 36V cob over 48.77W per cob, but it can run a 72V cob up to 104.293W

is it magically running the cob at more than twice efficiency because the wattage is more than doubled? NO!
its just running it at a higher wattage which is less efficient

if you used a different driver you could run the 36V cob at 2.8A and look at the table... bingo! the same 104.293W

is it "more efficient" at double the current because it magically "creates" more than double the watts? NO! it is *less* efficient. it CONSUMES *more* than twice the watts and puts out *less* tan twice the light

chomp on that and ill come back and answer sir ttystikks specific question

BOBBY_G said:

OK everybody slow down, lets look at some fundamentals:

a 72V cob run at a given current looks identical to a driver as a 36V driven at double that voltage. the arrays real limitation is WATTAGE.

from CREE PCT for CD bin 3590, 72V on right, 36V on left:

View attachment 3613936

look at 3.6A on the left vs 1.8A on the right. see how the wattage and efficiency (lumens per watt) are basically identical, and the voltage is exactly doubled?

thus is the nature of a 36V vs 72V cob- how its wired internally, its really the same array jsut wired differently.

advantages to the higher voltage cob are primarily they can be run hotter on a given driver. meanwell doesnt make drivers (in the popular constant current versions) over 1.4 or 2.1A depending on model. were leaving a lot on the table by only runing them so soft. so in the example above, a 1400ma driver cannot run a 36V cob over 48.77W per cob, but it can run a 72V cob up to 104.293W

is it magically running the cob at more than twice efficiency because the wattage is more than doubled? NO!
its just running it at a higher wattage which is less efficient

if you used a different driver you could run the 36V cob at 2.8A and look at the table... bingo! the same 104.293W

is it "more efficient" at double the current because it magically "creates" more than double the watts? NO! it is less efficient. it CONSUMES *more* than twice the watts and puts out *less* tan twice the light

chomp on hat and ill come back and answer sir ttystikks specific question

Click to expand...

In other words, 'yes' lol

But I already knew efficiency would improve with double the number of chips running half the watts on the same drive current. The whole point of the exercise would be to take advantage of that efficiency gain.

So how does current droop compare with temperature droop?

ttystikk said:

Okay, I think I know what I'm trying to ask; can I run 8 CXB3590 36V chips on this driver and replicate his setup?
View attachment 3613917

Click to expand...

Yes, you sure can.

ttystikk said:

Okay, I think I know what I'm trying to ask; can I run 8 CXB3590 36V chips on this driver and replicate his setup?

Click to expand...

you most certainly can. the voltage at 0.7A should be about 32.72V per supras estimation (as you see in the CREE PCT it doesnt go that low, they only provide modeling down to 1.4A)

8*32.72 = 261V, driver goes up to 286. boom!