DiY LED - Cree CXA3070

Bueno Time

Bueno Time

Well-Known Member
Finished up the red and blue LED bars and tested them out. Holy crap are they ever bright. At least twice as bright as I thought they were going to be, I am very impressed.

They are two 24" heatsink bars with a total of 10 Luxeon ES EX6 bin 660nm, 4 Luxeon ES M4R bin 450nm, 4 Cree XP-E P3 bin 630nm powered by a fasttech 650mA driver.

I am not doubting these LEDs should help kick the ladies up a notch this run. Cant even capture it well on camera but took a couple pics anyway.





 
epicfail

epicfail

Well-Known Member
stardustsailor said:
Option A ) Simple Voltage regulator 10V pwm ,from Arduino .

1x L7810 Voltage Regulator .
1x Electrolytic cap 47uF / 16 V
1x Ceramic cap 100nF
1x 2N3904 transistor
1x 4700 ohm resistor .
......
& 12 VDC power supply

View attachment 3202937
View attachment 3203119
Don 't forget :

* Connect Ground (-) of the 12 VDC supply with Arduino's ground ,if Arduino is powered by a different
power supply ...Always "Common Ground" between different circuits ......
(In this case : ! Does not apply for the gnd pwm output ! )

*Total cost: probably less than $5 .

+10 VDC out goes to " Dim +" of driver ,likewise GND PWM out goes to " Dim- " of driver ...


With Arduino 's 'pwm library ' and some code lines one can replicate very easily ,with plenty of parameters set , dimming control like " Dawn" ,"Dusk " Noon Peak" , " Cloudy weather / gimme a break for five - " ,etc ...

{ -Set a operational Pwm frequency .Usual range is 100Hz - 1000 Hz.Check led driver's spec sheet for detailed manufacturer's recommendation / guides .

-Dimming is achieved by adjusting the duty cycle of the pwm ....
From 0% up to 100% of duty cycle.

-Of course both manually or automatically dimming ,or both ,
can be programmed with Arduino ...
( Interrupts and manual overide if switch ..blah ,blah ...Manual control also can be done
from using a pot ,to utilising touch sensors ... Or even a touch screen ..!!! 8-) )
Click to expand...

Thanks SDS, I built it the next day with 4 channels for independent control over colors. It works great, but I did come across a problem. With the Mean Well 3-in-1 dimming function of the HLN series (I think its the same in the HLG) I had to bridge dim+(blue) and dim-(white) with a 1K ohm pull down resistor in parallel with your circuit. I was stuck on it for a bit until I took a voltage meter to them and realized there was 10v running through. I should have figured that out since that's how dimming with just a resistor (or pot) works. Once I did that I could fade from 0-10v (9.88v) PWM very smooth.

I am now looking into your other thread about thermal protection. You my friend are one smart man.
 
stardustsailor

stardustsailor

Well-Known Member
epicfail said:
Thanks SDS, I built it the next day with 4 channels for independent control over colors. It works great, but I did come across a problem. With the Mean Well 3-in-1 dimming function of the HLN series (I think its the same in the HLG) I had to bridge dim+(blue) and dim-(white) with a 1K ohm pull down resistor in parallel with your circuit. I was stuck on it for a bit until I took a voltage meter to them and realized there was 10v running through. I should have figured that out since that's how dimming with just a resistor (or pot) works. Once I did that I could fade from 0-10v (9.88v) PWM very smooth.

I am now looking into your other thread about thermal protection. You my friend are one smart man.
Click to expand...

If that's the case ,then reversing the polarity will probably render the 1K resistor in parallel useless.

If Dim+ = +10VDC and Dim- = Ground / 0 VDC ,
then connect GND PWM with Dim+ (blue wire ) and +10V Out with Dim- (white wire .)

Now ,with 0% duty cycle of PWM signal ,Io of driver = max .
When PWM duty cycle = 100% ,Io= min .

Although ,if it works for you the way you already have it ,just leave it as is ...

Cheers.


EDIT: Actually the way you already have it ,is the best .
As the 1K resistor is the "min Io Limiter" ..
It can be used to set the minimum Io of the Driver ...
(For example 350mA ....)
Refer to driver's spec sheet ,in order to set a min .Io ,via the resistor in parallel .

( 1K resistor is 1% of Io max .For the Cree CXA3070 ,better set Io min = 320-350mA .)
 
Last edited:
tenthirty

tenthirty

Well-Known Member
The seller of the bad LED drivers received my order back on Tuesday and said nothing.
I escalated the case with ebay and received the refund today, not counting the 60 bucks that it
cost. to send the things back.

Looks like I'm buying a bunch of Meanwell drivers.
 
Bueno Time

Bueno Time

Well-Known Member
So I will let you all know the results from my less than perfect first run with COBs.

The only lighting source for veg was a single CXA3070 3000K Z2 bin COB at 1400mA ~52w. For flower the lighting was 1 CXA 3070 3000K Z4 bin and 1 CXA3070 3000K Z2 bin both at 1400mA a piece, ~104w LED dissipation.

From four plants in half gallon grow bags vegged for 3-3.5 weeks 24/0 lighting from seed and 70 days 12/12:
Pineapple Express x Cheese n Chaze 23g @ 60% humidity in jar
Blackberry Kush x Chemmy Jones pheno1 26.4g @ 61%
Blackberry Kush x Chemmy Jones pheno2 31g @ 61%
Blackberry Kush x Chemmy Jones pheno3 24.3g @ 62%
+1.1g misc. mixed mini buds I tossed into the bottom of the drying tub
total 105.7g

So around 1GPW, not bad. Definitely my personal best and accomplished some goals I had set for the COB setup.

I wanted to match or beat my yields with 150w HPS with ~100w LED and I was successful in that. Beat my best 150w HPS yield by about 2g with 2/3 the wattage.

I wanted to increase the bud quality and density, and that was also successful. The bag appeal of the buds and frost is very good on these buds. I was impressed by the density of the little popcorn nuggets further down the stalks, with the 150w HPS they were super airy and these were nice dense little nuggets with the COBs. Calyx to leaf ratio was good not really hard to trim, took me a little under 4 hours to trim it all very well. I am anal about any little leaf sticking out so I manicured the buds really well.

I think anyone running these COBs should be able to hit 1GPW of top quality buds fairly easily with the efficiency of these babies.
 
Last edited:
Bueno Time

Bueno Time

Well-Known Member
Positivity said:
Gram a watt on first try, congratulations!

And probably still room for improvement on yield?
Click to expand...
Ya for sure could have been better yield. A larger volume of medium per plant could have easily helped. There was going to be 6 females flowering and the two OG s1 x SWT#4 IBL hermied early flower so I pulled them out and finished with 4 so if I had the two other plants they yield would have been higher.

SupraSPL said:
Nice work BT and thank you for documenting your results it is appreciated!
Click to expand...
No problem, I tried to refrain from posting a ton of pics in here throughout, didnt want to hijack the thread just enough updates to show the effectiveness of the COBs.

It was a fun test, now Im running almost double the wattage and higher efficiency bins too so about double the light roughly I think. Not going to be a huge yield this run though with the Bubba crosses I have going this round. Quality of the buds should be supreme though I really cant wait for this next run.

I'll get some pics up of the finished product eventually here too.
 
Bueno Time

Bueno Time

Well-Known Member
Bueno Time said:
Finished up the red and blue LED bars and tested them out. Holy crap are they ever bright. At least twice as bright as I thought they were going to be, I am very impressed.

They are two 24" heatsink bars with a total of 10 Luxeon ES EX6 bin 660nm, 4 Luxeon ES M4R bin 450nm, 4 Cree XP-E P3 bin 630nm powered by a fasttech 650mA driver.

I am not doubting these LEDs should help kick the ladies up a notch this run. Cant even capture it well on camera but took a couple pics anyway.
Click to expand...
Did the Kill A Watt testing on them today:
.99PF 30.4w 0.25A 30.6VA

The total power consumption of flowering tent with the new lighting setup on 12/12 lighting is 92.3 KWH per 30 day month.

Last run in flower was 57.3KWH per 30 days for 70 days of flowering plus 25 days of 24hr lighting under 1 COB module and a 120mm pc fan for exhaust (literally a couple watts so we will ignore that for simplicity). 36KWH for veg and 133.5KWH for flower = 196.5KWH used to harvest 105.7g about .5g/KWH. This is including exhaust fan and heatsink fans.
 
Last edited:
Bueno Time

Bueno Time

Well-Known Member
Pulled out a little popcorn nug to "sample" of the Pineapple Chaze, its only been curing for a few days so far though. Smells like captain crunch cereal, creamy or milky, slightly lemony almost tropical fruity. Pretty nice smells so far gotta stay out of the jar for a while let it get a nice cure on it. Not too green tasting for being so fresh still.

Its smell seems like a mix of the Pineapple Express mother I used in the Pineapple Chaze cross and a DNA Lemon Skunk fem freebie I grew out a few years ago.



 
R

reasonevangelist

Well-Known Member
do you think just one cxa3070 would be enough for a single mainlined plant?

By "enough," i mean if my soil is sufficient and nothing goes wrong, i would like to harvest enough to not run out of smoke (at a rate of 1g/day) before the next harvest would be ready (let's assume 4 months, 7g/week, 1g/day = 120g total). I have 1.5' x 3' horizontal space, and enough room for 1m of height above surface of medium, and below the maximum height of the lamp. Planning on a 15gal pot.

Also: what's the penetration like on these? How many canopy levels are worth allowing to consume resources? I would guess "maybe two (including the top)," but no idea really.
 
Bueno Time

Bueno Time

Well-Known Member
Thread went a little dead in here I see. How bout some CXA bud porn.

Curing bud pics of a little popcorn bud from the Blackberry Kush clone only x CSG Chemmy Jones male pheno #2 from last run. Shit is super frosty definitely frostiest bud Ive grown yet, the trichome stalks are longer in length than usual for me.

For a refresher of which pheno this is, here is the plant in flower day 56 12/12.




And the curing bud pics at about 11 or so days curing in the jar.





 
You must log in or register to reply here.
Top