DIY led grow

Isn't the wide yellow band desirable? What do you mean "makes up" for? What is the best spectrum?

stardustsailor said:

Now ...
If their high irradiance ,'makes up ' for the spectrum ...

Click to expand...

I think he was talking about a light source with a higher cri. Like a cmh for example. The 3000k spectral curve is nice no doubt...but I'd think a high cri version would be even better if it could maintain efficiency the same.

Like the spectral dot type cobs. High cri with bold claims about the performance even with lower efficiency...then say a cxa 3070.

churchhaze said:

Isn't the wide yellow band desirable? What do you mean "makes up" for? What is the best spectrum?

Click to expand...

Positivity said:

I think he was talking about a light source with a higher cri. Like a cmh for example. The 3000k spectral curve is nice no doubt...but I'd think a high cri version would be even better if it could maintain efficiency the same.

Like the spectral dot type cobs. High cri with bold claims about the performance even with lower efficiency...then say a cxa 3070.

Click to expand...

Regarding the red /amber/yellow/green ....=> click!

"Makes up ' ..Maybe wrong expression ..

At low irradiances ( 200-300 umol/sec/Sq.m ) Warm white 3000K ( 80 CRI ) ,if used solely ,
will grow some bud ...Thing is that at such irradiances ,the green wls percentage will envoke
Shade Avoidance Syndrome ... Not necessarily with a negative impact in yield ,but neither with a 'positive' ..
But with a certain impact at the morphology/physical appearance of the plant(s) ...
(Extended leaf stems ,for example ...Large ,thin (thus 'sensitive' ) leaves ...Etc ..)

At such low irradiances ,green wls are not necessary ..
Better that % of power ,to've been at red or amber wls instead of green ..

Blue & red led combinations,still are the most efficient at those flux densities ...
Given the fact that they are used as supplementary illumination to greenhouses and
with bedding / flooring growing techniques ...
And output power spreaded over large surface / area ...
Nothin' can beat them ,at this type of horticulture ...
Still ...
They are used for plants that they do need more than 300umol/sec/m^2 to grow...
(peppers and tomatoes are some exceptions ,but then B&R combos are used as supplemental illumination,
to natural sunlight ,in nothern countries ...)

If leds are to be used as the solely source of illumination and at high irradiances ...
Then green wls ( 520-550 nm ) are mandatory ...
Yellow wls(550-590 nm ) are also mandatory ..
Amber (590-610 ) wls are mandatory ...
Low reds (610-640 )are mandatory ..
Deep reds (640-700) are mandatory ...

Mandatory ,if we 're still referring to the highest photosynthetic activity possible ,under i.e. 700umol/sec/m^2 ,
for 12 continuous hours ,at same power (irradiance ) levels ...

You can still supply 700 umol/sec/m^2(or more ) of red & blue wls ...
And ,of course saturate or even fry the plants ...
(it won't even take half of the 12 hour exposure duration,for photosaturation to 'kick in ' )

( Do not expect high yields at this exact situation described...
Test it ...Get the best blue & red leds ,available on the market ...
And try to grow with 700umol/sec/m^2 of blue and red ... Try it ..
I dare you ...And then compare it with plants grown under same quanta flux density
of white light ,coming from a CXA 3000K ... )

..Anyway ...Warm white led light ,probably can grow plants ,in the best possible manner ,
achieved by solid state light ..

It takes the right leds/arrays (efficient ) and plenty of power ,to make it work ...

And here it gets 'complicated' ....

In my fixture ,to give you an example ,4x CXA3070 3000K Z4 are driven up to ~2A ...
For an average -stable-operating Tc =45C ,the leds efficiency is ~33% ...
(Slightly better radiometric efficiency than a HPS 400W ) ..
The leds have a total electrical power of 312.3 Watts.
To keep Tc=45C ,the fixture has active cooling ..
So total (@Plug ) power of fixture is about 340W (fans at lowest setting ,CXAs @2 A )
(with fan psu,cc led drivers,MCU,LCD display,Voltage reg along with fans ..)

Output irradiance is about 103 Watts ..
15-17 Watts lower that the output of a 400W hps ...
But with same or even more quanta flux (umols/sec ) ,than the 400W HPS ..
(Due to more output in red wls ,of the CXA 3000K than in HPS )

So 4x CXA 3070 driven @2A ,one can assume that is the SSL replacement of a 400W HPS ..
(with slightly better el efficiency,way longer service life,slightly lower output power in W ,better spectrum and better
heat management of a 400W HPS..)
And all that in a pretty reasonable price tag ...
.....

Now ....
What about if ...
If instead I had used 16x CXA ,but driven at 500mA ?

For the same Tc temperature ..
Efficiency of arrays would have been 46% !!!
The el.power (total ) would have been 272 Watt.
Output light power (total) would have been ~125 Watt !
Given the -slightly- better spectrum ....

At this case ,I've surpassed by far the growing efficiency of a 400W HPS ...
Fan Power for keeping the TC at 45 C ,is also lower ..(more arrays ,lower driving current )
Thing is ...I would have needed 16x CXA arrays ,16x CC drivers ..more heatsink mateial ..
More space ,thus multiple panels or even larger case ..And so on ...
And ,of course ,I would 've needed another ..'cash budget' ..Way bigger/larger ....

Anyway ...
No matter what or how ...
The CXA3070 array is trully a 'breakthrough' in led horticulture ...
(At last ...)

It can be used either hard driven -and few parts needed - to grow like a HPS does ..
( With Very closely 'alike' characteristics ..Spectrum,efficiency,el.power ,irradiance ,etc )

Or it can surpass ,in every characteristic the average 400-600-1000 W hps ..
If used as 'many and soft driven ' and not as ' few but hard driven '
Trully utilising less energy ,to grow more efficiently,better & higher yielding plants ,than the HIDs ...
And-of course- still ,price/total cost is the greatest 'discouraging' / limiting factor ,at such case ..

Secret with CXA3070 array is ...are ..two of them ...
-Good cooling ...
-High irradiances ...

You won't believe in your own eyes ,what this array can do ,
if these two ..' conditions' are fulfilled ..

( solely by itself..Without any other type of leds added )


Cheers.

stardustsailor said:

You can still supply 700 umol/sec/m^2(or more ) of red & blue wls ...
And ,of course saturate or even fry the plants ...

Click to expand...

We're on the same page here! Once the top layer of canopy has reached saturation from red and blue wavelengths, there's no other way to increase irradiance without using "lower quality" colors. (lower quality by my definition means it causes lower %Pfr, 730nm being the lowest "quality")

No other way.... unless your plan is to burn a hole in it...

stardustsailor said:

In my fixture ,to give you an example ,4x CXA3070 3000K Z4 are driven up to ~2A ...

Click to expand...

Seriously? I've been thinking ~1A is a good sweet spot. I put together a chart showing the efficiency of the lights at each current, and it looks like the theme is lower current means higher efficiency, even down at 200-300mA.

I think 500mA would be a great balance point if the price comes down.

I want to get 16 CXAs and run them around 800-1000mA, in 4 strings. I think you've seen my thread with the simulations. (gives you an idea of the currents I'm looking at running)

Good info..wish I could contribute more but my brain is fried eggs right now. Time for a break!

This is the final setup..

Two 22" 24w Arcadia t5 UVB floros. Was initially aiming one across canopy for uneven distribution..did work well enough though. Now two fixtures for a better coverage..starting on a 4hr midcycle high noon setting.

Cxa main light now has 2 365nm in the middle with no reflectors. They have a beam angle of 70* so that should be good. Light is up to 400w now...I'll be keeping power at around 360w this run...maybe even less in the future. More power than I need but it's there if I want it.

And flower trigger. 3 min before lights out and 3 min after.

I will copy this good info for the future ref., for now I will have to use what I have gotten. Let this newbie crap wear off a bit as I get my hands dirty (so to speak). Learning so much all over again, seems to never end when growing this fabulous plant.

only1realhigh said:

I will copy this good info for the future ref., for now I will have to use what I have gotten. Let this newbie crap wear off a bit as I get my hands dirty (so to speak). Learning so much all over again, seems to never end when growing this fabulous plant.

Click to expand...

Enjoy your setup realhigh...your leds are plenty good. These new cobs will be developing even further no doubt. The good bins of the current lineup still aren't widely available. Mine are a few notches down from the ones I wanted.

What you've got is a nice base built up that can be upgraded whenever you want...hard works done..

Positivity said:

Enjoy your setup realhigh...your leds are plenty good. These new cobs will be developing even further no doubt. The good bins of the current lineup still aren't widely available. Mine are a few notches down from the ones I wanted.

What you've got is a nice base built up that can be upgraded whenever you want...hard works done..

Click to expand...

I like those reflectors ...
I guess that if a certain number of CXA arrays are to be used ,
to cover a certain area-like on Positivity's fixture- ,
then those reflectors might prove to be really useful ..

Anyway ..
Here's how it illuminates ,without lens or reflector ...








Cheers.

I really like the Angelina wide so far. I thought 52* degrees would be too tight but in use they still have a decent amount of spread. I could see using even a tighter version like the 40* or 28* in the right circumstances...like an apache 600 sized fixture running them at low current and getting an obscene amount of lumens.

Bad news for my light. I've been running it and keeping a close eye on things.

In a large open room the heatsink literally stayed very cool. In my 2 x 4 tent it's steadily getting hotter than I like. Room temp about 5*f over ambient. So from 90*f ambient to 95*. Not that bad..but not where I want it.

So I'm going to take it apart and reassemble a bit. The middle diodes are going to need to get centered so the heat can distribute more evenly. And my black box will be center mounted with two fans added over the sides. Doubling the cooling and spread the cxas evenly

I figure if I can cool the hotspots down a bit I'll have less buildup to deal with

So your temp inside the tent is 90 degrees and heatsink is at 95 degrees?

Im a bit curious to know my heatsink temps but have no infrared thermometer.

Hope you can figure out your cooling situation better to your satisfaction cuz the light looks awesome and no doubt it puts out a ton of light with all those COBs.

Didn't measure the heatsink itself just the temperature in the tent. I'd say my heatsink averaged 40*c in a big room but once confined to the tent probably closer to 60*c

Maybe I just need to exhaust more air? Think I need to try and cool her down a bit more first...

Positivity said:

I really like the Angelina wide so far. I thought 52* degrees would be too tight but in use they still have a decent amount of spread. I could see using even a tighter version like the 40* or 28* in the right circumstances...like an apache 600 sized fixture running them at low current and getting an obscene amount of lumens.

Bad news for my light. I've been running it and keeping a close eye on things.

In a large open room the heatsink literally stayed very cool. In my 2 x 4 tent it's steadily getting hotter than I like. Room temp about 5*f over ambient. So from 90*f ambient to 95*. Not that bad..but not where I want it.

So I'm going to take it apart and reassemble a bit. The middle diodes are going to need to get centered so the heat can distribute more evenly. And my black box will be center mounted with two fans added over the sides. Doubling the cooling and spread the cxas evenly

I figure if I can cool the hotspots down a bit I'll have less buildup to deal with

Click to expand...

Pos,try to 'channel' the airflow....
You'll probably need a ' lid' of some sort ,over the fins of the heatsink ..
Actually you've to cover the fin side totally ,since you went the forced convection way ...
(Aluminium plate - sheet -with openings for fans..)
You can attach (screw ) the fans also on the lid ...
As everything else ,also ...
Keep the fans at center ,they way they are already placed .

Right now ,except from CXA placing ,a lot of the airflow is not "guided ' as it should've been ..
(lots of airflow is ' bouncing' upwards ,on the heatsink and
not flowing -horizontally-towards the fin length -both ways ..)

Actually thought about the lid design. The air pretty much follows the channels and rushes out the ends...so I didn't see a huge advantage to it.

Hmmm...it's maintaining levels 2*f over ambient in the tent now that's its nightfall. My room/tent situation does suck big time too...really shitty airflow. Hot summer temps might be magnifying problems with my low exhaust airflow setup

To work on it or to run it...that is the question..lol. Probably tweak it a little. Maintain ambient temps at max power and cooling of the hotspot cxa cluster on ends will be the goal tomorrow

If your Ta = 32- 35 °C (90-95 °F )

Then Tj = Τc + ( Rcxa * Heat Pwr)

Tc = Ta +{ ( Rtim +R fc heatsink ) * Heat Pwr }

As you can see if { ( Rtim +R fc heatsink ) * Heat Pwr } =0..... (feasible only in a dreamworld ),
then Tc=Ta ...


1) Yes,doing whatever possible to lower the Ta ,is desirable ....
( A/C - Bigger CFM air exchange fan - Continuous air exchange -air circulation -etc )
........Still,it's pretty limited the range of actions, in this part of the equation ...

2 )while on the other part ....

{ ( Rtim +R fc heatsink ) * Heat Pwr }

1) Rtim ....(pads or solder paste ? )

Pads approach the effectiveness of thermal paste only if grit of heatsink surface is less
than Ra=2microns ...(With that grit ,heatsink surface would be so glossy & reflective , like a mirror ! )


10W/mk pad vs 7 W/mK paste ..
At best possible DIY /amateur procedures Ra of 5-10 microns ,that translates at an average +5°C in Tc ....
Now ..
What it would take to lower Ta 5 °C ?
Way easier to subtract 5C from Tc ,
by changing from pads to a good & thin thermal paste compound layer ...

2) R fc heatsink
There's no way ,that at this case ,the restrain lid won't make such a difference.
I trust otherwise ...Better examine once more -closely- the cooling design ...
I bet you'll find some weak spots here 'n' there ...


A good designed forced convection ( fc ) could lower the thermal resistance of a naturally convecting ( nc ) heatsink by a 'x10' factor or even more...

(I.e. from a mere 0.8 C/W when nc , to a whopping 0.08 C/W when fc ...
But airflow has to be ' guided ' in a really efficient way,to cool fins / base of the heatsink ...)

3) Heat Pwr...
Drive the CXAs lower level ?
(No...Not really ......)

stardustsailor said:

Pos,try to 'channel' the airflow....
You'll probably need a ' lid' of some sort ,over the fins of the heatsink ..

Click to expand...

This would be a good idea. It's called a shroud in the world of CPU and GPU coolers. As a quick experiment you could tape up the whole upper part of the heatsink.

Not sure where you pull the intake air supply from, but I pull mine from under the home. Coolest point in the summer time and warmest area in the winter. Always fresh air for the girls.
Another help for the plants in warmer temps is to keep the air moving around them. I used a log of muffin fans in my cabinet grows for this. Push the air from down low up thru the plants towards the light, push the air across the canopy between the light and bud, and a nice exhaust fan is in order. Keeps mold down just the same and wind movement helps create stronger stems for those bigger buds you grow.

Where did you get the big metal sheet/sink to mount all the cobs/leds on? Sorry if i missed that part. I'm a dunce when it comes to metalwork. I don't want to under or overengineer my design, and I'm already planning on paying out the ass for the cobs.

If your ambient temperature is higher, you should expect the sinks to function poorly. In that case, I'd worry more about keeping the ambient temperature a bit lower than improving the efficiency of the sinks further. (assuming it works fine when ambient temperature is nominal)

churchhaze said:

Where did you get the big metal sheet/sink to mount all the cobs/leds on? Sorry if i missed that part. I'm a dunce when it comes to metalwork. I don't want to under or overengineer it, and I'm already planning on paying out the ass for the cobs.

If your ambient temperature is higher, you should expect the sinks to function poorly. In that case, I'd worry more about keeping the ambient temperature a bit lower than improving the efficiency of the sinks further. (assuming it works fine when ambient temperature is nominal)

Click to expand...

You've got it somewhat wrong ...

I'' give you a quick example ,to help you realise some things ....

Ta =35°C
Heat power =100Watt

Case 1)

Pos is trying the best to lower the Ta ..
Ok ...It takes a whole A/C unit to lower 5°C the ambient temperature of a 5 m x5 m x 3m room
...So ....
Let's say that Pos ,comes up with the most crazy ideas and mods ,
just to lower 5C the Ta of his grow site ...

What ?
You do not like the 5C figure ?
Ok ..Make it minus 10C ...

So ,initially ......

Tc =35 + { ( Rtim +Rheatsink )* heat .. }

Rtim with pads (250 microns/ 10W/mK ) = 0.00025 /( 10 * 0.000748 ) = 0.334 C/W
R heatsink with poor "airflow-escaping-from-everywhere" design = say 0.5 C/W

Initial Tc = 35 + ( 0.834 *100 )=> Tc = 35+ 83.4 = 118.4 C

Click to expand...

Pos puts a small A/C unit ,in his grow site (there's no other way,he'll manage a -10C Ta decreasement ... )
And Tc now is 25C !!!!
Great ?
No, not so great ....

Case 1 Tc = 25 + 83.4 = 108 .4 C

Click to expand...

???


Case 2) :


SDS ,pays a visit to Pos to share some spliffs ,
and " BTW "changes the pads with a Prolimatech thermal paste ..
10.2 W/mk / 50 micron thickness...

Rtim with paste = 0.00005 /( 10.2 * 0.000748 ) = 0.006 C/W

SDS ,also helps Pos to make a decent lid for his fixture ...
Airflow now ,is restrained only to the fins and heatsink base ...
R heatsink with good "airflow-exausting -only from sides " design = say 0.1 C/W

Case 2 Tc = 35 + ( 0.106 *100 )= 45.6 C

Click to expand...

It took only a thermal paste and a lid ,to lower 72.4 C the Tc of CXAs ....
On the other case it took a whole Ac unit and plenty of electric energy spend ,
to lower Ta and the CXAs by a mere of 10 C ...

I hope that now you can clearly realise ,
how desperate,unsuccessful and unefficient would be trying to lower the Ta ,
in order to have the CXAs operate cooler ....

" If your ambient temperature is higher, you should expect the sinks to function poorly. "

And that is somewhat a false statement ...

The ' function' of a air convecting heatsink is to transfer the thermal energy (heat )
from a (heat )source attached ,
as 'quick' (forced convection / airflow -speed & volume )
and/or
as 'massively' (natural convection / material,surface & mass ) possible,
to the ambient atmosphere ...

This function is / can be 'impaired' or ' enhanced' or affected whatsoever ,
from plenty of other factors,but I'm afraid , not from Ta ...
Ambient temperature is just the lower possible (and totally theoritical ) minimum value ,
that the heat source can obtain,
while the heatsink convects the heat with the ambient environment (atmospheric air ) ....

Neither if Ta = +35 C ,neither if Ta -35 C ,the heatsink's function is affected ....

What is changing really ,is the lower -theoritical - value (lower limit ) ,
obtainable by the heat source ,via the heatsink's function ....

Case 3)

Pos can still do nothing with the fixture and move to South Pole ,
set a grow -site wherever he wants ..(literally speaking ..)
And still have his CXA's operating relatively hot at 53.4 C ,even if Ta = - 30C(-22F)
( Initial Tc = -30 + ( 0.834 *100 )=> Tc = -30+ 83.4 = 53.4 C)

While if he was not so ' stubborn' ' dumpass ' ,
he could have those CXA's operating at 45.6C ,while Ta was a hot 35 C ..
Instead he moved to S.Pole ....
.....I just hope that he'll let everybody know if those CXAs are able to support the growth ,
of some odd polar , sub-zero photosynthesizing organism(s) ...


The air convecting heatsink's (+TIM layer ,+ Fans included ) 'function' ,
is / can be affected by plenty of other facts ,but certainly not from the Ta ...

Cheers.