C.O.B.S.S.L. News and Articles

SupraSPL said:

The CXA series excels when underdriven. So if the CXA is more efficient than Vero at 2.1A, it is a lot more efficient at 1.4A and more still at .7A. It is also slightly cheaper. According to my math (minimum figures), at .7A the CXA is 49% efficient and the Vero 29 is 37.5% efficient. At .7A the CXA cost $4.05/PAR W and the Vero cost $4.24/PAR W

Because we tend to run our COBs relatively cool and we prefer a wide spread of light, the superior thermal resistance of the Vero package does not come into play, but the superior current droop characteristics of the CXA do. This makes the CXA more suitable to our application IMHO and it shows up very noticeably on paper. For example I have demonstrated that it is cheaper

Click to expand...

Some real -life measurements with a PAR meter also ,can easily reveal which one ,actually excels...
..
Also:

Thing is ,that in real life applications ,Vero 29 runs cooler ( 5°C less ,at average ) than the CXA3070 , at same If and at same cooling system.
That alone,makes Vero 29 somewhat superior ,
as it operates at lower Tc ,thus outputing more light.

Still,if one wants to see which Array is more efficient,
all it takes is a reference cooling & driving scheme and then
some radiant power measurements.

I was pretty suprised myself from the outcome.

" Cheaper,easier to find ,easier to install and more efficient..."


Cheers.

stardustsailor said:

There's no such graph in the new pdf of the VERo 29.
(Revised spec sheet :Bridgelux Vero 29 Array Series Product Data Sheet DS33 REV E (9/12/2014) )

Values for Vero 29 are not from pulsed operation ,but from after reaching thermal equilibrium.
(Steady state ).

Click to expand...

SDS, you are right I was looking at the sheet from February (google is linking to the outdated PDFs on Bridgelux site somehow) BUT, the new sheet still clearly states:
"Pulsed Measurement Data (Tj = Tc = 25C)" and they refer to the output as "typical pulsed flux" and it states in the notes:
"Parts are tested in pulsed conditions, Tc = 25°C. Pulse width is 10 ms at nominal drive current"

Other than the contradictory Vf data, nothing has changed, still 8920 lm minimum at 2.1A Tj 25C. That is not enough to outgun the CXA3070 AB

You may be correct, if you crank them up to 2.1A (a questionable economy) the Vero might be 5C cooler than the CXA. But that does not make it more efficient. A 5C difference is an almost negligible advantage. It cannot make up for a lower efficiency to begin with.

Anyway, this is your thread and you have been generous to let me share my viewpoint. So if you prefer to go on to another topic I will happily shut my trap Otherwise I will reconfigure my sheets based on the typical flux values and with the new Vf data and post the results here.

Yay, suddenly my vero 29 3000k setup is more efficient than cxa3070 3000k 80cri ab!! horray!!!!



My neighbors' grass is not as green anymore!!!

@stardustsailor , did you just cover all your CREE shorts?

This must be a reverse pump and dump!!!

For the record, I have no interest/stocks/shares in CREE and I never have. But it sure is getting a smack down this past year while underperforming the NASDAQ very badly. I am surprised by this because I see so many more LED lights for sale in Home Depot/LOWES and even CREE in headlights coming from Asia. Maybe they are not able to maintain high expectations because of growing competition?

Good disclosure.

I used to short CREE in swing trades, but that was in 2011 and 2012. I haven't played the stock market since 2012 really. None of it makes any sense. I was shorting a company I thought would ultimately do well, but its stock would stagnate for systemic reasons.

One time I accidentally went long AAPL when I meant to go short, and after I realized my mistake and closed the position, I had made 100 bucks. How lucky (dumb) is that?

Now this begs the question, how does the Vero 18 stack up compared to the vero 29?

I just measured the Vf of my Vero18 at 350mA 700mA and 1050mA and it agrees with the new Vero PDF thankfully, I love it when things makes sense

Does this change your spreadsheet calculations on the vero 18?

SupraSPL said:

I just measured the Vf of my Vero18 at 350mA 700mA and 1050mA and it agrees with the new Vero PDF thankfully, I love it when things makes sense

Click to expand...

Yes the new data gives all the Veros a slight boost. I always go by worst case scenario so I was using the Vf figures from the graph/curve that Bridgelux updated in September. I am in the process of updating my spreadsheets and will post them on my thread and here if SDS wants to look them over.

SupraSPL said:

Yes the new data gives all the Veros a slight boost. I always go by worst case scenario so I was using the Vf figures from the graph/curve that Bridgelux updated in September. I am in the process of updating my spreadsheets and will post them on my thread and here if SDS wants to look them over.

Click to expand...

Of course!
After all ,this is what the thread is all about !

Cheers !

Nice numbers on the Vero29, I have been liking the look of them for awhile... Could some one run the numbers on the Vero 29 4000k? The BXRC-40E10K0-L at 2100mA run at 127lm/W I feel like this increased efficiency as compared to the BXRC-30E10K0-L at 2100mA run at
120lm/W

Shit I just realized I am looking at the pulsed VS DC stabilized... even DC vs Pursed we see 114lm/W vs 108lm/W and the 4000k has about 2x the relative blue compared to the 3000k. How does that all play out?

I would be interested in a comparison on the BXRC-40E10K0-L if possible.

I also feel that the bridgelux built in mounting system and ez mate connector makes them much more user friendly.

I read most of this thread, and Stardustsailor as always you bring the big data about the gonja, props. The build you are showing in progress with the different phases of color during different periods of growth is exactly what I have been imagining.

I came across this earlier today and you guys might find it interesting:
https://ag.arizona.edu/ceac/sites/ag.arizona.edu.ceac/files/10 R. Hernandez Plant Lighting.pdf

Some interesting information about light diffusion in green house glass increasing yields over direct lighting. I wonder how this relates to LEDs and COB distance from the plant. It seems to me that it would follow that placing the LEDs farther from the plant but having good reflectivity, white walls, white floor, white buckets.... Would have a similar effect to diffusive green house glass?

Feel free to call me stupid if I am way off base here...

Using the updated PDFs and the "typical" figures, at Tj 50 C the 3000K is 117lm/W (36.5%) and the 4000K 123lm/W (37.8%). SDS numbers may vary from mine, it is tricky to derive these numbers from the PDF. But they are a good for qualitative comparison at the very least. The 4000K may be able to flower well on its own, especially for sativa dominant hybrids. According to Mr Flux estimation, the 3000K has 8% blue and the 4000K has 13% blue.

It's always nice when a new calculation can make your light more powerful.

I am not sure if this is the right forum for this, but it relates to some things people have mentioned here.

First, about growing in Alaska, one of the books I was reading mentioned briefly passing that some growers in Canada are able to use 18 hour grow cycle, with 6 hours of dark and 12 hours of light during flowering. If this does work I would be interested in seeing how that works in conjunction the flowering initiator, creating the possibility of a 14/6 hour flower cycle?

I also want/need more information about how this flower initiation works... and what the simplest ways to do it are because that seems like low hanging fruit to me.

Second, I am also getting information that indicates that you only need about 1/5 the light during veg and that it is better for it to be blue for seed/veg. Lets say I am using 12 4000k Vero29 for a flowering room I could get away with 2-3 5000k for the veg room?

coolbreez1 said:

I am not sure if this is the right forum for this, but it relates to some things people have mentioned here.

First, about growing in Alaska, one of the books I was reading mentioned briefly passing that some growers in Canada are able to use 18 hour grow cycle, with 6 hours of dark and 12 hours of light during flowering. If this does work I would be interested in seeing how that works in conjunction the flowering initiator, creating the possibility of a 14/6 hour flower cycle?

I also want/need more information about how this flower initiation works... and what the simplest ways to do it are because that seems like low hanging fruit to me.

Second, I am also getting information that indicates that you only need about 1/5 the light during veg and that it is better for it to be blue for seed/veg. Lets say I am using 12 4000k Vero29 for a flowering room I could get away with 2-3 5000k for the veg room?

Click to expand...

I've been using the 18/6 method while vegetating for three indoor grows now (not entirely a bunch), and it works great and seems popular among my fellow growers.

You may be aware that during the flowering stage, 12/12 is encouraged to persuade the plant(s) into thinking it's naturally time to flower.

The flower initiator provides infrared energy/light to the plant, thus converting a particular matter within the plant's structure into another form of matter, in short allowing the plant to 'jump ahead' during the night phase. It's popular because you can add up to around two extra hours of daylight, giving users the option to roll with 14/10 (the plant thinks it's 14/12).

I cannot explain the scientific details in great length, as I am still relatively new to the subject but this link should help answer at least some of your curosity:

https://www.boundless.com/biology/textbooks/boundless-biology-textbook/plant-form-and-physiology-30/plant-sensory-systems-and-responses-184/the-phytochrome-system-and-red-light-response-701-11926/

I'm currently using infrared, while running on a 13/11 timeframe. It works well, but sudden stretching came as a surprise!

A link to my infrared setup, which accompanies a single lady:
https://www.rollitup.org/t/diy-led-frankensteins-crown.848100/page-2#post-10997653

As far as choosing primary lighting sources, I'd consider running 5000k COBs for vegetation (anything higher would be better), and 3000k-4000k COBs for flowering.

It is just a foot not in this book: Marijuana Horticulture: The Indoor/Outdoor medical growers bible, Jorge Cervantis

page 165:

"I talked with Dutch and Canadian growers who claim their plants FLOWER under a 6-hour dark and 12 hour light photoperiod. This expedited, 18 hour photoperiod regimen is supposed to work, but I am not solid on it. Growers say that their harvest is undeminished, and that they are getting 25% more marijuana in the same time"

coolbreez1 said:

some growers in Canada are able to use 18 hour grow cycle, with 6 hours of dark and 12 hours of light during flowering. If this does work I would be interested in seeing how that works in conjunction the flowering initiator, creating the possibility of a 14/6 hour flower cycle?

Click to expand...

As far as I know, the plant keys on the length of the night cycle. So in theory you could run 6on 12off, but 12on 6off I expect would present a problem for most varieties.

coolbreez1 said:

I am also getting information that indicates that you only need about 1/5 the light during veg and that it is better for it to be blue for seed/veg. Lets say I am using 12 4000k Vero29 for a flowering room I could get away with 2-3 5000k for the veg room?

Click to expand...

Yes you can get away with relatively low light levels and still get decent growth in veg. You can increase light levels if you need to veg them up quicker or dial it back if they are ahead of schedule. It is very handy to have your veg setup with a dimming driver. I use about 110W of LED vegging power to keep pace with a 1300W flowering setup.

Some COB "terminology " .....

Chip -On -Board tech is basically ,in simple words ,
many blue ' excitation' L.E.D. dies/chips ,under a phosphor doped silicone ' yolk' .
(Thus the term "fried eggs" ,'cause COBs look alike fried eggs ,white case with yellow/orange 'yolk '...
The 'yolk ' is also known as Light Emitting Surface ( LES ) .

For the white light from phosphor conversion LED tech ,terms as CCT ,CRI and LER ,
can form altogether a 'review' of a given COB .

Some color basics ...

Regarding visible light ,three are the 'basic ' colors ' that can form
'white light ' -if combined- ,to the human vision.
Blue,Green & Red.

So a led of 450 nm ,one of 550 nm and one of 620 mm ,
lit together ,will be appering as outputting white light ...
Nor blue,nor green ,nor red ....

Lets see their combos ....

blue + red = magenta (not violet )
blue+green= cyan
green+red= yellow/amber

So ....

Green + magenta= white
blue = yellow/amber=white
red + cyan = white .

The second combo looks familiar ....

But ...but ...
green+red= yellow/amber
Ain't that so ?

So blue + ( green+red ) = white

Now....
Since the excitation source of phosphor conversion COBs ,is somewhere in the blue range ,
they render blue colors very good ,or at least they could.
Their 'issue' in correct color rendering is the 'other side' ..
The reds ,ambers,yellows ,and in a lesser extend the greens.
So ,when you hear a " High CRI " PC led or COB ,then
this is what is-most probably 99% -happening
blue + ( green+red ) = white

Blue die and two different phosphors .
One red and one green.
That means lower efficiency than a single yellow or amber phosphor...
(for plenty of reasons: i.e. particle size,re-excitation,ect...)

Why not blue dies and yellow +red phosphor mix ?
Might also be the case ,in some leds/cobs ..
Or blue dies /yellow phosphor +red dies also...

HIGH CRI ,usually means: more than one phosphor used .
(or red dies are used along with PC white light .)

Caution :High CRI can be (very rare applications) ,
also Cyan + deep red (Cyan= blue die +green phosphor )
=higher rad.efficiency of rest High CRI light combos ,but poor on yellows-ambers ..
(meaning poor color rendering of 'skin tones ' , ' wood / natural fabric / leather tones' , 'bakery tones' and so on ...)
Thus ,not exactly fulfilling the term " High Color Rendering Index" light source ..



LER :

( Lumens / ( Vf*If ) ) / LER = Radiometric efficiency = Radiant Power / ( Vf*If )


Some LER values :
Bri Vero29 3K/80 : 315 lm/Wr
Cree CXA3070 5K/75 : 321-322 lm/Wr
Cree CXA3070 3K/80 : 325lm/Wr
Cree CXA3070 3 ? (3.5)K/95 : 275 lm/Wr