FLIP CHIP OPTO LUNA 300!!!

URSA LED

URSA LED

Well-Known Member
Hey stardustsailor,

Great info! Just what we've been looking for.
Just a heads up, the Duet 2400 was designed for plant grow in general (not specifically for cannabis), however we may decide to run with a white light with 660nm depending on market reaction.
We'll have a better idea after hempcon.

Our research team including me will be going through this data today.


stardustsailor said:
Thank ,once again ,for your answers .


The question is like that- ( rephrased ) :
Why 552.4 umols/sec of the ( red nitride ? ) phosphor output ( broadband 660 nm ) versus
1036,2 umols/sec of the 455 nm chips output ?
For most plants it should be the other way around ..
(Duet 2400 @ 6A )



- Why would anyone want to use white light to grow marijuana?

( Quick answer ? Cause white light grows plants ,better than any monochromatic combination .... Real Life Field tested 101% ! )

For quite a few reasons,to be more precise :

a ) Firstly to check upon plants for under or over fertilisation , pathogens and/or other plant health issues .
White light helps our vision ,for doing so ..
Furthermore white light makes a nice and relaxed "environment " for human vision ,
when attending the garden ...

b ) Green wavelengths serve as a "slow down " signal to the photosynthetic systems .
Very useful ,when growing under ( of non- variable characteristics ) artificial lighting,
for 12 hours or more .
Nobody wants oversaturated & exhausted chloroplasts .

c ) " The absorption level for green is extremely low on plants " ...
Rather false statement ,I should say ,If I may be excused ...

Lower than blue and red wavelengths yes ,but not " extremely low " .
Extremely low is the absorption over ~730 nm .

Furthermore ,the " Relative Quantum Efficiency " of green wls ,
is actually higher than of blue ones ...
( As also RQE of 600-620 nm amber-red wls is higher than of 660 nm )

d ) "Green window " ...Quite a few folks ,like to call it " light penetration " ..
As blue & red wls ,tend to be readily absorbed by the top layer foliage ,
something has to be "left " for the lower parts of canopy ,to be "fed "...
That "something " is the green and yellow wls ,that are being transmitted through the top layer leaves to the underlying leaves ...

During the early era of LED growing ,many had pointed out that "LEDS lack penetration " ,
when compared with the "trad" HIDs ...
HIDs do emit a good amount of green and yellow wls ..(power-wise )
Blue and Red monochromatic LEDs do not ...
Thus ,they lack ....penetration ..
( ...Damn ,those Kevlar-reinforced leaves ....:P ... )

e ) Blue wls output should not be more than 10-20% of the total output ,in power terms ...
(photobleaching & photoinhibition await somewhere around here ...check question #2 )
Cool white ( phosphor converted ) ,can easily provide the blue wls needed (in power terms ),
plus it also supplies the plants with the VERY useful green and amber wls ..

.....
Hope I was helpful...
More info > here < .

( You should consider hiring an experienced grower as an advisor ...
LOL!
Me for example ...:lol::lol::lol::lol::lol:... )

Cheers.
:peace:
Click to expand...
 
URSA LED

URSA LED

Well-Known Member
nevergoodenuf said:
@Alex Lau - Why aren't you at work? I was going to run down for 2 more Luna 300s. Can I get a Duet 250 to play with? Now I won't be able to get there unless your office calls me back soon with stock, otherwise I will get them Friday.
Click to expand...
Hi nevergoodenuf,

I was out of office yesterday.
I checked stock and we have the following on hand for purchase;

Luna 300 - 4080
Luna 300 - 2790
Luna 300 - 3080
Luna 300 - 3090
Luna 300 - 5070

I have about 15 pieces of each.

For Duet 250 - we don't have any on hand right now.

I'll be in the office the rest of the week. Just drop by anytime besides lunch time ;)
 
URSA LED

URSA LED

Well-Known Member
Hey Stardust,

Again, thanks for the info, it really help us develop the right product coming future;
Unfortunately some of the technical aspects I can't divulge at the moment (trade secret).

We totally understand the 640/660nm spectrum with cannabis, and right now we are working to perfect the spectrum colors, but this is a limitation in the industry I can't share additional information on at this time.

stardustsailor said:
Alex ,here's a rough idea ,but nevertheless worth of further testing ....

View attachment 3581676
A two (or four ) channel COB ...

The outside LES "ring" is using a broadband red phosphor ,with a peak of 640 nm .
Why not 660 nm ?

1) 640 peak red phosphor still emits plenty of 660 nm light ,but also " fills " better
the "lower " red region ,thus "taking care " for both the ChA nad ChB absorption peaks and supplying the highest RQE red light ( 600-620 nm ) .
2) Less power is wasted in the NIR region of over 730 nm ,than when utilising 660nm peak broadband red phosphor .
(as the latter is converting too much of excitation energy into unwanted and useless NIR radiation.)
3) Stokes shift losses are less with a broadband 640 nm phosphor ,than with the broadband 660 nm one.Thus a higher radiometric efficiency is achieved ,when utilising the former.

The inside LES circle is a mix of green and yellow phosphors*
This phosphor mix will emit a cool white light of ~ 5000 °K ,80-Ra .

(* can be GAL (Aluminate ) or NYAG ( Garnet) phosphors ,
but rather not Silicate ones, due to the relatively large particle size - D50(V)- of the latter )

If this COB outputs totally 1500 umol/sec ,then it could be "broken down " to something like that :

Channel #1 ( 5000° K )
400-499 nm : ~250 umol/sec (peak 450-455 nm )
500-599 nm : ~400 umol/sec
Channel #2 ( broadband 640 nm )
600-730 nm : ~850 umol/sec (peak 640-645 nm )

......
Just a rough idea ,there ...

With the use of dimmable drivers and this COB ,the grower can possibly achieve quite a few things ...

During seedling and early vegetative stages of plant growth ,the grower can set
the Channel #1 at 100% output and channel #2 at 20% output ...

~820 umol/sec totally ,is more than enough radiation for a vigorous vegetative growth ..
Plus the fact ,that the spectrum emulates closely enough ,the sunlight during early spring ,
thus resulting in a greater female:male ratio of plants and a "stocky " stature .
From late vegetative growth to midst of flowering ,the grower can have at 100% ,both channels ...
And then ,during late reproduction stage (towards harvest ) ,the grower can
set the channel #1 at 20% and channel #2 at 100% ( total ~980 umnol/sec ) ,
so to avoid the unwanted foxtailing and the degradation of the cannabinoids
(trichomes turning amber/brown ) ,due to high energy / short wavelength photons.

Throw in a microcontroller ,and then the " spectrum shift " will be handled automatically ,
during the whole growing cycle ..." Spring light =>Summer light => Autumn light " ....

Alex ....
I'm hoping that you and the rest of the FCO team ,will think it over ...
You can always make some "beta" CObs and let us do the actual testing ..
(:

Cheers.
:peace:
Click to expand...
 
BuddyColas

BuddyColas

Well-Known Member
Alex Lau said:
Hello There! This is Alex from Flip Chip Opto.

Let me know if there is any technical questions I may be able to answer!


-Alex
Click to expand...
Not a technical question...a sale question. Is the Luna 50 currently available? If so, where can I get my hands on one? Thanks.
 
stardustsailor

stardustsailor

Well-Known Member
alesh said:
@Alex Lau I'm quite baffled when going through your data sheets. Everything is specified at Ta=25°C. What's that supposed to mean? It's great to know that ambient temperature is 25C so you were comfortable but what's the actual temperature of case or junction during testing?
Click to expand...
And probably there are few mistakes ,too...

For the red phosphor part ..
If that graph was done while :
Vf=46,1 V , If = 6A ,per each red channel ...
Then at the graph the figures at the y axe ( uW/nm ) are missing one zero..
It should be 900000 uW/nm and not 90000 .
If that is the case then ,according to my calculations :
Each Red channel emits ~450 umol/sec (@ spec sheet : 276,2 umol/sec)
(600-630 nm :~ 55 umols/sec
631-660 nm : ~131 umol/sec
661 -680 nm : ~90 umol/sec
681 -730 nm : ~128 umol/sec )
Lumen flux is 6551,2 lm per red channel (@ spec sheet : 5320 lm )
output power per red channel is ~ 80 930 mW = 80,1 W ,
thus radiometric efficiency is about = 80,1 / ( 46,1*6 ) == 0,289= ~29%

Cheers.
 
Last edited:
nogod_

nogod_

Well-Known Member
That's a shame...I was looking forward to a fancy new avatar.
:P

stardustsailor said:
And probably there are few mistakes ,too...

For the red phosphor part ..
If that graph was done while :
Vf=46,1 V , If = 6A ,per each red channel ...
Then at the graph the figures at the y axe ( uW/nm ) are missing one zero..
It should be 900000 uW/nm and not 90000 .
If that is the case then ,according to my calculations :
Each Red channel emits ~450 umol/sec (@ spec sheet : 276,2 umol/sec)
(600-630 nm :~ 55 umols/sec
631-660 nm : ~131 umol/sec
661 -680 nm : ~90 umol/sec
681 -730 nm : ~128 umol/sec )
Lumen flux is 6551,2 lm per red channel (@ spec sheet : 5320 lm )
output power per red channel is ~ 80 930 mW = 80,1 W ,
thus radiometric efficiency is about = 80,1 / ( 46,1*6 ) == 0,289= ~29%

Cheers.
Click to expand...
 
bassman999

bassman999

Well-Known Member
I just want to know how efficient it is at wattages riven typically by cree. I wont be doing 150W or higher per cob and def not 300, but 100W or less is a possibility if I get a bigger grow space.
If they can be more efficient as 50-75W than cree its worth the price tag imo and thermal runaway in parallel configurations isnt a concern
 
stardustsailor

stardustsailor

Well-Known Member
The spec sheet of the Duet 2400 needs ..reworking ...
Digitized the graph with the blue section emission ,from 400 to 600 nm = x range .
( max Y value line = 6E+5 uW/nm = 600 000 uW/nm =600 mW / nm )
....
Then y data obtained ,passed on a spreadsheet ...

Total output 400-600 nm power = 12,68 W (from 400 nm to 600 nm )
PPF = 47,52 umol/sec
( 3,74 umol/sec/ Wφ)
Lum.Flux = 388 lm

It's obvious that the power plot was not made ,with the DUET 2400 @ 46,1 V & 6 A ...:evil:


So ,the previous red section plot excluded data are wrong,also ....

Update:
Output PPF / output power / lum.flux for the red section (calc. range : 400-800 nm ) : 45 umol/sec / 8 Watt / 655 lm .
( 5,625 umol/sec/ Wφ)

Both power plots most probably were made with the same Vf and Io .
But both are not stated and remain unknown **..

**( Seems that the DUEt 2400 was driven somewhere close to Vf=~ 39 V Io= 500mA / P=19,5 W .,in order for the explanation of plots values to be close to ...sanity ...
If that is the case then rad efficiency for the blue section is (at that current ) is about ~65%
(...not that great I've to admit ....
500mA / 12 rows = each diode of the 180 (per section )is driven at 41,6 mA...
Rad efficiency should have been close to 80% or even more -- like the Cree or Bridgelux diodes used in CXA,CXB / Vero series... )
and for the red section is ~41% )
Bummer ...:cuss:

Another assumption is that ,since the four different sections have same driving characteristics then :
We may assume 12,68- 8 = 4,68 W ...
That's the losses of conversion ...
Almost 37% of excitation energy is lost (turned into heat ) due to phosphor conversion ....
(At Tc =x ? Vf=y? Io= z? )
So the red phosphor overall efficiency is about 63%...
( 65% * 0,63 = 41% )

I'm sure there are few ways to increase that figure ...:wink:
Still...

Duet 2400 spec sheet lacks plenty of information ...
Unlike the spec sheets of Cree or Bridgelux ...:fire:
 
Last edited:
stardustsailor

stardustsailor

Well-Known Member
Hey Alex ...
Would not be better if each section had 15 rows of 12 diodes ,
instead of 12 rows x 15 diodes ?

Then at 6A Vf would have been close to 36,87 V and each diode will be driven
at 400mA (instead of 500 mA ) ...
That should gain some efficiency ...

Once again ...
The use of 640 nm or even 630 nm peaking broadband red phosphor ,will add to the efficiency gains ...

Lastly ,the use of highly efficient blue diodes ,remains the primary measure to be taken ,
in order to achieve high efficiency in phoshor conversion ...

Just some notes ,there.
From a smart-ass wise-guy ...:P

Cheers.
:peace:
 
don benosee

don benosee

Well-Known Member
Alex Lau said:
Hi don benesee,

We ran the Apollo 2400 with 4X mean well drivers + a smart control would allow independent control via a remote or even your smart phone. Our team is busy working on that right now.
Click to expand...
Now that's some good thinking. Thanks Alex.
 
SupraSPL

SupraSPL

Well-Known Member
alesh said:
@Alex Lau I'm quite baffled when going through your data sheets. Everything is specified at Ta=25°C. What's that supposed to mean? It's great to know that ambient temperature is 25C so you were comfortable but what's the actual temperature of case or junction during testing?
Click to expand...
I was wondering the same, very good question. COBs pulsed at room temp so that Ta, Th, Tc and Tj all equal 25C?
 
SupraSPL

SupraSPL

Well-Known Member
The older spec sheets used to list everything at Tj25C (junction temp) but the newer ones list lumens at both Tj25C and Tj85C. Recently the specs sheets have started using a confusing mixture of Tj and Tc (Case temp of the actual COB or solder point temp). Tj25C is unrealistic for the most part but Tj 35-40C is possible for large COBs running soft. Tj85C I would consider running hot for our application, but commercial LED fixture engineers consider that acceptable in many products. Tj 150C might be the theoretical maximum for LEDs, OK for short periods of time.

I understand they are trying to move toward using Tc in the spec sheets because it is theoretically possible for users to measure Tc rather than Tj BUT the Tj depends on more than the case temp, it depends thermal resistance of that model COB and the dissipation power. I think that is why they are using a mixture of both.
 
zangtumtum

zangtumtum

Well-Known Member
SupraSPL said:
The older spec sheets used to list everything at Tj25C (junction temp) but the newer ones list lumens at both Tj25C and Tj85C. Recently the specs sheets have started using a confusing mixture of Tj and Tc (Case temp of the actual COB or solder point temp). Tj25C is unrealistic for the most part but Tj 35-40C is possible for large COBs running soft. Tj85C I would consider running hot for our application, but commercial LED fixture engineers consider that acceptable in many products. Tj 150C might be the theoretical maximum for LEDs, OK for short periods of time.

I understand they are trying to move toward using Tc in the spec sheets because it is theoretically possible for users to measure Tc rather than Tj BUT the Tj depends on more than the case temp, it depends thermal resistance of that model COB and the dissipation power. I think that is why they are using a mixture of both.
Click to expand...
Tj25C it's realistic,in my opinion, only if you run softly in active cooling setup...
More interesting data,I completely agree, the situation from Tj35C to max Tj50C, not Tc.
Very interesting COB of course,I'm very interested to see possible model, maybe to test on spectra-lamp, but they need more information....
 
URSA LED

URSA LED

Well-Known Member
SupraSPL said:
The older spec sheets used to list everything at Tj25C (junction temp) but the newer ones list lumens at both Tj25C and Tj85C. Recently the specs sheets have started using a confusing mixture of Tj and Tc (Case temp of the actual COB or solder point temp). Tj25C is unrealistic for the most part but Tj 35-40C is possible for large COBs running soft. Tj85C I would consider running hot for our application, but commercial LED fixture engineers consider that acceptable in many products. Tj 150C might be the theoretical maximum for LEDs, OK for short periods of time.

I understand they are trying to move toward using Tc in the spec sheets because it is theoretically possible for users to measure Tc rather than Tj BUT the Tj depends on more than the case temp, it depends thermal resistance of that model COB and the dissipation power. I think that is why they are using a mixture of both.
Click to expand...
Hi SupraSPL,

Your explanation is relatively correct, although our datasheets could be much more clearer.
Everyone seems to be interested in our Tj and Tc. You have to understand our junction temperatures, our thermal resistance is TWO DECIMAL places LESS than CREE/CITIZEN on our Apollo Series, and very competitive on our Luna series.

The thermal resistance is what you are looking at when you are building your heatsink and fixture.
 
SupraSPL

SupraSPL

Well-Known Member
Sorry I wasnt referring to your datasheets as being confusing specifically, I was referring to the evolution all LED datasheets in general having a mixture f Tc and Tj figures. I felt like I had a better understanding and ability to compare different LEDs when everything was in Tj but I understand Tc has it uses.
 
You must log in or register to reply here.
Top