Astir Grow Led Panel Project...

Bumping Spheda

Bumping Spheda

Well-Known Member
Good reads SDS. Thanks for the links. It's too bad, but I'm not paying to read any of the full ieeexplore articles. At least the abstracts give you a pretty good idea of what's going on.

So I guess the "windowed" phosphor method decreases efficiency, but also decreases wide-angle CCT variation. I'll take efficiency, I think.

Seriously, a converging lens would be a really cool experiment, I'm surprised I haven't seen anything with one yet. Converge the light to a very fine point (think laser) where it enters the back of a cone reflector with a very small diameter hole at the vertex, then the light diverges again, then it hits the phosphor and only light that's reflected directly back out of the reflector is lost. And, assuming all diverging light excited the phosphor (albeit at different angles), what if the reflector was much larger than the phosphor in order to take advantage of the windowed method? A 1W diode sized entry hole carrying 20-50W of light directed at a phosphor designed for maybe 10W surrounded by a 10-20W sized flood lamp reflector. That's the premise, I guess.
 
stardustsailor

stardustsailor

Well-Known Member
Yeap..You got some nice ideas there ....
Although using collimator & cone + phosphor layer ,might have an impact in overall efficiency ...
Gain there ,lose here ...
But really cool idea ,nevertheless...


A 1W diode sized entry hole carrying 20-50W of light directed at a phosphor designed for maybe 10W surrounded by a 10-20W sized flood lamp reflector.

The solution here ,is increasing distance between phosphor layer and leds..
(more blue light is falling dispersed over the surface of phosphor layer and thus excitation "hot spots " on certain parts of the phosphor layer,are avoided....)


Another possible drawback is the cost of collimating mechanisms ,along with the cone reflectror .....


Check out this :

http://www.digikey.gr/search/en/CL-DEMOKIT-2D-PC/1079-1030-ND?recordId=2688355&keywords=intematix+

[url]http://media.digikey.com/pdf/Data%20Sheets/Intematix%20PDFs/ChromaLit%20Demo%20Kit%20Guide.pdf
[/URL]
 
Bumping Spheda

Bumping Spheda

Well-Known Member
What if the phosphor matched the LED's viewing angle so all beams would travel the same distance and strike perpendicularly? Harder to do with a die matrix chip, but possible, and to an acceptable accuracy once calculated, I'd assume. One of those patents was talking about vacuum forming the phosphor/epoxy mixture. You could make any shape you wanted that way.

Edit -
The solution here ,is increasing distance between phosphor layer and leds..
Are you talking like DLP projection? How much bigger?
 
stardustsailor

stardustsailor

Well-Known Member
Look ...
In the back of my mind ,I always keep the "simplicity"....
We lack -realllly expensive -high tech equipment like plating spinners ,centrifugal separators ,vacuum ovens,pulsed jet sprayers ,ect

...
We 've to settle with more simple manufacturing equipment ,like a DIY vacuum glove chamber,an high quality- with ultra thin needle-airbrush ,some clear polycarbonate / PMMA sheets ,a bit of solvent and begin with some phosphor samples (maybe green ,for sure yellow,orange & of course reds .. )...

Simple design ...Driver ,heatsink ,leds ,MCPCB,polymer mixing chamber ,reflective & diffusive material & remote phosphor screen ...
Each part will be re-evaluated and " adjusted " ,to work efficiently with the rest ....
For now ,heatsink,leds & mcpcbs are not of an issue ...
Heatsink is really great ,MCPCB s also and we can get some high quality blue leds ....
We are considering some better drivers ,to go with...
Polymer mixing chamber design and optical geometry and it's reflective coating is also , under investigation ....
What awaits for .."playin' with " is the Remote Phosphor Screen thing ....

We 're not aiming into manufacturing the state -of-the-art design ...
(At least ,right now ,we 're not...Maybe a bit later... )

We just try to make a decent & rel . cheap led (in relation with manufacturing cost,time & " trouble " )lamp ,
that can actually grow healthy and productive plants ..
 
stardustsailor

stardustsailor

Well-Known Member
My rough approximation is that a 24 led panel,of ours , if turned into Remote phosphor one, will have an additional cost of + 40-60 € ....
( polymer mixing Chamber + Remote phosphor screen + assembly with the rest of panel )

So say 130 -140 $ for each panel at worst case senario ....So ...Say at worst case is 0.2 W / $ ....
I know that this is a poor figure...But the whole concept of multiple panels is(probably) the most expensive one ,already...
For larger single panel with 3-Watters and Remote Phosphor / heatsink & fan modules ,yes it might be better than 0.4-0.5 W / $ .....


Right now we're weighing ,whether is more cost-effective , to make our own screens or have them made according to our specs ...
They are of quite large surface for 24 leds ... ( 200 x 160 x 1 mm ) ....
But leds -specially blue & whites- do get better and better....
So better have a large sized " HD* ready " screen .....
*High-(flux of light ,of course )- Density ........
 
stardustsailor

stardustsailor

Well-Known Member
This remote phosphor and blue/violet/uv leds combo sounds way ,effective...

Light that is produced ,could actually be the closest simulation/replication -of-natural -sunlight ,ever ,that leds can achieve / provide ...
Omnidirectional & diffuse both in power and spectral terms ,"Full " regarding spectral quality , plant growing -wise...
Almost of full homogeneity ...

Cooling-wise ,led devices work less affected with back-reflected light (less heat generated ) ...Easier cooling...

Electrical-wise ,led as devices work mainly(and more efficiently ) as light emitting ones and not as light harvesting ,"electric energy generators "....Plus the fact that they work cooler...
....( white ) Leds show increasements at the 30-60 % range , of final efficiency ,with remote phosphors ...

Stability/Durability-wise both led & phosphors work(and stay..) more stable and last way longer ....

Modularity ? Adjustability of spectrum ? ( different Dies used or just change screens ...)..
...
Way many advantages ....

Price though ,is one and serious drawback...

We'll see what we can do about it,though...
Nothing is final..
It's just the beginning....
 
stardustsailor

stardustsailor

Well-Known Member
...
ASL 24 Remote phosphor hot spot led rad.jpg

The above is the surface plot ,of 24 "excitation " led's , light "cone" bases ,arranged according to the MCPCB .
Distance from led die level to phosphor layer level is 20 mm .
This is just the light cone base of each led ("irradiation trace ") falling on the screen .Circle perimeter is half-power .( 2θ/2=120° ) .
( R= d x tan 60° .d=20mm => R= 34.64 mm = ~ 35 mm* )

Array consists of 4 ' vertical ' rows,each with 6 leds ( or 6 ' horizontal 'lines x 4 leds )..
Distance from led to led (die level ) in rows is 25 mm and led to led in a line is 30 mm ..

MCPCB is 182 x 142 mm .
Led Array(on PCB ..) is 125 x 90 mm .
For each side of array ,35* mm of light radius added....
125 + ( 2 x 35mm )=> 195 mm
90 + (2 x 35 mm ) => 160 mm

195 x 160 mm coverage with min of Power at 50% ....
Remote phosphor screen size is 200 x 160 mm ....
So...
2 cm distance between led dies and back of screen ,is ideal ,for direct excitation of screen from led light, covering 97.5 % of screen's surface ....

Caution: Only direct irradiation of screen,is studied .Light from reflections needs ray-tracing software ...
And a "mixing chamber " designed ,so to simulate ......(which doesn't happen ,right now...)

ASL 24 Remote phosphor hot spot led lut.jpg

Here is the summed overlapping ,of led light cone bases ,as irradiated 20mm away from dies ,onto the Rem. Phos. screen ....
Study of "hot spots " .....
.....

The following are rough approximation ,of different direct light of leds properties ...

( Half angle Irradiance at Rem phosp. screen - simulated emission of screen (only for direct light excitation) - same as before ,but more detailed )
....
Surface_Plot_of_ASL 24 remote 1.jpgSurface_Plot_of_ASL irradiation.jpgSurface_Plot_of_ASL remote 2.jpg
 
weedman420gr

weedman420gr

Active Member
Merry Xmas and Yields,

Day 23 today.yesterday, my little girl was fed with 1/2 table spoon of 20-20-20...

Temps; 26-28C ON....20-25C OFF...
Humidity; 30-45% (not bad!)

So,here is the first Comparison Post...more will follow...

First 3 pictures;the 44 watt LED plant at Day 23 from seed...lights 20/4
Last 3 ,the 70watt HPS plant at Day 24( think it was 4 days old already when i started the journal) lights 12/12
 

Attachments

Bumping Spheda

Bumping Spheda

Well-Known Member
Just thinking here:
LED chip Reflector
Royal Blue LED chip
50W cc LED driver
50W LED heat sink
75mm 2700K ChromaLit round

$76 for everything after shipping (not including thermal plaster, epoxy/hot glue for assembly, wall plug/cord, etc.), ~51W @ 34V*1500mA, ~.67W/$. The LED reflector has a 78mm diameter lens so the 75mm round ChromaLit should replace it nicely.

Again, given the price of the phosphor lenses I feel more Watts/cm^2 makes sense. I suppose you could sprinkle ~400nm LED's of lower wattage in somewhere if you had a custom (self made?) reflector built specifically for a different phosphor lens. That's a different project, though.

Epistar makes a 45mil 50W 445nm chip, but it costs twice as much.

I think it should -at the very least- sufficiently act as a veg/mother light. Then, again, fully assembled 30-50W LED flood lights can be had for ~1W/$, and induction is looking more and more enticing. :/

Thoughts? Concerns? Hell yea's or no's? Lmk.
 
Chronikool

Chronikool

Well-Known Member
Bumping Spheda said:
Just thinking here:
LED chip Reflector
Royal Blue LED chip
50W cc LED driver
50W LED heat sink
75mm 2700K ChromaLit round

$76 for everything after shipping (not including thermal plaster, epoxy/hot glue for assembly, wall plug/cord, etc.), ~51W @ 34V*1500mA, ~.67W/$. The LED reflector has a 78mm diameter lens so the 75mm round ChromaLit should replace it nicely.

Again, given the price of the phosphor lenses I feel more Watts/cm^2 makes sense. I suppose you could sprinkle ~400nm LED's of lower wattage in somewhere if you had a custom (self made?) reflector built specifically for a different phosphor lens. That's a different project, though.

Epistar makes a 45mil 50W 445nm chip, but it costs twice as much.

I think it should -at the very least- sufficiently act as a veg/mother light. Then, again, fully assembled 30-50W LED flood lights can be had for ~1W/$, and induction is looking more and more enticing. :/

Thoughts? Concerns? Hell yea's or no's? Lmk.
Click to expand...
The heatsink is way too small in my opinion.
 
You must log in or register to reply here.
Top