DIY Modular LED-Light

"-How hot did the heat sink get for your passive 30W?

was no Problem to go for 50°C
they are now working at 450mA and more or less 40°C ambient @ 20°C .."

That is impressive...

Would you be so kind ,sir,to please me also ,and
give off the full dimensions of the heatsink used ?
Or a link with this model ?
(Probably Fischer ,uh ? ...)

Edit: And one more ...
Since I think I've found the right man....

Well ,does it play a significant role ,for the heatsink to be black anodised ?
(And I don't mean for alum. corrosion protection...I mean for heat dissipation through
"black"('grey" in reality ) body radiation....

According to Stefan-Boltzmann Law ...
q = ε σ T[SUP]4[/SUP] A

σ = 5.6703 10[SUP]-8[/SUP] (W/m[SUP]2[/SUP]K[SUP]4[/SUP])
ε=
Because ...
According to this...

Net Radiation Loss Rate

If an hot object is radiating energy to its cooler surroundings the net radiation heat loss rate can be expressed asAnodising ,really starts playing a role ,when Th is over 100° C.(373.15°K )...
(T[SUB]ambient[/SUB] =Tc=25°C=298.15°K )
It seems so....

Plus ....

right on all, sailor
hope you correct the Error in the tables, if not see my post for that tool.

for all monochrom led the curves a good enough for all manufacturer.
real data for white Leds are hard to find, so we have to live the curves

lot of data from osram
http://catalog.osram-os.com/applications/applications.do?act=showBookmark&folderId=2&favOid=0000000200034c80025d0023&CMD=SELECT
the spectrum files are in something like this xxxx_ASCII.zip

in table All_spectra_101212 you can manipulate the Output-Power

col.M testobjekt
col.N the place for your radiometric led-data

right on all, sailor
hope you correct the Error in the tables, if not see my post for that tool.

Click to expand...

It was the first thing to do....

for all monochrom led the curves a good enough for all manufacturer.
real data for white Leds are hard to find, so we have to live the curves

lot of data from osram
http://catalog.osram-os.com/applications/applications.do?act=showBookmark&folderId=2&favOid=0000000200034c80025d0023&CMD=SELECT
the spectrum files are in something like this xxxx_ASCII.zip

Click to expand...

Many thanx...

in table All_spectra_101212 you can manipulate the Output-Power

col.M testobjekt
col.N the place for your radiometric led-data

Click to expand...

Ooooochhhh.....

That really hurts...
Yeah...
I can also go and play the next lotto and win those ....
(Many little papers,anyway......)
That would be a whole lot easier...

Dunno...

Maybe I 'll put an average -conservative- value....
based on bridgelux's die data sheets....
Say 10-15% minus from phosphor absorbance....
300mWatt...
Fairly low value...

But-probably- close to reality ,though...

But they are getting better...
-They ..."have the ball" ,now....
Neither of ...us ...

Would you be so kind ,sir,to please me also ,and
give off the full dimensions of the heatsink used ?
Or a link with this model ?
(Probably Fischer ,uh ? ...)

Click to expand...

FISCHER ELEKTRONIK - SK 168/50 SA
(Rthk) 0,9 K/W (Thermal performance... link below)
L:W:H(mm) 50,0: 226,5: 40,0

Well ,does it play a significant role ,for the heatsink to be black anodised ?

Click to expand...

black anodised stays 5 - 10% better(higher temp better performance), for a passiv one. 0% with aktiv cooling

----------
more on this here

http://sound.westhost.com/heatsinks.htm
great site also, look at he rest.

and here is the next devil you have to deal with

http://homepages.which.net/~paul.hills/Heatsinks/Heatsinks.html

4.6. Thermal performance .v. mounting attitude
--------------
and a last one.
raw pure aluminum weathered with time in a bad manner for cooling.

keep it cool, bro..

reaching for the sun



no bleaching

Σ'ευχαριστώ πολύ...
( Ι Thank you ,very much...)

Now...
I need your experties ,again....

Based on min (340 mW ) & max (440 mW ) of the dies of my whites....

( And after exchanging couple of e-mails ,with the manu/er of my leds,about technical details...)
I set an average value of 390 mW ...
-10% because of absorbance by phosphor/silicone/lens...(Manu/er's measurments )
I get an average radiometric power value of 351 mW....
Set it ,to 350 mW for Cools/Neutrals....
For Warms ,300mW...(Thicker layer of red phosphor,according to manu/er...)
And still being conservative...

I've adjusted the rest ,according to average values...
(rather conservative,also...)


(Deep red=660 nm / Cherry red =730 nm /"Ultra-Violet" ..my @ss...= 400 nm )


What do you think about the values I've set ?
I would really ,appreciate your opinion ,about the @350mA ,power estimations...

(Set the photometric to null...Afterall,it doesn't have any significant value or meaning for us (growers) & plants .
Neither it affects any calculation of the spreadsheet....)



When you find the time & and if you want ,of course,please have a look...

Edit: Do not mind for the UVa 370..Haven't found still any data for 350mA..
Haven't searched yet,to be honest....

stardustsailor said:

Come to think about it....
Don't you think that the values for the 700mA ,are a bit low ?
I thought that-nowdays-they emit at 220-240 lm range ...

Click to expand...

Cree XTE @700mA:
Cool white R3 : 217 lm
Cool white R5 : 247 lm
Warm white Q3: 167 lm
Warm white Q5: 191 lm

• black anodised stays 5 - 10% better(higher temp better performance), for a passiv one. 0% with aktiv cooling
  
  ----------
  more on this here
  
  http://sound.westhost.com/heatsinks.htm
  great site also, look at he rest.
  
  and here is the next devil you have to deal with
  
  http://homepages.which.net/~paul.hil...Heatsinks.html​
  
  
  
  

Click to expand...

Damn I wish I'd read that three weeks ago lol.

Well,it seems you didn't find the time (or you just didn't bother ? ) ,to answer to my queries...

It's really,ok ..
You are not obligated ,after all...
But I got something for you ,to examine....

(Got the time to find out that "ledengin" spreadsheet ,has some serious issues...)

a) It seems that,it doesn't measure the output flux power, correctly...

For example: If i set all the powers of all leds at ALL radiometric power columns ( E ,M ,N ) at tab " all spectra 101212 " to 100mW (0.1 W )
back to tab "Mix Spectrum",when I set 100% to any ,almost,led I get an total flux over 0.100 W ...

WoW ....Bonus Power from nowhere....

b) To measure (or more correctly ,to convert mW to umol/sec ) PPF ,it uses a math equation like :

( Ox *( Ax/10^9))/(R$20*R$21*R$22)*1000000...

Which is ...totally wrong....

To find out How many umols/sec is a given power in Watts ,this math. equation is
absolutely presice...

W /( umol/sec ) = 119.708 /wl (nm) ...=> PPF= Power (in Watts) * wl (in nm) /119.708


Thus ,for the spreadsheet,it should have been ..

Ox *( Ax/119.708 )
.....
So ,both ,the results about output power ,are wrong,from what it seems...

It's fun to play though and just mix spectrums...

...radiometric power columns ( E ,M ,N ) at tab " all spectra 101212 " to 100mW (0.1 W )

Click to expand...

did you understand this table really?



the columns A to I are the measurements of the Leds

the column M is the same as in E with a factor 1000
the column N is the Flux-Bin of the Led you actually use
the column I is the ratio of M/N

absolutely presice...

Click to expand...

119.708
ever look at the solution of this term
R$20*R$21*R$22*1000000

guod said:

did you understand this table really?



the columns A to I are the measurements of the Leds

the column M is the same as in E with a factor 1000
the column N is the Flux-Bin of the Led you actually use
the column I is the ratio of M/N


119.708
ever look at the solution of this term
R$20*R$21*R$22*1000000

Click to expand...

I know perfectly well ,what every column represent...
Still ,I get an output power over 100mW ,when I set the max.Output flux of led to 100mW (column N ) ..
How that's possible ?

R$20*R$21*R$22*1000000....
Solution of this term ?
No....
Can you help me a bit ?
Is it the same as 119.708 ?
?

Still ,I get an output power over 100mW ,when I set the max.Output flux of led to 100mW (column N ) ..
How that's possible ?

Click to expand...

same here,
I will check that out...

--------------------
and i allways check tools from the InterneT twice, before i start!

R$20*R$21*R$22*1000000....
Solution of this term ?
No....
Can you help me a bit ?
Is it the same as 119.708 ?

Click to expand...

first things first, that can also be wrong

pic of the week


highlighted Aera at 100%

salted popcorn everywhere
------------------------------
and the other one

reactor critical

Attention!
we are now 1 meter under the canopy.


..........
--------------------------------------------------------
for nerds only

Usage and control of solid-state lighting for plant growth

PDF-parts 3 & 9 for a nice start

https://aaltodoc.aalto.fi/handle/123456789/4514

Haha-ha...
"..for nerds only.."

Liked that....

Oh ,com'on now....
I have some other things to do than sit and read papers from 2005 ,considering leds...
Lots have changed in 7 years time, from then....


....

Ohhh.I can't resist...
..

Chlorophyll-a has a strong light absorption peak located
around 660 nm.

Click to expand...

When inside a tube diluted ,not on a living plant...
We have been ,through,this...

Luminous intensity, typically presented in millicandela, is the most measured
LED optical parameter. It is also the most misunderstood LED parameter.
Luminous intensity is defined as the visible flux per unit solid angle in a given
direction from a source. Luminous intensity measurements frequently involve a
point source, the inverse square law and the assumption of constant illuminance
across the detector. A point source need not be small. For example, a star, which
is not small, when seen from a far distance (as from the earth), can be treated as
a point source. The angular distribution of various point sources with rotational
symmetry can be approximated (Fig. 4.0) by:
I = Io cosg-1θ
Where Io is the intensity normal to the source itself and g ≥ 1. For example:
g = 1 for an isotropic point source
g = 2 for a Lambertian point source
g > 30 for an LED point source (LED lamp)
The inverse-square law (which is strictly applicable only for all point sources) states that the
illuminance E at a point on a surface (a calibrated detector) varies directly with the intensity I
of the source (as defined above for various types of sources), and inversely as the square of the
distance d between the source and the point (detector).
E = I / d2
If the surface of the detector is not perpendicular to the direction of the incident light, the
Cosine law comes into effect and the angle of incidence θ must be considered. The equation
may be expressed as follows:
E(&#952 = (I(&#952 / d2) * cos θ
A useful extension of the Cosine law is the cosine-cubed law. The Cosine-cubed law comes
into effect when the surface at the point (detector) is moved up and down in a vertical plane
perpendicular to the source. The cosine-cubed law can be expressed as:
E(&#952 = (I(&#952 / d2) * cos3 θ
Where “d” is the distance between the source and the detector at θ equal to 0°. For standard
lamps, the angular distribution is wide enough that uniform illuminance across the detector is
not an issue and these equations hold true within one-half percent when the source as viewed
from the detector is in the far-field. For a typical light bulb or luminaire, the far-field distance
"d" is defined to be at least ten times the maximum dimension of the source. For an LED,
according to industry experts, no such rule-of-thumb exists and, the far-field distance "d"
can only be determined by finding the 1/r2 irradiance fall off region through measurement.
Once the far-field region is determined, measurement errors can still occur if the illuminance
detector in use is "too big" in a relative sense, where the illuminance across the face is not
constant. The inverse-square law references E at a point on the surface of a detector. Simply
restated, the intensity of the source can only be calculated as the area of the detector goes to
zero. For LED lamps, the relative size of the detector in the calculated far-field can be too
great, making the angle subtended by the detector at the source a non-point. In such a
situation, the illuminance on the detector would not be constant and the detector would be
performing an average intensity measurement. In order to minimize or calculate the averaging
effect, an equation for the irradiance over the surface area of the detector can be written and
solved for, as the solid angle subtended by the detector is taken to go to zero. (Fig. 4.0.):
E(&#952 = dφ / dA = ∫I(&#952dw/πr2
Assuming the intensity pattern of an LED can be approximated by I = Io cosg-1 θ, E(&#952
can be approximated as:

E(&#952 = (Io / d2 ) • [1 - (((g + 2) / 4)•(r / d)2)]

In this equation the (((g + 2) / 4) • (r / d)2) portion of the equation accounts for the
averaging effect uncertainty with respect to the true intensity value. By setting this portion
of the equation equal to an acceptable chosen uncertainty, one can derive the required
relationship between the radius of the detector and the "far-field" measurement distance:
d ≥ r•√[(g+2)/(4&#950]
where ζ is the uncertainty in question. For example: if the LED to be measured has a half
angle of 15° (30° viewing angle), g can be solved by substituting the half angle of the LED
for θ and realizing that I(&#952 / Io at the half angle is 0.5 by definition, then:
cos (g-1) θ = 0.5 where θ = half angle
solving for g yields:
g = 1 + Ln(.5) / Ln(cos(half angle))

The area of the detector is
always 100mm 2
For an LED with a half angle of 15°, g = 21. If “g” = 21 and ζ is chosen to be 1% then the
radius of the detector must be = d/24. If “d” is chosen to be 100mm (reasonable for most
LEDs), the radius of the detector must be equal to or less than 4.2mm in order to maintain
the chosen 1% uncertainty. For such a geometrical setup, the plane angle formed by the
measurement setup would be 2.4° (tan-1(4.2 / 100)) and the solid angle subtended would
be 0.005sr ((area of detector/1002)).
Having discussed the far-field and equating a proper detector size within a chosen uncertainty,
the only issue left in calculating intensity is determining the actual location of the effective
light center of the LED. Within an LED, the light is produced in the LED chip. A portion
of this light goes directly from the chip and is refracted by the epoxy dome. The remainder
of the light is reflected by the reflector cup and then refracted by the epoxy dome. The light,
which is directly refracted, appears to come from a certain location within the LED, while the
light, which is reflected and refracted, appears to come from a different location. In addition,
because the LED chip itself has a physical size and is not a point source, the refracted light
does not appear to come from a single location, but a range of locations or a focal smear
(virtual image). To create the best approximation using a point source model, the center point
of the focal smear should be chosen as the location of the point source. Manufacturers often
provide the optical position of the point source. If this point is not provided, one can be
calculated by taking two different illuminance measurements, E1 and E2 in the far-field on
the same axis, at distances d and d+x apart. Since the intensity is the same at any point within
the same solid-angle within the far-field, then:
E1 • d2 = E2 (d+x)2
Since E1 and E2 are known measured values d can be solved for:
d = (x(E2 +√E1E2))/(E1 - E2)
and the intensity can then be calculated.

Click to expand...

That's more like it ,for the actual physics ...

As it happens in most of lighting applications the uniformity
of the light distribution over the surface should be as good
as possible. LEDs have half-isotropic(120° g=2....g=1 =ideal point light ) emission which
makes them directional emitters. Nevertheless some of the
emitted photons propagate in directions defined by large
dispersion angles. Depending on the mounting height
(YES! That's what I'm talking about ..Thank You!!! )
this results that in certain situations ?????.Now the good part starts....
YEs.Ok. "certain situations".
Can you imagine which ,that might be ?
When we want to illuminate from far above ? )
a significant portion of the light emitted is misused.
For those cases the radiation pattern of the light emitted by the LED should be more carefully
considered. The use of secondary optics or LEDs with
small viewing angle can be a valid option in order to more
efficiently direct the light towards the area to be illuminated. (Meaning when ?At which cases ? Dunno I'm stupid ... )
Collimator lens are not perfect devices although they have
a high optical coupling efficiency which can be up to 85 or
even 90% in some cases. Collimators are encapsulating lens
which can reduce the number of LEDs required to achieve
the desirable PPF level and therefore reduce the wiring
complexity.(Ahaaaaaaaa...! Now ,you 're talking ,dude..That's the problem....)

Click to expand...

".....The results of the simulations indicate that reductions of the
number of LEDs can be achieved by collimating the light
emitted with some expense on uniformity...."

Click to expand...

According to the type of use ,they utilised then ..
Yes,the man says no lies...
But...
Leds may..And ,I say just may,not work that way...

There is -always-the other "side"....
Which among others,mainly is based on the actual numbers of low current ,driven leds...
Not in a few numbered-but powerfull- led illuminaire...
Then,the whole thing ,changes...
Doesn't ?

..
Ahaaaaaaaa!!!!

Now,I've got you....!!!!

By using LED luminairies plants can be lit in different and
more flexible ways than ever before.Small luminaries may
be design in order to allow a distributed lighting strategy
closer.....
-...He/she means,no lenses,also..Depending on the mounting height ...Doesn't he/she .........?
..Neither he/she means ,"reductions of the number of LEDs "..Uh ..????)
....to the plants which makes the illumination more
effective.
Why more effective,dude ?
You wrote a whole chapter about lenses ,and nada -*not even a sub note.???..*- for what you 've just mentioned..!!!
Why ???
????
....
.
-w.t.f.............Wiring.????.
-Guh..Ehhh..Uhm..Ehm....
-Yes or no..?
..
Anyway...
.So,they took the other way,nevertheless...)
....The use of LEDs as an artificial light source in
horticulture sounds promising. The high potential of energy
conversion, small size, directional light(WTF you put lenses then,at first place ? When they are needed ? Wheeenn ? ),
spectral emission and chromaticity variety will make LEDs a viable light source
in artificial lighting for plants in the future. Until then, prices
reductions and increments on the total radiant output power
per device have to continue. Just in that way will be possible
that LED-based luminairies for plant growth compete with
conventional and low cost light sources such as HPS
lamps, in a more effective way.

Click to expand...

Blah,blah,blah......
Yes,thank you,we 've found out ourshelves ,also ,that this is possible....
But we do it ,in another way...
Totally different way,than the one you have described, in your paper..
More like the way,you characterised as " flexible than ever before"...

LOL....

...
#9,later...

Guod,nevertheless ,the pdfs' are worth taking a look...
I 'll do it when I 'll find some serenity time...
Thanx,anyway for the link...

Mythbusting the Titans.....

7. Conclusions
During the design stage of an LED luminaire for plant applications
several compromises need to be established. The
design of an LED-based luminaire for plant growth has to
have into account the maximum photosynthetic and photomorphogenesis
performance of the plant. This should be
achieved with the maximum energy efficiency possible.(note that !!!!)
Due to the high cost of LEDs(aha.....!!!!)and wiring complexity(???)the number
of emitters required should be reduced to the minimum
indispensable number.
(But if more powerful ...-25% energy efficiency,note that....)
The use of secondary optics, namely collimators can reduce the number of LEDs required to
achieve a certain PPF light level.

( Uh? What he just said ? To use something that reduces "energy efficiency possible." -10% at best case..If to use powerfull leds...Another -25% .?????.
Plus the losses from Inverse Square Law as applied for leds....

" By applying the inverse square law it is possible to obtain
the expression which calculates the irradiance level at the
calculation point. This expression is given by the Eq.4."...

plus less coverage...
Plus spectrum uniformity ,less than optimal...Uh ? )

This is achieved by redirecting
more effectively the light beam to where it is required
and reducing in that way the light waste.
(having already -35% at least,if you choose the option of 700mA or at least
-10% if you stick to the lenses,only...????...
WTF you 're talking about,dude ? )

Due to reduction
on the number of LEDs the required input power is less
resulting in lower running time costs.

(I'm going to faint.....)


Depending on the type
of optics used and their optical efficiency and commercial
price reduction on the initial costs of the luminaire can be
attained.(Aha! ) The initial costs include electronic drivers, LEDs
and secondary optics. Nevertheless a trade-off between
mounting height of the luminaire,costs, circuitry simplification
and lit uniformity is required considering the specific
lighting strategy and the application itself.
(Ahaaaaaaaaaa!!!!!!!!!!!)

By using LED luminairies plants can be lit in different and
more flexible ways than ever before.

Small luminaries may
be design in order to allow a distributed lighting strategy
closer to the plants which makes the illumination more
effective.
(Now,we can all faint...)
The use of LEDs as an artificial light source in
horticulture sounds promising.
(In what way ? How ? ....)
The high potential of energy
conversion, small size, directional light, spectral emission
and chromaticity variety will make LEDs a viable light source
in artificial lighting for plants in the future. Until then, prices
reductions and increments on the total radiant output power
per device have to continue. Just in that way will be possible
that LED-based luminairies for plant growth compete with
conventional and low cost light sources such as HPS
lamps, in a more effective way.
(*Conserning energy or cost ,dude ? I lost you .....)

Click to expand...

...has to
have into account the maximum photosynthetic and photomorphogenesis
performance of the plant........= prices,cost , low cost, high cost...

Get it ?



Who ..funded... the whole research ?
...
Just asking ....
Being ,purely,just curious..


What more to say......
...
Opening pdf #9 ...
It came all back to me ..I have read it in the past...
....
Things changed,from then...
A lot...

after reading this

stardustsailor said:

Haha-ha...
"..for nerds only.."

Liked that....

Oh ,com'on now....
I have some other things to do than sit and read papers from 2005 ,considering leds...
Lots have changed in 7 years time, from then....


....

Ohhh.I can't resist...

Click to expand...

and see your second big post over this PDfs

i have a very little smile on my face, ... very little!

thanks..

Now,you caught me!..
Don't know ,what you really mean...
Are you just being ,bad,with me ?

..
Or I misunderstood you ?

Oh! Com' on ...
What the heck ?
We can not be liked by everyone..
Can we ?
..
Nobody's perfect....

And ,I'm nobody...
ha-ha-ha-ha.....

I really do,though,admire your thirst for more ..
For that small,little detail...

I admire also,your excellent workmanship,with your led implementations ...
( Think you have the knowledge & experties by now ,needed for the making of the ultimate micro -panel..
Main problem really ,remains the whole "placing" thing...
Can Grow rooms or tents,be modificated to be like "space grow chambers "?
No...Wait..I'm joking...
But ,it's not that far from reality..
Truth is there are..."trade-offs"..
Ease of placing & adjusting...That's one ,really sh!tty...
So,to use the leds as spatially efficient, as it can be ..How ? How one can hang ,individually,10 or 20 small little panels ,in a tent ?
Wiring ?....... OMG !! Did ,I just said wiring ? ....See ? ..Nothing is perfect ....)

Oh,well...

I could not leave your suggestion,without reading it first...
..Twice......
Noway.I could not just let it drop on the floor...
Anyway...
We've gained a lot,from it..
All of us...