New Experimental Advanced DIY Designs.

stardustsailor

Well-Known Member
....

To go on with this design one has to master the techniques of making an Aluminium Circuit Carrier Heatsink ..

Thus has to know how (& be able) to :
-mix a thermal conductive epoxy formulation from low viscosity liquid ,2-part ,epoxy resin & a thermoconductive filler in small particle size .I.e BN
-Apply a thin layer of thermal conductive epoxy at heatsink and bond a copper layer over it .
-Etch a circuit onto copper surface .
More blah-blah about it here ....

For this example a heatsink surface of 200 x 160 mm is chosen .
35 pieces of Oslons SSL will be soldered .

osl 35 bcc heatsink & led.jpg

30 x WW ....
oslon warm.jpg
&
5 x NW ..
neutral white oslon.jpg
....

The circuit ...
osl 35 bcc heatsink & led & lines.jpg..


And then a new experimental passive cooling design ....

" Slug Direct Copper Cooling Circuit " ...or " CCC "

Since Slugs are electrically isolated ....
Then all slugs can be "circuited " into a large area of copper ,
which may act as a heat radiator ...

Conducting heat ,direct from soldered led thermal slugs and radiate it towards air from one side
....
and from the other side ,towards heatsink,conduct it(spread it more evenly ),in larger (epoxy thin -100um- layer covered) surface ...

osl 35 bcc heatsink & led & lines & cc.jpg...


If everything is done correctly , within the specs
(mainly regarding epoxy layer's thickness,consistence & thermal conductivity ),
this design probably will possess superior cooling characteristics
( No MCPCB,no pad/glue/paste in between pcb & heatsink ___"Copper Cooling Circuit" passive surface radiator )...

Moreover .....If an fan is added ,at back of heatsink .....
Cooling efficiency , is not going to be an issue ...
Not for this kind of panel ...
 

HiloReign

Well-Known Member
So where do I buy one exactly :hump:?

This is going to be great.

I have been itching to order these Osram LCW CQ7P.CC for a little while now! I think I may be near finished finding parts for a dimmable panel, but now I must learn to reflow solder!

I appreciate the work you share with us, my friend!
 

polyarcturus

Well-Known Member
its a very good idea in fact its so simple (relative to the tech these led companies have access to) that these are not being produced yet.

very innovative idea.
 

bicit

Well-Known Member
The amount of info you post in a very useful no bullshit format is amazing. I can't even comprehend a lot of it. It will certainly help me along my way to building an LED panel. Thank you for your hard work, you are both a scholar and a gentleman.

Is this a veg lamp or a flower lamp? Is this all theoretical at this point? By that I mean can you effectively grow plants under all white LEDs?
 

stardustsailor

Well-Known Member
I would go forthe 7V (as first choice ) or 7S (as second choice,if first is not available )..
Cause of their less blue ....each 35 Watt panel is going to have 30 x WW ...pretty good amount of blue all together,
even at the lowest power choice ....( 7V or 7S )

(Ok....You are pure gold ,my brother Guod ...Where that pic came from ? )


My God ! I love the curves of these Oslons SSL WW..
So Sexy !
 

PetFlora

Well-Known Member
SDS, so this is why I haven't seen you posting much on your original thread

I figured you must be busy playing Mad Scientist
 

stardustsailor

Well-Known Member
The amount of info you post in a very useful no bullshit format is amazing. I can't even comprehend a lot of it. It will certainly help me along my way to building an LED panel. Thank you for your hard work, you are both a scholar and a gentleman.

Is this a veg lamp or a flower lamp? Is this all theoretical at this point? By that I mean can you effectively grow plants under all white LEDs?
Search around the led section of RIU....All the answers to your questions and much more ,are all over ...

(-Full cycle lamp ...
-Nop...It's actually working ...
-Way lotta better than you might think...)
 

stardustsailor

Well-Known Member
SDS, so this is why I haven't seen you posting much on your original thread

I figured you must be busy playing Mad Scientist
Ahhh ! My favourest of games ....

A bit of that ...Hmmmm..And this goes there ....Oh...that has to be added,also ...Nope..this one,instead .......
....

I mean ...Science & Growing .....What can be more fun ?
( ok Sex ...But after the 3rd -fourth time in a row ,kinda stops being so fun ....)

Just stoney bablin' again ...

:-P
 

stardustsailor

Well-Known Member
Guod has hit a " hidden goldmine " ....
With his interesting question ...
Hmmmm.Let's see..

{ Guod I trust ,that I'll have your help along with this ...

For sure I'll be needing it ...}

First things first ...
For Psuagro ....

WW ssl vs CMH.jpg

If Absolute Output power is the same ,then relative power spectral graphs are directly comparable ...

(If for 120 Watts of light with CMH ,we need a 400 W lamp (CMH efficiency ? ) ,then for 120 Watts of light with leds we'll need (~0.4 eff. ) 300 Watts of WW Oslons ...
Then Relative power spectral graphs can overlap and be compared ...
So ....400 Watts CMH vs 300 Watts WW SSL 80 ...
See the above graph ...
They follow Kinda same "ups' n' downs " ...
Kinda same waveform...
But...
CMH is much more "gentle " and "full " at lowest part of wls (/violets/blue ) ...

While, reds from WW are way "superior" than CMH ...
Specially past the 630 nm mark ...
 

stardustsailor

Well-Known Member
Now ....
To Guod's question ...
Time for color choosing ...
Trimming 'n' color binning ....

Ok....
So...
These ...
spectra LCW Cx7P.jpg


Are here...


coordinate groups.jpg


The Cie 1931 standard :





And the color bins / groups ( for 2700 °K only ).....:
coord 2.jpg
....



Ok.....

From where shall we start ?
 

PSUAGRO.

Well-Known Member
the spectral distribution has changed quite a bit with the improved Philips elite agro 315w CMH....
it is a beautiful bulb...

id put up a graph......but my phone is uncooperative:-)
 

PetFlora

Well-Known Member
I never understood why no manufacturer jumped on replicating PAR for veg and flower once NW/WW/CW bulbs were viable

There is too much stubbornness in selling the old. I wouldn't be surprised if they have stupidly large inventory to dump first, so they don't want to acknowledge it

We are forced to think for ourselves
 

stardustsailor

Well-Known Member
18 different color groups for the 2700° K Warm white LCW Cx7P.
(each group having 4 different leds ...! )

72 different versions of the same 2700° K WW !!!!


Four " teams " : U-V -W-X from Cx =~0.44 up to Cx =0.48

Each team having four "grades " : 5-6-7-8 from Cy =~0.39 up to Cy =~0.43

Now...
Let's see ..
Which we would like most ...
Or which combination of them,maybe ....


Cx Coordinate .....
0.44 to 0.48 ....
Cy coordinate
0.39 to 0.43

group choice 1.jpggroup choice 1cup.jpg


It seems that the one with the most deep reds- ( High Absorption Rate Reds 640-660 nm ) is X5 .....
(
X5 group Cx / Cy
0.454 / 0.393
0.459 / 0.402
0.465 / 0.404
0.459 / 0.394


And maybe some few from X7 group to fill the "yellow -orange " ,High -RQE reds 600-630 nm part more...

(RQE=1 from ~ 600nm to ~ 625 nm )



( I would avoid the groups closer to " absolute white " , U and V ,as this means "more blue " in those leds ....
Better the groups W or/and X ....)


I wonder ,If a brother here will be able to find and provide some of the ' related 'spectral graphs ...
( X 5 to 8 and W 5 to 8 )
 

stardustsailor

Well-Known Member
Now ,back to cooling design ...
Or more precisely "thermal management " ....

Which of course is efficient cooling of leds ..
But also affects physical placement & stability (thus overall operation and longetivity ) of led onto cooling mechanism / electric circuit .

So....ssl s 1.jpg


Regarding the requirements for good
thermal management of the OSLON LEDs,
the copper surface around the thermal pad
should be kept as large as possible
. This
serves for distribution and spreading of the
heat
and is typically covered with solder
resist.
....
Copper Cooling Circuit ....
All slug solder pads," circuited " into large copper surface .........

osl 35 bcc heatsink & led & lines & cc.jpg
If an MCPCB or IMS (Insulated Metal
Substrate)
is used, the difference in the
coefficients of thermal expansion (CTE)
between the OSLON and the IMS PCB
creates a stress on the solder joint; Cu is
therefore preferred over Al as base plate
material, because of the lower CTE.

IMS -Insulated Metal Substrate-...Aka Circuit Carrier Aluminium Heatsink ....
No MCPCB !

Ceramic Package of Oslons SSLs is made of Aluminium Nitride ( AlN ) with CTE= 4 ppm/K
(Oslons are made of Alumina Al2O3 )
Cu= 16 ppm/K
Al = 24 ppm/K


Cu is used for solder pad material of IMS of the project .....

So far so good ....

ssl i rt.jpg


If for example an OSLON LED is soldered
onto an aluminum MCPCB, temperature
changes
will result in thermo-mechanical
stressing of the soldered joint,
due to the
difference in the CTE and the material
strength.
This leads to gradual fatigue and
aging of the joint and – depending on the
level of stress – sooner or later to a
functional failure, owing to loss of
mechanical, electrical and thermal contact

ssl ty.jpg

New lead-free solder materials, e.g. INNO
solder (SAC solder with additives such as
Ge, Bi
) or IMS carriers with specially
adapted dielectric strengths
, offer further
potential for improvement.


So ...IMS carrier heatsink and extended thermal pad surface with the " CCC " design ,will offer-amongst others :

-Great ease at panel design .

-Superior cooling efficiency.
(.IMS design : led=>(copper pad +solder) =>dielectric=>heatsink=>air
& CCC design : led=>copper+solder =>air )

-Superior mechanical, electrical and thermal contact durability.

.....

Now ...

Some more info for the crucial dielectric layer .....

ssl  1.jpg


IMS with Enhanced Dielectric :
75 um thickness & 1.3 W/mK thermal conductivity ....

Rth Juction-Ambient = ~33 K/W


High Performance IMS :
38 um !!!! thickness & 3 W/mK !!! thermal conductivity !!!

Rth Juction-Ambient = ~28 K/W !!!!

( Those are really good specs !!!! )
Wow!!!


Before , aiming for more like 100um thickness and 0.8-1.3 W/ mK th.conductivity ...
Now ,things are getting pretty serious ...

Hexagonal (HP grade ) BN (Boron Nitride ) is the ceramic filler of choice
- powder particle Dia =~ 2um
-Thermal Conductivity =600║; 30 ┴.
Thermal expansion = −2.7║; 38┴ ppm/°C (10[SUP]−6[/SUP]/°C)

**Thermal Conductivity/expansion of h-BN differ within the basal planes (║) and perpendicular to them (╧) **

West Systems 105 resin + 205 (or maybe 209 ) hardener epoxy, still remains the choice for epoxy ...
( λ=The normal range of polymers is from 0.17 to 0.35 W/mK.
Have to find thermal expansion ....)

Relatively Low viscosity of 1000 centipoise (cp) at 72 F (22C) .
(Super easy-flow,easy wetting,easy adhering & easy covering viscosity should be around 500-600 cp )

105 Epoxy Resin® is the base material on which all West System 105 System epoxy compounds are built. The resin is a clear, pale yellow, low-viscosity liquid epoxy resin. Formulated for use with West System hardeners, it can be cured in a wide temperature range to form a high-strength solid with excellent moisture resistance.

105 Resin is formulated without volatile solvents and does not shrink after curing. It has a relatively high flash point and no strong solvent odor, making it safer to work with than polyester or vinylester resins. Resin viscosity is approximately 1000 centipoise (cp) at 72 F (22C),

105 Resin is designed specifically to wet out and bond with wood fiber, fiberglass, reinforcing fabrics and a variety of metals when mixed at the proper ratio with a West System hardener. An excellent adhesive, 105 mixtures will bridge gaps and fill voids when modified with West System fillers and can be sanded and shaped when cured. With roller applications, it has excellent thin-film characteristics to flow out and self-level without “fisheyeing.” 105 Resin/207 Hardener mixtures cure clear so you can achieve a natural finish when coating with varnish.
http://www.westsystem.com/ss/typical-physical-properties/

( used along with Alum 860 Surface Prep -Alodine preparation/passivation )
 

stardustsailor

Well-Known Member
...And back to spectrum again ....

So ....
Say ....
5 x NW 5000 °K Oslon SSL 80 LCW CQ7P.PC Group H4
4 x WW 2700 °K Oslon SSL 80 LCW CQ7P.CC Group X7
and ...
26 x WW 2700 °K Oslon SSL 80 LCW CQ7P.CC Group X5

....osl 35 bcc color.jpg...NW color groups.JPGWW color groups.JPG

Now,that's something I like even better ....
 
Top