LED DIY : Making of a "DJ SET "-style grow led light .

guod

Well-Known Member
best is to spread the soldermask over the complete PCB, only Ledspads are free.

some post before i show up my led-layout, here is it again, now with Beam-angles.

light-cone.jpg

this is only 2D, in s.up with transparent 3D cones, could maybe helpful to find a layout.
 

PetFlora

Well-Known Member
Haven't seen before such material ...
....
It looks really good for making a lid ..

I want your opinion on something ..
Shortly ...
I'm thinking for 3x (custom ordered) mcpcbs' ..
And I'm experimenting ...
What do you think about something like that ?
.....View attachment 2514573......

Ok....And then ...I thought ...Since Osram is doing it .....
(just a "sketch" ,not a carefully designed one .....===> View attachment 2514572.....

Any remarks or comments on that ?

Edit : Another idea ...
Leds more spread ...
View attachment 2514578
Pic 3: If you invert one of the lower horizontal rail, you would have Merkaba
 

stardustsailor

Well-Known Member
Guod ...
For the analog dimming of the drivers ...
We can use a pot of 10 K and and a 1206 SMD 2.2 K resistor for the Vref ..
It sets a bit higher the limit ( V ~ 4.09 +/- 0.5 V ) ...

Output mA =700* ( 5-V [SUB]analog dim input[/SUB]/5 ))....


V[SUB]analog dim input[/SUB] = 5 ....or 4.95 - ( R[SUB]vref[/SUB] / ( R[SUB]vref[/SUB] +R[SUB]pot[/SUB]) * 5 ...or 4.95 )
When pot at "0" ,total resistance is is R[SUB]vref[/SUB] + R[SUB]pot[/SUB] ....
Meaning ( if 10K pot & 2.2 K for Vref Resistance ... )

V[SUB]analog dim input[/SUB] = 5 - ( (12.2 K / (12.2K +x)) *5 )= 0,0000000000......practically zero ....
So then we have 700mA at leds ......

If pot is at full (max ) position ,then ..

V[SUB]analog dim input[/SUB] = 5 - ( (2.2 K / 12.2 K ) *5 )= 4.098 Volts ...


Output mA = 700*(( 5-4.098 )/5)= ~126,2 mA at leds as lowest setting .....

Is there any other problems with that config of resistace and pot ?
(It is just that these are handy .... )
 

guod

Well-Known Member
the Problem here is the current we need to drive the analog-inputs
for a loaded voltage divider the current across the pot has to be 5 to 10 times higher then the current that goes out.
at higher current across the pot the output V is more stable.

some numbers
the input-current for the analog-input is 0.2mA*. 3 of them we have, so 0.6 mA for all.
going for a 10k pot + 2.2k R at 5V the current is 0.42mA. that´s lower than the current for all 3 a.input. no go!

with a 2.2k pot + 420 R the current is now 1.9mA. this is a little bit critical because the factor is about 3.
factor seems low, but in this case we can do it.

with a 1K pot + 180 R the current is now 4.2mA. factor here is 7.
current is also below the max. output of Vref(5mA).
this will be the best combination.

brainfuc...
http://denethor.wlu.ca/pc200/lectures/vdivbeam.pdf

*this current is constant on different Voltages.
 

stardustsailor

Well-Known Member
You're great !
Ok ...
So best combo there would be 1 K pot and a 180Ω 'resy' ....

Hmmm..

Also ...I'm thinking something different considering the fan issue ...
Since I do not have any sound disturbance issues ....
I'll go for the 4418 F .
(48 Volt - 170 cm/h - 43dB -2900 rpm - ~37.500h at max)..
...
Double the noise and almost half service life than the 4414FL ...
But stronger and directly driven by the main PSU ...
...

Dunno ...
it is approx 5.5 Watts ...

While going the 4414 FL way ....

1 watt is fan power ...
And ~2 Watts of heat at the LM317 power regulator circuit ..
Right ?
(fan needs ~0.083 A to work ...from 48 V down to 24 V is 24 Volts difference (reduction )....
Which ,end up as heat at the LM317 ...0.083 x 24 = 1,992 Watt ....Am I wrong ? )
...

Not decided yet ..Can't really end up to one side ...
The " small" 4414FL or the "strong and loud " 4418 F ?
 

stardustsailor

Well-Known Member
And I need a logical explanation for these :

This is NES-350-48 : http://gr.mouser.com/ProductDetail/Mean-Well/NES-350-48/?qs=/ha2pyFaduiI%2bF2y6zP5DSAMAC/sTgeLnI%2bLSmIBmJRbTyQbUlohUKioJWrrp39C

.....48,28 €

And this is SE-350-48 : http://gr.mouser.com/ProductDetail/Mean-Well/SE-350-48/?qs=/ha2pyFadug7HmKgbt3swuVYu6aumWxUqjjj18QWqDQ=

.......82,25 €


Now ...Check out their specs ....

"Spot the differences " ....

I didn't find ANY!!!

So ...For what the - ..~double...- price difference ?

( Because of the tag "NEW ! " ? )
 

stardustsailor

Well-Known Member
And a new thermal design mcpcb for 12x OSlons SSLs ...

Somewhat better than the previous ....

led pcb new thermalpath.jpg...

(DipTrace has some amazing pcb designing features ! .)
 

stardustsailor

Well-Known Member
Ok ....

The top lid of the box , is 200x200 mm ...

Say ...
What about 2x 80x80x25 mm fans ? At 24 V each ...

Like 2 x 8414NGML

http://www.ebmpapst.com/en/products/compact-fans/axial-compact-fans/axial_compact_fans_detail.php?pID=53884

Look...:
- 2 x 1.1 Watt =2.2 Watt ,all at the fans ..
-2 x 45 cm/h
-2 x 19 dB (I do not think it makes 38 db ,here ...Not sure though ..)
-2 x24 -in series- = 48 V
-2x 80 = 160 mm (One side is just from the other two ,the 25x25 angles need to be reworked ..
A 5x 80mm section from central part of each of angles ,needs to be removed for the fans to fit ...
(5x 40mm from center to each side of angle ) .....

What do you think ?
Does it stand as an idea ?
 

stardustsailor

Well-Known Member
Guod...
What will happen if say one puts two fans in series co-axial ?
One behind the other one ...
Both pushing air at same direction (that was not necessary to be mentioned ... ) ?

I know from boats ,that they use that ...
Double propeller ..Co-axial ....Working as one ,pushing water behind ..
Kinda sharing the "load " ......

Would it work if two 4414 FL s ( 2x 24 V ) were "stacked" on top of each other ?
2 Watts ,no heat loss from power regulator ,no power regulator at all and probably a not so beatiful box ...
Hmmm.I've to see how it looks ...
Wait ...
 

guod

Well-Known Member
on the Meanwell site they look different, nes can have dual; triple output, Miniature size with complete protection, etc
but in the datasheet i miss all these points...copy and paste i think.
 

stardustsailor

Well-Known Member
22.01 db !!! and 2 x 45 cm / h with 2 x 80 mm ,24 V 84ngml ...
Side to side ....

Or 2x 4414 FL on top of each other ...
See ..
double fan 1.jpg...double fan 2.jpgdouble fan 3.jpgdouble fan 4.jpg
 

guod

Well-Known Member
Would it work if two 4414 FL s ( 2x 24 V ) were "stacked" on top of each other ?
gives more pressure, do we need this?... NO!
airflow is what we after

not all Fans work easy in series.
if the Motor is from EC-Typ* it would be tricky. see schematic in my thread/first post.
i use a capacitor for start up.

*)http://en.wikipedia.org/wiki/Brushless_DC_electric_motor
 

jamesvagabond

Well-Known Member
This is great! I have a lot of catching up to do, once I set up my atomizer aero system, LED arrays are next for sure. Are you adding a UV-B flourescent bulb in the panel like the california lightworks solar storm?
 

guod

Well-Known Member
if this design is ready for production, we will go for a UV-B, IR(730nm) add up based on Leds. but it will not cheap!:-P
 
Top