Need help with design

[Zer0Ry0k0]

My brain hurts after spending three days sorting out what LEDs and all kinds of details but I am having trouble figuring out what I need to buy to do what I want. I have no experience designing a circuit.

I need a LED driver.
I need more than 1.
each one must either be variable or set to 720hz @ 50% duty.

I have 7 different colors of LEDs, I was assuming I would have each color on it's own driver. The bulbs will be 1w each.

I am going to fill a 2ft x 2ft square with LEDs with gaps for air to get through.

What driver can I buy to do what I need? or maybe someone could design something that would be able to accomplish what I need? I would prefer to not have to build a circuit at this phase.

 

[hempfield]

A LED driver will assure a CONSTANT CURRENT on the circuit. This means constant, with no variation, no interruption and no gaps.

If the driver is PWM (pulse width modulation) enabled, you cand use it together with a 720Hz square ware generator (a PWM generator with adjustable frequency and duty factor).

How many watts (or 1w leds) are you going to power this way ?

The PWM circuit is easy to build, with just a few components, and can be build without soldering on a prototype breadboard.

 

[Zer0Ry0k0]

Excellent! Thanks Hempfield.
I am not sure how much space I should put between the LEDs, I figured because they are 1w and I will be using a copper plate (1/16"), so heat should not be an issue.

Because I don't have the LEDs in my hand I can't see how close I want them, but as close as I can get them with room for the wires. Maybe every 1 inch by every half inch? or 1x1? so 400 lights? ouch, this is just a prototype, haha.

I am planning on using 50% neutral white (3500k-5000k),
some 420nm, some 450nm, less 660nm, and a bit of 560nm, 610nm and 300nm.
I think this will be the magic combination to get full PAR and no wasted energy.

 

[Zer0Ry0k0]

Also, could someone please check my work here...
I was trying to convert these chart values to some type of LED light output.

http://www.soda-is.com/maps/world_uv_ab.png
http://www.cpc.ncep.noaa.gov/products/stratosphere/polar/gif_files/uv_dosage_world_est.gif

I thought it came out to ~10w (over 2 sq ft) of UV radiation to mimic the highest levels that hit the earth. But maybe I am way off? Maybe it is closer to 180watts for 2sq.ft.?
and then I need to take like 25% of that value to equal the UV-B specifically.

http://earthobservatory.nasa.gov/Features/UVB/uvb_radiation4.php
This link is specific to UV-B, saying that we got 8000 J/m.sq. in some parts of the world in November of 2000.

8000 Joules per square meter means what in terms of Watts per sq.ft. or....???


and why do I care to damage my plants?
http://www.youtube.com/watch?v=lfiI78uN3Ks
This guy explains everything I have been reading on my own and adds more I didn't think and know about. (I was trying to find the scientific articles I was reading but I can't find the page right now... I have too many internet tabs and windows open right now!!!)
As he says in the video, these lights add to your daily UV, so turn them off when you go inside the room to work, but I personally don't get enough sun so I will keep them on.

 

[Zer0Ry0k0]

blah, there was a paper I was reading that was talking about how early life cells would have had to deal with UV exposure as they came towards the surface of the goop and most of them would have died, then some would have adapted and that process of the UV-B hitting the proteins and joining them together, then the cell had to learn to split the protein.

It also said that this may have been the foundation of all cell mitosis because when a cell divides, it uses the same technique... and I forget all of these wonderful words because I can't find the paper! split or splice or.... I will try to find it again... I meant to bookmark all of them but I guess I missed some.

 

[Zer0Ry0k0]

These are different, but still relevant.
http://www.ncbi.nlm.nih.gov/pubmed/12733652
http://www.jipb.net/pubsoft/content/2/3521/x030412.pdf

They say basically that you can have O3 if you have CO2 to balance it, and other stuff I was reading was showing that you can even boost the CO2 usage when you have O3 present. And the damage from the UV-B is mitigated by the increased CO2. The plant gets the energy and not the damage is my hope.


Okay okay. Enough talking from me, haha. I need to get this light built first!

 

[Zer0Ry0k0]

I am trying to decide how many 1watt bulbs to use...

I am thinking of the CREE XP-E series.
http://www.cree.com/~/media/Files/C...d Modules/XLamp/Data and Binning/XLampXPE.pdf

The canopy space will be 2ft x 2ft x (18 - 24 inches). The light array will be approx 20"x20" of LEDs. The LED I have chosen are 3.45mm in Length and Width. If I space them out to 4mm apart, leaving 1mm gap I could do as many as 64,516 on this one pannel?!??? I don't know what is overkill. Heatsink size and air flow are not a limitation for me. Please help!

 

[hempfield]

Ha ha ha , 64,516 LEDs ???? Are you insane man ? What are you going to build ? A micro 'super nova' ?

The XP-E are small indeed, but to be able to use them, they must be placed at least on a small (10,14,16 or 20mm) alluminium PCB, because their connecting points are placed under the led itself .


In ideal condition you need 50W of LED light per square foot. So 200W (real power) will be more than enough for a decent grow.

The XP-E could be pushed up to 3.5W with proper cooling. But why don't you use XM-L or even the latest generation XM-L2 , to simplify the construction and for more efficiency (and just enhance the full white spectrum with some blue, red or even UV-B).

 

[Zer0Ry0k0]

yeah, I will be using many specific wavelengths and I was just reading about the benefits of infrared (730nm) just after lights out during flowering. you get to shorten the dark period if you want.

No need to overdrive these. For this specific application the plan is to have a super flat SCROG with these 1w bulbs approx 4 inches from the top of the finished plant. the plant will be grown extremely short. Space is extremely limited! only 2'x2'x1.5-2' above the screen.

If I can find a distributor, I will try to get XP-E2's.
add: Cost is an issue, this is still just a prototype stage. Many changes are expected.

 

[Zer0Ry0k0]

AHAHAH! I just looked at some prices for UV-B (300nm) LEDs. The cheapest I have seen so far is $180.