What's new
  • As of today ICMag has his own Discord server. In this Discord server you can chat, talk with eachother, listen to music, share stories and pictures...and much more. Join now and let's grow together! Join ICMag Discord here! More details in this thread here: here.

I need an Electical Engineers help for LED's

DeadlyFoez

Active member
But the thing ur missing is that you dont need to project the light, you can dangle led's into the middle of your plan and under it and everywhere else in between. You dont need that intensity. It's better off that way cause when you have to keep ur HID over a foot away from ur plant then you have already lost a lot of that precious light, but with LED's you can keep them very close and not have to worry that much.
 

DeadlyFoez

Active member
So do you think I'm better with luxeons, or should I just stick with regular type LED's and get them at the exact correct wavelegth? I've had my mind set on luxeon because they are higher power but I know the wavelengths are slightly off.
 

DeadlyFoez

Active member
Right now I'm considering not using the luxeons because of what I read. Although I strongly believe that I could handle the heat factor of them. I'm pretty much prepared to have pcb boards built and whatever else I have to goto to not make my setup cheaply done.

Though, I haven't done the math to figure out if the luxeons are more efficient than others when you compare light output with light intake by that plants at the effective spectrums. I dont have the patience for that type of math. But if the case was that the luxeons would be more effective then I would like to use those for at least the spectrums that they are close to and use the lower powered for the spectrums that they dont hit.

The obvious logic is more light emmitters per sqr inch means better chance to provide enough to grow a plant, our plant of choice.

Right now I'm looking through all the threads and web pages trying to find the peak wavelengths for chlorophyl a and b and carotene.So far I have found
Chlorophyl A 430nm and 662nm
Chlorophyl B 465nm and ???
Carotene 612nm

Whats the other red to fill in the spot for Chlorophyl B in the reds. And are those correct?

And on this one chart I found it saying that Phycoerythrin and Phycocyanin use specific wavelengths too. Are those ones we should be focusing on also?

Knna, What have you found for U.S. distributors that provide LED's in our desired wavelengths?

I have read a lot of the stuff on led arrays from the links you guys provided.

Een, or anyone else reading this, if I were to come up with a final design for this, would you be willing to design up the schematic for me or even do the math and tell me how it would need to be to work as far as the resistors and different colored leds in strings?

I was also wanting to know more info or a link on those current limiting drivers you talk of. Would one be used as apposed to a computer psu? Are there any benefits to one or the other?
 
theese lamps are prebuilt and bought from www.ledgrowlights.com . i know that guy who runs that grow. as i spoke to him, they are just 470 + 630 nm (only 1 type of clorophil)
othervise you need 635 + 455 for one type and 660 + 425 for another.
if you want to design your own array, i think the best way is to use 5, 8 or 10mm normal leds or maybe superflux that knna mentioned. i wouldn't use luxeons. i don't know if i remember this right, but i think that at OG some guys in that led thread mentioned thet the lifetime is only a half of lifetime of normal leds. even for wiring i think that normal leds are much easier to wire than luxeon or superflux.
be carefull with power supply. pay attention on voltage drops when the PS is loaded, what output you use,... measure as much as possible, use fuses,...
 

DeadlyFoez

Active member
That driver that you described is pretty much exactly what I want to do. Yeah if you can come up with the schematic then that'll be great. Hell, if you can get one made for me I'll pay you for it.
 

DeadlyFoez

Active member
Ok, trying to keep this thread alive here.

I'm doing my research and trying to figure out all my math, but the past few days I've been a little too busy then a little too high...or not high enough?

So anyways. If anyone else has some good informational links that that I can learn from please post them. I'd also like to learn about circuitry so links for basic components and how to use them would be superb. Thanx.
 

DeadlyFoez

Active member
Thanx for the link man. Good info for beginners like me. I already had a few pages for led calculators, but that page also has all the info I wanted like ohms law and converting ohms amps voltage.... all the stuff I needed.

Well through that website I went on through links and I found this, a 3 channel pwm controller. It says that it can handle 50 watts per channel. It seems to be pretty much what I was looking for...I think. What do you guys think of this??? Considering I'm going to have so many arrays and different led's going on it will help me to be able to control all the different wavelengths to the desired output that I want, like UV which I wont use until flowering.
 

DeadlyFoez

Active member
Oops, I forgot the link to it. Its on this page. It's listed as a 3 channel pwm controller.
http://www.quickar.com/tkit.php?session=SOfINulI

I'm thinking I could use about 3 of these, but I'd also want to be able to controll how bright they pulsate instead of just controlling the pulse timings. It would be great if I could overdrive the leds also. I'm sure that I would just have to use a lower rated resistor on one end and a variable resistor nob on the other polarity end. Will this work???

I'm aware that these aren't actual power modules and only something to make the LED's pulsate.

I wish I had a way of just turning a nob and having some sort of a display to be able to see the numbers of how it would pulsate so you could adjust the pulsate to exact specs for what it would be worth. Is that possible??
 

DeadlyFoez

Active member
Isn't just the use of resistors current limitting enough? Do we really need special drivers for any reason? Please explain.

<---Not an electrical engineer....yet
 
Last edited:
i control my arrays with voltage. but i calibrate it with current. i buy or make a DC powersuply and then connect my arrays with some resistors in that circuit. i always make same strings, so that they run approximately at the same voltage/current. then i calculate how many mA should go into the array and then choose the resistor that gives me that current. at the end i find out, that the voltage is too low if you believe what it says in led datasheets.
for example: i make 4red + 1 blue led series, then connect 10 of that series paralell. each string runs at 20mA, then the whole array should run at 200mA. that is 'normal condition'. if i check the data sheets, maximum continuous current for blue is 30mA, for red 60mA, so it is no problem if i get 190-220mA / array - maximum before blue leds start to get too hot is 300mA/array. but i always try to get the nominal current. then i turn on the lights, let the ampermeter in the circuit for some days, to see if there are any changes. if not, it is ok, i take the ampermeter out of the circuit and i leave my lights 24/0 for a week. if there is no problem, the lamp is ready to go.
but that is my way of doing it. i'm sure most of you wouldn't agree with that. but i couldnt find a current regulator, so i used voltage regulated power supplies.
 
Last edited:

DeadlyFoez

Active member
Hey Mr LED,

what other sites do you participate on?
Just wanting to know where else I can get great info.

Thanx
 

DeadlyFoez

Active member
Ok, I read alot of stuff. It seems the easiest way is just to find a power supply 18-24 volts would be great and capable of 400 watts. I'm going to use the highest power leds for whatever wavelengths I can, including the luxeons. I am going to make a heat sink for it for it. <--[EDIT: I must be high]

I'm actually thinking using a zalman heatsink.
http://www.frozencpu.com/products/4...Dual_Heatpipe_Graphics_Card_Cooler.html#blank
Something along the line of this would be perfect. Just use a copper plate for all the high power led's.

The hard thing is to have all these led's mixed together, high and med powered ones and not have the heat sink be in the way of your wiring and stuff. But I'm thinking that if I am to go as far as to have pcb boards made, I could also have some copper machines to have some 1/4" pins that go through holes in the pcb board to touch on the back of the high powered led's. I'd use some good artic silver 5 for it.

Really, this would be incredibly sick and work so well. I will have a freakin awesome grow system in the end.

But it all comes with a price....and for me every cent of it is worth it. I'm ready to drop $6k on this as the money comes in, so it's just time. I'd even pay for the engineers to do it up for me if someone is willing to step up to the plate for me.... save me some time of learning too much.
 

DeadlyFoez

Active member
Is it possible to digitally and accuartely control pulsating and brightness?? Like I want to be able to adjust light on and light off in milliseconds and be able to view the numbers and change them to what I exatly want?????
 
i remember we talked about that at OG. the point is, that if you drive your leds with pulse curret, you can use much much more current if the on-periode is short enough. with that you get much more intensive light. they said at og, that plant doesn't need light always on and that it is better to give it pulsing light. but not sure what timing, pulse durations,... they recommended. canna, do you remember that maybe?
on the other hand, pulsing generator can be very expensive.
 

DeadlyFoez

Active member
LED_experiments said:
the point is, that if you drive your leds with pulse curret, you can use much much more current if the on-periode is short enough. with that you get much more intensive light. they said at og, that plant doesn't need light always on and that it is better to give it pulsing light. but not sure what timing, pulse durations,... they recommended. canna, do you remember that maybe?
on the other hand, pulsing generator can be very expensive.

Yeah, that is exactly what I was wanting to do, and I've got the means of beable able to do it. If I had $600 or $800 on hand right now I could make it, but it would be controlled by nobs and you'd have no way of knowing exact timings.

I dont know the best timings for pot, but a little research and experiementing will work great.
 
Top