LED light

[thejuggla1]

Ya in the diagram they are, I'm not that great with the schematics. There is no LEDs yet, I'm asking all the questions I have before putting in a order. I think I got all the important ones answered now.

I thought 2 LEDs should never be ran in a series though?

 

[audioguru]

LEDs should never be connected in parallel unless they are tested and matched.

LEDs are fine in series because they all use the same current.
My LED Christmas tree lights have about 60 in series.

 

[thejuggla1]

Ok so let me make sure this is correct, using example voltages ect.(Lets worry about shutting off certain colors after this) left UV and Infra R out of this for now, assume all the LEDs need 20MA:

https://i233.photobucket.com/albums/ee268/Allan-2greasy/diagram3.gif?t=1248574766

everything correct?

 

[audioguru]

Your circuit is fine only if your red LEDs are actually 2.0V and your blue LEDs are actually 3.0V.

But LEDs are not made accurately. A 2V red one could be 1.7V or 2.4V.
A 3V blue one could be 2.9V or 3.6V.

If your supply voltage is higher then the difference in the voltage of the LEDs has less effect.

 

[thejuggla1]

Although the guy before said the lights don't need to be dimmed, my friend wants it, so in the end more money for me since I can charge him more for this, unless he does change his mind, in that case I'll just add a switch into each color.
How do I go about adding a POT with a transistor for each color to this circuit?

 

[marcbarker]

Add pot-adjustable current regulator? based on a 317 (k?).

LEDs should never be connected in parallel unless they are tested and matched.

Have you tried this with a pre-pack bag of LEDs as they normally come? You may be pleasantly suprised.

 

[thejuggla1]

Add pot-adjustable current regulator? based on a 317 (k?).

A LM317? I'm not sure how I wire it into the circuit though.

 

[thepaan]

Plants only use the red and blue.

Not quite.

**broken link removed**

Chlorophyl-a has a small peak in the blue end at 410nm (hence the UV LED) then the major peak at the blue end at 430nm. Then we have Chlorophyl-b peaking at 460nm in the blue end followed by the carotenoid second peak at 480nm. The carotenoids pass energy from that spectrum to the chlorphyll in low light situations. Finally down at the red end you have Chlorophyl-b again peaking at 635nm and Chlorophyl-a peaking at 660nm.

Plant spectrum use drops off sharply past the peaks and less sharply behind them. Thus a 670nm red will be far less effective than a 650nm red and, in fact, spectrums into the orange range have been proven effective. Similarly a 440nm blue is much less useful than a 420nm blue. A plant without blue light will begin to grow spindly whereas a plant without red will almost not grow at all. Thus red around 660nm is the most important but none of the peaks, save the carotenoid, are trivial. Studies have shown (google if you are curious) that anywhere from 1% to 20% (8% nominal for lettuce) blue light is most effective. Lastly, plants absorb no light in the invisble IR but some in the invisible UV.

 

[RODALCO]

schema

Attached is a simple schema with ¼ or ½ Watt potentiometer driving via a resistor through the base of a BD139 transistor.
This can control a bank of parrallel strings of leds with their own current limiting resistror.

If you want to control the colours independantly make this circuit up 4 times.

VR1 is 2k5
R1 is 470 ohms
Q1 is BD139 (NPN)
D1-D8 LED's
R2-R5 series resistor ( values as per earlier post )

 

[Torben]

So I need A resistor for the voltage then one for the current? I am confused.

No, you only need one to limit the current. The LED provides the voltage drop.

An LED is a semiconductor so does not behave like a typical resistive load. It will draw as much current as it can if it has enough forward voltage applied, and if that current is large enough, it will happily draw enough current to kill itself. You use the resistor to limit the amount of current it can draw.


Torben

 

[marcbarker]

Move some (or all*) of the Ohmage of R2, R3 etc to below the transistor emitter. Improves light power stability as transistor warms up. (How much heatsink is required?)

* connect the LEDs in parallel, with a small ohmage in series each LED chain if required to 'balance' LED voltage drops

PS plant's light spectral power requirements change according to their growing, flowering, and seeding periods

 

[Hero999]

Chlorophyl-a has a small peak in the blue end at 410nm (hence the UV LED)

You can buy violet LEDs which will give you 410nm.

then the major peak at the blue end at 430nm. Then we have Chlorophyl-b peaking at 460nm in the blue end followed by the carotenoid second peak at 480nm.

A blue an cyan LED will cover this part of the spectrum.

The carotenoids pass energy from that spectrum to the chlorphyll in low light situations. Finally down at the red end you have Chlorophyl-b again peaking at 635nm and Chlorophyl-a peaking at 660nm.

Plant spectrum use drops off sharply past the peaks and less sharply behind them. Thus a 670nm red will be far less effective than a 650nm red and, in fact, spectrums into the orange range have been proven effective.

You can get red LEDs that produce light of the above wavelengths.

 

[HarveyH42]

Not quite.

**broken link removed**

Chlorophyl-a has a small peak in the blue end at 410nm (hence the UV LED) then the major peak at the blue end at 430nm. Then we have Chlorophyl-b peaking at 460nm in the blue end followed by the carotenoid second peak at 480nm. The carotenoids pass energy from that spectrum to the chlorphyll in low light situations. Finally down at the red end you have Chlorophyl-b again peaking at 635nm and Chlorophyl-a peaking at 660nm.

Plant spectrum use drops off sharply past the peaks and less sharply behind them. Thus a 670nm red will be far less effective than a 650nm red and, in fact, spectrums into the orange range have been proven effective. Similarly a 440nm blue is much less useful than a 420nm blue. A plant without blue light will begin to grow spindly whereas a plant without red will almost not grow at all. Thus red around 660nm is the most important but none of the peaks, save the carotenoid, are trivial. Studies have shown (google if you are curious) that anywhere from 1% to 20% (8% nominal for lettuce) blue light is most effective. Lastly, plants absorb no light in the invisble IR but some in the invisible UV.

The plants will grow well enough with just the red and blue, since it's consistently getting the same levels of light. Outdoors, under the sun, it changes from day to day, so the plant needs to take advantage of what it can get. The plant will do the same with the artificial lights you provide as well. Only bought it up to save time, money, and complexity. Depends on the project. I just wanted to get some plants started, before moving them outdoors (Florida sun is a little intense for young plants). My sprouts are in a window sill, but don't get much direct sunlight.

 

[thejuggla1]

Attached is a simple schema with ¼ or ½ Watt potentiometer driving via a resistor through the base of a BD139 transistor.
This can control a bank of parrallel strings of leds with their own current limiting resistror.

If you want to control the colours independantly make this circuit up 4 times.

VR1 is 2k5
R1 is 470 ohms
Q1 is BD139 (NPN)
D1-D8 LED's
R2-R5 series resistor ( values as per earlier post )

Will these values change if I add more LEDs, or use a different voltage LED?

 

[RODALCO]

resistors

The R2, R3, R4 values only change when the colour of the LED's changes.

For 5 Volts dc

2 red leds in series (2 x 1.7 = 3.4 Volts). 1.6 Volts to get rid off. at 20 mA U=IxR ... R = 80 Ohms, use 100 Ohms 1/4 watt.

2 orange leds in series (2 x 2.2=4.4 Volts). R=30 ohms, use 47 ohms.

The UV blue and white leds are have usually a 3.0 - 3.6 Volts drop.

design it for 3 Volts so 2 volts to get rid off. then a 100 ohm resistor is needed here. use 120 ohms to keep current below 20 mA.

 

[thejuggla1]

Alright I read this page(Scroll down to LEDs is series)

Light Emitting Diodes (LEDs)

it says I need to keep an extra 2V for the resistor? I'm confused why is that? Then I won't have enough volts to power 2 Leds?

 

[audioguru]

You don't know the exact voltage of your LEDs. It is a range of minimum to maximum voltages.
You must have about 2V extra for the resistor to allow for the variation in voltages your LEDs might have.

If you don't want to waste 2V in the resistor then you could measure and sort your LEDs then calculate a suitable resistor for each one or for each series string.

 

[Garibaldi]

If you're not worried about spending a little more money (or passing it along to your client), I'd suggest using either a Xitanium or Meanwell constant current power supply specifically designed for driving LED's. Both make dimming models.

I just did this to replace my Aerogarden light and my miniature chillies seem to enjoy it much more than the original CFL's. Though I used six 5W reds, one 5W neutral white from Ledengin, and 60 10mm blues (20mA) from AND/Purdy. I didn't use one of the dimming models, though. The current for the 10mm's was limited by using Supertex CL2 chips which are wired in series just like another LED.

Simple, easy to put together, and very safe to run because of the Xitanium driver.

 

[thejuggla1]

You don't know the exact voltage of your LEDs. It is a range of minimum to maximum voltages.
You must have about 2V extra for the resistor to allow for the variation in voltages your LEDs might have.

If you don't want to waste 2V in the resistor then you could measure and sort your LEDs then calculate a suitable resistor for each one or for each series string.

But in this case it will be fine to run 2 LEDs and not have to check them all?

 

[audioguru]

But in this case it will be fine to run 2 LEDs and not have to check them all?

Simply calculate the "worst case current" of LEDs with minimum voltages then calculate with maximum voltages.

If the resistor does not have enough extra voltage and both LEDs are minimum voltage then they might burn out.
If both LEDs have max voltage then they might be too dim.