Building a Full-Spectrum LED Light Array for My Grow Station

In Summary, Kamko is working on a grow station that will use full-spectrum LED lights to grow plants. He has three panels made of aluminum, and plans to mount three LED lights in each panel. He plans to use different colors of LED lights to produce a full-spectrum light pattern. He has purchased LED lights and is trying to figure out how to power them. He also needs help figuring out how to grow plants under full-spectrum LED lights.
  • #36
Why have you got wires glued to the metal plate atop the ICs? (Photo is not clear.)

See all those return wires that run back beside each row of LEDs and get soldered each to a green wire? Could you have done away with all that, and instead run a single heavy-gauge wire to connect together each last LED of each row and run that straight back to the power-supply?

What is the model of the IC? Do you set the current level with a single resistor?

Are the LEDs a push fit? If one LED is sick or dead, it may affect all the others in that row. So can you swap one at a time a LED from a crook row into a good working row to see whether the fault lies with that LED? If all LEDs test OK, then you'll have to conclude that you have got the IC wrong somehow.
 
Engineering news on Phys.org
  • #37
Kamko - do you have a el-cheapo multimeter? They're only $9.95 at Walmart, i like the little yellow GE in hardware better than the Keltec in automotive because it has one more ohms range and it's fifty cents cheaper (at my walmart anyhow).

Anyhow - Nascent is more skilled than i at posting pictures and talking you through tests...

Simple voltage readings across each LED will take you quickly to the trouble.
Probably about two volts across the ones that are lighting properly.
A single bad solder joint will make a string not light at all
One LED in backward will probably make a string dim, and so would a bad constant-current regulator.

This is how ALL projects go -
first you build it,
then you get it sort of working
then you get it working,
then you get it working GREAT !
and how we learn from our mistakes !

You're doing just fine. Where did you buy your LED's ?

old jim
 
  • #38
jim hardy said:
Anyhow - Nascent is more skilled than i at posting pictures
Eh? I don't think I've ever posted a picture! But I'm working up to it ... :wink:
One LED in backward will probably make a string dim,
But will they all glow, even the backwards one?
You're doing just fine. Where did you buy your LED's ?
+1

Didn't he give a link to the receipt from his purchase of the LEDs? (I can't open it, but nevertheless, he said it's the receipt.)
 
  • #39
kamko15 said:
...The electrical tape is only temporary...

Nice heat shrink tubing. Always good to have around.
 
  • #40
Hey guys, I am very appreciative of the fast and focused replies. I have received some questions from you all and will be pleased to answer all of them.

NascentOxygen, you asked

Why have you got wires glued to the metal plate atop the ICs? (Photo is not clear.)

there is no metal plate, the "IC" housing is all some type of polymer plastic. the Buckpucks I am using have he option to dim. Currently I do not wish to use this feature, so I glued them on the top of the plastic Buckpuck just in case I would ever want to include a dimmer.

all those return wires that run back beside each row of LEDs and are each soldered to a green wire? Could you have done away with all that, and instead run a single heavy-gauge wire to connect together each last LED of each row and run that straight back to the power-supply?

No the drivers need to complete a series in each series of LEDs. They have a set input and output witch only allows them to power so many. If I where to use the method you described I do not think that the panel will perform the right way, or even work at all.

What is the model of the IC? Do you set the current level with a single resistor?

Im not sure what the term "IC" stands for so I am assuming your talking about the drivers.They are http://www.luxdrive.com/download/?dltf&dmid=1109. (let me know if this link doesn't work)

Are the LEDs a push fit?

I'm not familiar with your terminology, please explain and I will answer best as I can.


JimHardy, Thanks for the tip about the multimeter I went out and purchased one For $21. A little pricey, but it has many functions and I am all about investing in tools. you asked where I purchased my LEDs.

Heres the invoice, all the info you may want to know is there. They arrived within a week, no damage, extremely well packaged, and shipped DHL. Let me know if your curious about anything else.

QxUDG.png
 
  • #41
Eh? I don't think I've ever posted a picture! But I'm working up to it ...

MY MISTAKE ;;; that was Digoff - my apologies to both of you !


Thanks for the tip about the multimeter I went out and purchased one For $21. A little pricey, but it has many functions and I am all about investing in tools...

$21 wow, i guess you got a digital?

okay , time for old fashioned troubleshooting.

Set meter to DC volts

on a string that works, measure Buckpuck volts between LED + and LED -
and measure volts across each LED in that string so you get a feel for how close they match.

On a string that doesn't work, make same measurements. What's different?

What are those black & yellow wires on top of Buckpucks for? Resistor for control-ref pins?
What's value of resistor? I didnt quite see in datasheet how to use that feature... probably i just didnt look hard enough.

its late - i'll take better lok at your pictures in the morning.

Lastly i notice you have seven LED's in your rows and Buckpuck only shows six in their diagrams... they recommend input voltage be 4 volts higher than what is needed to light LED's , can you tweak that supply up a bit? Its datasheet says it'll go to 27 volts...
http://www.meanwell.com/search/CLG-150/default.htm

Subtracting that 4 volts"headroom" from 24 volt supply leaves 20, which shared by seven LED's is 2.85 volts each

but subtracting 4 from 27 leaves 23, or 3.28 volts each.
just thinking ou loud.
 
Last edited by a moderator:
  • #42
The sight of all those IC drivers adding their heat to your grow box makes me think we should have suggested that you house the drivers separately from the LEDs. You could perhaps use their heat underneath a sprouting tray.

Keep this in mind if you find there is too much heat for the plants under the LEDs.
 
  • #43
NascentOxygen said:
The sight of all those IC drivers adding their heat to your grow box makes me think we should have suggested that you house the drivers separately from the LEDs. You could perhaps use their heat underneath a sprouting tray.

Keep this in mind if you find there is too much heat for the plants under the LEDs.

The drivers do not produce any heat themselves. Do you mean they make the LEDs run hotter?
 
  • #44
kamko15 said:
The drivers do not produce any heat themselves.
I just looked at their data sheet. They are a switching regulator, so indeed produce very little waste heat. Some of them can even be powered from AC or DC, that's neat!
Do you mean they make the LEDs run hotter?
No. The LEDs are being well looked after.

How are you going with the fault-finding? All working yet?

Have you been able to gauge whether your LED array will produce too much heat for good growing conditions, such that a fan is needed?
 
  • #45
All working yet?

Indeed, keep us posted.:approve:Here's something that might be worth a try, just to rule it out:

Paragrapph "Connections" in BuckPuck datasheet suggests for DC versions a capacitor near the DC input terminals.
With high frequency stuff that's sometimes important.
Twist the + and - leads together too, that minimizes the area(and therefore inductance) of current loop to the switchers .
 
Last edited:
  • #46
kamko15 something caught my eye in the second of these photos: https://www.physicsforums.com/showpost.php?p=3788447&postcount=34
Where you joined each heavy gauge orange wire to the green, the solder joint is exposed and itching to make contact with the aluminium plate. If more than one exposed joints touch the aluminium at the same time, there may be damage to the drivers, I can't say. But even if they work okay, intermittent contact like this may be a source of radio interference on nearby radios. I'm wondering why you have not insulated that joint?
 
  • #47
Plants grow best may only use in the mixture of red and blue spectrum. And red is good for promoting the fruiting and flowering ,while blue is good for enhancing vegetative growth.
 
  • #48
kamko15 said:
I recently began making a cabinet into a grow station. It's almost finished, but I'm having trouble with the lighting system.

Well, Kamko15, it has been 5 months since you got this going. We're waiting for a report on its success. https://www.physicsforums.com/images/icons/icon6.gif

How is your first crop of tomatoes coming along? :smile:
 
Last edited by a moderator:
Back
Top