Building a 1 watt LED light

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
You can do that if you want...kind of hard to piss me off over the internet, while being stoned, talking about stuff I know very little.
BUT...this thread may deter people from actually building a light. Your thread, to me, is the better option to show people how easy it really is. No need to make a box, circuit board, blah blah. Just straight forward, solder wire to star pad, mount on heatsink. hook up power and done. There are even lenses for some of the extrusions.

On my light, I should be able to change some of the output of the leds in my design. Whether by dimming or turning off some of the leds. It should be versatile. Will more blue light help? Will more red? What if I only have one little plant, can I turn off a couple rows to save a pennys worth of energy?

Okay...about ratios. I've been reading tons and read everything from 2:1 red:blue to 8:1 red and blue. I'm guessing most of these are being geared toward flowering. I'm going for a 1:1 ratio with it slightly leaning more to the blue side. If you look at plant chlorophyll and spectrum charts you don't see a 2:1-8:1 ratio in favor of the reds but actually slightly more action in the blue spectrum.

That said, I'm no scientist so take this with a grain of salt.

View attachment comparison_legend.png.cf.png

View attachment PAR2.jpg.cf.jpg
 
Too me, its all either confusing or I cant find the information I need to understand. So Im hoping I'll be able to dim something or turn off or on a different spectrum. This is just for my veg cabinet. Overkill but changeable until I understand a little more.
 
You just need to put diffent leds on different drivers if you want to turn some off and leave some on. You'll need dimmable drivers to turn the brightness up and down.
 
It looks like its coming together or at least I can see that it might work.

Here my thinking, I have a LPF-60D-24 constant current power supply. That is 24 vdc and 60 watts of power. Hope its enough. "D" is for dimmable, I can control the output of this power supply. That will power the buck board. That board is used to distribute power and control the different strings of leds. The LDD-1500L, is a buck circuit that with an output rating of 1500 mA. More than what I need for this project but will still work. My strings are only drawing 350mA. The board also allows me to control the individual strings of leds. There is a pin on the chip that senses a pwm signal. All the outputs of those chips will be wired to the led board. The led board has all the surface mounted leds. They have the "built in" lens so I don't think a hood or secondary lens will be used.

View attachment 20170121_155552.jpg

View attachment 20170121_155412.jpg
 
It looks like its coming together or at least I can see that it might work.

Here my thinking, I have a LPF-60D-24 constant current power supply. That is 24 vdc and 60 watts of power. Hope its enough. "D" is for dimmable, I can control the output of this power supply. That will power the buck board. That board is used to distribute power and control the different strings of leds. The LDD-1500L, is a buck circuit that with an output rating of 1500 mA. More than what I need for this project but will still work. My strings are only drawing 350mA. The board also allows me to control the individual strings of leds. There is a pin on the chip that senses a pwm signal. All the outputs of those chips will be wired to the led board. The led board has all the surface mounted leds. They have the "built in" lens so I don't think a hood or secondary lens will be used.

Looking really good my friend I really can't wait to see your finished product
 
Well ?????

Heres plan A. My original idea was to mix the strings with different colors. Add the voltage drop for each series led in the string, add a couple volt cushion and parallel the six strings. I have the dimmable power supply, but this thing isn't that wicked powerful. I think I can raise it a couple of inches to get the same results as dimming.

Do I really need to control a few extra watts of leds in a small light?
Do I really need to separate leds, to control a single color on a string?
I'm leaning towards, no, no.
I'm leaning towards, KISS.

View attachment 20170122_124556.jpg
 
Gentlemen, would you mind sharing who your searching for and purchasing components through? Ive done some searches but it would benefit all to get some feedback on the suppliers you're using. Though my background is in Electricity, I've haven't been involved in electronics too much since classes in high school, where are you gaining the technical backup for your circuit designs? Much appreciated.
 
Gentlemen, would you mind sharing who your searching for and purchasing components through? Ive done some searches but it would benefit all to get some feedback on the suppliers you're using. Though my background is in Electricity, I've haven't been involved in electronics too much since classes in high school, where are you gaining the technical backup for your circuit designs? Much appreciated.

The easy answer is google. The long drawn out answer is life experience and how to apply that experience to any new experience. I can go on but I'll have to google the rest of my thought.

Circuit design? Not much of a circuit, + to -.

I am just using the software because the leds I bought are surface mounted and will be on fr4 board. I didn't want the traces to cross over each other. Makes it easier to have less solder junctions and it gets me thinking about TRON. I bought them off of amazon, from a company in china. Long wait, poor quality control in manufacturing, maybe not even the correct spectrum, but cheap.

Many different ways to design stuff, there are comprises, requirements and oops in all my designs.
 
Feel free to correct me.
This is an option for a LED array. Not a complete drawing, but enough to start.

In the first circuit there are 6 blue LEDs and 2 red that I want to power. The power supply that is being used is LPF60D-24. The voltage is 24vdc and up to 60 watts of power. Its a constant voltage power supply, it always wants to put out 24vdc whether the load is .3 amps or 2 amps.

I'm not going to bore you with the details of Ohms Law. The only thing you have to know now is that it clearly states, smoke'em if you got'em.
So, the first circuit is switched on and running within spec. Everybody's happy.
That's just a fantasy world. Well, it will work. Still a fantasy world. Problem comes when you try to find that resistor. What is it? In this type of circuit, I call it a current limiting resistor. Its used to determine the amount of current flowing in the circuit. In this case, I want 350mA to flow. And following Ohms Law, smoke break.....
I'm back....
I see somebody calculated 1.14 ohm resistor needed for this specific circuit of LEDs. If we change the value of the resistor higher or lower the current will change. I want 350mA, not 320, 400. So a 1.14 ohm resistor is needed. In my perfect circuit, all the LEDs are lit. Why is 350mA so important? Its not really super important but I want the LEDs to perform up to the manufacture specs. You can get the forward current rating from the datasheet. In my case, the 1 watt LEDs that I am using all have the same foward current rating, but different foward voltages. The numbers above the LEDs and resistors are the voltage drops across each component. Remember, I only have 24 volts. When you add up all the voltage drops you get 23.6. So the difference between what the power supply always puts out, 24, and all the LEDs drops, 23.6, is the voltage drop on the resistor. Again, Ohms Law break.........
I love perfection.
Wait I got these LEDs from blah, blah, blah. The forward voltages vary between each LED. Maybe not much but some. Those little numbers add up. That also means I can't figure out the exact value for the resistor until I know them. Maybe I can use a potentiometer and adjust it until I read the current that I want.

OK, I got the LED voltage drops, the correct value of resistor, the power supply.

Then heat happens.

What if one LED fails?

In second circuit, the LED fails by opening. I know the manufactures claim 50000 hrs of life. Blah blah 50000. That's NOT being driven at 100% of its current rating. Its life is alot shorter driven at 100%. Then add the ill effects of heat and the life becomes even shorter. Anyway, most of the time when they fail it will open the circuit. You loose that circuit of lights, the other LEDs aren't affected but there is no light. You troubleshoot circuit and repair. Not a big problem.

But what if the LED shorts out.

In the third circuit the LED shorts out. The power supply will still put out 24vdc. So the voltage of 24 still gets distributed to rest of the circuit. But the datasheet spec said 3.2 volts for a blue LED. It doesn't add up. Well, that 3.2v is when it is driven at 350mA. Now that one LED is shorted, its voltage drop is distributed to the rest of the circuit. So each LED and resistor gets a little more volt drop onto it. For the resistor, it heats up. The LEDs, the more voltage allows the LED to conduct faster (current rises) so they have more power and they will be brighter. Not a big problem, the LEDs are brighter thats what I want!

Back to reality.....

That little voltage in this case, according to Ohms Law, smoke break....

Missed it.

Yeah, its gets brighter for a little while then the cascade effect happens because of the increased current flow with each failure. They all short out by the time reality sinks in. Imagine if it was your whole light, poof.

I does happen on occasion.

In my light there are multi strings of LEDs. So it may only take out a portion of the overall light. The other strings would be in parallel to the example circuit and wouldn't be affected by that circuit, hopefully.

For your homework, calculate the value for the current flow in the third circuit when the blue led shorts out. Hint: the extra 3.2v is not distributed evenly amongst the components.

Valid circuit to use but the failure mode is a concern and to be able to find the correct resistor, also. I didn't even get into the heat issue on this circuit so I'll put it on the back burner for now.

View attachment circuit.png
 
Gentlemen, would you mind sharing who your searching for and purchasing components through? Ive done some searches but it would benefit all to get some feedback on the suppliers you're using. Though my background is in Electricity, I've haven't been involved in electronics too much since classes in high school, where are you gaining the technical backup for your circuit designs? Much appreciated.

EBay or LEDsupply
 
Feel free to correct me.
This is an option for a LED array. Not a complete drawing, but enough to start.

In the first circuit there are 6 blue LEDs and 2 red that I want to power. The power supply that is being used is LPF60D-24. The voltage is 24vdc and up to 60 watts of power. Its a constant voltage power supply, it always wants to put out 24vdc whether the load is .3 amps or 2 amps.

I'm not going to bore you with the details of Ohms Law. The only thing you have to know now is that it clearly states, smoke'em if you got'em.
So, the first circuit is switched on and running within spec. Everybody's happy.
That's just a fantasy world. Well, it will work. Still a fantasy world. Problem comes when you try to find that resistor. What is it? In this type of circuit, I call it a current limiting resistor. Its used to determine the amount of current flowing in the circuit. In this case, I want 350mA to flow. And following Ohms Law, smoke break.....
I'm back....
I see somebody calculated 1.14 ohm resistor needed for this specific circuit of LEDs. If we change the value of the resistor higher or lower the current will change. I want 350mA, not 320, 400. So a 1.14 ohm resistor is needed. In my perfect circuit, all the LEDs are lit. Why is 350mA so important? Its not really super important but I want the LEDs to perform up to the manufacture specs. You can get the forward current rating from the datasheet. In my case, the 1 watt LEDs that I am using all have the same foward current rating, but different foward voltages. The numbers above the LEDs and resistors are the voltage drops across each component. Remember, I only have 24 volts. When you add up all the voltage drops you get 23.6. So the difference between what the power supply always puts out, 24, and all the LEDs drops, 23.6, is the voltage drop on the resistor. Again, Ohms Law break.........
I love perfection.
Wait I got these LEDs from blah, blah, blah. The forward voltages vary between each LED. Maybe not much but some. Those little numbers add up. That also means I can't figure out the exact value for the resistor until I know them. Maybe I can use a potentiometer and adjust it until I read the current that I want.

OK, I got the LED voltage drops, the correct value of resistor, the power supply.

Then heat happens.

What if one LED fails?

In second circuit, the LED fails by opening. I know the manufactures claim 50000 hrs of life. Blah blah 50000. That's NOT being driven at 100% of its current rating. Its life is alot shorter driven at 100%. Then add the ill effects of heat and the life becomes even shorter. Anyway, most of the time when they fail it will open the circuit. You loose that circuit of lights, the other LEDs aren't affected but there is no light. You troubleshoot circuit and repair. Not a big problem.

But what if the LED shorts out.

In the third circuit the LED shorts out. The power supply will still put out 24vdc. So the voltage of 24 still gets distributed to rest of the circuit. But the datasheet spec said 3.2 volts for a blue LED. It doesn't add up. Well, that 3.2v is when it is driven at 350mA. Now that one LED is shorted, its voltage drop is distributed to the rest of the circuit. So each LED and resistor gets a little more volt drop onto it. For the resistor, it heats up. The LEDs, the more voltage allows the LED to conduct faster (current rises) so they have more power and they will be brighter. Not a big problem, the LEDs are brighter thats what I want!

Back to reality.....

That little voltage in this case, according to Ohms Law, smoke break....

Missed it.

Yeah, its gets brighter for a little while then the cascade effect happens because of the increased current flow with each failure. They all short out by the time reality sinks in. Imagine if it was your whole light, poof.

I does happen on occasion.

In my light there are multi strings of LEDs. So it may only take out a portion of the overall light. The other strings would be in parallel to the example circuit and wouldn't be affected by that circuit, hopefully.

For your homework, calculate the value for the current flow in the third circuit when the blue led shorts out. Hint: the extra 3.2v is not distributed evenly amongst the components.

Valid circuit to use but the failure mode is a concern and to be able to find the correct resistor, also. I didn't even get into the heat issue on this circuit so I'll put it on the back burner for now.

Screwdriver, So looking just at the voltage drops or Vt / Et (where Vt/Et is total voltage drop)of the third circuit you'd drop 3.6 volts across your resistor. Using then simply ohms law I=e/r you'll have I= 3.6/1.14, that's I = 3.157 amps, much higher than the capacity of the driver power supply. imho your resistor should be 10.4 ohms and also at least capable of withstanding 2 watts of power. At the 1.14 ohms , you'll be subjecting it to over 24 watts of power and of course much higher amperage (for a short time). Not having any knowledge/experience with the drivers, I'm curious if they have any overload protection for just this sort of fault?
 
What are your guises thoughts of running something like this on a light similar to the one your building

https://www.amazon.com/dp/B00C498MGY/?tag=skimlinks_replacement-20


78ae3bd710fbfc021f45a6ee1a03a71d.jpg



Sent from my iPhone using Tapatalk
 
Last edited by a moderator:
The above way is only an option, whetever the values. I really want to control the current at 350 mA and let the voltage drop be.

You may be able utilize the ebay board but I have not had good luck. Not for this type of project but bought a motor controller and was not what the rating was.

I'll be using the meanwell buck for a constant current of 350 mA with two LEDs or 6. It should work, dont know never used one.
 
Did a horrible job of etching but still works. This board distributes the power to each string of LEDs. Even though these are dimmable I am not going to use it. Just leaving this pin open.

You're way over my head dude. I'll have to look into the ldd. I've only used the constant power supply drivers.
 
I think those are the same thing, different package. Do you plug those into the wall?
 
I googled. I remember.
Mine is just split into two components. Different route.
 

Latest posts

Back
Top