LED basic theory

[Naptalene]

Hi guys,

 

I've had a DIY rig on my nano for about 3 years containing Cree RB,CW, Cyan and UV.

I originally had a grid layout and then rebuilt with Neutral Whites and dropped the numbers of LED's and did a bit of a circular pattern with the stars.

 

I also felt that I was blasting the corals before and figured I'd drop the Amps on the ELN drivers so everything sat down at 700mA. I figured that If I was running the controller at 50% anyway I'll get a better resolution on the PWM fade by using all the steps up to 700mA instead of only half(ish) of them up to 1.3A (this is kind of the point that I'll get back to)

 

This was working well but I tried droppping the PWM signal again.

It's now siting at 40% for the RB, 40% for the Whites and 20% for the colours (looks a bit nasty but my SPS growing like beasties) The SPS have also coloured up even more. So I figured that it must have been too much intensity.......

 

But then I was thinking - which is dangerous at the best of times - the pwm signal doesn't adjust the "strength" of the lights does it? If I'm not mistaken It just cycles them on and off for a given period?

So when we are adjusting the "strength" on our controllers are we actually adjusting the relative light cycle? Obviously our eyes pick up the dimming effect so we think that channel is brighter or duller but is it actaully?

 

I think the crux of my question is this....

 

If you are running an LED string on a PWM controllable channel and you change from say 100% to 50% (and lets ignore any inefficiencies and assume it is exactly half) do you get 50% less "power" or 50% less illumiation time?

Does the answer to this transfer to an analogue controlled PSU?

 

Please put me out of my misery on this one.

Thanks

[no694terry]

I was always under the impression you changed the amperage which in turned increased the wattage because of the watts = amps x volts thing. I would think if your running a led down to 20% pulse that you would see and obvious flicker so i would think the lights are adjusting intensity, not on/off time

 

and you reminded me that warmer light grows some things faster, i may have to look into that again and make some adjutments to my light cycle controller so im closer to 10k while im at work and 18k again while im at home before the lights turn out

[Naptalene]

Led's are constant Amp so that's doesn't change with the controller on a PWM system. Not sure on a 5 or 10v one.

 

So you set your amps on the driver then it adjust voltage to how many led's you have. That's why I get cleverer people than me to tell me how many led's I can run on a string.. If I have it correct lol.

[Nano sapiens]

http://afrotechmods.com/tutorials/2011/11/28/pulse-width-modulation-pwm-tutorial/

 

The first 1/3 explains the concept.

[evilc66]

That explains pwm, but doesn't really answer the OP's question.

 

With the ELN drivers, they are constant current output (more specifically, constant averaged current). There is no chopping of the output via pwm. The pwm signal that comes into the driver from your controller only tells the internal control logic to adjust the current output based on the duty cycle of the signal. This makes for flicker free lighting even at lower duty cycles and frequencies.

 

This is the way that most drivers work. There are some exceptions to this, but for the most part, the drivers we all deal with at this time work this way.

[Naptalene]

Thank you guys,

 

I'll definitely look at the link

 

It looks like I have a major misunderstanding of the pwm functionality. I thought it flickered the lights on and off at varying speeds and we just saw the change as a dimming/brightening.

Back to the books for me then

 

Something new every day

[Naptalene]

Embarrasingly I only just managed to go through that link now.

That was a brilliant explanation of duty cycle and the fact that it's giving an average voltage not a digital on/off as such - which is what I was thinking in my head.

I had also assumed that "Constant Current" meant the current sat at what I dialled in on my ELN with the multimeter and the voltage was controlled by the PWM to adjust the brightness. So thanks for those two answers, serious coin drop moments.

 

Question2 WRT Colour choices......

I have read the "Colours and what they do" sticky, and realise that LED has a much tighter bandwith on their wavelength.

Having said that......

 

There is a local supplier selling 10W "12-15K" chips.

LED's also use Phosphours to get the colours correct?

Is there a difference between a MH running 15K and an LED running 15K?

If there's a good link to this answer please let me have it.

 

The reason I ask:

 

I setup a small tank (20L) to test the viability of CHEAP NASTY 10W CW Led floodlights for frag propogation.

It is the most neglected tank you have seen.

So far growth is good and not a single loss of Plating Monti, Elkhorn or Pink Birdsnest. I also added a beaten up piece of Acro that had been stung and it has healed and is doing well.

They are also very brown.

 

So I was wondering about trying one of these 12-15K chips (I assume that variance is less than strict QC) and supplementing with RB and NW and see if I get growth and colour. Its very shallow water so I don't think any lack of power compared to "Proper" LED's is too much of an issue. They are extremely well priced here (+/_ $5 but I dont have to worry about shipping/import duty/VAT) and so far look like a surprisingly viable option.

[Naptalene]

I was always under the impression you changed the amperage which in turned increased the wattage because of the watts = amps x volts thing. I would think if your running a led down to 20% pulse that you would see and obvious flicker so i would think the lights are adjusting intensity, not on/off time

 

and you reminded me that warmer light grows some things faster, i may have to look into that again and make some adjutments to my light cycle controller so im closer to 10k while im at work and 18k again while im at home before the lights turn out

You were right

[Matteo]

This was a really informative read:)

[evilc66]

Embarrasingly I only just managed to go through that link now.

That was a brilliant explanation of duty cycle and the fact that it's giving an average voltage not a digital on/off as such - which is what I was thinking in my head.

I had also assumed that "Constant Current" meant the current sat at what I dialled in on my ELN with the multimeter and the voltage was controlled by the PWM to adjust the brightness. So thanks for those two answers, serious coin drop moments.

 

Question2 WRT Colour choices......

I have read the "Colours and what they do" sticky, and realise that LED has a much tighter bandwith on their wavelength.

Having said that......

 

There is a local supplier selling 10W "12-15K" chips.

LED's also use Phosphours to get the colours correct?

Is there a difference between a MH running 15K and an LED running 15K?

If there's a good link to this answer please let me have it.

 

The reason I ask:

 

I setup a small tank (20L) to test the viability of CHEAP NASTY 10W CW Led floodlights for frag propogation.

It is the most neglected tank you have seen.

So far growth is good and not a single loss of Plating Monti, Elkhorn or Pink Birdsnest. I also added a beaten up piece of Acro that had been stung and it has healed and is doing well.

They are also very brown.

 

So I was wondering about trying one of these 12-15K chips (I assume that variance is less than strict QC) and supplementing with RB and NW and see if I get growth and colour. Its very shallow water so I don't think any lack of power compared to "Proper" LED's is too much of an issue. They are extremely well priced here (+/_ $5 but I dont have to worry about shipping/import duty/VAT) and so far look like a surprisingly viable option.

 

To begin with, color temperature, or CCT (correlated color temperature) is pretty subjective. Then you have the fact that any CCT can be created many different ways. In short, a 15K MH setup could look different than an LED 15K setup.

 

LEDs do use phosphor (not phosphorus) to convert certain wavelengths of light into other wavelengths, but they are typically limited in use to white LEDs (blue die with a yellow/green phosphor applied to create white light). Of late, there have been other "phosphor converted" LED colors such as PC amber, lime, and magenta. All other colors (violet, blue, royal blue, cyan, green, amber, red, deep red, etc...) are all direct color. That means that the physical makeup of the die material is what creates the specific wavelength of light.

 

High CCT LEDs are not really desirable, as they generally have low color rendering. You would be better off starting with the cool white LED you currently have and add LEDs to it to improve the overall color of the light and the color rendering. Lack of blue light (440-480nm) is the primary reason why you aren't getting great color, but it also could be a lack on intensity as a coral will increase the amount of zooxanthilae (which are brown) at the surface of its tissue to maximize it's light gathering capabilities.

[Naptalene]

I'm back.......

Basic question again

 

• Assumption1..... If I take an LED rig - it has an area of high intensity that it emits out in a very directional manner, that is commonly demostrated as a cone shape.
• Intensity will drop the further away from the point source you move ie vertically and horizontally.

Does the distance that the LED is ABOVE the water affect the strength of the light that reaches the bottom of the tank?

So...

• LED is 3" off the water line.
• I get 50 PAR on the substrate in a 4" area

If i lift the LED to say 12" above the water line, will I still get 50 PAR (or very close) in a wider area, or will the PAR drop as well?

 

Is there a standard formula you guys use to determine the footprint of LED's to cover a given surface area at a certain PAR?

 

Thanks again.

[evilc66]

Of course PAR will drop as the fixture is moved away. The upside here is that even though the intensity is reduced, the area of coverage is increased, so you would get a more uniform light pattern.

 

What you are looking for is the inverse square law. It won't help you determine the coverage area though. You can figure that out with basic geometry.

 

https://en.wikipedia.org/wiki/Inverse-square_law

[Naptalene]

But moving higher, you are spreading out over a less dense medium (air)which should equate to less power lost? Or do you still have the same amount of photons spread out any way?

 

So is it better to add more LEDs to spread out the coverage or raise the light..... Or depends lol.

 

I do apologize for the recurring stupid questions but I try to understand what's going on.

[Naptalene]

Ok, looking at that link...... My dumbass recap is:

 

The LED is like a shotgun.

Little light particles come flying out (birdshot)

The more of that birdshot in a certain area the higher the par.

 

Therefore the further away, the more spread out the buckshot is, hence a lower par value. Same amount of birdshot, just less concentrated

 

Taking that further..

 

Can I equate the size of the buckshot to wavelength?

[evilc66]

In a way, yes. Smaller shot would be longer wavelengths (orange, red, etc...) which would lose their energy quicker and won't penetrate as deep as larger shot would (shorter wavelengths, violet, blue, etc...)