Epistar 3W LEDs are that bad?

[DrPhoton]

For the ratio of blue, red, green. Take into consideration both the electrical efficiency of LEDS and their efficacy for plant photosynthesis. For reference, use the mccree graph i posted in my earlier post. As you can see, red has a higher efficacy for the plant. Less photons are required to complete the same cycle of photosynthesis compared to green or blue. Red is slightly less efficient electrically compared to blue, how much will depend on the manufacturer. Green used to be really bad and this is the 'real' reason why green is never used for LED grow lights. If they had a higher efficiency they would be used, exclusively. Now days they are improved, but research shows that the lack of green does not have much if any photomorphological processes and physiological benefits, so adding green doesnt sound like a logical choice when its still less efficient compared to other colors. White falls into the same category as its just a combination RGB to make white.
Personally i would use more red than blue, for that very reason that red has a higher efficacy for a plant. Blue should only really be present for the photomorphological properties to be satisfied. If blue has a significant electrical effiency margin between red. Then i would bring the ratio of red and blue closer.

 

[DrPhoton]

This is not entirely related but i wanted to put this conversation here as it was good information that i would like to use to refer people to in the future.

Hello,

I've read several of your posts and you seem to have a passion for light technology. Do you have a COB LED unit that you would recommend?

I don't have the skill to build a DIY unit, so I'm looking for a commercial unit.

Thanks.

Hey there dragon, i have a passion for many areas of science. Personally i have not decided to choose LED over HPS because as far as looking at it from an electrical engineering point of view, they are not quite better...yet. But they are not far off. Really its not the technology that prevents LED from being better than HPS, but the manufacturers and cost factor preventing it from being a ideal.

Electrically, LED is more efficient watt for watt. But not by much, they are between 10-20% Max. But grow light manufacturers dont utilize this potential because of the cost limitation. It costs a lot to get that efficiency level.

Saying that, it wont be long when it will be indefinetly better. Semiconductors follow a law called haitz law and eventually it will completely replace all other light technology. As an electrician, LED'S have been a residential and commercial product for nearly 20 years. However its only of the last decade that they have been a more efficient and superior technology compared to other light technologies such as fluorescence, incandescence, compact fluorescence.

For grow light products available today, the only light i have seen that shows a real benefit over HPS is the products by spectrum king. Many like the kind LED but from memory i felt they were not very efficient.

I tell you what, i will have a good look at popular LED products and see if there are any that are decent.

Talk soon.

Thank you for your reply, I would appreciate your opinion.

I had a look around and could not see any other definitive brands with comparable products compared to kind LED and spectrum king. Although virpaspectra has some nice products with some seemingly more reliable marketing. They appear to be a good bang for buck brand and i am sure i have seen a couple of threads about them over the last few months. One particular model is the vipraspectra 700 with individual dimming controls for three color bands. This gives great flexibility and control for the user to define the lights color balance. Specially for me because i would use them quite a lot during germination and early vegging so i can ease them into full power. Giving them more blue than red in veg and then full power red in flower with a slight rollback of the blue.

But manufacturers dont provide any specs that you can use to 'actually' know what its power output actually is. Whats really needed is radiometry data. The typical PPFD at specific heights is completely crap on its own and says practically nothing about the lights capacity.
So look at vipraspectra, kind LED and spectrum king. See if you can find any youtube videos with people testing the lights, like how monster garden does with their famous light bult tests. Multiple PPFD readings over the entire canopy and then averaged out. This is the only way outside the manufacturers standard provision to know what a light actually puts out.
For comparison, 600W HPS has around 400umols canopy average, 1000w just over 520umols. With the best reflectors out there, this can be pushed to 800umols. A lot of energy is wasted from most mid range reflectors.

Compare those to other LED test results (if any) and compare the input power to output energy ratio to work out the efficiency difference.

Danny.

Thank you very much for looking into this. I had a look at Spectrum King and their SK600 looks awesome but more than I am willing to spend on my hobby. I was told I could write off costs as a medical expense since I have a recommendation.

I will have a look at Viparspectra, they seem to be an affordable choice. Any thoughts on CMH?

CMH has been tested quite a bit and unbeknown to most growers, there are heavy studys comparing light technologies. The most popular one is the economic analysis of greenhouse lighting.

They used an integrating sphere which is highly accurate and shows the entire radiant flux (total light output), so not a half assed or a cheap study. It shows that CMH is barely better than 1000w HPS. However not better than double ended HPS. DE is better by a significant margin. I would even argue that 600w HPS is probably as good or even better than CMH as the 600w bulbs are known to be more efficient than their 1000w counterparts due to chemistry. Which the study had not applied to their study.

Really, comparing CMH to HPS to LED right now, is really not about, which is more efficient. Because at this time, they are all very close. The question is, which has the best fixture cost per umol. Which HPS is far better by a large margin. Pay close attention to 'photon efficeincy' and 'fixture cost per umol' in the provided study. This will show you the important details between the technologies.

Danny.

 

[Tazard]

Here is a circuit very close to the one you are showing me. There is current limiting resistor to protect the leds. I think if you forward this to him that it'll get him on track
http://www.reuk.co.uk/wordpress/lighting/led-dimmer-circuit/

from my dad

 

[danbwc]

Hey man! Thanks for showing this to him. I'm not using these resistors cause I already have a driver with Q1+Q2+R3. In the circuit of that pag he has only the PWM so he needs those resistors to limit the current. But if this is the only thing he told you, I think my LED will work fine. Thanks a mill!

 

[danbwc]

Thanks again dan!

So, I have to use many drivers cause I'm limited by my power sources, if I wanted less strings I would need a higher power source and I don't want to buy one.

I thought about the idea of changing the driver for a 1200mA current and put two strings togheter in the same driver, but if one LED from one string burns, it would instantly burns one from the other string and I would have a problem even bigger. These drivers won't cost expensive, so no probs.

Again, the power source are my limits. I'm adding the maximum of LEDs the power source allows me. But I'll check again if I can make some changes to get a better result.

About the yellow (590nm), do you think would be a good idea to change some of the reds to yellow?

 

[Tazard]

No problem. He asked what the LED's were supposed to do. 😂 I told him "make light" but I see the reference is from a mj based source. He isn't dumb, he gave me my first HPS lol.

 

[danbwc]

Lol hahah good luck with your plants man, thanks for the support

 

[Tazard]

Anything anytime

 

[DrPhoton]

Only if yellow is more electrically efficient, otherwise not imo. A plants response to different wavelengths of light comes under photomorphogenesis. If you satisfy the photorecepters that have morphological properties such as cryptochromes and phytochromes, any additional wavelength energy is just that, more energy. It would likely not add to a plants overall health because the photorecepters responsible for physical change are already being satisfied. Stick with wavelengths between 400-500 for blue and 600-700 for red. Using the most electrically efficient between those ranges. UVB can be used and does appear to have photorecepters and provide improved plant health and quality. But they provide no additional support for photosynthesis.

 

[DrPhoton]

Looking at it again closely, i think the topology you have chosen is the most ideal. Although i still think a single driver is more efficient and simplistic, you do have to compromise with what you have to work with.
I suppose any improvements from here would be from the ratio of serial to parallel devices. Adjusting this ratio to suit the drivers and power supply could possibly improve efficiency.

Perhaps when you decide to build your final light, you can aquire a big powersupply and drive units that could implement as many chips as your budget allows.

 

[DrPhoton]

One thing to also think about that we have not covered yet but will likely not be importent until the final build is the optics. Optics play a big role in the control and dispersion of light. Its analogous to regular lights with reflector hoods. The great thing about LED lights is the ability through optics to focus the light more. Light hoods loose energy through reflection loss and their often poor design in small grow areas adds to that when they are often reflecting off walls or directing light beyond the canopy and completely missing the plants. With optics, light is made to collimate more and shows more control with light, with very little energy loss (the energy lossed through optics is far less compared to reflection).

So the decision on the chips and their optics is important so that you can control light to be used as efficiently as possible. This comes down mostly to the optics beam angle but also how you arrange the LED's, spacing and placement all have an effect. The goal is to create a well focussed light that does not leak outside the intended canopy area but also have great uniformity. Uniformity both horizontally and vertically is the key.

However this is likely going to be an even bigger struggle compared to just building the light itself. It really shows the complexity involved in building a light.

 

[Tazard]

Will dimming them cause a loss of intensity? Thinking about penetration. Or will the additional LED's allow the lights to be closer?

 

[DrPhoton]

Yea exactly that, dimming the lights means you just have to bring them closer to have the same penetration and vertical uniformity. But any kind of reduction in radiant flux means less total output capacity and yeild. Penetration is only important for distribution of energy, spreading the load for more efficient use of the energy provided.
Thats why the height of the light does not dictate yeild, radiant flux is.

 

[Tazard]

Yes but having the light a greater distance (with same reflector) disperses the photons over a larger area decreasing their density at any point. Lowered PAR. Right?

 

[DrPhoton]

EDIT.

If the light is directed down without wall reflections, the same number of energy in the form of photons will reach the canopy that has left the source. Although the density decreases, this is only the spread of light through more space. This is what improves uniformity, more light spreads and more of the energy is shared among the space.

The energy is not destroyed or decay during its travel.
This is why 1 watt in radiant flux is the same as 1 watt in irradiance and 1 watt in radiant intensity. Because when all energy leaves a source and is recollected, nothing changes.

 

[Tazard]

Yeah I actually said that wrong should have been density of ppf not par as par is constant irrelevant of location.

 

[Tazard]

Little off topic but it's been shown that white light is more efficient to plants than blue light at converting light to energy.

 

[DrPhoton]

Yea white light probably does have a higher efficacy than blue light because depending on the color temperature of white, green and red are both more effective than blue. Which white LEDS are more heavily concentrated with red and green. However, white is electrically less efficient. Especially when white LEDS have a lot of green which is again less electrically efficient than both blue and red.

 

[DrPhoton]

Yeah I actually said that wrong should have been density of ppf not par as par is constant irrelevant of location.

Well actually PPF is a measurement technique of total light output, measured with such devices as an integrating sphere. When you take PAR readings from a unit of area, thats PPFD.
Thing is, a single PPFD measurement with artificial light, does not show you your lights output capacity. In actuallity, it only shows a PPFD reading over the diameter of the sensor only, not per square metre. Unless you are measuring a light source such as the sun where light is uniform in any direction.
Thats why measurement systems without integrating spheres will take readings from multiple locations accross the entire canopy and calculate a figure called the canopy average. Although no where near as accurate as a true PPF reading, its the best we have without having to fork out a lot of money.

To answer your question. Although if you take the same measurement say from the center would be lower after you raised the light. You would find that other readings outside the center would actually increase. Due to the spread of photons. So as i have said previously, its just the spread and uniformity that changes but the overall capacity doesnt change unless it losses its energy through conversion to heat (hitting at object).

-------------------------------------------------------------------

My apologies tazzard i believe i misinterpreted your question. You were asking if PPF changes due to the lower PPFD meadurement when raising a light.

As previously said above, the uniformity increases but the PPF stays the same.

 

[Tazard]

PPFD is photosynthetic photon flux density. PPFD measures the amount of PAR that actually arrives at the plant, or as a scientist might say: “the number of photosynthetically active photons that fall on a given surface each second”. PPFD is a ‘spot’ measurement of a specific location on your plant canopy, and it is measured in micromoles per square meter per second (μmol/m2/s). If you want to find out the true light intensity of a lamp over a designated growing area (e.g. 4’ x 4’), it is important that the average of several PPFD measurements at a defined height are taken. Lighting companies that only publish the PPFD at the center point of a coverage area grossly overestimate the true light intensity of a fixture. A single measurement does not tell you much, since horticulture lights are generally brightest in the center, with light levels decreasing as measurements are taken towards the edges of the coverage area. (Caveat Emptor: Lighting manufacturers can easily manipulate PPFD data. To ensure you are getting actual PPFD values over a defined growing area, the following needs to be published by the manufacturer: measurement distance from light source (vertical and horizontal), number of measurements included in the average, and the min/max ratio). Fluence always publishes the average PPFD over a defined growing area at a recommended mounting height for all of our lighting systems.