Discussion:
True performance of coloured leds ?
(too old to reply)
a***@gmail.com
2016-05-02 01:30:11 UTC
Permalink
Raw Message
I am looking around for a best combination of leds to illuminate my planted aquarium. I was looking for some mid/low power SMD leds (< 1W) since those used to yield better efficiency, less heat and thus less cooling equipment, while being suited to my moderate lighting need (freshwater plants require less light intensity than corals). To my surprise, when i browsed the web page of a prominent lighting led maker, ranking all their led family by efficacy, the most efficient appear to be the most powerful ones: 5W upward to 30W, peaking at 140lm/W and they are white leds, are those figures realistic ? Does it still make sense to choose under driven low power leds to achieve best efficiency ?

I am seeking to combine red, blue and white leds to obtain the optimal photosynthetic spectre while keeping a good colour rendering to human eye. How to choose the best efficiency for coloured leds since, as far as I know, the efficiency expressed in lumen/w depends on a convolution function of the own power spectral distribution curve of the subject led with a somehow idealistic human eye friendly power spectral distribution curve, so that blue or red monochromatic spectres generally have poor lumen efficiency.
Andrew Gabriel
2016-05-02 07:17:57 UTC
Permalink
Raw Message
Post by a***@gmail.com
I am looking around for a best combination of leds to illuminate my planted aquarium. I was looking for some mid/low power SMD leds (< 1W) since those used to yield better efficiency, less heat and thus less cooling equipment, while being suited to my moderate lighting need (freshwater plants require less light intensity than corals). To my surprise, when i browsed the web page of a prominent lighting led maker, ranking all their led family by efficacy, the most efficient appear to be the most powerful ones: 5W upward to 30W, peaking at 140lm/W and they are white leds, are those figures realistic ? Does it still make sense to choose under driven low power leds to achieve best efficiency ?
I am seeking to combine red, blue and white leds to obtain the optimal photosynthetic spectre while keeping a good colour rendering to human eye. How to choose the best efficiency for coloured leds since, as far as I know, the efficiency expressed in lumen/w depends on a convolution function of the own power spectral distribution curve of the subject led with a somehow idealistic human eye friendly power spectral distribution curve, so that blue or red monochromatic spectres generally have poor lumen efficiency.
One thought - can you use an occupancy detector to switch between optimum
growing lighting (usually mainly red, plus some blue for most plants) to
white lighting for viewing when people are around to see the plants?
The white could be formed just by adding a green light source (which
plants can't make use of), and maybe rebalancing the red and blue
intensity to achieve the white colour temperature required for viewing.
Of cource, such white light will not be a continuous spectrum, but neither
are white leds.
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
a***@gmail.com
2016-05-02 15:11:52 UTC
Permalink
Raw Message
Post by Andrew Gabriel
Post by a***@gmail.com
I am looking around for a best combination of leds to illuminate my planted aquarium. I was looking for some mid/low power SMD leds (< 1W) since those used to yield better efficiency, less heat and thus less cooling equipment, while being suited to my moderate lighting need (freshwater plants require less light intensity than corals). To my surprise, when i browsed the web page of a prominent lighting led maker, ranking all their led family by efficacy, the most efficient appear to be the most powerful ones: 5W upward to 30W, peaking at 140lm/W and they are white leds, are those figures realistic ? Does it still make sense to choose under driven low power leds to achieve best efficiency ?
I am seeking to combine red, blue and white leds to obtain the optimal photosynthetic spectre while keeping a good colour rendering to human eye. How to choose the best efficiency for coloured leds since, as far as I know, the efficiency expressed in lumen/w depends on a convolution function of the own power spectral distribution curve of the subject led with a somehow idealistic human eye friendly power spectral distribution curve, so that blue or red monochromatic spectres generally have poor lumen efficiency.
One thought - can you use an occupancy detector to switch between optimum
growing lighting (usually mainly red, plus some blue for most plants) to
white lighting for viewing when people are around to see the plants?
The white could be formed just by adding a green light source (which
plants can't make use of), and maybe rebalancing the red and blue
intensity to achieve the white colour temperature required for viewing.
Of cource, such white light will not be a continuous spectrum, but neither
are white leds.
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
Thanks, I has thought about that but is it that easy to drive the green led to render white in conjunction with the blue and red green, what wuld be the optimum choice ?
a***@general-cathexis.com
2016-05-03 21:49:07 UTC
Permalink
Raw Message
Post by a***@gmail.com
I am looking around for a best combination of leds to illuminate my planted aquarium. I was looking for some mid/low power SMD leds (< 1W) since those used to yield better efficiency, less heat and thus less cooling equipment, while being suited to my moderate lighting need (freshwater plants require less light intensity than corals). To my surprise, when i browsed the web page of a prominent lighting led maker, ranking all their led family by efficacy, the most efficient appear to be the most powerful ones: 5W upward to 30W, peaking at 140lm/W and they are white leds, are those figures realistic ? Does it still make sense to choose under driven low power leds to achieve best efficiency ?
I am seeking to combine red, blue and white leds to obtain the optimal photosynthetic spectre while keeping a good colour rendering to human eye. How to choose the best efficiency for coloured leds since, as far as I know, the efficiency expressed in lumen/w depends on a convolution function of the own power spectral distribution curve of the subject led with a somehow idealistic human eye friendly power spectral distribution curve, so that blue or red monochromatic spectres generally have poor lumen efficiency.
You do not want to measure efficiency in lumens/watt but in watts/watt within the PAR spectrum. Apparently the most efficient LEDs are blue which are up to 53% efficient. White LEDs are phosphor coated blue LEDs. You can buy similar LEDs with special phosphors for plants from Chinese Ebay stores, e.g., http://www.ebay.com/itm/10pc-1watt-full-spectrum-led-chip-1w-400nm-840nm-Plant-grow-lights-on-20mm-pcb-/262000853230?hash=item3d0077c0ee:g:oLIAAOSwd0BVyamK
There is just enough green in these so that plants look green. Initially, the lights look purple but, after your eyes get used to them, more of a warm white. Anyway, it's not going to cost you a fortune to find out.
Andrew Gabriel
2016-05-04 06:46:47 UTC
Permalink
Raw Message
Post by a***@gmail.com
I am looking around for a best combination of leds to illuminate my planted aquarium. I was looking for some mid/low power SMD leds (< 1W) since those used to yield better efficiency, less heat and thus less cooling equipment, while being suited to my moderate lighting need (freshwater plants require less light intensity than corals). To my surprise, when i browsed the web page of a prominent lighting led maker, ranking all their led family by efficacy, the most efficient appear to be the most powerful ones: 5W upward to 30W, peaking at 140lm/W and they are white leds, are those figures realistic ? Does it still make sense to choose under driven low power leds to achieve best efficiency ?
I am seeking to combine red, blue and white leds to obtain the optimal photosynthetic spectre while keeping a good colour rendering to human eye. How to choose the best efficiency for coloured leds since, as far as I know, the efficiency expressed in lumen/w depends on a convolution function of the own power spectral distribution curve of the subject led with a somehow idealistic human eye friendly power spectral distribution curve, so that blue or red monochromatic spectres generally have poor lumen efficiency.
You do not want to measure efficiency in lumens/watt but in watts/watt within the PAR spectrum. Apparently the most efficient LEDs are blue which are up to 53% efficient. White LEDs are phosphor coated blue LEDs. You can buy similar LEDs with special phosphors for plants from Chinese Ebay stores, e.g., http://www.ebay.com/itm/10pc-1watt-full-spectrum-led-chip-1w-400nm-840=
nm-Plant-grow-lights-on-20mm-pcb-/262000853230?hash=item3d0077c0ee:g:oLIAAOSwd0BVyamK
There is just enough green in these so that plants look green. Initially, the lights look purple but, after your eyes get used to them, more of a warm white. Anyway, it's not going to cost you a fortune to find out.
They may be tuned for plant growth, but plants don't want or need
full spectrum lighting for growth - chlorophyll can only use 680nm
and 700nm for energy generation used in plant growth, and blue is
used to trigger phototropism (growing towards the light). Light
inbetween these is not useful, which is why plants look green - they
don't want the green light. No spectrual distribution of the LEDs
is given to justify the claims of plant growth or full spectrum
(indeed, the two terms are fairly mutually exclusive).

Furthermore, when you see the term 'full spectrum' in a lighting
description, think 'snake oil', as it's such an abused/misused
term in the less professional side of lighting advertising
(usually being confused with high colour temperature, which again,
plants don't want).
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
a***@general-cathexis.com
2016-05-04 21:00:41 UTC
Permalink
Raw Message
Post by Andrew Gabriel
...
They may be tuned for plant growth, but plants don't want or need
full spectrum lighting for growth - chlorophyll can only use 680nm
and 700nm for energy generation used in plant growth, and blue is
used to trigger phototropism (growing towards the light). Light
inbetween these is not useful, which is why plants look green - they
don't want the green light.
...
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
You are wrong, mostly because of accessory pigments. The photosynthetic action spectra of many different kinds of plants have been empirically measured, e.g., see

Loading Image...

Loading Image...

Loading Image...

And, for no plants is your statement correct.
a***@gmail.com
2016-05-05 06:40:36 UTC
Permalink
Raw Message
Post by a***@general-cathexis.com
Post by Andrew Gabriel
...
They may be tuned for plant growth, but plants don't want or need
full spectrum lighting for growth - chlorophyll can only use 680nm
and 700nm for energy generation used in plant growth, and blue is
used to trigger phototropism (growing towards the light). Light
inbetween these is not useful, which is why plants look green - they
don't want the green light.
...
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
You are wrong, mostly because of accessory pigments. The photosynthetic action spectra of many different kinds of plants have been empirically measured, e.g., see
http://www.general-cathexis.com/images/ActionSpectra001.png
http://www.general-cathexis.com/images/ActionSpectra002.png
http://www.general-cathexis.com/images/ActionSpectra003.png
And, for no plants is your statement correct.
Hum,,

It is amazing because I had the same opinion as Andrew. So basically a peak some where between the blue and red is still useful to some plants...However, I am skeptical as for the spectrum extent of those purportedly full spectrum leds.

Besides, I read in datasheets recommendations of soldering temperature profile. My basic soldering iron cannot follow the profile, how much is the risk of damaging the led (or faltering its emission spectrum) when exceeding the max soldering temperature and not following the profile ?
a***@general-cathexis.com
2016-05-05 14:02:54 UTC
Permalink
Raw Message
Post by a***@gmail.com
Post by a***@general-cathexis.com
Post by Andrew Gabriel
...
They may be tuned for plant growth, but plants don't want or need
full spectrum lighting for growth - chlorophyll can only use 680nm
and 700nm for energy generation used in plant growth, and blue is
used to trigger phototropism (growing towards the light). Light
inbetween these is not useful, which is why plants look green - they
don't want the green light.
...
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
You are wrong, mostly because of accessory pigments. The photosynthetic action spectra of many different kinds of plants have been empirically measured, e.g., see
http://www.general-cathexis.com/images/ActionSpectra001.png
http://www.general-cathexis.com/images/ActionSpectra002.png
http://www.general-cathexis.com/images/ActionSpectra003.png
And, for no plants is your statement correct.
Hum,,
It is amazing because I had the same opinion as Andrew. So basically a peak some where between the blue and red is still useful to some plants...However, I am skeptical as for the spectrum extent of those purportedly full spectrum leds.
Besides, I read in datasheets recommendations of soldering temperature profile. My basic soldering iron cannot follow the profile, how much is the risk of damaging the led (or faltering its emission spectrum) when exceeding the max soldering temperature and not following the profile ?
You should only solder AFTER the LEDs are thermally connected to a heatsink and, in my experience, there is never damage. Incidentally, use thermal glue instead of wasting time with screws and screw holes. I use this:

http://www.ebay.com/itm/10g-HC-910-Thermal-Adhesive-Glue-Gel-Tube-Heatsink-Plaster-CPU-GPU-Mainboard-/311414023150?

Also, I just noticed that 3W chips are almost the same price as 1W chips therefore you may as well use 3W chips with 1W drivers (300ma). And, there are cheap, effective, plastic reflector/collimators that you can use with these chips,

http://www.ebay.com/itm/10pcs-60degree-led-Lens-White-holder-white-screw-for-1W-3W-High-Power-LED-/251745268162?hash=item3a9d2ff1c2:g:syAAAOSwD0lUgdgg
a***@gmail.com
2016-05-06 05:38:35 UTC
Permalink
Raw Message
Post by a***@general-cathexis.com
Post by a***@gmail.com
Post by a***@general-cathexis.com
Post by Andrew Gabriel
...
They may be tuned for plant growth, but plants don't want or need
full spectrum lighting for growth - chlorophyll can only use 680nm
and 700nm for energy generation used in plant growth, and blue is
used to trigger phototropism (growing towards the light). Light
inbetween these is not useful, which is why plants look green - they
don't want the green light.
...
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
You are wrong, mostly because of accessory pigments. The photosynthetic action spectra of many different kinds of plants have been empirically measured, e.g., see
http://www.general-cathexis.com/images/ActionSpectra001.png
http://www.general-cathexis.com/images/ActionSpectra002.png
http://www.general-cathexis.com/images/ActionSpectra003.png
And, for no plants is your statement correct.
Hum,,
It is amazing because I had the same opinion as Andrew. So basically a peak some where between the blue and red is still useful to some plants...However, I am skeptical as for the spectrum extent of those purportedly full spectrum leds.
Besides, I read in datasheets recommendations of soldering temperature profile. My basic soldering iron cannot follow the profile, how much is the risk of damaging the led (or faltering its emission spectrum) when exceeding the max soldering temperature and not following the profile ?
http://www.ebay.com/itm/10g-HC-910-Thermal-Adhesive-Glue-Gel-Tube-Heatsink-Plaster-CPU-GPU-Mainboard-/311414023150?
Also, I just noticed that 3W chips are almost the same price as 1W chips therefore you may as well use 3W chips with 1W drivers (300ma). And, there are cheap, effective, plastic reflector/collimators that you can use with these chips,
http://www.ebay.com/itm/10pcs-60degree-led-Lens-White-holder-white-screw-for-1W-3W-High-Power-LED-/251745268162?hash=item3a9d2ff1c2:g:syAAAOSwD0lUgdgg
A question comes to mind: how does the underdriven emission spectrum compares to the emission spectrum at the nominal current / voltage ?
aruzinsky
2017-03-25 14:05:04 UTC
Permalink
Raw Message
Post by a***@gmail.com
Post by a***@general-cathexis.com
Post by a***@gmail.com
Post by a***@general-cathexis.com
Post by Andrew Gabriel
...
They may be tuned for plant growth, but plants don't want or need
full spectrum lighting for growth - chlorophyll can only use 680nm
and 700nm for energy generation used in plant growth, and blue is
used to trigger phototropism (growing towards the light). Light
inbetween these is not useful, which is why plants look green - they
don't want the green light.
...
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
You are wrong, mostly because of accessory pigments. The photosynthetic action spectra of many different kinds of plants have been empirically measured, e.g., see
http://www.general-cathexis.com/images/ActionSpectra001.png
http://www.general-cathexis.com/images/ActionSpectra002.png
http://www.general-cathexis.com/images/ActionSpectra003.png
And, for no plants is your statement correct.
Hum,,
It is amazing because I had the same opinion as Andrew. So basically a peak some where between the blue and red is still useful to some plants...However, I am skeptical as for the spectrum extent of those purportedly full spectrum leds.
Besides, I read in datasheets recommendations of soldering temperature profile. My basic soldering iron cannot follow the profile, how much is the risk of damaging the led (or faltering its emission spectrum) when exceeding the max soldering temperature and not following the profile ?
http://www.ebay.com/itm/10g-HC-910-Thermal-Adhesive-Glue-Gel-Tube-Heatsink-Plaster-CPU-GPU-Mainboard-/311414023150?
Also, I just noticed that 3W chips are almost the same price as 1W chips therefore you may as well use 3W chips with 1W drivers (300ma). And, there are cheap, effective, plastic reflector/collimators that you can use with these chips,
http://www.ebay.com/itm/10pcs-60degree-led-Lens-White-holder-white-screw-for-1W-3W-High-Power-LED-/251745268162?hash=item3a9d2ff1c2:g:syAAAOSwD0lUgdgg
A question comes to mind: how does the underdriven emission spectrum compares to the emission spectrum at the nominal current / voltage ?
The phosphor colors do not change with driving current and the exact frequency of the blue LED is unimportant. See https://store.yujiintl.com/products/standard-phosphors?variant=968982261 .
Loading...