Diffused light

[naoki]

I happen to see these slides, and it talks about effects of diffuse and direct light on plants (about 1/3 from the beginning).

https://ag.arizona.edu/ceac/sites/ag.arizona.edu.ceac/files/10 R. Hernandez Plant Lighting.pdf

It talks about different glazing technology for greenhouses, and describe the advantage of high haze factor. With more diffused light, plants can receive light more evenly (less shadowing from the top leaves), and can increase the amount of light without causing the leaf burn (or photo inhibition).

I thought that it is kind of interesting (especially the real data about Tomato yield), and it might be useful for people who is going to build a new greenhouse. It's not really directly applicable for artificial light growers, but it has some implication about DIY LED light design.

 

[Brabantia]

Thank you Naoki for this paper. A good summary of the question, very useful.

 

[ALToronto]

Too bad they tested only blue/red LEDs against full(er) spectrum lights. Interesting point about diffuse light, but what about light loss due to light reflecting back towards the source? It's fine for sunlight in the summer, but the rest of the time we need all the light we can get.

 

[Ray]

I've been toying with the idea of using Solexx to recover my greenhouse, as it's a lot more energy efficient than the multiwall polycarbonate, but the shade/diffusion factor concerned me. maybe I shouldn't worry so much...

 

[terryros]

That was a very nice summary and contained most things I have been able to find. What is lacking are specific studies in different orchid genera. Not surprising since there is a bit more money in tomatoes than in orchids! If we could just do controlled comparisons of different light sources, spectra, and diffuse versus collimated light in different orchids. We have pieces of this with a few genera, but not complete stories. We are left with extrapolations and observational anecdotes by various growers. Most of us have difficulty controlling just one of the growth variables (light) keeping everything else the same. We always have uncertainty about attributing what we observe just to a change in light.


Sent from my iPad using Tapatalk HD

 

[TyroneGenade]

Great, thanks! I like slide 43. It is exactly what I expect to see in my aquarium plant experiments. Pity CWF isn't graphed on that chart...

Now I can dream about my green house with a little more precision.

Bye

 

[naoki]

Interesting point about diffuse light, but what about light loss due to light reflecting back towards the source? It's fine for sunlight in the summer, but the rest of the time we need all the light we can get.

Well, I came across it to look some info about point source vs scattered LEDs. My lazy COB type LED is less diffused (4x 50W COB to cover 4x4' area). Your type of using many, smaller LEDs scattered all over had the advantage. Similarly large light emission surface of florescent light has this advantage of scattered light. I mentioned this a couple times, but I was trying to see how much effect it could have (from experiments), and I came across this PDF. Also, using lens to make the beam angle narrow has the issue of direct light. For our eyes, direct light looks brighter (because it creates strong contrast).

I've been toying with the idea of using Solexx to recover my greenhouse, as it's a lot more energy efficient than the multiwall polycarbonate, but the shade/diffusion factor concerned me. maybe I shouldn't worry so much...

Is this the one?: http://www.solexx.com/index.html
I guess that it is pretty complex to make the choice of the materials for greenhouse (I've never done it). For the house, we can easily get the energy evaluation, and they will tell us how long it will take to cover the extra cost of insulation. I'm guessing that you have to do the calculation by yourself.

That was a very nice summary and contained most things I have been able to find. What is lacking are specific studies in different orchid genera. Not surprising since there is a bit more money in tomatoes than in orchids! If we could just do controlled comparisons of different light sources, spectra, and diffuse versus collimated light in different orchids. We have pieces of this with a few genera, but not complete stories. We are left with extrapolations and observational anecdotes by various growers. Most of us have difficulty controlling just one of the growth variables (light) keeping everything else the same. We always have uncertainty about attributing what we observe just to a change in light.

Yeah, I guess food is more relevant than flowers.... I think Ray has been mentioning whether photosynthesis of shade plants differ from sun plants. The typical response curve is from crop plants growing in full sun. I'm also interested in it, but I haven't looked into it.

I like slide 43. It is exactly what I expect to see in my aquarium plant experiments. Pity CWF isn't graphed on that chart...

Dry mass vs percent blue figure? CoolWhiteFlorescent is in the picture, but I don't know which point it corresponds to. I'm a bit surprised (or maybe I learned it wrong) to see that pure blue is better than pure red. I've always thought that pure red causes etiolated (elongated) plants, and that's why NASA added blue to make the plants more compact. But this slide shows opposite effect. hmmm.

 

[Bjorn]

Thank you Naoki for this post. Very interesting. What I could miss is something on white LEDs since I am playing around myself with those.

 

[terryros]

I found this article several weeks ago which I think makes the issue of diffuse versus collimated light (what is coming out of directionally focused LEDs) more uncertain. My experience with Phalaenopsis and LEDs with 40 degree focusing lenses suggests that PAR photon flux greater than about 30 micromoles/M2/sec result in compacted/thickened leaves and either shorter flower spikes or perhaps even reduced spiking. Do photons coming in at angles farther and farther away from vertical get reflected off to some degree? Maybe we measure these photons with our PAR meters but they don't end up being as efficient with the leaves. All guesses.

http://onlinelibrary.wiley.com/doi/10.1111/j.1365-3040.2007.01751.x/full

 

[TyroneGenade]

Phytochrome is also stimulated by blue light. It could be that the NASA experiments used a red diode with too much far red light in it? Who knows... but the graph says that the more red, the more biomass. The 100% blue data point does look odd but clearly plant growth has more to it than simply the number of photons. Spectral input matters. The blue is needed to trigger stomatal opening as well as certain aspects of phototropism. The high blue might just mean that more CO2 entered the plant, removing CO2 limitation. I assume there was ample water... It would be more interesting to use a C4 or CAM plant in the experiment to avoid the issues of water and CO2 limitations.

 

[naoki]

Bjorn, are you playing with LED for the winter supplement light in the greenhouse? For this, HPS might be still better?? But LED is getting better and better. In the last couple weeks, both Cree and Bridelux announced the improved 2015 version of COB LED (CXA2=CXB and Vero, respectively). I think about 10% or more increase in the efficiency.

Terry, that paper looks really interesting, I need to read it. Thanks!

Tyrone, do you mean cryptochrome (not phytochrome)? Blue light induction of stomata opening (and green light inhibition) is well studied, but these experiments usually isolate the stomata, and study the response. Paphs are used for this kind of research, because Paphs are weird that they don't have chloroplasts in the guard cells. In real life, guard cells use more integrated cues (humidity, CO2 level, temperature, circadian rhythm) to control the stomata opening. I'm sure you know about this. I'm not completely sure that blue light translate to more CO2 entered into the plant.

My understanding is that the growth of most strict CAM is limited by CO2 stored in vacuole (as Malate) under cultivation. But it would still be interesting to see the response to different quality of light. Related to this, I thought that CAM orchids (e.g. Cattleya) may grow faster if we use shorter day. For example, instead of 12h day 12night, I wonder if they grow faster with 6h day+6h night. Anyone tried this?

Related to Terry's comment #5, I recently saw that the ratio of chlorophyll a and chlorophyll b is correlated with the light habitat (sun vs shade plant). So this suggest the shade plants like orchids could have slightly different response to wavelength of light than crop plants.

 

[TyroneGenade]

No, I mean phytochrome. Phytochrome is also stimulated violet/blue light. The cryptochromes play roles in phototropism as well as stomata opening. I vaguely recall what you mention but I'm obsessed with issues of CO2 spectra so it didn't even pop into my mind when writing the response.

Incidentally, here is the Paph paper: http://pcp.oxfordjournals.org/content/43/6/639.full
The paper reporting a blue sensitive phytochrome response is http://www.ncbi.nlm.nih.gov/pubmed/16661489 but the blue response is only seen under certain conditions. Why blue-light stimulation may be important is due to the distortion of the absorbence by the plant tissue (http://www.ncbi.nlm.nih.gov/pubmed/24413936). In 100% blue the phytochrome response, coupled with maximaly cryptochrome stimulation could enhance growth? I am now speculating wildly but it is fun. Blue should shorten hypocotyl extensions and this is the opposite to what is seen on pg 44 of the slide show so it might have zero to do with Phytochrome at all and is 100% Cryptochrome + lots of blue photons so 100% modification of the photosystems to receive light of one color? That plant also looks a little more yellow than the others so perhaps more carotenes for photon capture and anti-oxidant defense?

From my understanding, the benefit of CAM and C4 is that the compensation points are much lower so there is less photorespiration and thus more effective use of the CO2 and light energy available (never mind water conservation aspects).

I would appreciate that chlorophyll paper. I have data on aquatic plants grown in shade and bight light and the ratios of chl a: chl b: beta-carotene are pretty much stable even though the absolute amounts change.

Bye

 

[naoki]

Sorry for the nerd talk for the other people... Please ignore this message (except Tyrone, who is a biologist)!

The paper reporting a blue sensitive phytochrome response is http://www.ncbi.nlm.nih.gov/pubmed/16661489 but the blue response is only seen under certain conditions.

This is an older paper from the time when blue receptor wasn't found. Check high irradiance reaction (HIR) section of more updated info here:
http://www.photobiology.info/Shinkle.html
The HIR is mediated by Phytochrome (the far red part) and by Cryptochrome (blue part).

Why blue-light stimulation may be important is due to the distortion of the absorbence by the plant tissue (http://www.ncbi.nlm.nih.gov/pubmed/24413936).

Is this the correct link you meant:
"Distorted phytochrome action spectra in green plants"?
I don't have the access to it at home, but it seems a bit unrelated.

I don't know enough about the topic, so I can't quite comment on your explanation of p.44 of the slide. It sounds interesting.

But the blue stomata opening is not related to Cryptochrome. The last time I checked this topic, one candidate was Zeaxanthin (related to Carotenoid).

From my understanding, the benefit of CAM and C4 is that the compensation points are much lower so there is less photorespiration and thus more effective use of the CO2 and light energy available (never mind water conservation aspects).

Are you sure you are not confusing mitochondorial respiration with photorespiration? Usually the compensation point is defined as the point where the balance between mitochondorial respiration and photosynthesis cause the net CO2 production of 0. On the other hand, photoresipriation is the wasteful process when rubisco binds to O2 instead of CO2. During the light rxn, P680 of Photosystem 1 receives electron from hydrolysis of water molecule, which causes the production of O2 and H+. Rubisco can binds to both O2 and CO2, and when it binds to O2, it creates a wasteful product, and the plants has to convert it back to a useful carbohydrate. During this process, the plant lose C as CO2 emission, and plants waste energy. This is photorespiration. C4 and CAM plants try to reduce this wasteful process of photorespiration by separating the light reaction from carbon assimilation reaction (Calvin cycle, where rubisco is an important enzyme). CAM and C4 both uses C4 pathway. C4 pathway requires extra energy, so under the ideal situation (enough water, high humidity so that stomata can freely open, not high temp which causes higher O2/CO2 ratio). C3 should be more efficient.

I would appreciate that chlorophyll paper. I have data on aquatic plants grown in shade and bight light and the ratios of chl a: chl b: beta-carotene are pretty much stable even though the absolute amounts change.

I don't remember where I saw it originally, but p.359 of this:
http://www.moodle.ufba.br/file.php/10674/artigos_folhas/Boardman_1977.pdf
discuss about it. The same page describe something (orientation of grana) related to what Terry mentioned above. Also there are quite a lot of papers describing the change in Chl a/b as the acclimation to light environment within a single species. I don't know much about this topic.

 

[Bjorn]

Naoki, right, its as supplements for the HPS. I have been wondering whether or not it would be feasible to exchange the HPS as well, because my LED design seems less bulky than the HPS I currently use. Guess its better to clean the equipment though and use the LED as supplementary blueish source. I tried to boost my heatsink (approx 30x10cm) to a whooping 300W, burt ended up frying all the LEDs. It takes 200W with proper fan though. The good thing with the LEDs is that they are less dependent on reflectors, actually I run them without.

 

[TyroneGenade]

Hi Naoki,

Yes, I mean the "Distorted phytochrome action spectra in green plants" paper as it indicates that that the typical test tube action spectra for Phytochrome might be experimental BS. But as the earlier blue effect is now explained by Cryptochrome it could be meaningless. (The danger of speculating wildly.)

Yes, I mean compensation point. It is lower in C4/CAM as opposed to C3 for CO2 concentration. As consequence, the plant runs into fewer issues with energy waste and wasteful photorespiration. Blue-light usage isn't as efficient as red-light usage so perhaps the stronger growth is simply because the photosystems don't get saturated and never supply too much ATP + NADPH and that there is then slow turnover in the Calvin Cycle and photorespiration doesn't become an issue? But I do see, from your explanation, that I'm making a mess of things with my garbled thinking (I'm not a plant physiologist by trade) so I'm going to rather shut-up from this point.

Thanks for the link.

 

[DavidCampen]

I tried to boost my heatsink (approx 30x10cm) to a whooping 300W, burt ended up frying all the LEDs. It takes 200W with proper fan though. The good thing with the LEDs is that they are less dependent on reflectors, actually I run them without.

I use 8 of these SA-LED-176E, 76 mm high x 70 mm diameter, without fans, to dissipate about 140 watts.

http://www.mouser.com/ds/2/303/s-sink-23139.pdf

 

[naoki]

I found this article several weeks ago which I think makes the issue of diffuse versus collimated light (what is coming out of directionally focused LEDs) more uncertain. My experience with Phalaenopsis and LEDs with 40 degree focusing lenses suggests that PAR photon flux greater than about 30 micromoles/M2/sec result in compacted/thickened leaves and either shorter flower spikes or perhaps even reduced spiking. http://onlinelibrary.wiley.com/doi/10.1111/j.1365-3040.2007.01751.x/full

I finally get a chance to read it. It is pretty interesting, and that's something I didn't know. It make sense that shade plants utilizes more diffused light and sun plants like direct light. According to the review paper I linked to the reply to Tyrone, the structure of chloroplast could be the reason for the difference of shade vs sun leaves. Simply put, sun plants have more neatly arranged chloroplast since most of the light is coming from the top, and shade plants have the sensors facing more random direction for scattered light. It is also interesting that within an individual, chloroplast structure could change by acclimation. So if I put plants, grown in diffused light green house, under point source, directional LED (like COB), then the plants may take for a while to be happy with the new environment. I wonder if this effect is what you are seeing.

My LED doesn't have the narrow focus lens (120 degree or so without reflector). But my species phals seem to be happy with 80-100 micromol/m^2/s. But lots of other environmatal factors (RH and temp) would influence the optimal light.

Do photons coming in at angles farther and farther away from vertical get reflected off to some degree? Maybe we measure these photons with our PAR meters but they don't end up being as efficient with the leaves. All guesses.

With reflection of light, if the light is hitting a surface perpendicular to the surface, least light is reflected as you said.
http://www.physicsclassroom.com/getattachment/reasoning/refraction/src41.pdf
But I'm not quite sure how this explains why your Phals prefer < 30 micro mol/m^2/s (it seems to be rather at a low end).

Normal PAR meter is measuring the density at a flat surface; number of photons going through 1m^2 flat glass regardless of the angle. In other words, if you compare the PPFD of light hitting the sensor perpendicular vs 45 degree angle (let's say you use the same light source from the same distance), the PPFD of 45 degree light should get about 1/sqrt(2) of the perpendicular light in theory. This is an ideal response of the sensor, and many meters try to achieve this response when it receives off-axis light (this is called cosine correction). Apogee (I think this is what you have) does this correction, but another cheap meter, Seneye, doesn't respond correctly.

Leaves are relatively flat, but they are not completely flat. Under the scattered light, leaf can get more photons than PAR meter suggests. Indeed there is a meter which measures the light from all direction (see p.3 of
http://www.licor.com/env/pdf/light/Rad_Meas.pdf,
which describes Photosynthetic Photon Flux Fluence Rate). Instead of flat surface, PPFFR is a measurement of light which is hitting a single point from any direction. So in plants, the actual light what they receive is in between PPFD (from normal PAR meter) and PPFFR.

Bjorn, supplementing HPD with blue (or blue dominated white) LED (most efficient conversion from electricity to light energy) seems to be a good idea. That's too bad that your LED got burned. I don't know too much about heatsink, but mine is just cheap (salvaged) CPU heatsinks+fans. 5x5x4cm can easily handle 60W COB. But if you are trying passive cooling like David, I can see that 300W could be too much. If you are going for efficiency, you want to underdrive (you might already know this), though.

 

[ALToronto]

In indoor conditions, a big risk from direct light is heat buildup. I think if we're dealing with sunlight, diffusing the light a little may not be a bad idea. But with artificial lighting, we need all the photons we can get - how many people here have maxed out their light meters with artificial lights?

Another interesting tidbit in the paper referenced by Terry that caught my attention - green light isn't useless! It penetrates deeper into the leaf than blue light. Another reason not to use the hideous red-blue LEDs.

 

[Bjorn]

Naoki, David,
Lots of good info in this thread, thanks a lot. The heatsink issue is my attempt to get as much as possible out of as little space as possible. What I do is that I floodlight my greenhouse with this extra capacity, an since 'my canopy' is already quite occupied, space is an issue. Passive cooling was a dream, but its simply not possible with these restrictions. As I use the light to heat up the greenhouse, the efficiency is not that big an issue, although I must turn of the light when the sun starts shining these days. But that is done automatically. I do all the viring myself, and come across quite a few issues with respect to ground currents(wonder if thats the word?). Our electrical system is more or less foolproof, cuts the main supply when there is an imbalance between the leads of more than 25mA. Prevents me from getting fried, but is a nuisance what concernes functionality. But then again, safety first!
Back to issue, I need to make something compact, the frying of chips is just a part of the development cost. The cost of the chips are not that bad, its all the other things that add up, but luckily, the destruction confines to the chips so I just have to modify the design, replace the chip and start over.
Currently it seems to work in a way, but I am still looking for improvements. Water cooling could be an option but then the heat would be lost.

 

[naoki]

I think that in the US, we call it Ground Fault Circuit Interrupter (GFCI).

With white LEDs, if you want to get compact, strong light sources, I would probably design around Bridgelux Vero 29 or 18. As I mentioned, a new version was announced, and should be available pretty soon. Here is the data sheet for the ver 2.0 (=2015 version).
http://www.bridgelux.com/wp-content...ero-29-Data-Sheet-with-increased-efficacy.pdf

The older version is discounted now (around $28), but the price for the new version will be $30-40 per piece. Since you want to have more blue, 5000K 70CRI would be the best (part# BXRC-50C10K0-L-24). The nominal current is 2.1A, and Vf is 36.8-38V, so about 80W. Vero can be still good at higher current than the nominal (i.e. reduction in efficiency isn't so bad).

Edit: hmm, I noticed that in 5000K, the efficiency increase is not as big as in 3000K. Interesting.

I'm using 2014 Vero 29 with 1.2-1.4A (for efficiency and for the super cheap, decent efficiency $10 driver). The 2.1A driver has been a bit more pricey, but there is $30 water proof driver on eBay now (around 90% AC-DC conversion efficiency, which is decent). I recently got a 2.1A driver, but I haven't had time to test it out. If 1.4A is OK, 1x MeanWell HLG-185H-C1400, a fancy high-efficiency driver, can drive 3x Vero 29 in the serial connection (David doesn't recommend this type of driver with high voltage, though). About 150W in this configuration, I believe. Actually with this driver, 4x Vero 18 (a smaller version of Vero 29) is a more cost effective combination (about $14 ea). I think this Vero 18 set up might be pretty good with the heatsink you have.

Most CPU heatsinks (with fans) can handle 1x Vero 29 (80W). Noctua water proof fan would be nice on the heatsink, but it is 120mm or 140mm, so it isn't so great for normal CPU heatsink. It's an Austrian brand, so it may be cheaper there.

Well, if you put the radiator for the water cooling outside of the greenhouse, the heat would be lost, but if you keep it inside, it is same as air cooling, right? I have thought of water-cooling, too, to keep my grow tent cooler. But it was much more expensive. Maybe you can use motorcycle radiators from junk yard?

Electric heat isn't your primary heat, is it? It seems like a really expensive option (at least in the US, where fossil fuels are pretty cheap). So it seems to be better for the light to focus on the light part (high efficiency) and not the heat part, doesn't it?