LED and BUD QUALITY

[Rocket Soul]

You can read significance in that graph. The standard error is at 95 % confidence interval. If it overlaps there is no statistically significant difference (at 95 % CI).

The point im making is that the significance is not in the flatness of the line, but yes, its calculated and included. The low uv condition that gave 15% effect/increase in cannbinoids is at p:0.50. So basicly you could conclude that the 15% increase has a 50% chance of being random noise in the numbers, 50% chance of being dependant on the uv treatment. Or at least thats how i understood things when it was taught to me a long time ago. Point is looking at the line and spotting some blips and discounting it as insignificant is not the right way of understanding this. You can very well have a very low effect size of just a couple of percent but it can still be statisically significant if youre variance is low and you have enough measurements.
Point is aswell that its a bit bad faith to say nothing happened closed case, when you did find effect.

The buds in a bucket: i didnt think about that and tbh i dont think they mentioned exactly how they went about this. The logical approach would be cutting down the grow and test the most apical bud of every plant. But i dont think bugbee expands on this. But im glad someone else sees this study for what it is, not an end all discussion for ganja growers

I might be imagining this but i do think bugbee ran tests on high thc cultivars aswell but i might be mixing up papers. Point is if your questioning setting up uv for ganja then this paper is not going to help you figure out anything.

 

[driver77]

For more stretch, Far red on a separate channel.

I agree. I added both ir and uva to my main and noticed more internodal spacing during stretch running both ir and uva on the same schedule. If I increase the uva time over the ir it isn't as noticeable but still more than I got before....I assume it was because of the uva counter acting the ir. Studies be damned my buds are stronger, more dense, and tastier since the addition of ir and uva. The improvement was so noticeable I added ir and uva to all my tents.

 

[Orange's Greenhouse]

The point im making is that the significance is not in the flatness of the line, but yes, its calculated and included. The low uv condition that gave 15% effect/increase in cannbinoids is at p:0.50. So basicly you could conclude that the 15% increase has a 50% chance of being random noise in the numbers, 50% chance of being dependant on the uv treatment. Or at least thats how i understood things when it was taught to me a long time ago. Point is looking at the line and spotting some blips and discounting it as insignificant is not the right way of understanding this. You can very well have a very low effect size of just a couple of percent but it can still be statisically significant if youre variance is low and you have enough measurements.
Point is aswell that its a bit bad faith to say nothing happened closed case, when you did find effect.

The buds in a bucket: i didnt think about that and tbh i dont think they mentioned exactly how they went about this. The logical approach would be cutting down the grow and test the most apical bud of every plant. But i dont think bugbee expands on this. But im glad someone else sees this study for what it is, not an end all discussion for ganja growers

I might be imagining this but i do think bugbee ran tests on high thc cultivars aswell but i might be mixing up papers. Point is if your questioning setting up uv for ganja then this paper is not going to help you figure out anything.

p = 0.5 means that is has a 50 % chance of being random, that is correct. And no, they did not find a scientifically relevant difference *in this experiment*. That is what they are saying and that is correct. You have to keep in mind that they are not a business trying to improve production/profit but a university writing scientific papers.

The need for more measurements is also acknowledged: "Further research is needed".

The bucket story in an analogy to illustrate what statistical significance means. Utah state university has no license to grow high THC weed.

 

[Ca++]

Ca, this is a bit silly, youre kinda showing off that you dont know how to read or understand one of these studies. This is not a graph indicating significance, its indicating measured effect. Significance is not pictured its calculated, its that little P value. And nowhere are they measuring how much buds was grown. You got it wrong, g/m goes with the graph on the right side and refers to canabinoids /m, the left measures concentration in %.
What graph does show is an increase in cannabinoidconcentration of 15% while applying the lowest UV dose, and then the concentration goes down as you apply more.

As expected, uv both stimulates and break down cannabinoids. Especially if you target uvb while giving no UVA.
What does bugbee do then? Try to make a linear correlation out of it when the data indicates that its actually a curve rather than a line.

Is it a curve though. If we plot between the data points, I don't see one.

Measured effect, significance, I don't see what you are trying to separate. The graphs are the data. It is what it is.

The right hand ones are yield, and the left concentration. You can't have a lowering yield, with increasing concentration, unless there is less bud. This is significant data from the graphs. The amount of cbd harvested from a meter, only went down with UV treatments. The THC went up at one data point only. I'm not seeing the full study, just the highlights, so there is less to get lost in, but also less detail than I would like. As you fight in the UV corner though, I guess the 15% gain is in concentration, not yield, as concentration is the harder hitting headline figure. The yield blip looks half as high. If a 7.5% yield increase, accompanies a 15% higher concentration, then we can see the plant weight drop. The 3 are inseparable. I'm not reading the data wrong, I'm talking about the significance, as you put it.

If I could be guaranteed a 15% increase in my weeds THC, if I took that 6% drop in bud weight, I would be interested. It would mean adding to my lighting bill though, and buying a UV meter to actually get an increase, not a decrease. A decrease being the more likely outcome of a hit&hope approach. Which I'm no stranger to. However, all plants are different, and we are looking at a study of hemp most likely. So I have to take the downward trend marked up the graphs, as being the average result. I'm not fishing for what I want to hear. I have looked at many UV studies, and these are common trends. Suppressed growth is the common factor in UV studies, and even blue light studies.


I'm going to set a silly example here, just for those not quite following.
If I half my bud yield, but still have the same resin yield, than my bud has become twice as strong. Because I have as much resin, now on half as much bud. So twice as much resin on each bud. My concentration has gone up 100% but my resin yield has remained the same, and my bud halved.
These numbers can't be separated. So if the concentration goes up 15% and the resin 7.5% then the kid at the back of the class, can see yield dropped about 7.5%.
This is a net gain, but it's only seen in the THC element of the study, at one data point.
Is that worth chasing, when there are 8 data points. Offering an overall downward trend.

ca++ = silly lol

 

[CannaT]

Reality check: the sun is far more powerful then any HID or LED light yet indoor weed is better.

Its all about stress the key is to ensure that the stress is carefully managed and appropriate for the specific strain and growing conditions. Overdoing it can have the opposite effect, causing harm and reducing overall plant health and yield.

 

[kro-magnon]

I agree. I added both ir and uva to my main and noticed more internodal spacing during stretch running both ir and uva on the same schedule. If I increase the uva time over the ir it isn't as noticeable but still more than I got before....I assume it was because of the uva counter acting the ir. Studies be damned my buds are stronger, more dense, and tastier since the addition of ir and uva. The improvement was so noticeable I added ir and uva to all my tents.

What type of bars did you get to add the IR and UV ?

 

[CannaT]

On avarage with no polution and clear sky sun for example (Sahara desert) hits the earth at noon on around 3-5% UV 43% visible light spectrum and around 52-55% Nir and IR.
With around 700W /m2.
It will be perfect led board spectrum.

HPS is like UV around 1%
Visible light around 85%
Blue 10%
Green 40%
Red 60%
And IR around 15%

CMH is UV 3%
Visible 90%
Blue 25%
Green 45%
Red 25%
IR 7%

LEDs are like
100% visible spectrum
Blue 30%
Green 20%
Red 40%
Far red 10%
UV 0% IR 0%

 

[CannaT]

What type of bars did you get to add the IR and UV ?

You can use tungsten incandescent lamps are thermal radiators that emit a continuous spectrum of light extending from about 300 nanometers, in the ultraviolet region, to about 1400 nanometers, in the near infrared region.

 

[driver77]

What type of bars did you get to add the IR and UV ?

I got 2 sets of the old style uv/ir bars from MH and 1 set of the SF bars for a 3x3. I would recommend the new style MH bars though.....they are all independently timed where the old bars can not easily.

 

[greyfader]

just thought i would throw this in here. it shows how closely a tungsten incandescent follows the action spectrum above about 550nm

 

[Orange's Greenhouse]

On avarage with no polution and clear sky sun for example (Sahara desert) hits the earth at noon on around 3-5% UV 43% visible light spectrum and around 52-55% Nir and IR.
With around 700W /m2.
It will be perfect led board spectrum.

HPS is like UV around 1%
Visible light around 85%
Blue 10%
Green 40%
Red 60%
And IR around 15%

CMH is UV 3%
Visible 90%
Blue 25%
Green 45%
Red 25%
IR 7%

LEDs are like
100% visible spectrum
Blue 30%
Green 20%
Red 40%
Far red 10%
UV 0% IR 0%

LEDs are whatever you want to have. You can build any spectrum you can imagine and have as diffuse or directional light as you design.

I don't know where you get your numbers from but assigning any single spectrum to LED is wrong.

 

[CannaT]

LEDs are whatever you want to have. You can build any spectrum you can imagine and have as diffuse or directional light as you design.

I don't know where you get your numbers from but assigning any single spectrum to LED is wrong.

I know but most of led boards for growing of various producers are like that...arent they ?
400-700 nm and thats all.
If led board you can buy are perfect and so diverse why people like rocket soul will diy boards by they needs...most store buyed led boards have shity spectrum.

One of the best in EU and most popular atm...you see UV IR ??
Other one also the best

Uv IR do you see it ???

Led is about electric eficiency....by some experts and scientist plant dont need UV IR for grow....

 

[Orange's Greenhouse]

I know but most of led boards for growing of various producers are like that...arent they ?
400-700 nm and thats all.
If led board you can buy are perfect and so diverse why people like rocket soul will diy boards by they needs...most store buyed led boards have shity spectrum.
View attachment 19043544
One of the best in EU and most popular atm...you see UV IR ??
Other one also the best
View attachment 19043545
Uv IR do you see it ???

Led is about electric eficiency....by some experts and scientist plant dont need UV IR for grow....

Last time I checked this thread is about using LED to get the best out of them. And not buying the most popular LEDs and convincing yourself that they suck. What is they point of your post? If you want to discuss how bad LEDs are make your own thread.

Rocket Soul made his own fixture, by using Grow Light Australia's strips that have a true full spectrum, including UV, IR and cyan.

btw, that first spectrum you posted has far-red which is in the IR range. Atleast according to all the textbooks.

 

[CannaT]

Last time I checked this thread is about using LED to get the best out of them. And not buying the most popular LEDs and convincing yourself that they suck. What is they point of your post? If you want to discuss how bad LEDs are make your own thread.

Rocket Soul made his own fixture, by using Grow Light Australia's strips that have a true full spectrum, including UV, IR and cyan.

btw, that first spectrum you posted has far-red which is in the IR range. Atleast according to all the textbooks.

constant questioning leads to progress. The forum is for discussions, not for your marketing and worship of the golden calf...if my arguments offend you, it's because you have no answer to them.

 

[Rocket Soul]

Is it a curve though. If we plot between the data points, I don't see one.

Measured effect, significance, I don't see what you are trying to separate. The graphs are the data. It is what it is.

The right hand ones are yield, and the left concentration. You can't have a lowering yield, with increasing concentration, unless there is less bud. This is significant data from the graphs. The amount of cbd harvested from a meter, only went down with UV treatments. The THC went up at one data point only. I'm not seeing the full study, just the highlights, so there is less to get lost in, but also less detail than I would like. As you fight in the UV corner though, I guess the 15% gain is in concentration, not yield, as concentration is the harder hitting headline figure. The yield blip looks half as high. If a 7.5% yield increase, accompanies a 15% higher concentration, then we can see the plant weight drop. The 3 are inseparable. I'm not reading the data wrong, I'm talking about the significance, as you put it.

If I could be guaranteed a 15% increase in my weeds THC, if I took that 6% drop in bud weight, I would be interested. It would mean adding to my lighting bill though, and buying a UV meter to actually get an increase, not a decrease. A decrease being the more likely outcome of a hit&hope approach. Which I'm no stranger to. However, all plants are different, and we are looking at a study of hemp most likely. So I have to take the downward trend marked up the graphs, as being the average result. I'm not fishing for what I want to hear. I have looked at many UV studies, and these are common trends. Suppressed growth is the common factor in UV studies, and even blue light studies.


I'm going to set a silly example here, just for those not quite following.
If I half my bud yield, but still have the same resin yield, than my bud has become twice as strong. Because I have as much resin, now on half as much bud. So twice as much resin on each bud. My concentration has gone up 100% but my resin yield has remained the same, and my bud halved.
These numbers can't be separated. So if the concentration goes up 15% and the resin 7.5% then the kid at the back of the class, can see yield dropped about 7.5%.
This is a net gain, but it's only seen in the THC element of the study, at one data point.
Is that worth chasing, when there are 8 data points. Offering an overall downward trend.

ca++ = silly lol

Sorry dude, im being to hard on you. Its just that weve been over this before with the same arguments. Im glad you at least admit that you dont quite understand what i mean or what the study is saying. When i say a curve rather than a line: since thc and trichomes seem to be a protective mechanism against uv: uv both stimulate the production and also break down thc its fairly rational to expect that there is an ideal amount of uv for maximizing thc. So you would expect the results go from no uv- low thc to some uv and more thc and then too much and lower thc. Like an inverted U. This is actually what we see in the results if we only look at the measured effect.
Measured effect: take a cannabinoid sample for each plant (or even a few) add them all together and divide by number of samples to find an average. Compare your uv sample to the no uv sample. Thats the measured effect, fairly straight forward.

But when ever you measure things you always have some variance and error. If youve measured effect you need to figure out if its due to the different conditions or due to the variance in your samples. You can do this with statistic analysis. You use some math and tests in order to evaluate if the measured effect is based on the two conditions or if its a spurious correlation. Basicly what you do is to try to calculate the chance of the difference being due to uv or just due to "noise" and variance in the measurements. In order to say it was significant you need to be 95% sure that its not spurious.


Bugbee states in his paper that he got up to 13% increase (not 15 as i said before, my apologies) in cannabinoids in the uv treatment group over the control. Its in the first paragraph of the paper, please look it up if you dont believe me.

From looking at the graphs you posted we can determine that this was in the THC concentration; between the no uv group and the lowest uv treatment (first dot on the left in the graph labeled C, and the second dot in the same graph, its were you see the highest difference between left most dot and any other dot in the two graphs on the left side, the ones of concentration).
As we move further to the right in that graph we see lower values for thc concentration. It actually describes (although not in a super clear way, looks like blips as you say) what im saying: thc concentration goes up with a little uv, then down with too much, as an inverted U. Bugbee tries to find a linear correlation in these values, it doesnt work very well since its not linear. That flat line your looking at, it means nothing pretty much, only that bugbee would have done better trying to limit the uv exposure somewhat, if thc goes up at 0.3 then down at 0.6 we can pretty much determine that the sweet spot would be somewhere between those numbers, and measuring 0.9 and 1.2 doesnt really add anything if the concentration behaves as i argue.

It goes up and then down. Still, not much can be made out of this as theres no statistic significance, nowhere do we see a p-value of 0.95. But we can still look at the numbers and make an idea. What we see is
1: the strongest increase cannabinoids was in thc and not cbd, and this happened in lowest UV supplement
2: this was also where we had the strongest statistical significance; 0.5 (other p values where all lower, 0.39, 0.19 and 0.11)
-this means plain and simple that the 13% increase has a 50% chance of being due to uv, 50% chance of being random. This is the closests he came to prove in this experiment. For me this, the way i read it is a clear indication towards "further research needed" rather than "uv has no effect".
Its easy to look at the lines and dots and say nothing happened but you need to look further than just those graphs. But for me i take it as "lets try this out" rather than "no point whatsoever". Maybe im a glass half full kinda guy

Now, i cant say that my uv weed is stronger thc, its just to close and too much placebo. Nor can i realistically do a study myself without measuring equipment; youd need about a dozen samples of each side @50euro each and that is not within my budget. But i can say that every person ive made a blind test with is able to identify that the "no uv bud" as the lowest quality as far as smell and taste.
There are statisical tools for testing this without going the way of thc measuring and i hope to do some soon, i just need to finish a crop with clones of several genetics from the same room with uv and no uv, and then get people to try to smell them. Ideally i would like to do this with 3 conditions:
White no uv.
White and red no uv.
White with horticentric uv/red supplement.

But i should be able to do this with plain white and horticentric in a few months; next harvest is on thursday and we have 5 strains with several phenos of each on each side, one white one horticentric +uv.

I stand by my word when i say ill send you the uv diodes to try, ill even solder them up for you. All you would have to do is to find a way to attach them to your lighting frame or similar and a 48V / 700mA driver. Ill give you the diodes if you pay shipping. Hit me up on pm if youre interested.

Sometimes its hard to gauge you, sometimes i get the feeling youre already decided that it doesnt work and enjoy being a contrarian (no harm meant, im another one ) sometimes i get the feeling you actually would like to find out if it didnt mean to much work or money. Im leaning towards the second "good guy" option, as i say im a "glass half full" kinda guy and i also see you helping a lot peeps here with a very genuine interest to help. If i can spend a few bucks on sorting this out and then never have to argue with you about this again im all for it
Peace

 

[Rider420]

No matter how good your lights are hemp is hemp and won't get you high.

Lighting is just one factor in the quality of cannabis. All knowledge is local, since cannabis enjoyment is subjective and requires different lighting depending on the strain indica vs sativas there is no one right answer.

So enjoy your life and stop worrying about how many angles can dance on the head of a pin.

 

[Rocket Soul]

LEDs are like
100% visible spectrum
Blue 30%
Green 20%
Red 40%
Far red 10%
UV 0% IR 0%

Please tell me where i can buy this led, i actually like those proportions quite a bit. They are quite far away from the market standard; standard white based leds have much more green, about 18-23% blues and nowhere near 10% far red.

Last time I checked this thread is about using LED to get the best out of them. And not buying the most popular LEDs and convincing yourself that they suck. What is they point of your post? If you want to discuss how bad LEDs are make your own thread.

Rocket Soul made his own fixture, by using Grow Light Australia's strips that have a true full spectrum, including UV, IR and cyan.

btw, that first spectrum you posted has far-red which is in the IR range. Atleast according to all the textbooks.

CannaTs example is not even close to standard leds of todays market. Todays leds are all about efficiency which means basicly mean low cri cold whites + 660 reds which in turn means high green content. I tried adding uva and uvb to that type of base spectrum and didnt like the effects. At first i couldnt really get any difference. With some experience and tweaking, along with a longer cure you can feel some different in taste and smell though not as intense or bright as the terps of other lights we got going. Rainbow chip went from bright fruity/minty smell with a subtle chocolate after taste, to a more subtle yet complex lavender and licorice smell and taste. I get something similar but not as pronounced from the GLA strips. My working hypothesis is that its down to the higher green content in these lights. In the end i like both but the lower green content lights have a more general comercial appeal: bright and loud vrs subtle and complex. Not too many people take the green response of plants into account: it creates fiber and stick, kinda opposite what you want if youre after fragrant flower. Thing is that its also seems to create really dense flowers and its really hard to convince people to sacrifice that.

Im going to post this article again cause its really interesting and detailed and it speaks volumes about spectrum influence on ganja. Its also the first time ive seen someone study whats basicly been my working hypothesis for some time of how to design a light that grows great weed. I rec it for anyone who has any interest, its also gives an idea why todays standard spectrum may seem lacking in quality produce.

Frontiers | The role of red and white light in optimizing growth and accumulation of plant specialized metabolites at two light intensities in medical cannabis (Cannabis sativa L.)

The cultivation of medical cannabis (Cannabis sativa L.) is expanding in controlled environments, driven by evolving governmental regulations for healthcare ...

www.frontiersin.org

 

[kro-magnon]

In my LED there is some diodes at 660nm, is it useful to get a bar of IR or just a UV will be enough ?

 

[driver77]

In my LED there is some diodes at 660nm, is it useful to get a bar of IR or just a UV will be enough ?

Both...they work together imho....660 is more far red......ir is further up the scale. Depending on how much 660 you have it can be supplemented also.

 

[chilliwilli]

Hi folks sanlight got some uv tests running the last 2 years. They tried 365nm or 420nm(not uv like they say) @80μmol/m2/s on top of their led @900μmol/m2/s. The seedrun saw a 15min increase of uv duration per week till 2h/d. The clone runs got UV from flowerday 21 on and was running all day.
They used this wavelenghts because they found some decent leds for a possible future product.

But their findings were strain dependent and marginal so their conclusion was they can't say if uv has an effect.

Serious happines and london mint cake were grown.

From seeds they saw
@365nm slightly more thc and less terps on both strains.
@420 SH got slightly more thc and terps but LMC got less thc and no change in totall terps.
They do say that terp composition changes. Some terps got a bump from uv others were lessened.

They also did 3 runs with clones from those strains.
@365 SH +0.6% thc, LMC +0.2% thc
@420 SH +0.7% thc, LMC -0.7% thc
Terp levels were about the same but also change in terp composition.



In the video description there are links to the other 4 videos. Video 2 shows the seedrun preharvest, video 4 a clone run. Video 3 is findings of seedrun.

Their blog about the experiment has maybe better infos for most here since google translate is way better than yt subs.

https://www.sanlight.com/news/einfl...ihr in,Ertrag, Terpene oder Cannabinoiden hat.