-
ICMag with help from Landrace Warden and The Vault is running a NEW contest in November! You can check it here. Prizes are seeds & forum premium access. Come join in!
Investing in new lights LED or HID?
- Thread starter repoocbd
- Start date
S
secondtry
Hey SSPL,
I wouldn't say QY is an educed guess, it's based upon the work of McCree, but I would say it's not very accurate for HID or cannabis. But the PAS (which effects QY of lamp) while not specific to cannabis is close to what the cannabis PAS will be. Most higher plants have similar PAS, that is why the work of K.McCree and K.Inadna is used for many types of plants.
That's what I mean by KNNA's data is not really that valid, it's based upon watts, lumens, etc. We need to use a quantum sensor; there is no accurate way to convert from Lux (lumen/m^2) to PPFD, and there is no accurate conversion from watt/m^2 (what Sunmaster incorrectly calls PAR watts) to PPFD. Thus KNNA can't really suggest 300-400 PPFD because he doesn't know the level of PPFD...
I too use a whole canopy height of about 18-24" generally.
And like I wrote on page 12, if a grower chooses a lamp with a SPD rich in green the lower section of the whole canopy will be better irradiated then if using a regular HPS with the same canopy PPFD of 1,500.
To me not reaching peak Pn is a waste of energy...if I am going to use energy to grow cannabis I want to grow the most efficient cannabis I can, and that means high Pn.
We want to offer irradiance just below the saturation point over the whole day, otherwise IMO we waste energy by only offering lower Pn. (ideally we would start the day with lower irradiance but that requires some kind of dimmer)
I can see why that would happen. PAS needs the whole PAR range, with (proportionally) more red than blue and green. Mostly because there are less red photons per second than blue or green, and that red drives Pn better than blue and green under monochromatic lighting. Most studies with LED arrays I have read found a mix of blue, green and red to be a good mix for increasing Pn, yield, growth, etc (of lettuce). One study found when 25% of PPFD comes from green light (500-600 nm) plant growth and Pn increases (see my next post).
All the best
P.S. One thing I wish KNNA would offer is references to his info on why 300-400 PPFD is ideal, it seems like those figures he pulled out of thin air (tho I could be wrong).
I wouldn't say QY is an educed guess, it's based upon the work of McCree, but I would say it's not very accurate for HID or cannabis. But the PAS (which effects QY of lamp) while not specific to cannabis is close to what the cannabis PAS will be. Most higher plants have similar PAS, that is why the work of K.McCree and K.Inadna is used for many types of plants.
Great, good luck. My problem is when using LED arrays on a large scale (that is relative I know, I am referring to greater than about 4'x4') one would have to spend a lot of money considering the size of the footprint of LED arrays.
If you can make that happen I would be very surprised and impressed (assuming the same canopy area and that all other growing environ and methods are the same).
What is "ue/w"? I assume it's a measure of efficiency of the LED for converting watt to lux?
That's what I mean by KNNA's data is not really that valid, it's based upon watts, lumens, etc. We need to use a quantum sensor; there is no accurate way to convert from Lux (lumen/m^2) to PPFD, and there is no accurate conversion from watt/m^2 (what Sunmaster incorrectly calls PAR watts) to PPFD. Thus KNNA can't really suggest 300-400 PPFD because he doesn't know the level of PPFD...
I too use a whole canopy height of about 18-24" generally.
Two things about that:
1) distance from canopy greatly effects incident irradiance (at the leaf) which will be different than what you have calculated at the diode. It seems you calculated the energy emitted at the diode?
2) if I correctly understand what you mean there is no accurate conversion from uE/W to uE/m^2/sec (assuming you mean PPFD; as in "uE" would be "umol"). Thus KNNA can't state 300-400 PPFD becuase he has no accurate method of finding PPFD from the data he used.
2) if I correctly understand what you mean there is no accurate conversion from uE/W to uE/m^2/sec (assuming you mean PPFD; as in "uE" would be "umol"). Thus KNNA can't state 300-400 PPFD becuase he has no accurate method of finding PPFD from the data he used.
Thanks for that, but I did find some misconception of KNNA which I think account for his suggestion of 300-400 PPFD. I also noticed he doesn't have any studies showing the reasoning for his claims about "disminishing returns" [sic]. Below I quoted from his post you provided (from C-W) to show why his ideas are wrong:
The goal I have laid out of 1,300-1,500 PPFD is below the light saturation point of cannabis, over 1,500 PPFD is where light saturation starts. Thus if we keep the PPFD as I suggest there is no reduction in what KNNA calls (incorrectly) "photosynthetic efficiency", there is in fact an increase! (vs. lower irradince like 300-400 PPFD).
And like I wrote on page 12, if a grower chooses a lamp with a SPD rich in green the lower section of the whole canopy will be better irradiated then if using a regular HPS with the same canopy PPFD of 1,500.
To me not reaching peak Pn is a waste of energy...if I am going to use energy to grow cannabis I want to grow the most efficient cannabis I can, and that means high Pn.
Yes!
We want to offer irradiance just below the saturation point over the whole day, otherwise IMO we waste energy by only offering lower Pn. (ideally we would start the day with lower irradiance but that requires some kind of dimmer)I disagree. To increase Pn offer irradiance of 1,300-1,500 PPFD, use a light mover, a good reflector (re: PL) and LST or prune plants. Offering lower PPFD than what I have laid out will lower Pn...
Yea I read his most recent post at GardensCure in this spreadsheet thread and he seems to be claiming (like I am) SPD (vis a' vis McCree) is less important then PPFD (irradiance).
Your very welcome. I am very impressed and happy to be able to have a polite and constructive conversation with you. I find most people who like LEDs will not even consider what I write, while you may not agree with me I really like your style.
I can see why that would happen. PAS needs the whole PAR range, with (proportionally) more red than blue and green. Mostly because there are less red photons per second than blue or green, and that red drives Pn better than blue and green under monochromatic lighting. Most studies with LED arrays I have read found a mix of blue, green and red to be a good mix for increasing Pn, yield, growth, etc (of lettuce). One study found when 25% of PPFD comes from green light (500-600 nm) plant growth and Pn increases (see my next post).
All the best
P.S. One thing I wish KNNA would offer is references to his info on why 300-400 PPFD is ideal, it seems like those figures he pulled out of thin air (tho I could be wrong).
Duh, LEDs are going to be more expensive for initial buy in for years, maybe forever until they start phasing out HID bulb factories.
No one argues that, because if you're looking long term it doesn't matter at all. Todays market is literally a non factor.
HID is a very developed technology. It probably wont get much better in terms of efficiency and it'll never be as versatile to package for different growing methods and less single point heat.
And while changes in HID lighting spectrum will probably happen, MH, HPS, CMH amplitude vs frequency plots don't really look that close to chloryphyll peaks or RQE plots at all anyway.
LEDs will get better, maybe for decades. And they get cheaper all the time. Industrial and civil lighting is converting over to LED, if not for efficiency for reliability and spectrum.
Obviously if you want bang for your buck right now you go HID. Trying to compare absolute light output over a given range, which happens to be pretty much all of visible range, of HID and one LED solution as some sort of scientific evidence is ridiculous.
If you know the spectrum that works, and obviously you don't if you have to go by a generic industry standard that makes manufacturing of lights in a consistent manner possible, then you just build that spectrum and run whatever amplitude makes the bio-optical people happy.
Doing absolute power levels in an bandpass with the wrong spectrum LEDs is putting the cart before the horse, its proving what most people already know from experimental experience, and really doesn't help much.
I can't stress enough how silly these absolute power arguments are. These are LEDs. You need more optical power you just put more. They'll get cheaper and more efficient, and it'll probably happen fast.
So you mean all those stories about ninjas learning to jump high by hopping over a growing cannabis plant every day is a lie?!!?
Sucks. =(
uE per watt would have to be stated as absolute amount of light produced or as measured in a given set of circumstances, like uE in a given area at a given distance over a given amount of time.
uE per watt (uE/W) as I understand it doesn't make much sense, as a watt describes the amount of power dissipated an a given moment, and a mole is an actual physical amount of something, like a coulomb, so one could assume uE/W means uE per watt per period of time, most likely a second.
In any case, you can turn absolute power output into a power output in a given area at a given distance by knowing the focus pattern of the radiation. Since focus pattern of an LED is pretty much whatever you need it to be for an application, absolute power output is probably a better meter stick as the derived or measured uE/m^2/s can be manipulated by changing distance, and LEDs and HID wont be run at the same distance. Reflectors and lenses change uE/m^2/s to our advantage, but the light output of the emitter device, being a diode or an HID tube, should not change given the same power input (drift from changes caused by heat aside) is best measured in absolute values, not in a given area.
Also uE/m^2/s doesn't sound like a complete set of parameters anyway, you NEED a distance to make that number useful. The device that measures this would be no different than a microphone. If you dont know how to set the mic up to take measurements, and don't state your test parameters and environments, your speaker data will be misleading (and you'd just be another shady Pro Audio company, but whatever i digress...)
uE/W(/s) is a measure of absolute light output and efficiency, assuming they mean watts through the diode, and absolute values of uE/m^2/s can be derived from it given a power level, radiation pattern and distance.
This stuff is trivial if you know about EM or acoustic wave propagation.
Also, even spectrum equalized and bandpassed measurements of light can be converted, you just have to do the math to get the absolute values at every frequency in your test domain and compare the summed values.
It's called calculus. I hear they are getting pretty good at it -- it might find its way into science and industry pretty soon.
It's called calculus. I hear they are getting pretty good at it -- it might find its way into science and industry pretty soon.
S
secondtry
@ All
I have placed the member "renz" on ignore so I can't see what she/he is writing. I wanted to mention that people should not pay attention to renz because it's obvious she/he has no clue about photobiology and plant photosystems, nor how to quantitate incident light (irradiance) used by plants. In renz's previous posts she/he made lots of claims about radiance power measurement, not irradiance photon measurements (which is what we need to use).
@ renz:
Please go away becuase I see you are still posting each time I post which to me says you are trolling. I think that violates the TOS of ICmag. If you keep trolling me I will ask you get a week ban from ICmag. STOP POSTING BULLSHIT, IT'S OBVIOUS YOU DON'T KNOW WHAT THE HELL YOU ARE WRITING ABOUT!
Renz, on a side note you should notice how SSPL acts even tho he likes LEDs just as much as you do...
I have placed the member "renz" on ignore so I can't see what she/he is writing. I wanted to mention that people should not pay attention to renz because it's obvious she/he has no clue about photobiology and plant photosystems, nor how to quantitate incident light (irradiance) used by plants. In renz's previous posts she/he made lots of claims about radiance power measurement, not irradiance photon measurements (which is what we need to use).
@ renz:
Please go away becuase I see you are still posting each time I post which to me says you are trolling. I think that violates the TOS of ICmag. If you keep trolling me I will ask you get a week ban from ICmag. STOP POSTING BULLSHIT, IT'S OBVIOUS YOU DON'T KNOW WHAT THE HELL YOU ARE WRITING ABOUT!
Renz, on a side note you should notice how SSPL acts even tho he likes LEDs just as much as you do...
You post like every other post -- by your logic everyone is trolling you. You obviously think everyone who disagrees with you is trolling you, which is coincidentally funny as I'm agreeing with you on HID being the current best choice for anything other than research and curiosity.
Perhaps you should start your own thread instead of jacking this one, which was just about helping people pick LED vs HID, not arguing about academics at a level which helps no one interested get a practical answer in a reasonable amount of time. Results from experimentation with different methods does, even if not to a degree which most of us would like at this point.
Actually, he seems to like them more for this application -- I won't be convinced until I see an ocean of repeatable results from controlled side by side experiments. I plan on contributing to the effort.
But I'm into LEDs for more than just plants. Electronics is my life, I can't really go far before I'm stepping on one of the little guys...
S
secondtry
@ ALL,
I see renz has posted twice since my last post, just like I thought she/he is trolling me. I am now going to ask she/he gets a week ban from ICmag considering she/he joined JUST to post (i.e. cause trouble) in this thread and troll me. Even worse is renz could actually convince less well informed people of her/his idiocracy [sic].
I see renz has posted twice since my last post, just like I thought she/he is trolling me. I am now going to ask she/he gets a week ban from ICmag considering she/he joined JUST to post (i.e. cause trouble) in this thread and troll me. Even worse is renz could actually convince less well informed people of her/his idiocracy [sic].
Yeah better not let me convince people to make the same decision you're telling them to make...
Getting someone banned for agreeing with you. Classic.
lol don't sweat it Renz, he's just making a drama queen show of it. Eventually he will calm down and grace you with his acknowledgement. He's got me on super triple irrevocable lifetime and beyond extra thorough but peeking and pretending not to ignore. But I had to really "troll" him - caught him disparaging a soil professional who does not frequent this board, and I don't agree with attacking people who can't offer a defense, politicians excepted since they constantly offer a defense.
don't take it badly, he's just a bit weird. I love 'em anyway. He's like a brilliant and irrational teddy bear.
don't take it badly, he's just a bit weird. I love 'em anyway. He's like a brilliant and irrational teddy bear.
SupraSPL
Member
True the in initial cost would be significant. If we are correct and LED can outyield HID by 40%, when you break it down in terms of electricity cost and Hortilux bulb replacements, LED is significantly cheaper. Even more so when total system efficiency is considered (cool tube fans + AC)
Sorry I should be more specific. In this case ue/w is based on the dissipation power using these exact bins at specific current levels and with sufficient cooling (125-150 cm2/watt). This explains why he can recommend 28-33 watts per sq ft. It only applies to this given system. If a different lamp were used, many factors would affect the required dissipation power (bin used, SPD ratio, current level, lensing, cooling). We use the 28-33 watts per sq foot range to help select heatsink. The taller and/or denser the heatsink fins, the more concentrated the irradiance will be. It is best to spread it as evenly as is possible. In the case of vegging lamps, shorter heatsink fins allow a wider spread and less irradiance density (if desired).
I understand your concern here, this really gets to the heart of our inquiry. From what I understand he has derived this theory from studies of other plants. He has been experimentally growing with LEDs since at least 2006 along with the Spanish team conducting experiments. That experimentation may support the theory.
This grow netted approx 1.7 gr/watt yield (see page 41 of thread) in soil with a crop of White Russian. (428 grams/~240watts/66 days). The grower, Mr. X has 6800 posts on CC.net and is regarded by some as the best soil grower on CC.net, so is not a fly by night type.
SupraSPL
Member
RE ue, Avenger and Secondtry:
Sorry fellas this is a technical issue that I lack full understanding to address properly, but I will try to explain my thoughts.
In an effort to compare output, KNNA has often quoted 400w HPS as emitting 650ue. My 142 watt (dissipation) LED emits approx 151ue. Unless used vertically, the HID will suffer significant reflector losses as well as loses through cooltubes or glassed hoods. The LED is designed for a specific footprint and a specific height above the canopy. The emitters are evenly spread across the heatsinks which cover almost the same area as the canopy it is illuminating.
KNNA wrote in a post from 2006 (post #19 on thread):
"...we must use photons number data to compare. For the minimun standard, a 400w HPS per sq meter, it equals to about 650 micro mols of photon (uE) (per sq meter), or 1000 uE (600w HPS)=1 mili mol of photons.
The usual growing setup is a HID lamp in zenital position using reflector, wich lose about 20-25% of the bulb's light emited (because of this, vertical cooltubes without reflectors achieve better gr/w results). Leds are packed often in a directional way, so only a 75-80% of photons emited are neccesary.
660nm leds have about 25% more photosyntetic action than the average HPS spectrum, and from 15 to 25% better leaf absortion. So using this wavelenght, the best one can expect to obtain same yield is using half the photons than a HPS emit. A minimun of 325 uE, then.
But, contrary as many people thinks, leds are actually less efficients than high watage HPS converting energy into light. Including ballast loses, a 600w HPS convert 35-38% of its energy input into PAR (400-700nm) light. Currently, best leds in the market are below 35%, and very often the mass leds are below 20%. (Supraspl: Keep in mind this post is from 2006)
A way to increase led efficiency are underdrive them, but this lead to raise the initial cost largely.
Supposing we have a 660nm led with a efficiency of 30% (very good), and 5,35uE per watt emited, to achieve a irradiance of 325uE (the target to get similar results than a reflectored 400w HPS), we need 60,74 watts emited, so 202watts consumed. Well, it is good, half the watts required, but is it profitable pay a lot for it?"
Sorry fellas this is a technical issue that I lack full understanding to address properly, but I will try to explain my thoughts.
In an effort to compare output, KNNA has often quoted 400w HPS as emitting 650ue. My 142 watt (dissipation) LED emits approx 151ue. Unless used vertically, the HID will suffer significant reflector losses as well as loses through cooltubes or glassed hoods. The LED is designed for a specific footprint and a specific height above the canopy. The emitters are evenly spread across the heatsinks which cover almost the same area as the canopy it is illuminating.
KNNA wrote in a post from 2006 (post #19 on thread):
"...we must use photons number data to compare. For the minimun standard, a 400w HPS per sq meter, it equals to about 650 micro mols of photon (uE) (per sq meter), or 1000 uE (600w HPS)=1 mili mol of photons.
The usual growing setup is a HID lamp in zenital position using reflector, wich lose about 20-25% of the bulb's light emited (because of this, vertical cooltubes without reflectors achieve better gr/w results). Leds are packed often in a directional way, so only a 75-80% of photons emited are neccesary.
660nm leds have about 25% more photosyntetic action than the average HPS spectrum, and from 15 to 25% better leaf absortion. So using this wavelenght, the best one can expect to obtain same yield is using half the photons than a HPS emit. A minimun of 325 uE, then.
But, contrary as many people thinks, leds are actually less efficients than high watage HPS converting energy into light. Including ballast loses, a 600w HPS convert 35-38% of its energy input into PAR (400-700nm) light. Currently, best leds in the market are below 35%, and very often the mass leds are below 20%. (Supraspl: Keep in mind this post is from 2006)
A way to increase led efficiency are underdrive them, but this lead to raise the initial cost largely.
Supposing we have a 660nm led with a efficiency of 30% (very good), and 5,35uE per watt emited, to achieve a irradiance of 325uE (the target to get similar results than a reflectored 400w HPS), we need 60,74 watts emited, so 202watts consumed. Well, it is good, half the watts required, but is it profitable pay a lot for it?"
S
secondtry
Hey SSPL,
Nice to see you again
I also think that could very well be his method because he's not the type of person to just pull a number out of thin-air (I was joking, sorry if it didn't seem that way). The figures of 300-400 PPFD are for crops like tomato (tho they can use higher PPFD) and other veggies, etc. There is a list on LEDGirls website, see the PPFD page.
This is my main point, I referenced studies using cannabis for my data. I found four different studies, each independent of each other using different testing variables, goals, variates and species of cannabis, etc, and each one found about 1,500 PPFD is ideal for cannabis to reach peak rate of photosynthesis (3 found about 1,300-1,500 PPFD and one found about 1,400-1,600 PPFD).
I have to head out now, I will respond to the rest of your messages soon.
All the best.
Nice to see you again
I also think that could very well be his method because he's not the type of person to just pull a number out of thin-air (I was joking, sorry if it didn't seem that way). The figures of 300-400 PPFD are for crops like tomato (tho they can use higher PPFD) and other veggies, etc. There is a list on LEDGirls website, see the PPFD page.
This is my main point, I referenced studies using cannabis for my data. I found four different studies, each independent of each other using different testing variables, goals, variates and species of cannabis, etc, and each one found about 1,500 PPFD is ideal for cannabis to reach peak rate of photosynthesis (3 found about 1,300-1,500 PPFD and one found about 1,400-1,600 PPFD).
I have to head out now, I will respond to the rest of your messages soon.
All the best.
Don't under estimate yourself Supra. you seem to understand it pretty well.
The data you have for µE is as you said above "output" of the lamp/led. This is a radience measurement or in some cases "calculation". It is the photosynthetic photon flux, or absolute total photons provided by the lamp in the PAR range.
What Secondtry has been preaching, 1500 uE/m^2/sec (PPFD) is not output of the lamp, but rather the light recieved at a certain distance from the lamp. This is an irradience measurement. It is the photosynthetic photon flux density, or the number of the photons that is recieved by the plant at a given distance from the lamp per unit time(second).
This is an approx average PPF(absolute total photons) for quality 400w horticultural HPS lamps.
He came to this minimum because that is a commonly quoted minimum standard for indoor gardens.
Basically if a newbie asks how much light he needs for a closet that is one meter by one meter, it is often said you need a minimum off 400 watts HPS for a garden one meter square.
This is a bit of an old theory. I am sure KNNA knows that you can not grow anything of quality under monochromatic 660nm light.
And this is the crux of the LED claims. That since you can use certain LEDs with narrrow bandwidths to tune the spectrum, that you can make a light that is heavy in the bandwidths that are more effective/efficient for photosynthesis, thus get by with lower irradeince which equates to lower wattage lamps.
And that may be, probably will be true. To what extent or percentage the tuned spectrum can reduce the required PPFD is more than a guessing game, and all the claims by LED light manufacturers so far have been way too optimistic and not based in fact.
But even with the perfect spectrum, current LEDs can not produce this irradience throughout a normal plant canopy unless they are used on top and as side lighting at very close proximity. And the use of side lighting is not conducive to large densely planted gardens. It works well with smaller gardens.
i belive LEDs will soon take over the closet/cabinet garden light market as well as the supplemental garden/greenhouse lighting market.
They simply must come up with an LED with more radiant flux/PPF before thay can take the place of the large wattage HID in the garden lit completly by artificial light.
S
secondtry
Hey Avenger,
I was going to respond to SSPL's other message but your answer I think is better than what I could have written. Thanks.
I was going to respond to SSPL's other message but your answer I think is better than what I could have written. Thanks.
What happens if you light a plant from underneath? Not just from underneath but at a lower ratio of amplitude so the bulk of the growth still projects upwards?
That's the only 'side' lighting i can think of that could install into a large dense grow. But if they just start to grow into the lights and growth medium, that maybe doesn't work. A setup in trays with large rockwools (6"+) would probably work because of the offset from where you could place the LEDs. For example, moving the lights down as the plant grows, like you move them up for the top canopy.
LEDs aren't hard to package as water resistant and even with high ppm solutions directly shorting a string with water may have little or no effect do to the relatively low impedance and low voltages involved. I have to see how it corresponds to ohms to really know. ( I know the info is available i just haven't got to the stage of design where I'm pouring nut solution all over my arrays just to see what happens).
But yeah I think side lighting and spectrum tuning (like 660nm red assist for HPS) is definitely a good application for LED arrays. I designed the hardware for my extruded channel arrays so you could hang them at basically any useful angle and from each other or HID reflectors. Like you could chain them together to make a wall, but they could easily slide underneath plants to light them from the bottom up:
(Note: I did that design, don't steal. Unless you are a normal person and not a company selling LED lighting... then yeah, steal design, make 1000 if you want, take pics.)
You can easily use this format to create walls, use parallel sets to create 'panels' for canopies like most people are using, or just slide them in between plants if bottom lighting actually works.
(i posted links to this image before in this thread but not with the forum image display tags, in a previous post there is also links to 75% complete but 100% optically functional prototypes)
If you can increase the output of the bottom sections of the plant, that's more untapped yield that's harder to accomplish with HID, even with light movers.
It would be a goofy setup, though (!). Maybe place the arrays on 1" blocks so you could adjust height from bottom canopy. Again this is a situation where LEDs might be better assisting HID as opposed to directly competing. If secondtry is all about better quality buds, I think even he/she could appreciate getting more light to places that otherwise could not be fed the proper energy levels.
If your ballast wont be in the closet, consider going magnetic instead of digital. A 400w HPS ballast kit is only $50. Definitely look into 400w CMH too.
Though if you can, a MUCH better upgrade would be to somehow increase your space and fit a 600W in there. After this harvest I'm going with a 600w and a new fan.
Though if you can, a MUCH better upgrade would be to somehow increase your space and fit a 600W in there. After this harvest I'm going with a 600w and a new fan.
