Best K (Kelvin) light value for positive (landrace) sativa effects?

[RenaissanceBrah]

I've heard that for growing equatorial landrace sativas indoors, people have recommended high Kelvin lights (10,000K, 6500K, etc) to bring out the most positive effects / high in the genetics.
I'm chasing the most positive effect possible, I don't mind low potency and low yield.

Anyone have experience with varying Kelvin values, and how it changed the effect of the sativa?

(Was going to buy LED lights soon, I saw 6500K Kingbrite lights I was thinking of getting. I would get the whole spectrum of lights, but on a budget, was going to buy two 250W 6500K Kingbrites).

 

[TPFTFW]

I’ve never heard this

would be interested to see what others have to say

 

[troutman]

For the longest time I used 600 Watt MH with a 7,200 K spectrum.

I used to start the Sativas under my Quad T-5 used a pair of 6,500K and 10,000K bulbs.

 

[bleepboop]

If only it was as straightforward as that. Sadly the higher Kelvin (4000+) will drop off quite sharply in the red/deep red area and have minimal far red which are all super helpful for triggering phytochrome response points.

There's an element of truth in that there's more near uv (400-440 nm) or so with higher Kelvin CMH lights for instance. This is true of some LEDs (95+cri, Nichia and Seoul Semiconductor) but most cri 70/80/90 LEDs drop off below 440nm quite sharply.

check some of the YouTube videos by Dr Bruce Bugbee, of Apogee instruments (best light measurement sensors) for the best explanation of up to date science on photosynthesis and light spectrums.

 

[MadMac]

hello,
well interesting topic...
but you found the source of information as well...
all important is found in the videos by Bruce Bugbee...
most other information are guess and not science proof...
use LED panel 3500K + Red & add far red & UV led...
do it like he said in the videos... and you will see the wonder...
M.

 

[RenaissanceBrah]

For the longest time I used 600 Watt MH with a 7,200 K spectrum.

I used to start the Sativas under my Quad T-5 used a pair of 6,500K and 10,000K bulbs.

Thanks Troutman for your response - did you ever see any differences between your own (and perhaps others) grows, growing with a high K light?

I know the node distance will be shorter, but wondering if you saw any difference in effect of smoke when you changed K values.

 

[RenaissanceBrah]

If only it was as straightforward as that. Sadly the higher Kelvin (4000+) will drop off quite sharply in the red/deep red area and have minimal far red which are all super helpful for triggering phytochrome response points.

There's an element of truth in that there's more near uv (400-440 nm) or so with higher Kelvin CMH lights for instance. This is true of some LEDs (95+cri, Nichia and Seoul Semiconductor) but most cri 70/80/90 LEDs drop off below 440nm quite sharply.

check some of the YouTube videos by Dr Bruce Bugbee, of Apogee instruments (best light measurement sensors) for the best explanation of up to date science on photosynthesis and light spectrums.

I watched some of the Bugbee videos, it seems that he mostly focuses on yield however - do you happen to know (or have heard) of the effect of the smoke being different, when using higher K lights? (6500K, 10000K, etc).

 

[RenaissanceBrah]

hello,
well interesting topic...
but you found the source of information as well...
all important is found in the videos by Bruce Bugbee...
most other information are guess and not science proof...
use LED panel 3500K + Red & add far red & UV led...
do it like he said in the videos... and you will see the wonder...
M.

Thanks Mac - I watched some of Bugbee's videos, but it seems his focus is yield and health of plant, I was wondering about a change in effect when using higher K lights. I know there's probably no science behind it, but wondering if there's any anecdotal evidence out there among the IC community

 

[chilliwilli]

I got the impression that nld do better with cmh light. Yield and quality. I only used Osram Powerball HCI-T 150W NDL 942 G12 because they were a very cheap way to try cmh. Most change i saw with the trainwreck cut in austria. Growing only under hps i got that typical foxtail buds. Under cmh hps mix the buds were more roundish with only a little foxtailing at the end. My best 2 grows with 2x150w cmh + 250-400w hps dimmed i got the magical 1g/w. Normal with hps was more 0,5-0,6 g/w and i didn't change my growstyle or improved my other growskills.

 

[bleepboop]

Effect with "UV" on essential oil production and trichromes is well documented. It's mostly come around to near UV/a little UVA being the most useful while not being as detrimental to plant & person health as uvb/lots of uva

The classic points about shorter nodes are valid. Far red/Emerson effect is useful though, shade avoidance behaviours stemming from it means more cell division and bigger leaves aswell as the stretch aspect.

Check the Feb 2021 UV Bruce Bugbee video too as linked.

That's not to say higher Kelvin won't grow but 630nm (bright red), 650-680nm (deep red) and 720-740nm (far red) peaks of light would be very useful as far as actual photosynthesis goes. Without them NLDs would probably develop a bit more squat and possibly bushier too. But you're still missing out on a lot of potential photosynthesis.

There's a good bit of discussion around the development threads elsewhere of the newest "highlight v2" led boards by Grow Lights Australia and LED Teknik. Including some good side by sides of UV supplementation.



The combination of spectrums used for that is fantastically tailored.
​​​​​​
120x LT-3030 Narrow Band 3000K (CRI 95)

120x Nichia 3030 NF2W757V3-F1 “Vitasolis" 3000K (CRI 80)

90x Nichia 3030 NF2W757HT-F1 5000K (CRI 70)

30x LT-3030 PC Blue (1x 405nm die + 2x 420nm dies)

40x LT-3535 Deep Red (660nm)

20x LT-3535 Infrared (730nm)

That near UV PC blue would hopefully do everything required for boosting terps and cannabinoid production without triggering stress responses like proper UV. that particular diode is pretty much unique atm though.

The high Kelvin CMH route might work out cheaper for you. Straight 6500k Samsung 301b boards are what I have for germinating without stretch and they do that fine but for flower I use 5000k CRI 98 Nichia optisolis strips at a 2:1 ratio with led Teknik/cutter electronics strips of broadband deep/far red aka PC red which has a wide peak centred at 680nm. These particular 5000k strips also start at 420nm (purple/near UV) and are pretty flat all the way to 630nm (bright red) rather than the traditional grow light spectrum which starts at 440nm (royal blue/deep blue).

Just one single strip at a certain spectrum doesn't cover EVERYTHING but there's also plenty of people out there growing wonderful plants under 1800k, 2200k, 2700k, 3000k, 3500k, 4000k, 5000k and 6500k strips.

​​​​​​​We are also learning so much more genuine science year on year as the rate of new videos from Bugbee goes to show.

 

[RenaissanceBrah]

Effect with "UV" on essential oil production and trichromes is well documented. It's mostly come around to near UV/a little UVA being the most useful while not being as detrimental to plant & person health as uvb/lots of uva

The classic points about shorter nodes are valid. Far red/Emerson effect is useful though, shade avoidance behaviours stemming from it means more cell division and bigger leaves aswell as the stretch aspect.

Check the Feb 2021 UV Bruce Bugbee video too as linked.

That's not to say higher Kelvin won't grow but 630nm (bright red), 650-680nm (deep red) and 720-740nm (far red) peaks of light would be very useful as far as actual photosynthesis goes. Without them NLDs would probably develop a bit more squat and possibly bushier too. But you're still missing out on a lot of potential photosynthesis.

There's a good bit of discussion around the development threads elsewhere of the newest "highlight v2" led boards by Grow Lights Australia and LED Teknik. Including some good side by sides of UV supplementation.



The combination of spectrums used for that is fantastically tailored.
​​​​​​
120x LT-3030 Narrow Band 3000K (CRI 95)

120x Nichia 3030 NF2W757V3-F1 “Vitasolis" 3000K (CRI 80)

90x Nichia 3030 NF2W757HT-F1 5000K (CRI 70)

30x LT-3030 PC Blue (1x 405nm die + 2x 420nm dies)

40x LT-3535 Deep Red (660nm)

20x LT-3535 Infrared (730nm)

That near UV PC blue would hopefully do everything required for boosting terps and cannabinoid production without triggering stress responses like proper UV. that particular diode is pretty much unique atm though.

The high Kelvin CMH route might work out cheaper for you. Straight 6500k Samsung 301b boards are what I have for germinating without stretch and they do that fine but for flower I use 5000k CRI 98 Nichia optisolis strips at a 2:1 ratio with led Teknik/cutter electronics strips of broadband deep/far red aka PC red which has a wide peak centred at 680nm. These particular 5000k strips also start at 420nm (purple/near UV) and are pretty flat all the way to 630nm (bright red) rather than the traditional grow light spectrum which starts at 440nm (royal blue/deep blue).

Just one single strip at a certain spectrum doesn't cover EVERYTHING but there's also plenty of people out there growing wonderful plants under 1800k, 2200k, 2700k, 3000k, 3500k, 4000k, 5000k and 6500k strips.

We are also learning so much more genuine science year on year as the rate of new videos from Bugbee goes to show.

Thanks for sharing your experience.

Damn, it's a tough subject, since yield and bud density is the most important thing to most people, I'm only focused on purely effect though, hard to find any datapoints centered around that, especially for landrace sativas.

I was thinking of going pure 6500K and experimenting with that, but don't want to drop $400 on lights to find out it's a dud, will probably go with a 4000K and 5000K light probably to be safe.

I wonder if Kingbrite makes a light that has a mix of all the K values, that would be ideal. Still trying to wrap my head around how these light setups work.

I would build my own with the combination of spectrums you mentioned, but time is a constraint. Though maybe it's not as hard as it seems.

 

[bleepboop]

Aye there's some mixed spectrum strips/bars and boards on there these days that are lm301b/h 3000k + 5000k/6500k + 660nm.
good things to look out for are:
Seoul UV Cree xpg3 660
Cree xpe2 730
Osram 660/730 if genuine would probably be decent chips too.

I've very little experience with AB manufacturers. I had a Folux Meijui lm561c qb288 in 4-5yrs ago which developed several very intermittent strings. Could have easily been sugary knf foliars that caused it. I think lm301H have protective coating for that these days. It was a disappointing experience though which led me to try understand things a bit better. Kingbrite seem dependable though and I'm sure you'll be diligent in your research.

 

[RenaissanceBrah]

Aye there's some mixed spectrum strips/bars and boards on there these days that are lm301b/h 3000k + 5000k/6500k + 660nm.
good things to look out for are:
Seoul UV Cree xpg3 660
Cree xpe2 730
Osram 660/730 if genuine would probably be decent chips too.

I've very little experience with AB manufacturers. I had a Folux Meijui lm561c qb288 in 4-5yrs ago which developed several very intermittent strings. Could have easily been sugary knf foliars that caused it. I think lm301H have protective coating for that these days. It was a disappointing experience though which led me to try understand things a bit better. Kingbrite seem dependable though and I'm sure you'll be diligent in your research.

Thanks Bleep - I was on the fence with getting the waterproof coating but I think I will, seems worth the extra $10 or whatever to avoid those type of issues.

Not sure if Kingbrite would be willing to do for example 75% 6500K LEDs and 25% 3500K LEDs, but I asked them about it, I was thinking that might be a good mix for a board.

Think that would be a good, or bad idea? To run that mix for the entirety of the grow?

 

[bleepboop]

6500 + 3000/2700 would be fine. I do not agree that 75% 6500k would be conducive.

Also at the other place you got two genuinely spot on answers re:
the difference between actual daylight 6500k vs 80/90cri led 6500k
&
Better off with normal 3000k/3500k + 450nm and below supplements (aswell as 660/730). Like a mix of 400 uva + 420 near UV purple + 440 royal blue.

Hats off for wading through all the info. I know it doesn't feel tailored to tropical sativas but I think a tailored spectrum can be more important in that regard.

 

[RenaissanceBrah]

6500 + 3000/2700 would be fine. I do not agree that 75% 6500k would be conducive.

Also at the other place you got two genuinely spot on answers re:
the difference between actual daylight 6500k vs 80/90cri led 6500k
&
Better off with normal 3000k/3500k + 450nm and below supplements (aswell as 660/730). Like a mix of 400 uva + 420 near UV purple + 440 royal blue.

Hats off for wading through all the info. I know it doesn't feel tailored to tropical sativas but I think a tailored spectrum can be more important in that regard.

Thanks Bleep - trying to keep with just buying two 240W LEDs, already stretching my budget on that, so wanted to avoid supplemental lighting if possible.

I know outside is best, just too complicated to replicate sunlight it seems.

I'll probably get a 3500K Kingbrite and a 4000K Kingbrite 240W. I've heard people say using a 4000K still leads to good results, hope that's true.

PS - I've attached the 3500K Kingbrite LM301H + CREE XP-G3 test results. Do you think this type of spectrum profile would be good to run the full grow for equitorial landrace sativas?

 

[bleepboop]

Looks good. Xpg3 is the current best 660nm diode out there.

I would be tempted more by the following kingbrite which is in both bar and panel format.

LM301H 3500K+CREE XP-E2(660nm+730nm)+LG UV

 

[RenaissanceBrah]

Looks good. Xpg3 is the current best 660nm diode out there.

I would be tempted more by the following kingbrite which is in both bar and panel format.

LM301H 3500K+CREE XP-E2(660nm+730nm)+LG UV

Thanks Bleep. As full spectrum is more important, looks like that spectrum map on the light you mentioned could be best. Some have mentioned those IR and UV LEDs don't really do anything, but at a few extra bucks, it would be worth a try, I don't see the harm.

I'm wondering if I could custom order that light with 5000K LEDs and those IR and UV LEDs, and if that might be a good idea. I'm guessing not, but wondering about it.

I'll attach a comparison pic of all the spectrum maps I got from the Kingbrite Alibaba rep.

I'm digging through old threads that talk about Kelvin and I saw a post from the Rev, back from 2015. In it he mentions how you need higher Kelvin for landrace sativas, here is what he said:

"Most since I moved out west I see cut before its even ripe good in THC and why people get slow healings and some negatives in the head and have to spend more for the junk oops. H.p.s. is the worse thing to happen too cannabis. It will not allow the tree too ripen through its stages and gives even poor THC. There is no red/orange high spectrum were they come from. Were cannabis in its cursed form was scattered by the Adams and evolved there is high uva/uvb and blue spectrum. High blue hid or 4100k and above t5 or all white led or white/blue led will accomplish this. The tree has to die through these chemical stages to be the best healer and high pressure sodium's make them want to live, because of the high red nanometers. Takes even an 80% sativa 3ft tall under a hid 16 weeks to even get truly THC ripe in first stage under a 1000 watt. So most cut dreadfully early. Acapulco gold for example is not darn gold. Its green however the further ripening in the old days turned the bud gold and also when decarbed in sun, which is still the best. Allot of tests done on Acapulco golds and other landraces over the years showed them not too potent, however the bud tested was cut way early. Too see this tree do what it truly does, depending on genetics, you need 20-40+ weeks too achieve. Is it worth the patience for those cannabinoids?"

 

[bleepboop]

Hey, I've read the same thing and rev is completely right about the 530-640 orange to bright red abundance part which is absolutely what we see with hps. This is about deep red and far red though which weren't as readily available until just around 2015-2016 in horticultural led form and I have been glad to have them present

Equally his point about the blues in tube lights is totally valid - the Samsung LEDs, no matter what Kelvin you pick, all peak around 450nm or so, which leaves the cyan above it and the deeper blue/violet lacking.

The only broadband white 3030 size LEDs that really cover that region are the Seoul semiconductor 'sunlike' range (6500k is based on a 415nm pump) and Nichia 'optisolis' (5000k has a 420nm pump).

As mentioned, it's cheaper generally to not stress about getting a white led that covers all as the blue peak in Samsung lm301b/h is certainly useful but the broader bandwidth of blue rev alludes has valid uses.

405nm, 420nm and 440nm mono supplementals are readily available here in Europe, usually for the aquarium markets. They tend to be Epileds and in 1w/3w/10w format.

Uva below that (365/385/395nm) has been documented to be useful but 400-440 supplementation just as helpful but much easier to reproduce and less harmful to eyesight.

​​​​​​I can see the temptation towards 5000k of course but you do lose out on some useful reds. As we've discussed, it's nothing like daylight.

 

[chilliwilli]

Hi folks is there any advantage of using the lm301h one? The spectrum pic is from led-tech.de
thx chilliwilli

 

[bleepboop]

Hi chilli, the lm301b/h are very standard these days. Led tech de has a few really nice options including strips with Osram mono colours supplementing the Samsung white. (+451nm, 660nm, 730nm)
These mono colours by Osram are higher quality and efficiency than 660/UV/ir on alibaba.
they're also great value for money. As are the 'butterfly' boards. Butterfly appears to have space in the centre for supplementary mono colour boards too. But my German is not great (I can manage for work there but about it!)

For the more DIY option ledrise.eu/lumitronix have some great Nichia and SSC (Toshiba) strips and cobs.

here's one I made in 2019 with parts from ledrise and an old qb288 heatsink.
this uses 5000k Nichia optisolis 98cri and 2700k SSC Sunlike 97cri. Makes a very happy bonsai mother's light.




For flowering I have strips from cutter electronics in Aus:

Nichia optisolis 5000k strips by cutter (112x 3030 size leds/56cm)



And Led Teknik PC red broadband reds on cutter strips. (64x 5050 size leds/56cm)



I've had this built 1yr + already and happy with them. It's practical for small spaces but for larger I would probably try and get the 240w GLA highlight V2 boards as I mentioned in my first reply to this thread.
It is not the most efficient option but I have only 60x60/2x2 and 60x90/2x3 to flower in.