UV-B/C LED's

So I know I found one of you guys deep in another thread debating UVC/B LED characteristics. Ive been searching around a lot looking at components for a booster board design I *might* make, and my research seems to have led me to this LED - Optical output power looks solid (min around 30mw) and spectrum for the XBT-3535-UV-A130-CC280-01 should be 280-285nm. Based on my reading of the post I now cannot find, I understand the peak we are looking for to promote resin production is around 285nm? Thinking these could do quite nicely, and price drops to $16/ea over QTY10, so seems to be the most cost effective option Ive found, just trying to see if there's something Im missing.

Heres the datasheet - Luminus_XBT-3535-UV_Datasheet.pdf

I also plan to incorporate 390/420nm UVA, as it seems there is a correlated affect where UVA can protect from the damage of UV-B (well 280nm is right on the line of C/B) and even increase the effects. As well, it seems most the research Ive found indicates best practice would be to run UVA during all lights on, and 280nm only for a short duration. (so that 280nm DLI isnt too high)

Anything anyone would like to add is welcome, I dont think there's really a UV LED thread, unless Ive missed it (as the post I found was tacked onto a thread on a different subject)

Samples. A control and the test plant(s). Lab tested against each other.

Control group of 10 plants(?)

I want to see proof of increased resin levels. Higher THC profile.

285 is one peak for uvr8, a lot of spectrums will activate it we just assume 285 is what we need. Everything from 280 to 285 will satisfy the peak and good to cover that spectrum due to the peak shifting on absorbance.

You can add UV to many good lights and I don't think the difference is worth it, LEDs are busy doing a lot of things maybe someone can notice more difference compared to not using some.

The advice of those selling it is to get close to where you see damage then back off and your about right. I don't think UV is linked to increase in yield just quality.

Beehive said:

Samples. A control and the test plant(s). Lab tested against each other.

Control group of 10 plants(?)

I want to see proof of increased resin levels. Higher THC profile.

Click to expand...

In the very near future, experiments like this will be possible and undertook! (and completely legal and buyer funded lol)
I agree 100% we need real comparisons, for now just looking for input as to spectrum and in theory what may be best.
Bruce Bugbee and a few others are doing the "real" science on this, but he hasnt really published much on UV yet, most his research lately has been on far-red.

You may wanna stalk around some of @nachooo 's posting, succesfully uses uvb in similar wavelengths. 5mW per square foot a few hours a day seems to be enough to increase thc but noth enough to do damage to the plant, although you may have a little bleaching or pale green.
30mW seems like a lot to me from one point. 15mw seems to be alright for 12" distance.

These leds dont last very much, keeping current and temps down is essential. A good idea to keep the uvb and mainlight on separate heatsinks and get a pro to mount the leds on stars/pcbs as they seem to come diode only.

BBQtoast said:

You can add UV to many good lights and I don't think the difference is worth it, LEDs are busy doing a lot of things maybe someone can notice more difference compared to not using some.

Click to expand...

Yup Ive wondered the exact same, thats why I intend to do this more as an experiment - very soon I will have the space and license to conduct it well at full scale. Thanks for your input as well Im definitely trying to dig more into the research on various UV spectrums, but yes everything I see seems to indicate *potential* increased cannabinoid and terpene concentrations. IIRC it was actually Bruce's early research on UV that talked about intermittent scheduling, but I think he was talking about A/B days to essentially prevent the plant from becoming immune to the effects (shielding itself) - but Ill have to go back and read again, thats an oversimplification Im sure. At commercial scale, full panel testing is required anyways, so it gives a bit of a "buyer funded" way to research things like this.

Rocket Soul said:

You may wanna stalk around some of @nachooo 's posting, succesfully uses uvb in similar wavelengths. 5mW per square foot a few hours a day seems to be enough to increase thc but noth enough to do damage to the plant, although you may have a little bleaching or pale green.
30mW seems like a lot to me from one point. 15mw seems to be alright for 12" distance.

These leds dont last very much, keeping current and temps down is essential. A good idea to keep the uvb and mainlight on separate heatsinks and get a pro to mount the leds on stars/pcbs as they seem to come diode only.

Click to expand...

Oh man I think you found the user I was referring to! So yeah I assumed 30mw optical would be on the high end, but led's live best under driven, so these would give us some definite headroom.

As for UV LED life expectancy, thats interesting and something Ill have to research, didnt realize they differed so much from standard LED's in that regard, thanks for the heads up! As for mounting, these "strips" will have their own heatsinks, and are part of a greater light prototype. I can handle the SMD's, and am actually working on a board design for them. Thankfully found a pcb printer (jlcpcb) that's cheap enough to make toying around with various prototypes possible!

ndawg420 said:

Yup Ive wondered the exact same, thats why I intend to do this more as an experiment - very soon I will have the space and license to conduct it well at full scale. Thanks for your input as well Im definitely trying to dig more into the research on various UV spectrums, but yes everything I see seems to indicate *potential* increased cannabinoid and terpene concentrations. IIRC it was actually Bruce's early research on UV that talked about intermittent scheduling, but I think he was talking about A/B days to essentially prevent the plant from becoming immune to the effects (shielding itself) - but Ill have to go back and read again, thats an oversimplification Im sure. At commercial scale, full panel testing is required anyways, so it gives a bit of a "buyer funded" way to research things like this.

Click to expand...

It has been buyer tested and in a good full spectrum it had very little if any difference. If LEDs had a better spectrum I doubt it would be needed, as it happens some are reporting minor differences from adding some to their LEDs.

Anyone saying they get 20% more anything should stick to learning to grow.

ndawg420 said:

Oh man I think you found the user I was referring to! So yeah I assumed 30mw optical would be on the high end, but led's live best under driven, so these would give us some definite headroom.

As for UV LED life expectancy, thats interesting and something Ill have to research, didnt realize they differed so much from standard LED's in that regard, thanks for the heads up! As for mounting, these "strips" will have their own heatsinks, and are part of a greater light prototype. I can handle the SMD's, and am actually working on a board design for them. Thankfully found a pcb printer (jlcpcb) that's cheap enough to make toying around with various prototypes possible!

Click to expand...

Make sure to use alu pcb and not plastic. These chips are a good find but nowhere in the DS is there any indication of how long they last. I know nachooo uses a different brand with 150mA as max for approx similar price, but mounted already. These chips seems to be a really good find

BBQtoast said:

It has been buyer tested and in a good full spectrum it had very little if any difference. If LEDs had a better spectrum I doubt it would be needed, as it happens some are reporting minor differences from adding some to their LEDs.

Anyone saying they get 20% more anything should stick to learning to grow.

Click to expand...

The paper that was floating about here (which i cant find now) seemed to show about 15% increase for 5mW per foot. But not like 20% to 35%, like 20% to 23%. I also hear about changes in terp profile and generally great smoke tests.

Rocket Soul said:

Make sure to use alu pcb and not plastic. These chips are a good find but nowhere in the DS is there any indication of how long they last. I know nachooo uses a different brand with 150mA as max for approx similar price, but mounted already. These chips seems to be a really good find

The paper that was floating about here (which i cant find now) seemed to show about 15% increase for 5mW per foot. But not like 20% to 35%, like 20% to 23%. I also hear about changes in terp profile and generally great smoke tests.

Click to expand...

I studied and used it myself, the results were clear and explained above.

If it boosts led by 15% the problem becomes why it doesn't boost more fuller spectrums by the same and LEDs have always had a sort of problem they are still overcoming so one day you are going to notice less effect with better developments if that is the case.

Theory is one thing, 15% is worth a lot we would be using it already not waiting for LEDs to uncover the same experiments run for decades.

I think lots of grower disbelief is well founded, show me 15% I'll hand over triple what you ask.

BBQtoast said:

I studied and used it myself, the results were clear and explained above.

If it boosts led by 15% the problem becomes why it doesn't boost more fuller spectrums by the same and LEDs have always had a sort of problem they are still overcoming so one day you are going to notice less effect with better developments if that is the case.

Theory is one thing, 15% is worth a lot we would be using it already not waiting for LEDs to uncover the same experiments run for decades.

I think lots of grower disbelief is well founded, show me 15% I'll hand over triple what you ask.

Click to expand...

Could you maybe talk a bit about how you tested this? How did you add 285nm and to what base light for what thc %?

295nm is the sweet spot for a single source UVB solution. HY5 expression peaks around 300nm, while UVR8 absorption peaks closer to 285nm and UVR8 action peaks around 290nm.
At 295nm you stimulate both UVR8 & HY5. If using a single UVB source, and utilizing a chip with a peak lower than 295nm, you will more or less target only UVR8, and using a chip with a peak above 295nm may result in a disproportionate HY5 expression compared to the UVR8 action. Idk which is more important, or even the genuine implications of their overall impact (I understand a little bit, but I'm no biologist), personally, at this point I just know that the plant reacts to both UVB bands.
Combining a 285nm source with 310nm source is perhaps a cheaper more practical way of targeting both areas of the UVB spectrum as opposed to opting for a 295nm single chip solution due to 295nm price and spectrum width. The 310nm won't be as efficient at stimulating HY5 as the 280/285nm will be at stimilating UVR8, but its also very much less harmful than 285nm and you can use more of it to offset the poor efficiency without hurting the plants (to a point).

UVR8 absorption...


UVR8 action...


HY5 action...



Chiefrunningphist put out an excel spreadsheet that will tell you how effective your selected UVB source will be at stimulating UVR8 & HY5 respectively (amoung other things).

Rocket Soul said:

Could you maybe talk a bit about how you tested this? How did you add 285nm and to what base light for what thc %?

Click to expand...

I was in an adjudication recently, it wasn't upto me to disprove the agent but for the agent to disprove me.

This can be a hard stance to be in and common for LEDs, I'll buy in when everyone else shows the results.

Still doesn't answer the already proved evidence that UV has less effect the more the other spectrums are filled.

You get 15% from UV, how much from the new reds and were approaching levels of some fantasy, over time they will and always do come down in numbers.

It's like a broken dialogue, wait a few years you be saying other things and have a better understanding.

ndawg420 said:

So I know I found one of you guys deep in another thread debating UVC/B LED characteristics. Ive been searching around a lot looking at components for a booster board design I *might* make, and my research seems to have led me to this LED - Optical output power looks solid (min around 30mw) and spectrum for the XBT-3535-UV-A130-CC280-01 should be 280-285nm. Based on my reading of the post I now cannot find, I understand the peak we are looking for to promote resin production is around 285nm? Thinking these could do quite nicely, and price drops to $16/ea over QTY10, so seems to be the most cost effective option Ive found, just trying to see if there's something Im missing.

Heres the datasheet - Luminus_XBT-3535-UV_Datasheet.pdf

I also plan to incorporate 390/420nm UVA, as it seems there is a correlated affect where UVA can protect from the damage of UV-B (well 280nm is right on the line of C/B) and even increase the effects. As well, it seems most the research Ive found indicates best practice would be to run UVA during all lights on, and 280nm only for a short duration. (so that 280nm DLI isnt too high)

Anything anyone would like to add is welcome, I dont think there's really a UV LED thread, unless Ive missed it (as the post I found was tacked onto a thread on a different subject)

Click to expand...

Do not use UVC for anything in your grow. UVC is for sterilization and will hurt your plants DNA. UVA is good, UVB is good in small doses. I also don't think UVA will help prevent damage from UVB. UVA will cause the plant to throw more trichomes. UVB if used at proper wavelength and amounts can damage the plant and make it increase potency but must be done carefully as you are damaging the plant to do it. At least this is how I understand it. I run a UBA bar 12/12 along with my lights during flower. If I was using UVB I would run it in bursts. Especially the first 20 points of the UVB spectrum. In nature UVA is abundant but only 5% of UVB makes it through the atmosphere.

Preliminary efficiency comparison and investigation between an Arcadia D3 and generic UVB LED...

BBQtoast said:

and were approaching levels of some fantasy

Click to expand...

search for studies by scientist J. Lydon. if you're good, there's even a meta-analysis for UV-B + cannabis.
has 0 to do with led reds etc... greenhouse/ sun + artif. UVB suppl = more THC.

Doug Dawson said:

Do not use UVC for anything in your grow. UVC is for sterilization and will hurt your plants DNA. UVA is good, UVB is good in small doses. I also don't think UVA will help prevent damage from UVB. UVA will cause the plant to throw more trichomes. UVB if used at proper wavelength and amounts can damage the plant and make it increase potency but must be done carefully as you are damaging the plant to do it. At least this is how I understand it. I run a UBA bar 12/12 along with my lights during flower. If I was using UVB I would run it in bursts. Especially the first 20 points of the UVB spectrum. In nature UVA is abundant but only 5% of UVB makes it through the atmosphere.

Click to expand...

Sorry Im sure the top post was kind of confusing, so there is a lot of data that shows (as some others mentioned) uvr8 has a peak around 285nm. The UV LED's in this range are sometimes marketed as UV-C - in this case they have several PN that fall in the UVC range and like this one that borders on UV-C into UV-B (280-285nm) - hopefully that clears things up. I am interested in pulsed UV-C for IPM, but thats a completely different (albeit interesting) topic

Kassiopeija said:

search for studies by scientist J. Lydon. if you're good, there's even a meta-analysis for UV-B + cannabis.
has 0 to do with led reds etc... greenhouse/ sun + artif. UVB suppl = more THC.

Click to expand...

If I take away UV from sunlight I loose 15%?

Come on really.

ndawg420 said:

So I know I found one of you guys deep in another thread debating UVC/B LED characteristics. Ive been searching around a lot looking at components for a booster board design I *might* make, and my research seems to have led me to this LED - Optical output power looks solid (min around 30mw) and spectrum for the XBT-3535-UV-A130-CC280-01 should be 280-285nm. Based on my reading of the post I now cannot find, I understand the peak we are looking for to promote resin production is around 285nm? Thinking these could do quite nicely, and price drops to $16/ea over QTY10, so seems to be the most cost effective option Ive found, just trying to see if there's something Im missing.

Heres the datasheet - Luminus_XBT-3535-UV_Datasheet.pdf

I also plan to incorporate 390/420nm UVA, as it seems there is a correlated affect where UVA can protect from the damage of UV-B (well 280nm is right on the line of C/B) and even increase the effects. As well, it seems most the research Ive found indicates best practice would be to run UVA during all lights on, and 280nm only for a short duration. (so that 280nm DLI isnt too high)

Anything anyone would like to add is welcome, I dont think there's really a UV LED thread, unless Ive missed it (as the post I found was tacked onto a thread on a different subject)

Click to expand...

Looks like a decent chip. All UVB led are piss poor efficiency which correlates to higher operating temps (per watt input), and reduced longevity compared to visible spectrum LEDs.

Here they grew plants in a greenhouse, utilizing the suns full spectrum (minus any shortwave UV that the greenhouse glass blocks). They then added UVB by hanging fs40 lamps (with cellulose acetate filters attached; allow UVB through, block UVA+) above the target specimens. They turned on the UVB lamp for different durations and observed the results.


The longer the lamp was on the greater the UVB supplied, and also the greater the observed THC.

Greenhouses typically transmit all of the suns spectrum except the lower UVB wavelengths:




I guess what I'm trying to point out is that perhaps an expirement has already been conducted using a full spectrum base light while adding various levels of supplemental UVB to test the impact.