R:G:B ratios for different Kelvin White LEDs

GlueSniffer

Active Member
Haven’t been able to find a breakdown of Red : Green : Blue Ratios for different temp white LEDs (or for the black body radiation the colors came from).

I would guess 2700 or 3000 K looks like it might be 50% Red 600-700nm, 35% Green 500-600nm, and 15% Blue 400-500nm. That is around a 3 : 2 : 1 Red to Green to Blue ratio.

6500 or 7000K looks like it might be 33% Red, 33% Green, 33% Blue for a 1 : 1 : 1 Ratio.

Anybody have this info or know where I could find it? Looking for maybe 3000k, 5000K, and 6500K.

My light has 21% of total number of LEDs as 6500K White LEDs and curious how much blue that adds to the total spectrum. Wondering how much blue would be removed if those were replaced with a warmer white.

Thanks in advance.
 

Lockedin

Well-Known Member
Haven’t been able to find a breakdown of Red : Green : Blue Ratios for different temp white LEDs (or for the black body radiation the colors came from).

I would guess 2700 or 3000 K looks like it might be 50% Red 600-700nm, 35% Green 500-600nm, and 15% Blue 400-500nm. That is around a 3 : 2 : 1 Red to Green to Blue ratio.

6500 or 7000K looks like it might be 33% Red, 33% Green, 33% Blue for a 1 : 1 : 1 Ratio.

Anybody have this info or know where I could find it? Looking for maybe 3000k, 5000K, and 6500K.

My light has 21% of total number of LEDs as 6500K White LEDs and curious how much blue that adds to the total spectrum. Wondering how much blue would be removed if those were replaced with a warmer white.

Thanks in advance.
Not scientific answer at all but - a lot.

Maybe add a supplemental light to balance the spectrum instead of replacing?
 

Southernontariogrower

Well-Known Member
My light has a cri of 90 5700k plus Osram 660 deep reds. 1.5a leds bright as the sun at equator, sun in kelvins is 5777k at equator summer day, probably on June 20 ish as first day of summer. 3500k alone is well rounded spectrum, but above 4000k you need to add 660 reds to round them out. For growing. Hope this helps a little. Keep them they are good the way they are, maybe get board at 2700k if you want spectrum change. Or 3000k, 3500k your choice. The higher kelvins lean towards the blue, less stretching during flower. They say more resin with intense blue spectrum.
 
Last edited:

GlueSniffer

Active Member
So what I was looking for was integration of those curves. What Calculus?, I know... But calculus is what gives the answer.

To answer my question, you need to measure the area under those curves. How much area under the curve between 400 and 500nm, area under the curve from 500-600nm, and then area under the curve from 600 - 700nm.

You can look and guess, but not very accurately. You need software to analyze those plots if you want accurate B:G:R ratios.

 

sf_frankie

Well-Known Member
What’s your end goal here? Trying to build supplemental lights or what? I just recently started reading up on spectrum so I’m curious as well
 

GlueSniffer

Active Member
My current lights have...
  • 180 Samsung 3000K warm white LEDs
  • 92 Samsung 5000K white LEDs
  • 15 660nm deep red LEDs
  • 1 730nm IR LED
Is still like to know how much Blue vs Green vs Red is in a 5000K and 3000K spectrum. I want to be able to say “I’m giving them 20% Blue (300-400nm) now but want to try 25% so I’d like to replace X number of 3000K LEDs with 5000K LEDs”. Or something like that

Overall, it looks like it would be close to a 3500K spectrum
My light has a cri of 90 5700k plus Osram 660 deep reds. 1.5a leds bright as the sun at equator, sun in kelvins is 5777k at equator summer day, probably on June 20 ish as first day of summer. 3500k alone is well rounded spectrum, but above 4000k you need to add 660 reds to round them out. For growing. Hope this helps a little. Keep them they are good the way they are, maybe get board at 2700k if you want spectrum change. Or 3000k, 3500k your choice. The higher kelvins lean towards the blue, less stretching during flower. They say more resin with intense blue spectrum.
 

GlueSniffer

Active Member
Just want to know exactly what is in the spectrum I’m giving them and how I would need to alter the number or type of led to tune the spectrum.
I might pick a different light that has a different mix of LEDs.

I’d like to be able to look at the specs and say ok, this light is 50% red, 20% blue -that light is 40% Red, 15% blue.

Just trying to figure out the ratios so I can play around with crop steering. Maybe 30% Blue during 1st 2 weeks of flower slows down the stretch. Maybe 70% Red / 10% Blue in weeks 5 and 6 pack on the weight. But unless I know those ratios, I’ll never be able to get an accurate estimate of the %’s of Blue, Green, and Red that is in the spectrum.

What’s your end goal here? Trying to build supplemental lights or what? I just recently started reading up on spectrum so I’m curious as well
 

sf_frankie

Well-Known Member
Why not get some supplemental far reds for crop steering? According to Bugbee, far reds have good potential for crop steering. If you wanna get way deep into it you’re gonna have to build boards or strips with separate channels for each color. Seems like a lot of extra work for minimal proven results. But then again science is fun.
 

GlueSniffer

Active Member
Let’s say I add some red. Or Blue. That’s not a quantifiable addition unless I know what I started with.

Thanks for the input, but am only looking for specific numbers, so I asked a specific question.

Only interested in B:G:R ratios for different K temp spectrums. Not interested in how to crop steer. Not interested in how to alter spectrums with additional lighting

Unless somebody chimes in and says 5000K is x% 300-400nm, y% 400-500nm, and z% 500-600nm

or

3000K is “X” part blue : “Y” part green : “Z” part blue

then the question I asked remains unanswered.

Again thanks for the help, but only interested in the specific ratios or percentages for different temp spectrums. And yes I’m a nerd - so when geek out on a topic I tend to go deep.
 

bk78

Well-Known Member
Stole this from @bk78 but to me it looks like his pulse pro meter can give you the info you’re looking for.
I haven’t seen any third party testing on the spectrometer yet, but I have seen multiple with the par meter (which is within 5% of the apogee mq500

so take that for what it’s worth at this stage. It’s brand new so I’m sure there will be testing done soon.
 

GlueSniffer

Active Member
Thank you pulse meter pro seems to give me exactly what I am looking for.

I use a doctor meter lux meter, not a par meter. I took PAR measurements off a YouTube channel review of my light and then measured in the same fashion with my lux meter. Lux or lumens or whatever divided by 60 gives me PAR for my light.

60,000 lux = 1000 PPFD
30,000 lux = 500 PPFD
 
Top