The Ideal spectrum.

Rocket Soul

Well-Known Member
So im posting this here as its quite related and this thread already have the attention of some of the usual suspects.

Im helping a friend converting to led, currently at cmh and hps. Weve already done a couple of goes with some different spectrums but really very little difference: samsung 2700 and 3000 didnt really have very much difference though the intensity levels werent perfectly matched. We also tried bridgelux vestas, equal measure of 5000k and 2700k in 90 cri. Spectrum attached, against samsungs 3000k. Again, not really any difference, even in finishing time or end of transition stretch which i would have expected with a bit more far red. Were going to do some more tests and im looking for ideas on how to use the different components were getting. Heres a list of stuff thats getting tested:
Static spectrum:
- We have a lm301b board with 1/8th of the diodes epistar reds, 660nm. It will be tested next to the same board but all of the diodes lm301b. Expecting the added red board to be slightly less intense.
- 2700k against 2700k + about 15-20watts (about 5-9 radiant watts) over a tray, half and half. Infrared will be on day and night. Why: the 2700 fixtures we built have some realestate where the strips infra red strips would sit nicely. It would test white vrs white+infrared; the absence of infrared is to me the one big difference between HID and leds so wanna try it. Also interested in doing a limited run with infra red at night in case it makes the herm or something. Also, if i understand right: at night is the time when the grow, spending the sugars they built up during the day and usually its colder at night, a bit of extra heat could maybe boost the metabolism and growth. Well also be able to play with hanging height, maybe the extra heat make the plants able to stand more light.
In next post ill do the dynamic spectrums, boarding a flight now.
Any coments are wellcom, just wanna pick sone brains. @Rahz @jarvild @Randomblame and anyone else.
 

Attachments

Rahz

Well-Known Member
My opinion (which isn't entirely scientific) is that 3000K 90CRI alone is really good. When you add other spectrums with higher K values you're decreasing deep/far red percentages and boosting blue/green percentages. I can't provide exact percentages but for instance, what's the difference between 8:1 3000K 80CRI : far red -vs- 3000K 90CRI? Using the information in the math behind thread you could work out a spreadsheet to combine spectrums and intensities into a single data set and see how the percentages compare.

If I were to build a flowering lamp today I would look at CLU048-1212 in 3000K 90CRI running them 15-20w each or whatever it took to get close to 3 umol/j. Price it out and see how it compares to the boards.
 

Rocket Soul

Well-Known Member
Dynamic spectrums:
These combos will have spectrum enhancement on separate channels so we will be able to control it better. Combos are (not yet set in stone though):

-3000k 80 cri + solstrips from cutter: dual spectrum- 2000k and 390nm. The 2000k spectrum is a bit like bridgelux 1750, red peak around 630nm and not really very much else.

- 3000k 90cri + cutter redstrips (5 to 1 ratio of 660 and 730nm, incidentally more or less the same rate as the 2000k spectrum above)

- 3000k 80/90cri in 1 to 1 ratio + the same red strips as above.

All of them will have some infra red strips aswell, all of them one the same driver/dimmer. The all 3000k boards will in be driven in lines: the grow is in an aisle so we have 4 boards wide and each line has its own driver so we can get perfectly even spread, from side to middle to side.

There will also be another test with Cutter Br-solstrips but its more to try how the platform works; its a slightly different setup: 3 channel strips with high power 5050 leds, each 2 foot strips can apparently do over 200w but we will probablye keep them around 100-125 each depending on how they do on their heatsink and how the spread is. This combo will be:
- 3000k 80cri and 90cri, probably on the same channel + uv 390nm. Maybe well add some red strips and infrared on the same drivers as the strips that go with the boards above.

I got a few ideas but would still like to pick some brains out there:
- transition: the plants come from soft vegg and cant take high light levels straight away. Dont ask me why, its how he grows. Maybe even a few days of vegg under the new lights.
Anyone has suggestions on how to do pre and post strip? Low whites and higher output on red strips/2000k strips? With extra far red youd youd expect more stretch but when Rahz did his test a while back he found the opposite: his 90 cri plants went thru the stretch faster and thus stretched less if i remember right.
- UV- low level or just 15 mins at a time?
- infra red: suppose to cause a bit of stretch but also may be nice to have on from the start, simulating high sun summery weather.

Unfortunately i will probably not be able to post this grow as my friend is attention shy. We cant be totally scientific neither as this is comercial and cant risk to much yield to science. Were just really hoping to see something that we like stand out, sortas exploratory. If we find nothing then its 3000k 80cri all the way from now on, maybe some far red flower end of day switching down the line.

Anyone have ideas of what would be interesting to do during the cycle? One given is to increase uv as we get closer to the end, maybe decrease or remove the IR aswell towards the end.
 

Dave455

Well-Known Member
Dynamic spectrums:
These combos will have spectrum enhancement on separate channels so we will be able to control it better. Combos are (not yet set in stone though):

-3000k 80 cri + solstrips from cutter: dual spectrum- 2000k and 390nm. The 2000k spectrum is a bit like bridgelux 1750, red peak around 630nm and not really very much else.

- 3000k 90cri + cutter redstrips (5 to 1 ratio of 660 and 730nm, incidentally more or less the same rate as the 2000k spectrum above)

- 3000k 80/90cri in 1 to 1 ratio + the same red strips as above.

All of them will have some infra red strips aswell, all of them one the same driver/dimmer. The all 3000k boards will in be driven in lines: the grow is in an aisle so we have 4 boards wide and each line has its own driver so we can get perfectly even spread, from side to middle to side.

There will also be another test with Cutter Br-solstrips but its more to try how the platform works; its a slightly different setup: 3 channel strips with high power 5050 leds, each 2 foot strips can apparently do over 200w but we will probablye keep them around 100-125 each depending on how they do on their heatsink and how the spread is. This combo will be:
- 3000k 80cri and 90cri, probably on the same channel + uv 390nm. Maybe well add some red strips and infrared on the same drivers as the strips that go with the boards above.

I got a few ideas but would still like to pick some brains out there:
- transition: the plants come from soft vegg and cant take high light levels straight away. Dont ask me why, its how he grows. Maybe even a few days of vegg under the new lights.
Anyone has suggestions on how to do pre and post strip? Low whites and higher output on red strips/2000k strips? With extra far red youd youd expect more stretch but when Rahz did his test a while back he found the opposite: his 90 cri plants went thru the stretch faster and thus stretched less if i remember right.
- UV- low level or just 15 mins at a time?
- infra red: suppose to cause a bit of stretch but also may be nice to have on from the start, simulating high sun summery weather.

Unfortunately i will probably not be able to post this grow as my friend is attention shy. We cant be totally scientific neither as this is comercial and cant risk to much yield to science. Were just really hoping to see something that we like stand out, sortas exploratory. If we find nothing then its 3000k 80cri all the way from now on, maybe some far red flower end of day switching down the line.

Anyone have ideas of what would be interesting to do during the cycle? One given is to increase uv as we get closer to the end, maybe decrease or remove the IR aswell towards the end.
3000k 80cri plus arcadia d3 + 12% uv bulb brings it on.
 

ildiocane

Member
LM301 3500k + 660nm
2700k Nichia 757 + 4000k LM301B and dimmer
2700k Nichia 757 + 6500k Nichia 757 and dimmer

There are the ipotesis I made for the best spectrum based on what I read and tested some advise? what do u think about it? there is a way to solve this question in scientific way? there is some research or test?
 

ildiocane

Member
I love what Science can bring to the table but the ultimate test lies in what works for the individual grower for his situational growing style.
So I don't think there will ever be an Ideal Spectrum. We can get close though.
I love the science too and I also was a student, not gradueted yet but a big lover and I have the same opinion about the tests, they are important and thanks to test we can understand and use it for make other test and find new information and deduction.

I think we can realize the ideal spectrum if we do enought test and also analyze what other user did.
I find something could be helpful but my knowledge isn't enought to understand at all (is ok post link to other forum, I think yes but I want a confirmation)

All togheter we can make a change

https://www.icmag.com/ic/showthread.php?p=6718314
https://www.icmag.com/ic/showthread.php?t=295933
https://www.icmag.com/ic/showthread.php?t=293045
https://digitalcommons.usu.edu/cgi/viewcontent.cgi?referer=https://www.google.com/&httpsredir=1&article=1765&context=psc_facpub
https://books.google.it/books?id=sFMJAwAAQBAJ&pg=PA462&lpg=PA462&dq=mccree+action+spectrum&source=bl&ots=OazxY06fVw&sig=egMN6XMlFgW3bMu5okjJ-PmvLTo&hl=it&sa=X&ved=2ahUKEwjGx-_cr_DfAhXEzIUKHYG9BMQ4ChDoATAGegQICBAB#v=onepage&q=mccree action spectrum&f=false
https://www.karger.com/Article/Fulltext/489030
https://www.journals.uchicago.edu/doi/abs/10.1086/337342

to be contine..
 

Argentocobs

Well-Known Member
Nice read ! Im building a led module with 16 osrams LEDs , 10 square white 4500 k cri 70 and 3 630 nm and 3 660 nm, not sure if adding 730 nm with to main light with the way to go .
 

Moflow

Well-Known Member
Its a multichannel board, with wide red, wide blue/uv and white. ill give more details soon, meanwhile heres another example white mainly white and a bit of red and no extra blue. around 2700k :)
Multichannel as in a Chilled board type?
Both spectrums look good, the reds seem to peak in right place.
Are they top of the range diodes ie Cree, Samsung , nichia?
I've seen some spectrums with a lot more blue than what you've shown
Fluence for instance.
O1CN01Oqa3Qq1NdB23d8LV3_!!6000000001592-0-tbvideo.jpg
 

Rocket Soul

Well-Known Member
Multichannel as in a Chilled board type?
Both spectrums look good, the reds seem to peak in right place.
Are they top of the range diodes ie Cree, Samsung , nichia?
I've seen some spectrums with a lot more blue than what you've shown
Fluence for instance.
View attachment 4293838
Ill have more info soon, i just need to get another 2 boards added to my order and pay for them. I got them custom for these new tests were doing and yeah chilled boards was somewhat of an inspiration.

Blues: most of these guys are doing high K plus 660nm in a static spectrum. Think they go for higher k due to better efficacy. Im using 3500K white and red and blue channels- the blue is equal parts uva, 450 and 470. Red is going to have to be a surprise for time being...
 

TEKNIK

Well-Known Member
Dynamic spectrums:
These combos will have spectrum enhancement on separate channels so we will be able to control it better. Combos are (not yet set in stone though):

-3000k 80 cri + solstrips from cutter: dual spectrum- 2000k and 390nm. The 2000k spectrum is a bit like bridgelux 1750, red peak around 630nm and not really very much else.

- 3000k 90cri + cutter redstrips (5 to 1 ratio of 660 and 730nm, incidentally more or less the same rate as the 2000k spectrum above)

- 3000k 80/90cri in 1 to 1 ratio + the same red strips as above.

All of them will have some infra red strips aswell, all of them one the same driver/dimmer. The all 3000k boards will in be driven in lines: the grow is in an aisle so we have 4 boards wide and each line has its own driver so we can get perfectly even spread, from side to middle to side.

There will also be another test with Cutter Br-solstrips but its more to try how the platform works; its a slightly different setup: 3 channel strips with high power 5050 leds, each 2 foot strips can apparently do over 200w but we will probablye keep them around 100-125 each depending on how they do on their heatsink and how the spread is. This combo will be:
- 3000k 80cri and 90cri, probably on the same channel + uv 390nm. Maybe well add some red strips and infrared on the same drivers as the strips that go with the boards above.

I got a few ideas but would still like to pick some brains out there:
- transition: the plants come from soft vegg and cant take high light levels straight away. Dont ask me why, its how he grows. Maybe even a few days of vegg under the new lights.
Anyone has suggestions on how to do pre and post strip? Low whites and higher output on red strips/2000k strips? With extra far red youd youd expect more stretch but when Rahz did his test a while back he found the opposite: his 90 cri plants went thru the stretch faster and thus stretched less if i remember right.
- UV- low level or just 15 mins at a time?
- infra red: suppose to cause a bit of stretch but also may be nice to have on from the start, simulating high sun summery weather.

Unfortunately i will probably not be able to post this grow as my friend is attention shy. We cant be totally scientific neither as this is comercial and cant risk to much yield to science. Were just really hoping to see something that we like stand out, sortas exploratory. If we find nothing then its 3000k 80cri all the way from now on, maybe some far red flower end of day switching down the line.

Anyone have ideas of what would be interesting to do during the cycle? One given is to increase uv as we get closer to the end, maybe decrease or remove the IR aswell towards the end.
If possible can you add some extra blue 420/450/470nm and green to the mix for 3-4 hours in the middle of the cycle, I believe this will allow the plant to absorb the high level of red light you have better
 

TEKNIK

Well-Known Member
I came up with this spectrum using 4000 k and a mix of 630 and 660 . Used 4500 k but the spectrum simulator only had 4000 k . All using osrams square whites and oslons reds .
Cree has released thier new Etone COB range which fills in the 630nm, it would be great if they released these in a midpower chip so they could be easily combined with other chips such as 660nm
Screenshot_20190305-163843_Drive.jpg
 

its.always.420

Well-Known Member
Multichannel as in a Chilled board type?
Both spectrums look good, the reds seem to peak in right place.
Are they top of the range diodes ie Cree, Samsung , nichia?
I've seen some spectrums with a lot more blue than what you've shown
Fluence for instance.
View attachment 4293838
Isn't that the greenhouse spectrum? I think they added more blue to make up for the excessive amount of red in natural sunlight. Their indoor spectrum doesn't look like that.
 

Moflow

Well-Known Member
Isn't that the greenhouse spectrum? I think they added more blue to make up for the excessive amount of red in natural sunlight. Their indoor spectrum doesn't look like that.
Yea, just using that as an example
Yes , too much blue ratio in that spectrum, for me at any rate
Although I've some 2 channel boards with 90 Cri 3000k and 80 Cri 5000k I could tune to have plenty of blue in spectrum
 

TEKNIK

Well-Known Member
Ill have more info soon, i just need to get another 2 boards added to my order and pay for them. I got them custom for these new tests were doing and yeah chilled boards was somewhat of an inspiration.

Blues: most of these guys are doing high K plus 660nm in a static spectrum. Think they go for higher k due to better efficacy. Im using 3500K white and red and blue channels- the blue is equal parts uva, 450 and 470. Red is going to have to be a surprise for time being...
Hi Rocket, any updates here regarding how you plan to set all this up. I am quite interested in the results if you can push up the blue and green content for a few hours during the cycle.
 
Top