Page 2 of 2

Re: More SPAM from Prawn (Psst - buy my LEDs)

PostPosted: Fri Jan 11, 2019 6:38 am
by Prawn Connery

Yeah, that's right - that's why there have been concerns over the amount of UVB given off by twisty CFLs vs tubes and other lighting. In fact, LED companies tout the lack if UV as being an advantage for their indoor product over fluoros.

Re: More SPAM from Prawn (Psst - buy my LEDs)

PostPosted: Fri Jan 11, 2019 10:53 am
by deran
the diy panel i had in test only consumed 30w with 30 diods and it had a uv and fr diod, so of 30 diods , ionly 1 uv

funny thing , its less "bright" than the fr, you cant see both witgh the naked eye, the cmos tho cought the heat, but not the uv portion

i dont find uv destructive with this test, its the power/amount in ratio to all other bands that matters and that you dont overdo the plants limits

actually, bc tech is moving on, the leds became "too strong" nowadays, that you need to drive em seperatly only for 1 or 2 hours daily, which is enough, with the older weak leds you were able to prolong this time as the total amount of photons hasnt filled up in given time

its like with many things considering life, it can be deadly, but considering you know what you are doing, it will add to life (and fun)

Re: More SPAM from Prawn (Psst - buy my LEDs)

PostPosted: Sat Jan 12, 2019 8:41 am
by Prawn Connery
There are a few different theories on this. A lot of guys are running UVA and UVB for a few hours to simulate peak UV times during the day.

For Emerson effect, they are running far red (730m) towards the end of the cycle to simulate dusk. Both of these require separate channels to drive at different times.


The violet-based white phosphor LED we're using (6500K) has a strong peak at 400nm and a little bit of scattered UVA from about 380nm onwards. When mixed with the LEDs below, the UV peak is modest, but it is a high-energy wavelength that other white phosphor LEDs don't provide, so we believe the constant, low exposure to the plant will stimulate oil and terpene production. For food corps, for example, Beta Carotene and Chlorophyll A both have strong peaks below 450nm, which is where most blue-based white phosphor LEDs peak. Violet-based white phosphor LEDs have two peaks at 400-405nm and 450nm. They also emit a fair amount of energy in the 650-750nm range, which also coincides with the second Chlorophyll A peak. Most low CRI (70 80) blue-based LEDs peak below 600nm.


This is the CRI98 LED we're using - it has very strong peaks in the 620nm range and below 450nm. This SPD chart is the 3000K version, as we're using the 2700K version which has less blue. It actually has very good spectral distribution for plant growth on its own, but is not as efficient as the LED below.


This is the main LED used in the panel. Again, this is the 3000K chart and we are using the 2700K LED, which has less blue and green. There are 300x of these, 90x CRI98 and 60x near-UV LEDs for a total of 450. These Nichia CRI90 LEDs are also nicely weighted in the red spectrum and are the most efficient 3030 LED on the market, with tests showing they are actually more efficient than Samsung's flagship LM301B that a lot of other LED board manufacturers are using.

Once the boards have been assembled and we've tested the spectral distribution and total PAR output, we'll be in a better position to know how they compare to other boards on the market, but suffice it to say, it will have arguably the most complete spectrum of any single-channel, single driver board available - which is what were were aiming for. It will also have a nicely weighted red spectrum with enough blue and UVA to boost flowering and oil production, whilst bringing flowering times down with the help of a little far red boost.

Re: More SPAM from Prawn (Psst - buy my LEDs)

PostPosted: Wed Jan 16, 2019 1:42 am
by bentech
what do you see as the future of development in this area

driving cost down obviously
BUT the ability to build arrays that are 'tune'able is amazing

it seriously seem to me the only thing lacking now is a mike the plant can step up to and say what it would like more precisely...

More SPAM from Prawn (Psst - buy my LEDs)

PostPosted: Mon Feb 04, 2019 4:20 pm
by Prawn Connery
Success. Got a prototype up and running and it is performing exactly as hoped. You can see the UVA LEDs, but the high CRI LEDs are indistinguishable to the human eye.



It's not until you look at the spectrum that you can see the influence. Both these graphs below are of 3000K LED boards. The orange line is the 3000K Samsung LED that's in most popular panels, boards and strip builds. The blue line is the latest board. You can see how, even though they are both 3000K, the blue line has a lot more red in it and a nice little bump in the UVA and near ranges (380-420nm). The spectrometer only goes down to 380nm, so there's probably a small amount of UVA beyond that.

Not a lot of UV is required, as too much causes cell damage. The idea is to provide enough UV to stimulate terpene and THC production.

The orange line is also CRI 84, whilst the blue line is CRI 95 - very close to true sunlight colour rendition.


Here's what the spectrum looks like on its own:
original.png
original.png (33.62 KiB) Viewed 202 times


And matching heatsinks:


Re: More SPAM from Prawn (Psst - buy my LEDs)

PostPosted: Mon Feb 04, 2019 4:24 pm
by Prawn Connery
Heatsinks are optional if running lower currents. On a 30C (86F) day, the board without heatsink hit 47C (116F). The boards max out at 225W, but theoretically could run higher with active cooling. As long as they run below 100C they will last as advertised, but ideally you'd want no higher than 70C. LEDs are less efficient the hotter they run.

Re: More SPAM from Prawn (Psst - buy my LEDs)

PostPosted: Mon Feb 04, 2019 4:29 pm
by Prawn Connery
bentech wrote:what do you see as the future of development in this area

driving cost down obviously
BUT the ability to build arrays that are 'tune'able is amazing

it seriously seem to me the only thing lacking now is a mike the plant can step up to and say what it would like more precisely...

Tuneable is good, but it adds to complexity and costs. That's why these boards were designed, as they combine a spectrum closely matched to the McCreee curve but with added UVA and far red, all driven by one driver on the same channel. Just plug and play - couldn't be easier.

Once we have a few more boards and run a couple of flowering cycles, we'll know whether they truly live up to expectations, but the clones love it - ans these are dedicated flowering boards, not veg boards.