Programmable LED Lighting

Hash and Stuff Pre/Post-Processing
, ,
1

I have been researching LED grow lights, but have been unable to find what I am looking for. I have had this concept, in my head, for quite a while. Programmable LED lighting, beyond simple VEG and FLOWER settings.

What I have had in mind, and have been seeking, is a computer controlled, complete simulation of days and seasons.

Simulating a day or a season would requires dimmable LED’s, in a wide variety of frequencies. A computer would be needed to control them, because every bulb (or group of like frequency bulbs) would need to be individually addressable. This would allow each to come on, or fade out, slowly at different times, when needed. The light spectrum and intensity for a natural day changes constantly, starting well before sunrise, and lasting until well after sunset. Additionally, the overall daily light spectrum and intensity varies, throughout the seasons.

With microcomputers so cheap and available, I thought that something like this would be available. Is anyone aware of such a system?

3 Likes
2

I run a light with the system you seek. I have been working on it for some time now with the company.

1 Like
3

Gavita has a module for a wide assortment of settings. This is for their pro 1700e and other models I think.

Sunrise, sun set, and a bunch of other settings.

4

Scynceled I believe has what you seek

5

Heliospectre. At this point they are a little outdated but pretty neat for research. Conviron was using them in their research growth chambers for a bit.

6
7

What company?

My ideal system would be a mixture of traditional bubble LED’s along with the more current, full spectrum components, such as COB’s or LM301’s.

Bubble LED’s, including IR and UV ones, would be for ramp up and ramp down custom spectrums, simulating sunrise and sunset, and as supplemental spectrum, enhancement lighting. While the full spectrum COB’s or LM301’s, while also ramped up and down, would be used during the main lighting periods, when more intense light is desired.

8

I would also like to see topographical data incorporated along with lattitudinal data for recreating specific regions. Tracking each day of the year 24 hours, it would be sweet to recreate actual nightime conditions. Moon phases?

2 Likes
9

The California LightWorks system comes close, but it is only a 3 channel controller and it will only work on their lights. Additionally, their lights are extremely expensive.

It appears that they are controlling separate sets of blue, white and red LED’s. The intensity of all lights in each group can be adjusted, but only as a whole, not individually. While their controller allows spectrum intensity of each channel to be adjusted, the actual spectrum is not adjustable, it is determined solely by the type of LED’s that they employ.

I envision something using a larger (more channels) controller.

Using more channels would allow a lot more control, and custom programming. The light board could have an amazing variety of LED’s, allowing for actual spectrum adjustment.

2 Likes
10

Scynce led

11

That system, by Scynce, is a 4 channel controller.

They have more variability, but still limited.

Every channel on a controller allows for 256 levels of intensity, like a dimmer. You can mix LED’s, with different frequencies, on a single channel. The LED group intensity can be from 0-100%, but the frequencies emitted will not change, they are dependent on the diodes. To be able to manipulate the frequencies would require LED’s of as many different ranges as you want, and a multichannel controller with enough channels to control each frequency seperately.

12

I want the most DLI (daylight integral) possible for the plants. It’s the total time plants are absorbing photons that contributes to yield. Fading in and out lessens the amount of time the plant can absorb light before it gets harvested.

I use 3500k+660nm LED and CMH for a balanced spectrum in veg and bloom. They go on and off and that’s it. Spectrum going in and out seems like another thing that can go wrong in a cycle.

3 Likes
13

I was just thinking how much any of this actually matters. Kinda seems like over complicating things to me.

1 Like
14

There is absolutely nothing wrong with the current lighting styles, but like every aspect of canna-culture, things tend to get better, over time.

RSO was fine, but distillate eventually appeared.

Improving grow lights is just another aspect of gradual change and evolution. Without experimentation, we would still be using that old stuff that we threw away, years ago.

Plants are natural products. Everyone knows that plants grown outdoors are more robust (dare I say better?). Their more robust nature is due to the real world conditions that they experience outdoors, real dirt, insects, wind, sunlight, pollen, animals, etc.

Not all things outside are good, but the light is definitely good. Simulating that cycle, within reason, would be a challenge, but could provide advantages. Using a computer to set up everything, it would become a set and forget process. You would still need to adjust nutrients, at the proper times.

3 Likes
15

I feel like trying to replicate natural lighting only gets us 1/4 of the way there (which doesn’t necessarily dispute your point). What I would like though is to understand the mechanisms that make natural light better in some respects because there are others where it is certainly worse (IR raising leaf temperature being the big one that comes to my mind).

By understanding those mechanisms we can manipulate them to get the plants to do things they would never do naturally and that’s the cool part for me

5 Likes
16

Agreed, I’m curious if this is also a good way to start teasing out some minor cannabinoids.

17

Light spectrum can manipulate the cannabinoid ratios in growing plants.

https://www.karger.com/article/FullText/489030

18

Lots of people behind the scenes have been working towards your goal with lighting, purchase an “Arduino” controller for an affordable way to get involved with the world of building physical things that are controlled by a computer. Amazon has them cheap!

19

OK, got an Arduino Mega 2560, and a Raspberry Pi 4B. The Pi, was overkill, but is still fun.

Spent the past few weeks getting acquainted with everything. I am far from the most proficient programmer, but I do have some experience.

After puttering around for a while, I have learned how to program the LED’s. I am in the process of prototyping. Getting the LED’s to do what I want is primary, and has been accomplished, but I still need to integrate a real-time clock, and a display.

When I am done, I will have an 8 channel controller, that will independently regulate separate groups of LED’s. The light groups will be: IR (700nM), Far Red (680nM), Red (650nM),Warm White (3000K), Bright White (5000K), Blue (465nM), Royal Blue (450nM) and UV (400nM).

The clock and display are really not necessary, if a pre-programmed pattern is already loaded, but are nice additions, as they would allow re-programming without a computer.

I only chose to program 8 channels, but additional ones are available. I just felt that additional channels were not needed. If desired, the additional channels could be used to control secondary light groups, such as amber and green, or LED’s with different frequencies, near the blue and red spectrums.

I will probably continue programming. I should be able to integrate moisture, CO2, Humidity and Temperature sensors into the system. This will automate the fans and water supply. Automatic nutrient mixing would be possible, but probably not something that I will attempt.

2 Likes
20

So…looked at the prices on the LED types that I wanted to use. They are way too expensive to make this project viable.

Prototyping and programming were accomplished using readily available parts, including tiny standard LED bulbs. When looking at the high power COB light modules, it was determined that they were stupid expensive. A fully assembled, functioning, LED grow light fixture, containing the same LED’s, can be purchased for less than the cost of the individual components, to build this.

Found out that it would cost around $300 just to build a small 800W panel. Of course, the DIY panel would be way more controllable, but just I don’t think that the cost trade-off is worth it. Already built panels are cheaper to just buy…the power of mass production.

I have put some thought to re-wiring a commercial product, but believe that would be near impossible, considering that every wire is just a trace on the PCB. Sometimes I miss real wires.

1 Like