A Design Challenge for Anyone Interested

[Tyler520]

I work in the field of Architectural Sciences, and am relatively familiar with the nature of light and radiation, but am having trouble with a new project my firm is working on. I am hoping that some of you may be able to rise to this challenge, as it has been throwing me for a loop.

We are constructing a vertical greenwall inside a hospital to clean the air of the lobby. Some areas will receive natural light via southern windows, but some will not. the office lights will offer supplemental lighting. We have selected plants that require low levels of light to keep maintenance and power consumption as low as possible: We estimate a need for approximately 80 Watts/meter2 or 150 - 200 micromoles per meter squared per second at the surface of the plants - we need to deal in irradiance, rather than illuminance, as the 2 do not translate well. The greenwall is 20 feet tall, and 10 feet wide. we need to light the surface perpendicularly for even irradiance distribution. (no suspended floods or anything like that.

We are trying to figure out what would be the minimum number of lamp units necessary to make this greenwall stay alive, and need to provide multiple options for the client: LED option, HPS option, MH option and Fluorescent option. We are REALLY hoping that the LED option pans out, but nobdoy in the office has the knowledge to verify.

If anyone is interested, I can give more details if need be

Thanks in advance!

 

[Ken_McE]

Where/how can the lights be mounted in relation to the greenwall?
?

 

[Tyler520]

the LEDS would need to be right up against the plants, 8-12," and would probably be a "halo" ribbon LED strip around each plant every foot and mounted parallel to the wall (and on the wall) every 18 inches or so, but the more powerful bulbs (HPS, MH) can be up to 4 feet away mounted from the ceiling as pendants so that we can wash the greenwall with as much light as possible and as few lamps as possible.

we are trying to figure out what the proper unit of measurement is regarding what plants need to survive: Watts/meter2, micromoles/meter2/second...both? lots of contradictions and confusion right now... scientists aren't even fully certain

we know HPS puts out lots of both, so no problem there, but they are almost too bright, too hot, and somewhat combersome and ugly. We also know that fluorescent works well for growing plants, and has a much lower Wattage, yet a very low mmol/meter2/second output as well ( a 2,000 lumen, 4 foot 25watts T8 tube only puts out 27 mmols/meter2/s) and need. LEDs have an exceptionally low wattage, relatively speaking, but put out a near perfect mmol/meter2/second (150mmol/m/s at 670nm), but have such a narrow projection of radiation, that they cannot be very far from the plant, therefore I do not know how to quantify a number of diodes that would be sufficient for a particular area square footage of greenwall surface.

as you can see, these all work for growing plants, but work on different ends of the metaphorical spectrum (no pun intended). biologists tend to stick with mmols when describing how plants react, agricultural engineers use watts/meter2...illuminance doesn't really apply, even though there are some rough conversion factors floating around for various lamp types.

 

[LEDninja]

Plants do not use the full light spectrum of white light. Horticulturists use LEDs with only the colours the plants need to save energy. May make a living wall look weird though. Search "plants" (search entire posts) in this and the LED forum and you will come across treads like this:
http://www.candlepowerforums.com/vb/showthread.php?t=258734
http://www.candlepowerforums.com/vb/showthread.php?t=266740

 

[blasterman]

and need to provide multiple options for the client: LED option, HPS option, MH option and Fluorescent option.

Define 'options'. If you're just going to use low powered LED strips with a lifespan barely greater than fluorescent tubes then it doesn't sound like much of an 'option'. On the other hand, HPS and MH options are as simple as a reflector and ballast and fairly evolved for agriculture use.

we are trying to figure out what the proper unit of measurement is regarding what plants need to survive: Watts/meter2, micromoles/meter2/second...both? lots of contradictions and confusion right now... scientists aren't even fully certain

The only arguement with LEDs is determining what ratio of red/blue the particular plant prefers. Fruiting plants want red (or HPS), leafy/vegetative plants want more blue spectrum, house plants and indoor plants about a 1:3 mix of blue/red as a un-official rule. I'm propogating corn plants (Dracaena Massangeana) right now, and they could care less about blue light and want massive amounts of red LEDs. My pothos and philodendrons want a 3:1 red/blue mix.. Corals....massive amounts of blue only.

As a rough guide, a solid state LED fixture (good one) should have a 2x advantage over the best metal halide metal halide because of superior directionality and spectral efficiency. That should help with your immediate calculations. You're likely looking at 30-40watts per meter of LEDs at a 3:1 red/blue mix (or, what's in my living room).

If I were building this from scratch and wanting something to look 'cool' I'd use a grid of c-channel aluminum and bare 3watt emitters. If it's a high end or showy installation installation toggle the red and blues to alternate every hour over a 16hour period, or cyle slowly, and do some finishing work on the exposed side of the grid, but it sounds like you already have your design down.

Also, while we know that only red and blue or usefull here, adding a few white emitters will vastly help with aethestics. Thousands of lumens of bright pink (Red/Blue) LEDs get very annoying after awhile at these kind of power levels and adding a bit of white helps make things look alive.

but have such a narrow projection of radiation

Explain? Bare LEDs have very smooth 120degree emission cones than can easily be adopted down to 10degree with cheap optics. So, the spread and intensity can easily be customized. Can't do this with HPS and MH.

 

[James Jackson]

If it were me, and I were building this - from scratch - I would attach the 1 - 3 Watt LEDs directly to (or through the printed circuit board) copper tubing, and 'pipe' this tubing up and down the wall.

I would then plumb the tubing into the watering system for the plants, running cool water through the tubing - which would also help cool the lit LEDs.

A bit unorthodox - but would really work - and the LEDs would not need much beyond this for heatsinking.

Regards,

James Jackson