DIY LED lights

[naoki]

Martien, I guess that the names for the colors may be somewhat field-specific. You are right, low wave lengths in infra-red is also called near IR or IR-A. But in some biology (plant physiology), we also use "far red" to mean the same thing as IR-A (at least in the US). For example, phytochrome has 2 forms, and one form which absorbs 700-740nm is called P_fr (far red). Phytochrome - Wikipedia, the free encyclopedia Some people also use the abbreviate of P_ir. So we use it interchangeably. Now it may not be the convention among the engineering people, and they may not use the word "far red". Also, some LED makers seem to use "deep red" (different from far red) to indicate 670-690nm range.

So I think you are correct, and my usage is also accepted in biology (I didn't mean to be nit-picky, it's just for your info!). Similarly, horticultural people seem to use '4n' to indicate tetraploids. But we, plant biologists, usually use '4x' for it ('2n' etc has a different meaning in plant biology).

[DavidCampen]

As Naoki says; for example, this company uses "DeepRed" to describe their emitters that have a peak emission at 660 nm while they use "Far Red" to decribe their emitters that have a peak emission at 740 nm:
LED Engin

And, as Naoki mentions, while 700 - 740 nm light is not used for photosynthesis it does promote the conversion of the phytochrome form called Phytochrome-FR to the form called Phytochrome-R. This is significant because plants that are daylength sensitive use the ratio of the two phytochrome forms as part of the mechanism that controls flowering. Phytochrome-FR is quickly formed during the day from Phytochrome-R and at night Phytochrome-FR is slowly converted back to Phytochrome-R but light with a wavelength in the 740 nm range quickly converts Phytochrome-FR to Phytochrome-R and thus can be used to simulate long nights wrt flowering.

[ALToronto]

Thanks. This raises my next question - assuming I should supplement with separate far red diodes, what % of the total light output should the far red radiation be? If I'm using ten 5W white LEDs in one fixture (a mix of white colours, probably 5 @ 6500 K, 3 @ 5000 K and 2 @ 3000 K - any comments on the mix?), how many far red LEDs should I add to these, and at what wattage?

If I use them, I will get a separate driver for them, as I think that 5W would be excessive power for just the far red, and 1 W or even less would be more reasonable.

[DavidCampen]

The Far Red, 700-740 nm, you would only use at night and at low levels, to convert Phytochrome-FR to Phytochrome-R thus making the night seem longer to the plant. Since many orchids are not daylength sensitive there wouldn't be any general need for 700-740 nm in your lighting.

[ALToronto]

What about 690 nm?

---------- Post added at 07:45 PM ---------- Previous post was at 07:44 PM ----------

And any comment on my proposed mix of white LEDs? I asked for spectral analyses, so this will be a good way to test if this particular manufacturer knows what they're doing.

[DavidCampen]

Originally Posted by ALToronto

What about 690 nm?

---------- Post added at 07:45 PM ---------- Previous post was at 07:44 PM ----------

And any comment on my proposed mix of white LEDs? I asked for spectral analyses, so this will be a good way to test if this particular manufacturer knows what they're doing.

690 nm photons are very close to not having enough energy for photosynthesis. I would not want to use LEDs that had a specified peak emission wavelength longer than about 660 nm.

People seem to do fine growing with white fluorescent lights without having to worry about fine tuning the spectral composition of the white light. I would expect the same to be true for white LEDs.

[Ray]

Unfortunately, the white LED spectra I have seen show very low emission in the red end of the spectra, with a sharp blue peak and a big "dome" centering on the green in the 580 nm region.

[DavidCampen]

Originally Posted by Ray

Unfortunately, the white LED spectra I have seen show very low emission in the red end of the spectra, with a sharp blue peak and a big "dome" centering on the green in the 580 nm region.

When the manufacturer is trying to get the most lumens per watt then they use a lot of green because green has the greatest luminous efficacy. I believe that this color spectrum is called Cool White; it is a really horrid color IMO. An LED labeled as Warm White will give fewer lumens per watt but will have its peak emission shifted from green to red.

Compare this Warm White spectrum (page 11)
http://www.ledengin.com/files/produc...LZ4-00WW08.pdf

to this Cool White spectrum (page 11)
http://www.ledengin.com/files/produc...LZ4-00CW08.pdf

[ALToronto]

The spectral graphs I received are very similar to the ones posted by David. But doesn't cool white peak at blue? And isn't blue important?

I'm still planning to order a mix of 2700K and 6500K LEDs; I'm just not sure how I should proportion them - two warm for every cool maybe?

[lepetitmartien]

For IR after 690 nm you're better set with an halogen lamp a little every day. 690 on the other side is very beneficial David, it's still in the chlorophyl A red peak. Growers in France with White or Royal Blue/Red mixes found beneficial to have a few 690 in the mix.

Naoki, call me lpm No harm done, as long as we know about which wavelength we talk about. I did my homework before answering (those engineers… tsssss LOL )