LED grow lights / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

LED grow lights

previous 1 2

Move Thread

LAN_403

Sulaiman

Sat Nov 09 2013, 06:30PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I have a waterproof self-adhesive strip of 300 5050 RGB LEDs
(eBay Chinese, lucky @ GBP5 with IR controller, p&p)
and they seem fairly efficient and run only slightly warm

" 12 v " is because each three LEDs are in series with a resistor embedded along the strip
the LEDs turn on dimly at around 10 V and are probably over stressed at 14 V.

If you have a very stable mains supply (such as here in UK)
AND you give due care to insulation and safety (EVERY part will be LIVE)
then you could run many strips in series from directly rectified mains.
You could use just a bridge rectifier and no capacitance (possibly a little extra series resistance)
the LEDs will pulse at twice mains frequency, which in itself is not a problem
you could use a (smoothing, reservoir) capacitor and have smooth (with a little ripple) dc
which will allow maximum light output.

I've no idea what the actual light output is, quite a lot, but so cheap (if you bid patiently) must be worth a try. The strip I have has solid transparent silicone rubber front
(which if un-damaged I'd trust to protect me from occasional brief contact, you decide for yourself)
and 3M adhesive backing.
(probably ok at line voltages if dry and un-damaged).

For safety I'd prevent access to the assembly, e.g. height, glass, transparent plastic, netting etc.
P.S. for UK, 240 x sqrt(2) - ripple etc. is about 320 to 330 V, if Blue and Red pass 1A each,
then you'll have 650 W of LED striplight... fan assisted ventilation is required.
P.P.S. the LED RGB strips are arranged as three 5050 RGB packages in series with a resistor per colour,
100 cells in parallel, the strip can be cut at ant cell boundary, soldering is not difficult
So, 325/12 = 27 series strips!
Could do it for an experiment with one strip cut into 27 mini-strips of 3 cells each.
leaves 19 cells for 'playing'

klugesmith

Sun Nov 10 2013, 02:42AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Andy wrote ...
Reading up it says 50-90lm is a watt of light, 683lm seems abit high?.

Andy, your "reading up" taught you wrong. Let's try to set you straight.

One common stumbling block: the nominal power of a LED is the electrical input power.
That's different from the nominal power of a laser diode, which is its radiant output power.
Remember that even in the best LED's, less than half of the input power is converted to light.
Most of the electricity simply heats the apparatus, just as in an incandescent bulb. Except an LED works best when its active parts are cold.

- - -

You know that illumination imparts heat to things, just as irradiation by infrared does.
Radiant power at any wavelength is measured in watts, and the most fundamental measurement is to irradiate a practically "black" body and measure the resulting heat.

If the wavelength happens to be in the visible spectrum, we say the black body is illuminated (a special case of irradiated). Then we can measure the "luminous flux" in lumens, which are like radiant watts with a wavelength-dependent scaling factor that matches human vision. (search for"luminosity function").

One lumen is defined to be 1/683 watt of light at 555 nm. The sensitivity curve peaks at that wavelength, a yellowish green. So a radiant watt of that optimal green color is 683 lumens, just as Sulaiman said.
Near the edges of the visible spectrum, a radiant watt of 450 nm blue or 670 nm red would be 20x less potent, only 34 lumens. "White" light is roughly 280 lumens per radiant watt. A 100% efficient white lamp would produce that many lumens from 1 watt of electricity.

I'll bet that in serious work on colored grow-lights, they discuss irradiance in watts per unit of vegetated area. Not illuminance in lumens per unit of area. Because the plant response does not match our eyes's extreme changes in sensitivity within spectral ranges of a couple dozen nanometers.

Verstehen Sie?

Andy

Mon Nov 11 2013, 07:38AM

Registered Member #4266 Joined: Fri Dec 16 2011, 03:15AM
Location:
Posts: 874

Thanks
With the scaling, a plant uses more of the spectrum than our eyes, so would that mean 100lm would produce more watts to the plant than if its lighting a room....or 90watts electricity to a LED at say 75% efficiency(for the plant), than 10% for eyes?

BigBad

Mon Nov 11 2013, 04:24PM

Registered Member #2529 Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600

Yeah, lumens are the wrong measure, since it's scaled to how bright it looks to the human eye.

When I built my LED array for a different application, I had a similar issue; it was UV-LEDs, the lumens were quite low, but it's was actually reasonably high power; many watts. It was right pain because the manufacturers usually quote in lumens, when I wanted optical output power.

You want something like 'plant lumens' <- I made up a new term

Conundrum

Thu Nov 14 2013, 06:12AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Ha!
I did read somewhere that LEDs from flat panel TVs are now being made with RGBY because using separate diodes allows the colour to be adjusted over the entire screen which boosts contrast.
The technology to make separate diodes is more reliable than blue+phosphor because the phosphor discolours with use and for a Â£5800 TV this makes a lot of difference to the lifetime.

previous 1 2

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob