From: ronkean@juno.com
Date: Thu Feb 03 2000 - 21:33:36 MST
On Thu, 3 Feb 2000 12:06:06 EST EvMick@aol.com writes:
> In a message dated 2/3/00 10:44:50 AM Central Standard Time,
> butler@comp-lib.org writes:
>
> > Um, a 50% efficiency figure for LEDs sounds a bit on the high
> side. I'm confident that the efficiency is high. Last I heard, the
white
> light ones are about to push the efficiency of
> fluorescents/gas-discharge, somewhere around 18%(?), which is pretty
darned cool. Ultraefficient red LEDs might be quite a bit higher, like
30-35%.
> >
Lamps, including LEDs, are normally not rated in terms of efficiency
(watts of light per watt of electrical power input) because human
perception of light varies with the spectral character of the light.
Instead, LEDs are usually rated in millicandelas of luminous intensity at
the angle of peak intensity, at some recommended operating current. And
lamps in general, including LEDs, are rated in luminous efficacy, which
is in units of lumens per watt, at some recommended operating current,
voltage, or power.
A standard (non-halogen) 100 watt 120 volt incandescent bulb has a
luminous efficacy of 17.5 lumens per watt when operated at the
recommended voltage. The figure for an LED could be lower or higher than
that, depending on the specs of the LED. It varies a lot. Some LEDs put
out 10 or 20, or more, times as much light per unit of input power as
other LEDs. The luminous efficacy of an incandescent lamp or an LED can
be pushed higher by operating the lamp at a higher input power than the
recommended level, but that reduces the life expectancy. In practice,
LEDs are operated from a current regulator circuit, and the power
consumption of that circuit affects the net luminous efficacy of the
overall lighting system. The current regulator is usually just a
resistor, but that resistor can waste a lot of power. Incandescent lamps
operate directly from a constant-voltage power source and do not require
a current regulator circuit, as they are self-regulating.
To calculate the efficiency of a lamp from the luminous efficacy, one
needs to apply a conversion factor for the physical power equivalent of
light. That is 1.61 milliwatts per lumen at 555 nanometers wavelength
(green), which is the wavelength of the peak photoptic response of the
human eye. At other visible wavelengths the conversion factor is higher
(more milliwatts per lumen). So the maximum possible luminous efficacy
is 621 lumens per watt, and that would only be possible at a wavelength
of 555 nm.
Using the 1.61 milliwatts per lumen factor, and an efficacy of 17.5
lumens per watt, the nominal efficiency of a 100 watt incandescent bulb
would be 2.8%, but that would be technically correct only if the bulb
emits all of its light at about 555 nm. Since the bulb actually puts out
a yellowish white light, the actual efficiency, based on the physical
power equivalent of the light, is higher, probably somewhere in the range
5% to 8%. Specific information about the spectral content of the light
would be needed to calculate an exact efficiency figure. But an
efficiency figure can be a misleading expression of the performance of a
lamp, since the perception of light varies according to the spectral
content of the light. Comparing the efficiency of a red LED with the
efficiency of a soft-white fluorescent tube is like equating apples and
oranges, because of the color difference between the LED and the
fluorescent lamp, and its effect on the perception of light. Luminous
efficacy is a better practical standard of comparison than efficiency.
Ron Kean
.
>
> I dunno about efficeincy numbers but FYI LED's are replacing
> incadesants on trucks.
>
> My 48 ft RGN has all LED lites. They supposedly last almost forever
> and use a fraction of the power to produce more light. Very Bright
tail
> lites..
>
> The problem is they cost (a LOT) more. But the price is going down.
>
> EvMick
> Rogers Mn.
________________________________________________________________
YOU'RE PAYING TOO MUCH FOR THE INTERNET!
Juno now offers FREE Internet Access!
Try it today - there's no risk! For your FREE software, visit:
http://dl.www.juno.com/get/tagj.
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:26:36 MST