On Tue, 5 Feb 2002, Sid Washer wrote:
> I will not attempt to wrap those 2 poor neurons around linear or
> non-linear mixing theories since I do not possess the mathematical
> background;
Actually I think I partially misspoke. As I recall I mentioned "beat
frequencies" which would be caused by the waves summing linearly. This
would be an amplitude or envelope effect. Basically when the crests of
the waves line up you will get a maximum amplitude, and when a peak and a
valley are coincident they'd cancel out. So strictly speaking the
"beat" is the amplitude fluctuation that you hear when tuning musical
instruments and the notes are slightly offset from one another in
frequency.
If you combine two waves in a non-linear way things get much more
complex. The resulting summation is usually represented as a Taylor
Series expansion and has terms that show harmonic components. In other
words, some of the energy of the original two waves gets converted into
various multiples of each of the original waves. Basically you can create
new frequency terms beyond those of the two original signals.
So ... what I was thinking is that LED's are fairly monochromatic. In
other words, they only emit a fairly narrow band of frequencies. Simply
illuminating a space with red, green, and blue LED's should give you the
linear summation of each of the frequencies emitted by these light
sources. Varying the intensity of the LED's should only vary the
amplitude of each of the terms in the summation. There shouldn't really
be any frequency conversion, so if none of the LED's emit the frequency
you want, varying the relative intensities of the sources isn't going to
give you that frequency. The upshot is that this wont give you a nice
continuously tunable light source across the entire visible spectrum.
On the other hand, if the LED's are chosen so that they have overlapping
frequency spectra, it may be possible to modulate the intensity of bands
of frequencies by controlling the intensities of the LED's. Controlling
the intensity of the center of a the emission curve would be done by
varying the intensity of that specific color of LED. However, for
frequencies on the skirts of the emission curves you'd adjust the
intensity of both of the LED's whose emission curves overlap at this
point.
Hope this helps explain what I was thinking.
> I simply build, work and have fun doing both.
Yes, and often this is the most important thing of all.
> The neurons will shortly get a bath of my own good home brew cassis
> stout.
Sounds good - Enjoy! (Besides, one can always argue that natural
selection plays a role here. Only the strongest and fittest neurons
survive <wink>.)
- Wayde
([EMAIL PROTECTED])
--------------------------------------------------------
ISART 2002
International Symposium on Advanced Radio Technologies
http://www.its.bldrdoc.gov/meetings/art/index.html
--------------------------------------------------------
_______________________________________________
Cameramakers mailing list
[EMAIL PROTECTED]
http://rmp.opusis.com/mailman/listinfo/cameramakers
