Richard Clark wrote:
On Mon, 03 May 2010 18:28:25 -0700, Jim Lux
wrote:

Roy Lewallen wrote:
Jim Lux wrote:

Hi Jim,

Much of what you write looks like stream-of-consciousness writing.
Did/do you have a point?

The plastics degradation is definitely an "athermal" effect (because
adding carbon black to the plastic inhibits it, but doesn't change the
absorbed power very much.

UV radiation has migrated awary from electron/atom issues to
molecular/ionic bond issues. Calling it "athermal" seems to be
returning the discussion to the metaphysical.

All in a thread about temperature and heat..
That was actually in response to Roy's original comment
"When doing experiments with the sun's rays, you sometimes get
non-intuitive results, because there's a lot of energy (heat) at
wavelengths we can't see, particularly at the ultraviolet end."
and my response that there actually isn't much energy in the UV end.

Roy commented about sunburn, and I pointed out that the mechanism in
sunburn isn't thermal (and this is important to folks who worry about
RF exposure limits and regulatory compliance.. thermal effects have one
biological result, athermal effects are another..)

My comment was that sunburn (and Roy's example of plastic degradation)
are due to the energy of UV photons actually causing a chemical
reaction, as opposed to making something happen because of heat.





But..
note that the scale is in wavelength and the energy is "per nm" (because
that's how spectrophotometers work). the photons have less energy at
lower wavelength. (or, you could plot it in frequency, and then look at
the watts/Hz to integrate)

What is the significance of changing from wavelength to frequency?
(But?)

Roy's comment was about the amount of energy in the non-visible bands
(presumably in response to my comment that human eye sensitivity tends
to match that of the solar spectrum/ 5250K blackbody), and he cited the
very commonly seen graph in W/nm, with a scale linear in nm.

My point is that in the RF world, we tend look at power spectral density
in terms of W/Hz, so when you are looking at the graphs (with a linear
scale of wavelength or frequency, as apppropriate), a visual estimate of
the "integrated area under the curve" can be misleading. If you plot
the same data, but in W/Hz, and with a scale linear in frequency, you
get a very different looking graph.

Try it.. the equation is of the general form
power density (per hertz) =
constant1*frequency^3/(exp(constant2*frequency/T)-1)

power density (per unit wavelength) =
constant1/lambda^5 * 1/(exp(constant2/(lambda*T))-1)