Help me escape from prism!

I've started working on a simple diagram of a prism and the
spectrum of light it disperses. Please tell me if I've already
made some error, or if it just doesn't look right, or if there
is anything about it that could be better if done differently.

http://www.freemars.org/jeff2/prism2.htm

Thank you!

-- Jeff, in Minneapolis

Subtract 1 from my e-mail address above for my real address.
..

"Jeff Root" wrote in message
om...
I've started working on a simple diagram of a prism and the
spectrum of light it disperses. Please tell me if I've already
made some error, or if it just doesn't look right, or if there
is anything about it that could be better if done differently.

http://www.freemars.org/jeff2/prism2.htm

Thank you!

-- Jeff, in Minneapolis

Subtract 1 from my e-mail address above for my real address.
.

Looks OK (though angles are exaggerated presumably for illustration). You
have the blue refracting more than the red, which is better than some
drawings I have seen... You also have the increasing dispersion towards the
violet which is a nice detail.

Can you generate a violet-looking colour beyond the blue? Some sort of
purple? Then reduce the other angles a bit?

--
Mike Dworetsky

(Remove "pants" spamblock to send e-mail)

Mike Dworetsky replied to Jeff Root:

Looks OK (though angles are exaggerated presumably for illustration).
You have the blue refracting more than the red, which is better than
some drawings I have seen... You also have the increasing dispersion
towards the violet which is a nice detail.

I did it without using any references. The only prisms at hand
are 45/45/90, not 60 degrees, and I haven't pulled them out in
a few months. The wavelength ranges of the colors are written
down here somewhere, but I couldn't even find that to guide me.
To get it right I should decide what the prism is supposed to be
made of and calculate the angles, but I'm too lazy to do that
unless I have to or somebody shows me the formula and convinces
me it really isn't that much work. Short of that, I figured I
could do it by eye.

While making another drawing of just the colors spread out, I
realized that the distance from green to blue seemed to be about
the same as the distance from red to green, and the distance
from yellow to green seemed to be about the same as from red to
yellow. Which surprised me since the peak sensitivities of the
color receptors in the human eye wouldn't have to match the
spectrum like that, and my techique of making the colors look
right on my monitor screen is so crude.

But it fits perfectly with the two things I want to illustrate.
The first is how all the infrared light -- a band considerably
wider than the visible band-- is squeezed into a narrow wedge
beyond the red, which explains why a thermometer will register
the highest temperature if put into the region beyond the red.

Can you generate a violet-looking colour beyond the blue?
Some sort of purple? Then reduce the other angles a bit?

The second thing I'm trying to illustrate is what the colors
indigo and violet are, and how they fit into the spectrum.
My understanding is that there really is no distinguishable
indigo in the spectrum. Isaac Newton had his own reasons for
wanting there to be seven colors (much as Kepler wanted to fit
the orbits of the planets to the Platonic solids), and saw an
opening for indigo between blue and violet. The difference
between blue and violet is the fact that the red receptors in
the human eye are also moderately sensitive to light in the
part of the spectrum beyond the blue, so that both the blue
and red receptors are stimulated, resulting in violet.

I'll reduce the angles a bit, if it doesn't impair clarity.
If someone can point me to the formula and a reasonable value
for the index of refraction of my virtual prism, I may even
calculate the angles.

Thank you for your comments!

-- Jeff, in Minneapolis

..

Dear Jeff Root:

"Jeff Root" wrote in message
om...
Mike Dworetsky replied to Jeff Root:
....
I'll reduce the angles a bit, if it doesn't impair clarity.
If someone can point me to the formula and a reasonable value
for the index of refraction of my virtual prism, I may even
calculate the angles.

The index of refraction of all materials is a function of frequency, which
is why the different colors separate. How many iterations do you really
want to do to generate your rainbow?

David A. Smith

David A. Smith replied to Jeff Root:

I'll reduce the angles a bit, if it doesn't impair clarity.
If someone can point me to the formula and a reasonable value
for the index of refraction of my virtual prism, I may even
calculate the angles.

The index of refraction of all materials is a function of
frequency, which is why the different colors separate. How many
iterations do you really want to do to generate your rainbow?

I'd want to do the five primary colors I've already shown, plus
a small handful of additional reference lines beyond red at the
top and beyond blue at the bottom.

What are the frequency limits that a real material can handle?
Should I make my prism of diamond? Or what? Cost is no object.

Is it possible to have a single number for each transparent
material (in air, say), and calculate the resulting angle for
any frequency from that one number, using the same formula for
all materials?

-- Jeff, in Minneapolis

..

Jeff Root wrote:
David A. Smith replied to Jeff Root:


I'll reduce the angles a bit, if it doesn't impair clarity.
If someone can point me to the formula and a reasonable value
for the index of refraction of my virtual prism, I may even
calculate the angles.

The index of refraction of all materials is a function of
frequency, which is why the different colors separate. How many
iterations do you really want to do to generate your rainbow?


I'd want to do the five primary colors I've already shown, plus
a small handful of additional reference lines beyond red at the
top and beyond blue at the bottom.

What are the frequency limits that a real material can handle?
Should I make my prism of diamond? Or what? Cost is no object.

Is it possible to have a single number for each transparent
material (in air, say), and calculate the resulting angle for
any frequency from that one number, using the same formula for
all materials?

-- Jeff, in Minneapolis

.
A bit of info can be found at
http://www.astro.ufl.edu/~oliver/ast...asicOptics.htm

What you need is to get the index of refraction of your chosen glass
(maybe crown) at each of the wavelengths you consider typical of the
colors you want to depict. I show three at the URL above.

Hi Jeff:

"Jeff Root" wrote in message
m...
David A. Smith replied to Jeff Root:

I'll reduce the angles a bit, if it doesn't impair clarity.
If someone can point me to the formula and a reasonable value
for the index of refraction of my virtual prism, I may even
calculate the angles.

The index of refraction of all materials is a function of
frequency, which is why the different colors separate. How many
iterations do you really want to do to generate your rainbow?

I'd want to do the five primary colors I've already shown, plus
a small handful of additional reference lines beyond red at the
top and beyond blue at the bottom.

What are the frequency limits that a real material can handle?
Should I make my prism of diamond? Or what? Cost is no object.

Is it possible to have a single number for each transparent
material (in air, say), and calculate the resulting angle for
any frequency from that one number, using the same formula for
all materials?

http://online.cctt.org/physicslab/co...refraction.asp
down to the first graph...

A real material will have absorption bands (where n will perhaps be high)
but light is scattered internally to the prism. This is not shown on their
graph.

It appears a "simple" second order equation would do, as an approximation.
Perhaps you could leave the actual values to a user's imagination (I think
we're talking about a program of yours...). Perhaps a pull down selection
box, with "custom" as a choice.

If you're playing.

David A. Smith