Home made diffraction grating

Centro de Observação Astronómica no Algarve has a little program to make
diffraction gratings with a home printer. But it has some limitations. First
the resolution is maxed at 720 dpi and second the pattern is round wasting a
large amount of the transparency.

I have access to a HP 2200 laser jet printer that has a dpi of 1200. I
assume that is horizontal and not necessarily vertical but that should not
matter since you can make lines up and down just as easily as back and
forth, or for that matter set the printer to landscape. The idea appeals to
me because it is cheap, 50 cents for an 8 by 10 inch sheet of grating, and I
don't have to wait or pay for shipping. (Kind of along the lines of the
person who posted about frugal astronomy.)

So I have several question for the group.

1. Is the idea feasible?
2. Is 1200 dpi achievable or even desirable?
3. Would I be better off use 600 etching per inch in both quality and
defraction of starlight?
4. Are you better of putting the grating at the eyepiece or the objective?
5. Is there a simple way or producing this pattern in PhotoShop?

Thanks in advance and clear skies,
James King

On Thu, 26 Aug 2004 02:05:50 GMT, Lurking Luser
wrote:

Centro de Observação Astronómica no Algarve has a little program to make
diffraction gratings with a home printer. But it has some limitations.
First
the resolution is maxed at 720 dpi and second the pattern is round
wasting a
large amount of the transparency.

I have access to a HP 2200 laser jet printer that has a dpi of 1200. I
assume that is horizontal and not necessarily vertical but that should
not
matter since you can make lines up and down just as easily as back and
forth, or for that matter set the printer to landscape. The idea appeals
to
me because it is cheap, 50 cents for an 8 by 10 inch sheet of grating,
and I
don't have to wait or pay for shipping. (Kind of along the lines of the
person who posted about frugal astronomy.)

So I have several question for the group.

1. Is the idea feasible?
2. Is 1200 dpi achievable or even desirable?
3. Would I be better off use 600 etching per inch in both quality and
defraction of starlight?
4. Are you better of putting the grating at the eyepiece or the
objective?
5. Is there a simple way or producing this pattern in PhotoShop?

Thanks in advance and clear skies,
James King


1. I've made transparency gratings in the past, with mixed results.
Using a parallel-linear pattern on a 300 DPI printer, the finest
resolution was 150 lines per inch.

When placed over the objective, this caused a small spectrum angle which
was just right for a low power telescope ( 100x). The downside is that
the lines were very coarsely drawn, smearing out the spectra instead of
allowing a desirable resolution.

Two limiting physical constraints seem to be involved:
a. the ability of the Laser Printer to deliver smooth lines
(of whatever shape) at high resolutions.
b. the optical quality of the transparency film (also quite rough).

You might be able to do better if you use an optical cement to
glue the rough side of the transparency to some relatively
smooth plate glass. The cement should fill in the hollows
of the film, thus reducing the optical smearing introduced.

A very slightly curved set of lines may provide the widening
of the spectra to where you can make out emission and absorbtion
lines.


2. 1200 DPI is probably NOT desirable, unless you have a really
wide-field scope. Beside the finest grating you can print will be 1/2 of
the max resolution. You need gaps between every line pair to transmit
light!

3. Whatever resolution you use, the limiting factors will be your
telescopes magnification and how smooth the grating lines are.

4. I like the objective grating because you can then change eyepieces
to match the magnification and field size with the diffraction angle.

5. Instead of Photoshop, I used a CAD program. A more precise way might be
to send graphics commands to the printer, but that would
require a suitable knowledge of programming.

Cheers,
larry g.


--
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/

IMHO, a vector-based program such as Adobe Illustrator or Corel Draw
would be better for making a diffraction grating.
Vector-based graphics are much more scalable than raster graphics

Matthew Ota

Lurking Luser wrote:

Centro de Observação Astronómica no Algarve has a little program to make
diffraction gratings with a home printer. But it has some limitations. First
the resolution is maxed at 720 dpi and second the pattern is round wasting a
large amount of the transparency.

I have access to a HP 2200 laser jet printer that has a dpi of 1200. I
assume that is horizontal and not necessarily vertical but that should not
matter since you can make lines up and down just as easily as back and
forth, or for that matter set the printer to landscape. The idea appeals to
me because it is cheap, 50 cents for an 8 by 10 inch sheet of grating, and I
don't have to wait or pay for shipping. (Kind of along the lines of the
person who posted about frugal astronomy.)

So I have several question for the group.

1. Is the idea feasible?
2. Is 1200 dpi achievable or even desirable?
3. Would I be better off use 600 etching per inch in both quality and
defraction of starlight?
4. Are you better of putting the grating at the eyepiece or the objective?
5. Is there a simple way or producing this pattern in PhotoShop?

Thanks in advance and clear skies,
James King

IMHO, a vector-based program such as Adobe Illustrator or Corel Draw
would be better for making a diffraction grating.
Vector-based graphics are much more scalable than raster graphics

Matthew Ota

I think a CAD program is probably best, they are designed to print exactly to
the page and setting the line spacing is super easy. If everything is set
correctly, aliasing should not be a problem.

But the limitations of the printer are significant, dots are not squares so the
lines may have quite rough edges. 1200 DPI does not necessarily set the dot
size at 1/1200th of an inch, especially with inkjet printers. It just means
that there are 1200 dots of some size, probably greater than 1/1200th....

jon

Jon Isaacs wrote:

IMHO, a vector-based program such as Adobe Illustrator or Corel Draw
would be better for making a diffraction grating.
Vector-based graphics are much more scalable than raster graphics

Matthew Ota


I think a CAD program is probably best, they are designed to print exactly to
the page and setting the line spacing is super easy. If everything is set
correctly, aliasing should not be a problem.


Well, CAD programs are vector based....



But the limitations of the printer are significant, dots are not squares so the
lines may have quite rough edges. 1200 DPI does not necessarily set the dot
size at 1/1200th of an inch, especially with inkjet printers. It just means
that there are 1200 dots of some size, probably greater than 1/1200th....


Well, then you take it to Kinkos or another print shop that as the
technology.

Matthew Ota



jon

Actually, it really doesn't matter which you use. The first problem is
getting the printer to do equally sized lines and this is difficult to do
with a printer being driven by a windows type program. The only real way to
control this is to use the printer language directly to draw the lines.
Next is that the dots (sort of round but not entirely so) drawn by the
printer will not make a good line so you are reduced to drawing lines of
several dots wide in order to keep the percent error of width of the line to
a reasonable figure.
Much better if you plan on making accurate gratings is to make a grating
engine. This is basically a step and repeat machine that draws lines across
a material with a lot better accuracy. Drawing etched lines across a glass
plate and then filling the etched part with India ink will do a lot better
than any piece of acetate or mylar with dots of a printer on it will ever
do. You will also be able to do a lot of Ronchi gratings for your friends
and sell them to the community as well.
Another method is to make up a large image of the lines and photograph them
but that runs into problems with very large images as negatives get
relatively expensive as the size goes up although the printed circuit board
industry has some fairly low cost film available with the laser printers
that will allow for the smooth lines needed for the job.

--
Bob May
Losing weight is easy! If you ever want to lose weight, eat and drink less.
Works every time it is tried!

Well, CAD programs are vector based....

True but they are different in that they are designed to objects input as very
accurate dimensions. So one can easily draw 3000 lines across a page that are
1/600th of an inch wide and spaced at 300 lines per inch.

Well, then you take it to Kinkos or another print shop that as the
technology.

Matthew Ota

Still using dots....

Jon

"Lurking Luser" wrote in message
.net...
Centro de Observação Astronómica no Algarve has a little program to make
diffraction gratings with a home printer. But it has some limitations.
First
the resolution is maxed at 720 dpi and second the pattern is round wasting
a
large amount of the transparency.

James,

I believe that the Amateur Scientist column in Scientific American had at
least one article on actually ruling a diffraction grating. (Just in case
anyone is really ambitious.)

Clear skies, Alan

If the purpose is to make a ronchi grating you could try the DYI nylon
monofilament version

http://www.geocities.com/CapeCanaveral/9601/ronchi.htm

If you're interested, the April 1975 issue of Scientific American
magazine's "Amateur Scientist" column gave the details of a homemade
diffraction ruling engine. A fascinating read...

Clear nights!

Jim