Re: Long Prisms for homemade spectrometer

[Leigh Palmer <palmer@SFU.CA>]

>  > One doesn't simply pass sunlight through a prism to demonstrate the
> >  Fraunhofer absorption spectrum of the Sun. That works for Newton's
> >  demonstration of the phenomenon of colors, but for the narrow lines
> >  one needs lenses to collimate and focus the radiation, and an
> >  accurately aligned entrance slit as long as the desired spectrum is
>
> However, a cheap grating and slit brings out the lines just fine...

Yes. That is a superior way to demonstrate the phenomenon.

> I see faint lines in sunlight and cloud reflected sunlight through my
> el-cheepo $20 Project STAR Diffraction grating spectroscope.
> https://www.mailordercentral.com/aspsky/prodinfo.asp?number=OA%20160&variation=&aitem=8&mitem=9
> is where to see the device.

How large is this spectroscope? The distance from eyehole to
slit? I can't tell from the ad.

> I have been blissfully assuming these lines are fraunhofer absorption lines,
> an interpretation the STAR spectrometer folk have fostered in the
> accompanying literature that comes with the device.  If these are NOT
> Fraunhofer absorption lines, I'd like to know what they are.
>
> Otherwise, I assume the much greater dispersion of the grating overcomes the
> need for a slit any better than that provided with my toy...

Yes, the greater dispersion of a grating does diminish the need
for a slit. If no lens is used, however, the projected spectrum
will still be unresolved at the geometrical level represented by
the Sun's apparent disk, approaching a half degree. In your
spectroscope I expect you will find an eyepiece lens which moves
the virtual image of the slit to infinity. You are not looking
at a projected spectrum; you are looking through the grating at
a slit-shaped virtual source.Spatial coherence is not limited by
the size of the Sun's disk in that case.

If the spectroscope distance mentioned above is small (i.e. less
than 25 cm) this is the case.

No one has said anything about using a 45-45-90 prism for this
purpose.

Leigh