Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: homemade photometer



The n+1 method of removing the wavelength dependency is counter intuitive (I'm
missing something?), but won't matter if one is using a monochrometer.
Stopping is dangerous because the detector is likely to have a position
sensitivity dependence. It's often important the incidence beams have the same
size and position (WRT the detector).

bc

P.s. Better Q. Polaroid film is relatively neutral density, and if mounted on
goniometers, one may use Malus' law and only two films. If not neutral enuff,
one may use two uncrossed for reference (moon) and the other two crossed -- this
will make greater demand on the accuracy of the goniometer. Wrong!!! 89deg
59' and 55" gives a reduction of ~ 3E-5 while 50' gives ~6 E-5 rather poor
resolution and an order of magnitude insufficient reduction. An optical
attenuator is required -- uses a series of convex mirrors plus I think (~1962)
concave for collimation. It, thereby, is wavelength independent (aluminum
reflectivity limit). I think the exit beam was the same as the entrance for the
one I saw.


"John S. Denker" wrote:

At 08:56 AM 11/17/00 -0500, Gordon Smith wrote:
I'd like to take a slice of the spectrum (say, a bit of
yellow), and compare the intensity levels directly from the sun and
reflected off the moon as part of the lab.

Hmmm.

I suppose I could do this
directly on the white light version, but wouldn't that be more prone to
burning out the meters?

If burn-out is the main concern, then a prism is not the best line of attack.

First of all, if the _same_ slice of spectrum is used in both cases, then
it is unlikely that any reasonable detector can handle both signals,
because of the extreme difference in intensity. (And if _different_ slices
of spectrum are used, then all sorts of hard-to-control experimental errors
creep in.)

Suggestion: Assuming you have a detector that can measure the moon
(magnitude -12.5), then you can use the same detector to measure the sun
(magnitude -26.8) if you attenuate the signal properly. I figure if you
use the filter-elements out of cheap sunglasses (85% attenuation) then you
need something like 7 of them in series. You can work out the exact details.

http://web.missouri.edu/~physwww/astronomy_one/html/magnitudes.html

Or you could use an aperture stop in combination with attenuating filters.

If you want to be really professional about it, assume the filter contains
some attenuation plus some wavelength selectivity. Assume the wavelength
selection is idempotent. Then put one filter in front of the moon and N+1
filters in front of the sun. That gives a relative attenuation of
(something)^N, while establishing the same wavelength-dependence for both
measurements, to a decent approximation.