# ls14 - AST 3722C spring 08 class 14 tues april 15 We're in...

This preview shows pages 1–3. Sign up to view the full content.

AST 3722C - spring 08 - class 14. tues april 15 ---- We're in chapter 18. -- detectors. last time -- detailed stuff on statistics, getting S/N from a CCD. More bits: So we had the CCD equation: S/N = n * t / sqrt( n * t + npix(nsky * t + ndark * t + nrd^2)) Often the sky will dominate, so: S/N = n * t / sqrt( n * t + npix * nsky * t ) S/N = sqrt(t) * n / sqrt(n + npix*nsky) propto sqrt(t) In some cases the sky will contribute very little -- moonless nights, blue wvln where QE is low, but CCD has not so much dark current S/N = n * t / sqrt( npix * nrd^2 ) S/N = t * (n / nrd) / sqrt(npix) propto t In the general case, you can solve for t to get a specific S/N, just use algebra and quadratic formula. S/N = n * t / sqrt( n * t + npix( n_s*t + n_d*t + n_r^2 ) ) t = (-B +/- sqrt(B^2 - 4AC))/2A usually only 1 solution will be physical (real). A = n^2. B = -((S/N)^2)*(n + npix(nsky+ndark)) C = -((S/N)^2)*npix*nrd^2 How is S/N related to error in the measurement? Fairly simply. The fractional error is 1/(S/N). So if S/N=3, then the flux density is f +/- 0.333f. If S/N=10, then the flux density is f +/- 0.1f. How does that relate in magnitudes? Be careful: m = -2.5*log(f/f0) = -(2.5/ln10)ln(f/f0) sigma_m^2 = ( -(2.5/ln10)(f0/f)(sigma_f/f0) )^2 sigma_m = (sigma_f/f)(2.5/ln10) = (1/(S/N))*(2.5/ln10) = (1/(S/N))*1.086. so actually S/N is closely related to error in magnitude in a straightforward numerical way. S/N=3 --> error in mag is ~0.33 (really 0.362) S/N=10 --> error in mag is ~0.10 (really 0.109) S/N=100 --> error in mag is ~0.01 (really 0.011) This should give you a clue about quality of a measurement simply based on the magnitude error that's quoted. If uncertainty is 0.3-0.4 mag, it's basically a really poor detection. Last thing: some typical numbers about CCD -- handout #29 - page 45 from Howell. handout #30 - typical flat field-- dust donuts! -------------------------------------- Sec 15.7,15.8,19.9,

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Next thing - some info about spectroscopy. Taking pictures in astronomy is very useful but you get the _most_ info about things with spectra. Spectra tell you about - composition, temp, speed -- remember that from AST 2002? they can also tell you other stuff too: surface gravity, vertical structure, magnetic field strength. Spectra are extremely powerful data for galaxies, stars,
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 11/09/2009 for the course AST 4700 taught by Professor Fernandez during the Spring '09 term at University of Central Florida.

### Page1 / 5

ls14 - AST 3722C spring 08 class 14 tues april 15 We're in...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online