Previous: EXPOSURE TIME?
Up: EXPOSURE TIME?
Next: Time limits
Previous Page: EXPOSURE TIME?
Next Page: Time limits

Signal-to-noise

For stellar objects:

with

Table 1. CCD Parameters at JKT f/15 focus

Table. ~Observational data, Kitt Peak broadband filters.

(Details of current values of readout noise, dark count and cosmic ray events can be found in the performance folder for each chip compiled by CDP and available in the control room).

CCDs are commonly used to image very faint objects. Most faint objects beyond 20 mag are galaxies, but their extent may be similar to or smaller than the seeing disk, so that the stellar-object s/n calculation above still applies. Alternatively, if the extent of the object is known to be , the s/n can be calculated from (1) by ``modifying'' the seeing FWHM from to

For very extended objects the s/n as a function of time is given by:

and where is object surface brightness in magnitudes/arcsec.

Table 1 indicates pixel sizes and readout noises, values of N, and are given for the La Palma site in Table 2, and the geometric aperture of the JKT is 0.785 .

Some sample sums for stellar objects are shown in Fig 6. These curves may well provide adequate information to estimate exposure times for most JKT CCD observations. The dramatic effect of seeing is only crudely modelled by the approximation used in (1), which is that the usable signal is spread into an area occupied by the half-power area of the seeing disk. (Clever analysis might recover more signal at the higher values of s/n, but is unlikely to do so at low s/n.) Observations cannot be continued to infinitely faint limits. For short exposure times, s/n signal/readout, i.e. s/n t, the noise-limited case. But for faint objects and long exposure times, s/n , i.e. s/n , the sky-limited case. The point at which the observation becomes sky- limited, the ``knees'' in Fig 6, marks the point of diminishing returns.