We observed IP Pegasi at the RIT Observatory on two nights:
We used a 10-inch Meade LX200 telescope with f/6.3 focal reducer and SBIG ST-8 CCD camera with Johnson-Cousins V-band filter. Exposure times were 30 seconds on Oct 31 and (mostly) 15 seconds on Nov 01 (there were a few 30-second exposures taken on night of Nov 01, too). The telescope wasn't aligned very well, so there was often visible trailing in the 30-second exposures. I subtracted a median dark frame of the same length as the exposure from all images, but did not apply a flat-field correction because I'm still learning how best to create flats with this equipment. The comparison stars I used are close to the variable, but this lack of flatfielding might contribute to the errors.
The field of view was roughly 24 by 17 arcminutes, with pixels about 1 arcsec on a side. The FWHM
I used synthetic aperture photometry to measure the brightness of IP Peg and nearby comparison stars on each frame. I chose a small aperture (radius 4 pixels) in images with a good PSF, and a larger aperture (radius 6 pixels) in images with a big PSF (the late images on Oct 31).
I applied the method of ensemble photometry, as described in Honeycutt (PASP, 104, 435, 1992) to IP Peg and five nearby comparison stars. The observations were divided into 5 groups, and each group was analyzed as an independent ensemble. When I compare the relative magnitudes of comparison stars in the different groups, I find that they agree with each other, to within the estimated uncertainty in each group's mean relative magnitude.
The chart below is a subsection of an entire frame, oriented with North up and East left.
Below are some plots showing the relative V-band magnitudes of IP Peg and the comparison stars. The bright star "A" is set to magnitude zero in these plots, and in the data files below.
IP Peg and comparison stars, night of Oct 31. The graph is exactly 6 hours wide.
IP Peg and comparison stars, night of Nov 01. The graph is also exactly 6 hours wide.
IP and comparison stars, both nights graphed together -- but note that the second night's data has been shifted in time so that it is closer to the first night's data, just to make it easier to compare the two.
In this last graph, I have drawn horizontal lines near the mean value of each comparison star. They make it easy to check for systematic trends in the comparison star magnitudes. From top to bottom on the graph, the comparison stars are
star mean V mag relative to A typical scatter ------------------------------------------------------------- A 0.00 0.005 C 1.57 0.02 B 1.75 0.02 D 1.87 0.03 E 2.90 0.06
The 30-second exposures on Nov 01 show slightly smaller scatter than the 15-second exposures; evidently, the small amount of trailing didn't smear the signal out over a large enough area to decrease the signal-to-noise ratio very much.
The data are divided into 5 groups, mostly due to gaps caused by an Introductory Astronomy class using the telescope for a couple of hours each night.
The times recorded in these files are (Julian Date - 2,450,000) at mid-exposure, but there are caveats:
time of end of exposure +
time to download from camera +
time to convert to FITS and write to disk
I did a couple of experiments to estimate how long the extra
steps last, and then calculated the time of midexposure
accordingly. However, this is a major source of uncertainty.
My guess is that the times are good to +/- 2 seconds or so, relative to the proper Universal Time. I made no corrections for heliocentric or barycentric delays.
The data files are available below. Each one is a plain ASCII text file, with lines that look like this:
6 180.63 354.06 11 11849.54457 0.000 0.004 0.011in which
col 1 is an ID number for the star (not same on both nights)
col 2, 3 are (row, col) on chip
col 4 is index of frame within the set
col 5 is Julian Date - 2,440,000 at midpoint of exposure
col 6 is average relative magnitude of star in the set
col 7 is magnitude of star in that image, relative to A
col 8 is uncertainty in relative magnitude measurement
To make a light curve, one would plot col 7 versus col 5.