HD189733b
Millimag Photometry Techniques With Bright Stars
Maintaining a fixed stellar field on the CCD during an observing run is critical to getting high precision photometry measurements. This is normally very easy to accomplish when operating the SBIG ST10 camera self-guided. High precison millimag photometry also requires exposures sufficiently long enough to minimize scintillation noise.When stars brighter than ~ V=9.5 are being observed it becomes necessary to purposefully limit the light reaching the CCD detector to facilitate longer exposures. This negatively impacts autoguiding with a self guiding camera as exposure times increase. One possible solution is to construct a neutral density filter such that the autoguiding ccd has a clear opening allowing unfiltered access to the telescopic fov. Another similar alternative using a neutral density filter will be described below.
A successful observation of the ingress of exoplanet HD189733b at ~
3948.83. Conditions were typical in East TN for this time of year.
Clear "hazy" skies with absence of major clouds.Ambient temp ~ 82F
with very high humidity ( dew points in the low 70s) .
These observations were conducted unattended from my suburban backyard observatory located in Knoxville, TN to test the Astronomer's Control Panel( v. 4.2) automation software's ability to re-center the target and subsequent effects on differential photometric precision. Instrumentation used was a 0.36m Meade SCT with an Optec TCF-s focuser mounted on an Astrophysics AP1200 GTO mount with an SBIG ST10XME/CFW-8a CCD binned 3x3. An Agos universal focal reducer was used inline producing an effective focal length of 2261mm (f/6.37) with the current spacing. The resulting image scale is 1.86 arcsec/pixel with a fov of 15.2 x 22.5 arcmin. CCD operating temp was set to -15C, operating with water cooling assist. 30sec exposures were utilized with a Schuler Is photometric filter. The target is a very bright star, V~7.67 . A neutral density (ND 2.0) film was placed over the Is filter cell to further attenuate stellar flux. The scope was autofocused with Focusmax then the temperature compensation feature was enabled in Maxim. This automatically maintained correct focus position for the Optec TCF-s for the duration of the observing run.
The ND filter blocks a significant amount of light (99% reduction over visual wavelengths) rendering autoguiding with the ST10's self guiding ccd impossible.ACP was used to re-center the image as necessary. I am fortunate in that periodic error is extremely low for my mount ( ~ 3" peak-peak uncorrected). At this image scale with typical stellar psf's ranging 3.5"-4" PE is not even noticed. Polar alignment is such that stars drift ~0.01-0.02 arc-minutes/minute. ACP plate solves each science image then re-centers the target if drift >0.1 arc-min. This resulted in stellar positions deviating only a few pixels from their initial location over the course of the observing run. This can be observed in the animated gif below. Calibration flats were acquired at dusk. Dark and Bias frames were acquired at the same operating temp and integration time as light frames. Master calibration frames consist of 10 flats median combined, 40 bias median combined, and 40 darks sigma clipped combined. Acquisition and calibration performed with Maxim DL/CCD/ACP with AIP4Win v2 utiliized for the photometry analysis. Data manipulation and charting were performed with MS Excel 2003.
HD189733 is in the center of the fov. This is a sequential series over
a period of 25 minutes. 30 second exposures unguided.
ACP is re-centering the images if drift is > 6 arc-secs
I discovered that images corresponding to the large scatter in
the data had pixels of the target object that were clearly outside of the
linear regime for my detector running between 45k and 55k counts in the
measurement aperture. One sigma SD for the comparison star, HD 345459 @
magnitude 8.09 was .0044. I would expect higher precision with shorter
exposures or operating at 2x2 binning as this would result in pixels that
are not saturating.