Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/ftp/instrument_news/FOS/pub/fos_isr102.txt Дата изменения: Tue Oct 25 23:51:15 1994 Дата индексирования: Sun Dec 23 02:46:56 2007 Кодировка: Поисковые слова: п п п п п п п п п п п п п п

PRE-COSTAR FOS APERTURE THROUGHPUT VARIATIONS
DUE TO OTA FOCUS CHANGES

D. J. Lindler and R. C. Bohlin

FOS Instrument Science Report CAL/FOS-102
93 August


ABSTRACT

Since the throughput of the FOS entrance apertures depends on
the HST focus, estimates of absolute fluxes from observed count
rates require a correction for focus effects. For the larger
apertures, the throughput is nearly independent of wavelength,
while for the smaller apertures, the dependence on wavelength is
important. Differences in the red and blue side transmissions
are typically less than 1% for the same aperture and wavelength.
Since January 1991, the historic record of the effect of
desorption on the HST focus shows deviations of -25 to +15
microns from the nominal value. The 4.3 arcsec aperture
throughput observed by a 1.43 arcsec high diode array varies by
approximately 10% over this range. The 1.0 arcsec aperture
suffers the worst photometric throughput variations of nearly
20%. Throughput corrections are modeled using calculations of
the aberrated PSF at the FOS entrance apertures in conjunction
with a desorption model and a history of secondary mirror
movements.

This correction procedure will leave OTA breathing as a dominant
source of photometric uncertainty, which can still be almost 10%
in the 1.0 arcsec aperture.


THROUGHPUT VARIATIONS WITH FOCUS

Fiqure 1 is a plot of the HST focus position versus time,
computed using a desorption model by Hasan (1993). The focus
position is given in microns from the nominal focus and has
varied between -25 to +15 microns since January 1991. Since
January 1993 the focus has remained within 5 microns of the
nominal position. Extrapolation of the desorption model shows
that the focus will remain within 5 microns of nominal until the
end of 1993.

As the focus changes, the throughput of the FOS apertures
varies. The Telescope Image Modeling (TIM) version 27 software
package (Burrows, 1991) is used to compute a point spread
function (PSF) for three wavelengths at the positions of the FOS
entrance apertures for the red and blue sides. The throughput
for each aperture is computed by integrating the PSFs over the
area of the FOS apertures. For the 4.3 arcsec square aperture
and the 0.25 x 2.0 arcsec slit, the area is limited to only the
portion observed by a 1.43 arcsec high diode array. For the
blue and red sides, respectively, Figures 2 and 3 show the
 Page 2


computed throughput for a target centered in the 4.3 arcsec
square aperture, the 0.25 by 2.0 arcsec slit, and the 1.0 and
0.3 arcsec diameter circular apertures. The focus variations
for the 4.3 and 1.0 arcsec apertures are approximately linear
and show little wavelength dependence (less than 1 percent for
the 4.3 arcsec aperture and 3 percent for the 1.0 arcsec
aperture). The slit and the 0.3 arcsec aperture both show
curvature versus focus position and significant variation with
wavelength.

PROPOSED CORRECTION TECHNIQUE

Using average inverse sensitivity curves for the 4.3 arcsec
aperture and a model of time variations in the inverse
sensitivity curves due to changes in the efficiencies of the
detectors and optical elements (Lindler and Bohlin (1993), the
following process can be used to convert an observed net count
rate spectrum taken in an arbitrary aperture to absolute flux
units.

1) Correct the net count rates to nominal focus=0 position using
the throughput changes for the aperture as a function of
focus (see Table 1). Table 1 gives the computed throughput
variations for the 5 most used FOS apertures relative to the
nominal focus position. The focus position can be computed
for the observation date using the data plotted in Figure 1.

2) Compute the expected count rates for the 4.3 arcsec aperture
using the relative aperture throughputs at the nominal focus.
The relative aperture throughputs computed for TIM point
spread functions are tabulated in Table 2. However, relative
throughputs from observed data may be more accurate (Bohlin
1993).

3) Multiply the count rates computed for the 4.3 arcsec aperture
by the average inverse sensitivity curve for the 4.3 arcsec
aperture.

4) Correct for time variations in the sensitivity using the time
changes in the efficiencies of the detector and optical
elements determined for nominal focus (Lindler and Bohlin
1993).

A significant source of error in applying the above process is
the Optical Telescope Assembly (OTA) "breathing", which is short
term variation in the telescope focus. Monitoring of the
breathing shows focus variations up to 14 microns from the focus
position predicted with the desorption model (Hasan 1993). A 14
micron error in the predicted focus position can result in a 5
percent error in the throughput estimate for the 4.3 arcsec
aperture and almost a 10 percent error for the 1.0 arcsec
circular aperture. Analysis of the FOS sensitivity monitoring
data corrected to nominal focus with our calculated throughputs
show deviations from a smooth model of time dependent efficiency
variations that are on the same order as those expected due to
 Page 3


OTA "breathing" (Lindler and Bohlin 1993)

A second source of error, which we can not presently quantify,
is in the model PSFs computed by TIM. These errors may be
signficant in the computation of the actual throughput values
(Table 2). Since our correction process only uses the
throughput variations relative to the nominal focus (Table 1),
the errors in the model may not be significant when compared to
other error sources including OTA "breathing", errors in
centering the target in the aperture, and errors in the YBASE of
the spectral observations.



REFERENCES

Bohlin, R. 1993, FOS Instrument Science Report, in preparation.

Burrows, C. and Hasan, H. 1991, Telescope Image Modelling User
Manual, Version 7, Release 25 and updates through Release 27.

Hasan, H. 1993, Instrument Science Report, No. OTA 12.

Lindler, D. and Bohlin, R. 1993, FOS Instrument Science
Report, in preparation.



FIGURE CAPTIONS

Fig. 1 -- A plot of the history and extrapolation of the HST
focus position versus time computed using the desorption
model computed by Hasan (1993). The position is given in
microns from the nominal focus.

Fig. 2 -- A plot of the relative throughput observed with a
1.43 arcsec high diode array for the FOS Blue Side 4.3 arcsec
square, 0.25 x 2.0 arcsec slit, and 1.0 and 0.3 arcsec
circular apertures versus focus position computed from model
point spread functions. The focus position is given in
microns from the nominal position.

Fig. 3 -- same as Fig. 2 for the Red Side.