Previous Article
Next Article
Table of Contents

CCD News (1Sep95)
(from CTIO, NOAO Newsletter No. 43, September 1995)

In NOAO Newsletter No. 42 (p. 29, 31, 33) three new CCDs entering use
at CTIO were described. The implementation of these CCDs marks the end
of a decade of use of the "VEB" CCD controllers, designed by Bruce
Atwood. All Tololo CCDs now being scheduled are operated by Arcon
controllers. A brief update follows.

                         Loral 3K CCD

The Loral 3K CCD will have seen 80 nights of use on the 4-m telescope
by 31 July, with the R-C and ARGUS spectrographs, plus an engineering
night on the Echelle spectrograph (see other articles in this
Newsletter). Recent lab work has focused on attempts to reduce the
read time without compromising the read noise. Since this CCD has two
operative amplifiers that cannot be read out simultaneously, the
strategy chosen was to leave the lower-right (LR) amplifier alone, as
a point of reference, while making changes to the lower-left (LL)
ARCON video chain. This work has been successful, as illustrated in
the table below. The second and third columns illustrate that the
video gain of LL is almost double that for LR, yet for a given e-/ADU
the readout noise is the same.

Prior to this work, observers generally chose to use gain index = 4,
LR amplifier, which at 2.03 e-/ADU sampled the dynamic range (7.7 e-
RON, 80000 e- full well) satisfactorily, with read time a slow three
minutes. Now it is possible to instead use gain index = 2, LL
amplifier, which gives the same gain and read noise, but with read
time of two minutes.

        GAIN       1/G [e-/ADU]         RON [e-]    Time
        Index      LL        LR         LL    LR     [s]

          1        4.33     8.21       11.0  12.6   88.2
          2        1.99     4.17        7.7   9.3   20.5
          3        1.39     2.74        7.9   8.6   52.4
          4        1.03     2.03        7.3   7.7  184.4

                          STIS 2048 CCD

This CCD will have seen 96 nights of use on the Schmidt, from March 1
until July 31. Quad-readout has now been implemented, which makes a
great improvement to observer efficiency at this telescope, since for
most projects exposure times average only a few minutes. The linearity
of this CCD rolls-off above 90000 e-, so use of gain index = 2 or 3 is
normal. See the table that follows.

                   1/G [e-/ADU]          RON [e-]         Read Time
   Gain index    LL   LR   UL   UR     LL   LR   UL   UR      [s]

     1          2.9  2.9  2.7  3.1    4.2  4.2  4.2  4.2       32
     2          2.1  2.1  1.9  2.3    3.8  3.8  3.8  3.8       38
     3          1.6  1.6  1.5  1.7    3.8  3.5  3.7  3.7       47
     4          1.1  1.1  1.0  1.2    3.7  3.3  3.4  3.5       69

                            Loral 1K CCD

This CCD is dedicated for use with the 1.5-m spectrograph and was
tested at the telescope on 17 May (see report elsewhere). Due to the
block scheduling of instruments at the 1.5-m, the next use of the
spectrograph is not until November, for a one month period. Prior to
this time, some lab work will be scheduled to optimize the gain/read
time for this CCD, in much the same way as detailed above for the
Loral 3K.

                         Field De-Flatteners

As reported elsewhere in this issue, image quality at the 4-m prime
focus has been improved by careful attention to the alignment of the
primary mirror and the PF corrector. The well-known curvature of the
large Tektronix CCDs we use for imaging at this focus produces defocus
across the field, and observers normally compromise by focussing on
stars part way between center and corner of the CCD. This de-focus is
exacerbated with the sub-arcsecond images now possible at PF, so we
have ordered simple fused silica lenses that will, to first order,
compensate for the CCD curvature. These lenses will replace the dewar
windows for our two Tek 2048 CCDs, and should arrive in 2-3 months
time.

Alistair Walker

Previous Article
Next Article
Table of Contents