Previous Article
Next Article
Table of Contents

Updates on GoldCam and CryoCam (1Dec93)
(from KPNO, NOAO Newsletter No. 36, 1 December 1993)

We have undergone some minor changes with the detectors in GoldCam and
CryoCam. These are improvements to the CCDs previously in place, not
new chips.

In CryoCam, the detector is a thinned Ford 800 x 1200 CCD with 15
micron pixels (the same size as the old TI 800 x 800). We have recently
lowered the read noise to 14 electrons by careful adjustment of the
clock voltages. The chip has been coated with an AR coating which peaks
at 6000 . This choice of coating has the dual effect of reducing the
fringing in the far red and lowering the QE shortwards of 4000 A . We
decided to give up the near UV response because the optics in CryoCam
are not optimized for the blue anyway (the collimator is a crown-flint
doublet). The advantage in the red is dramatic. We have reduced the
fringing at 9000 A from about 25% amplitude peak-to-peak on a chip with
no coating or a blue- optimized coating to less than 5% amplitude
peak-to-peak. This facilitates sky subtraction and accurate fluxing of
faint objects. The chip is flat, of course, which also aids sky
subtraction. We certainly expect to put a similar device with lower
noise into CryoCam when we receive one.

As reported in the last Newsletter, we recently replaced the mediocre
Ford 3K x 1K CCD in GoldCam with one of much better performance. Users
of the old chip suffered from a large number of traps in the imaging
area which compromised regions of the spectrum and necessitated a
preflash, increasing the effective noise. The new chip solves these two
problems. It has no traps above pixel 600, below which the focus starts
to seriously deteriorate. The read noise is 8.5 electrons, and no
preflash is needed. We had also intended to remedy the fringing
problem, which was quite severe with the old CCD. Unfortunately, due to
a calibration error, the AR coating which was put on the new CCD was
too thin, and both the spectral response and the fringing were quite
similar to the old CCD. We have just recoated the chip with a coating
which should move the peak response somewhat further to the red and
reduce the fringing substantially. At the time of writing we have been
able to estimate from lab measurements that we have reduced the
fringing by about a factor of 3.  A determination of the new
sensitivity curve will have to await measurements at the telescope,
expected during the next few weeks. The one remaining problem is the
resolution. Since we have installed the first Ford 3K x 1K CCD in
GoldCam, we have never achieved better focus than about 2.5 pixels. We
have now changed the chip, the coating, and the field flattening lens,
each of which we thought might be responsible. None of these changes
have had any effect. We will continue to work on this problem.

Todd Boroson, Rich Reed, Di Harmer

Previous Article
Next Article
Table of Contents