ESO is working on the replacement of the present 2kx2k,
24µm pixels detector of FORS2 with a mosaic of two MIT 2kx4k,
15µm pixels red-enhanced chips with the first tests starting in
late 2001.
The goal is to get a much better sensitivity in the 700-1100 nm wavelength
domain. Furthermore a lower readout noise will be achived at much higher
readout speed then with the Site CCD used with FORS2 so far.
The detector upgrade including all software changes on the instrument side
will be ready for operation in April 2002 with a temporary installation of the
mosaic during the bright time in October/November 2001 for tests and
commissioning.
This page summarizes the changes expected to come with the
CCD upgrade without being very accurate since most parameters are still to
be measured but it should be accurate enough for proposal preparations.
There will be no changes for the observing block preparation of next
observing Period 68 (October 2001 to March 2002)!
Detector Characteristics
High Time Resolution - HIT modes
One important change in the new configuration will be that the CCDs are
read out in dispersion direction. As a consequence the high time resolution
modes will have to be modified such that spectroscopy will be done with
the cross disperser grisms. This cannot be offered in time for Period 69
and has to be followed up later on.
CCD Readout Modes
As a general rule only 2x2 binned readout will be offered (100kHz for
spectroscopy and 225kHz for imaging) to obtain homogenous FORS archive data
and to reduce calibration times. This will result in image scales of 0.25
arcseconds/pixel in standard resolution and 0.125 arcseconds/pixel in
high resolution mode. Unbinned readout modes for special applications
(225kHz) must be explicitely requested and justified in the proposals.
CCD Readout Time and Readout Noise
The new MIT CCD can be read out faster then the SITE CCD at lower RON.
As a consequence of the changed readout direction there will be some overhead
time required to rotate the images. Therefore we request users to consider
40s including the readout time and the so far unknown overhead time in the
P69 proposals for every single CCD readout.
observing mode
readout speed
binning
RON in e-
readout time
Site 2kx2k (old)
50 kHz 1-port
1x1
5.8
85s
Site 2kx2k (old)
50 kHz 4-port
1x1
5.8
21s
MIT 2x2kx4k spectroscopic
100kHz 2-port
2x2
<3
25s
MIT 2x2kx4k imaging
225kHz 2-port
2x2
<4
10s
CCD Response
New Field Geometry
With the standard resolution collimator the field will be restricted by the
geometry of the MOS unit which is permanently in the field of view and of
approximately the same size as the 6.8 arcminute field of view of the
previous configuration. In high resolution mode, the field of view will
increase from 3.4' x 3.4' to 4.2' x 4.2'.
The CCD mosaic is mounted off-axis to ensure that the target will fall
on the upper (master) CCD and to ensure that the gap will fall between
two MOS slits.
Even though the enlarged detector size will be vignetted in the focal
plane, there will be more freedom in selecting slits in a larger x-axes
range for a given wavelength range! In the example below the wavelenth
range was set to the nominal wavelength range (545 - 810nm) of grism 600RI
with the old SITE CCD. The inner blue box (between the two |X| |X|)
is the additional space within slits can be placed with the
MIT mosaic:
The new field geometry for observations with the high resolution collimator.
The field of view will be limited by the size of the two CCD (two green
rectangles).
The following photometric zero points were calculated with the ETC
for a zero magnitude star at zero airmass. The calculated performance can
be compared among the two CCD configurations and with data taken on the sky:
Filter
Calculated
Calculated
Measured
 
MIT
SITE
Aug.18, 2001
z
27.15
26.31
 
I
27.63
27.26
27.23
Rsp
28.20
28.13
28.13
V
28.00
28.00
27.95
B
27.60
27.64
27.58
Usp
23.96
24.70
24.71
FIMS - Mask Preparation Software
FIMS will only support the modes for which mask preparation is absolutely
required: MOS, MXU and imaging with occulting bars. All other observing
modes like imaging, long slit and Echelle spectroscopy are supported with
"fast mode" which should be more straight forward to prepare from the
point of view of an observer.
FIMS will display the CCD projected to the sky (green rectangles) and the
field stop geometry in the focal plane of the telescope (blue rectangle).
Observations and Target Acquisitions
All target acquisitions will be done based on acquisition images with the
upper "master CCD". As a consequences users will be forced to select
reference stars and reference slits only on the master CCD.
The lower "slave CCD" will be merged based on transformations calculated
from pinhole masks by the FIMS mask preparation software and stand alone
routines to be provided to the FORS users.
Wavelength Coverage
Grism name +number
Wavelength range (SITE)
Dispersion
Wavelength range (MIT)
Order separation
[nm]
[Å/mm]
[nm]
filter
GRIS_600B+22
345 - 590
50
330 - 621
none
GRIS_600I+25
690 - 910
44
663 - 939
OG590
GRIS_300V+20 (1)
330 - (650)
110
330 - (650)
none
GRIS_300V+20 (1)
385 - (750)
111
385 - (750)
GG375
GRIS_300V+20
445 - 870
112
445 - 870
GG435
GRIS_300I+21
600 - 1100
108
600 - 1100
OG590
GRIS_200I+28
560 - 1100
162
560 - 1100
OG550 (cemented)
GRIS_150I+27 (1)
330 - (650)
225
330 - (650)
none
GRIS_150I+27 (1)
385 - (750)
230
385 - (750)
GG375
GRIS_150I+27 (1)
445 - (880)
230
445 - (880)
GG435
GRIS_150I+27
600 - 1100
230
600 - 1100
OG590
holographic:
GRIS_1400V+18 (3)
469 - 573
20.8
456 - 586
none
GRIS_1200R+93 (3)
590 - 715
25.0
575 - 731
GG435
GRIS_1028z+29 (3)
790 - 930
28.3
773 - 948
OG590
GRIS_600RI+19 (3)
545 - 810
55
512 - 845
GG435
GRIS_600z+23 (3)
770 - 1036
54
737 - 1070
OG590
2nd order:
GRIS_600I+25 (2)
380 - 475
19
369 - 488
FILT_465_250+82
GRIS_600z+23 (2,3)
404 - 535
20
389 - 546
FILT_465_250+82
(1) If used without or with the listed order separation filter,
the orders will overlap above the given wavelength.