Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.arcetri.astro.it/irlab/fasti/doc/perugia_O1.ps
Дата изменения: Thu Jul 28 18:23:52 2005
Дата индексирования: Sat Dec 22 06:54:14 2007
Кодировка:

c
SAIt & IEPI 2001
Memorie
della Societa Astronomica Italiana
JOURNAL OF THE ITALIAN ASTRONOMICAL SOCIETY
The Fasti Project
C. Baa 1 ,
INAF, Osservatorio Astrosico di Arcetri, Largo E. Fermi 5, 50125 Firenze
e-mail: cbaffa@arcetri.astro.it
Abstract. The Arcetri Infrared Group has developed a new architecture for the
newer infrared astronomical array detectors, which is intended to overcome the
limitation of present-day commercial controllers. Named Fasti, it is designed to be
modular, exible and extendible. Fasti is also suitable to be used with very fast
optical detectors, as those used in Adaptive Optics.
Key words. astronomical detectors infrared astronomy adaptive optics
1. Introduction
Some recent technological advances enable
us to develop a controller architecture for
fast astronomical array detectors (for ex-
ample the 1024x1024 Rockwell's Hawaii in-
frared HgCdTe detector) which is intended
to fully exploit the capability of present-
day detector. Named Fasti, it is a light
electronic system, which is modular, ex-
ible, extendible, and avoids obsolescence as
much as possible. It is suitable to be used
with both DRO and CCD detector and it
is also well suited for very fast optical de-
tectors, as those used in Adaptive Optics.
The design phase for infrared is com-
pleted and we are starting to build
the infrared version of Fasti. Meanwhile
we started a collaboration with Bologna
Observatory for the development of a new
Adaptive Optics detector system, based on
the Marconi intrinsically intensied photon
counting detector, the L 3 CCD, needed by
Send oprint requests to: C. Baa
Correspondence to: Largo E. Fermi 5, 50125
Firenze
the new generation telescopes, the TNG
and LBT.
2. Project Description
Fasti is a controller architecture devel-
oped to be powerful and extendible. Its ex-
tendibility enabled us to plan an optical
version of it. As an added bonus, the de-
sign does not constrain the downhill con-
trolling computer: Fasti is seen as a periph-
eral through a network connection, with
easy integration in existing telescope con-
trol structure as in TNG case.
The basic ideas of Fasti design are:
{ to divide the system in several modules
with clear boundaries.
{ to use, as far as possible, commercial
parts at board level.
{ to use largely accepted standards as
boundaries, as a PCI bus, a parallel in-
terface or a serial.
{ to design the few custom parts as con-
cepts, not around a particular electronic
device, so that they could be imple-

2 C. Baa: The Fasti Project
Fig. 1. An example of Fasti waveform as seen by the development software.
mented by components from dierent
manufacturers, or by devices developed
in the future.
{ to be as exible as possible on waveform
generation. This part has proved to be
exible enough to drive both DRO and
CCD detector.
3. Fasti structure
Fasti is designed in modules, with clear-cut
boundaries. Its main parts are:
The Waveform Generator
The exible waveform generator is a
custom part of which we had already
built the prototype. It is based on
a specialized micro-controller, where
the waveform denition is realized by
means of a program in a pseudo assem-
bler language, greatly simplifying the
denition of new waveforms. We had
already developed all the support soft-
ware for waveforms design and test-
ing. This part, named SVB, can gen-
erate not only the standard waveforms
to read the full array, but also arbitrary
sub-array scan patterns. It can be re-
programmed in seconds, and hold up to
four dierent clocking schemes, which
can be selected on a per integration ba-

C. Baa: The Fasti Project 3
sis. The SVB is implemented in pro-
grammable chips, but, being a concep-
tual design, can be easily transferred to
newer devices.
The Global Controller
Inside Fasti there is a central con-
troller for start-up, general housekeep-
ing, global control of operations (start
integrations for example), data collec-
tion, formatting and buering, or for
data pre-processing when needed. In the
present design all this is realized with
a disk-less embedded computer, using
an Intel or Alpha family CPU and few
commercial boards. The parallel digital
acquisition board and the fast Ethernet
interface are hosted here.
The Serial Bus
All the low speed communication inside
Fasti are routed through a serial bus.
This approach gives us the possibility to
use a simple and widely accepted stan-
dard for communications, and to use all
commercially available parts. The bus
chosen is the old RS232, but routed on
Ethernet. This choice bring together the
ubiquity and the ease of use of RS232
and the exibility and addressability of
Ethernet.
The Conversion Subsystem
This part has to be custom developed.
This section mainly consists of a small
number (4 for NICMOS3 and Hawaii)
of analog to digital converters and some
glue logic. We will use high quality 16-
bits converters for Nics 1 version, and
very fast low resolution (8 or 12 bits)
converters for Adaptive Optics version.
The Analog Interfaces
This part consists mainly of the bias
levels generation, of digital clocks level
shifting and of detector output condi-
tioning. For Infrared version this part
inherits the NICS 1 design, and for
L 3 CCD uses a Marconi commercial
board.
1 Nics is the Infrared Camera Spectrometer
developed by arcetri Infrared Group for the
TNG (Baa et al. 2001; Lisi et al. 1999)
4. Project Status
At the date of this paper three of the four
boards for Nics 1 version have been designed
and are in fabrication or testing. The fourth
(the buer board) is in its nal design
phase, ad are expected to be completed in
a month. The system is designed to be able
to work (even at lower capabilities) so we
expect to be able to get rst acquisition be-
fore spring 2002.
In our present schedule, the complete
system will be read before summer.
For the L 3 CCD version, a collabora-
tion with Bologna University have been
started, and we are already actively collab-
orating. The expected laboratory version is
expected for the summer.
Acknowledgements. We would like to thank
the many people, in Florence, at Bologna and
at TNG which help this project come to reality.
Fasti development has been partially sup-
ported by a CNAA Special Projects Grant.
References
Baa, C., Comoretto, G., Gennari, S., et
al, 2001, A&A 378 722.
Baa, C., Comoretto, G., Gavrioussev,
V., Giani, E., Lisi, F., 1999, AstroTech
Journal 2
Comoretto, G., Baa, C., C., Gavrioussev,
V., Lisi, F., Sozzi, M., 1999, AstroTech
Journal 2
Lisi, F., Baa, C., Gennari, S., Oliva, E.,
1999, AstroTech Journal 2

4 C. Baa: The Fasti Project
Fig. 2. An Artist's view of Fasti-IR structure