Previous Article
Next Article
Table of Contents

Instrument Control and Data Reduction Upgrades...(1Jun93)
at the FTS
(from NSO, NOAO Newsletter No. 34, 1 June 1993)

At the Solar FTS, the DEC PDP-11/44 data and CAMAC interface hardware have been
retired and replaced with a 68030-based VMEbus system for data acquisition and
a Sun SPARC IPX workstation for reduction and analysis.

The acquisition system ("xx", pronounced "Dos Equis") contains a Heurikon
module with a Motorola 68030 CPU, 4 Mbytes of RAM plus 24 Mbytes to buffer the
incoming data, an Ethernet interface, a 1024 x 768 pixel color graphics display,
and input/output modules to interface with the existing FTS electronics. The
acquisition program, written in C, runs under the "VxWorks" real-time operating
system. The same type of CPU module and operating system are in use elsewhere
on Kitt Peak, as telescope controllers.

The acquisition system's user interface is menu-driven and can be accessed
through a terminal or through a window on the Sun workstation. The observer
enters setup parameters in a spreadsheet-style screen, which updates its
display of information (such as resolution and scan time) as new parameters
are entered.

The use of large memory buffers allows both the current scan data and the sum
of previous scans to be stored in RAM. This allows the current scan, if it is
defective for some reason, to be subtracted from the sum, possibly saving
several hours worth of data.  This has been a greatly desired feature at the
FTS and has already saved several scans.

The data and header files are written to the Sun workstation's disk through a
dedicated Ethernet link.  The workstation ("Corona") is a Sun SPARCstation
IPX with dual Ethernet ports, and 3 gigabytes of hard disk space on its three
drives. Running an updated version of "GRAMMY", Corona is capable performing up
to 4-million-point transforms in under 3 minutes. The resultant spectra can
then be displayed using "DECOMP" or an IRAF package known as "ISAAC". In most
cases the transform is done and ready to be looked at before the next scan
is ready to start.

Along with the 11/44, gone are the days of carrying nine-track tapes down to
Tucson to have them transformed and hoping for spectra to arrive in the mail a
few weeks later. With the new system, by the end of a run, the transforms are
done, the data are written to Exabyte tape, and the observers can go home WITH
THEIR DATA!

The project to re-archive all of the past FTS scans is in full swing, and
compact disks (CD ROMs) full of FTS data are being cut. Each disk contains
600 Mbytes of data, and we hope to get the entire archive, from 1976 to the
present, on about 36 CDs! The files on these disks are in FITS format and
contain the newest "type 6" header keywords. An IRAF program called FTSDBM has
been written that allows users to search all of the data in the archive using
parameter queries. This program allows the user to locate quickly particular
scan types in the archive. FTSDBM also keeps a low-resolution (2048-point)
image of each spectrum on-line to be examined by the archive searcher.

Any questions concerning the new FTS system should be addressed to Claude
Plymate (cplymate@noao.edu).

Jim Brault, Shelby Gott, Dyer Lytle,
Claude Plymate, Lourdes Ramirez

Previous Article
Next Article
Table of Contents