|
Sloan Digital Sky Survey
Review of Observing Systems and Survey Operations
Spectroscopic Observers Program Overview
Scott Burles
April 13, 2000
Introduction
The Spectroscopic Observers Program, or SOP, provides the
observers an interface to the spectroscopic systems and control of the
SDSS telescope during spectroscopic observing. SOP is entirely dependent
on IOP, and utilizes IOP commands to control the 4 cameras in the spectroscopic
system. The SOP interface is entirely command line based, and resides in
an "IOP" environment.
Goals:
-
Control Spectrographs
-
Communicate with Cameras
-
Acquire Spectroscopic data
-
Communicate with calibration lamps, screen
-
Communicate with TCC
-
Perform closed-loop guiding
-
Digest plugMap file for guiding and QA
-
Real-time feedback from guide images and science frames
-
Output of report files and informative file headers
Successes:
-
Full control of spectrograph mechanicals
-
Control of 4 cameras, including rebinning.
-
Routine data acquisition is complete.
-
Lamps and screen communication set.
-
Basic TCC communication
-
Stable guiding with 2 - 11 coherent guide fibers
-
Straightforward reading of plugMap files
-
Automatic monitoring of guide images for feedback
-
Correct log files and header entries are creates
Remaining Tasks:
-
Graceful recovery from hangs and crashes:
-
Currently, all critical information is stored in current
window.
-
Upon exiting, information is lost, and calibrations must
begin anew. This can cost up to a 30 minutes to recover lost position.
-
Solution: Critical information must be stored on disk or
crates, and with one command, the observers can recover the desired state.
-
Lack of exposure control:
-
pause, restart, abort
-
Fifteen minute exposures are currently best not interrupted,
even if no data is be obtained.
-
This uses an IOP command "goStare", and needs to be fixed
at that level.
-
Need to properly receive status from TCC
-
Many SOP tasks depend on completion of slews, offsets, focus
and scale changes issued by the TCC. Currently, only predicted delays are
used and which assume the previous action was completed. Instead, we should
receive confirmation from the TCC that the desired action completed successfully.
-
Quantitave feedback from:
-
Guide Images (exposure meter)
-
Science frames (SPECTRO-on-the-mountain)
-
Critical to decide whether data meets survey standards, and
to alert observers to any problems which may arise in the data.
-
Scope, speed, and platform of feedback needs to be decided.
-
Increase efficiency!
-
Auto-loading of fiber rotation angles (w/ index pins)
-
multiplexing of calibrations
-
Is the current spectroscopic paradigm optimal?
-
Documentation
-
Documentation of operating and checkout procedures (on a
yearly, run and nightly basis)
-
Documentation of procedures for recovery from failure
-
Special flat-fielding algorithm to obtain near-uniform CCD
coverage
-
"Rocking collimator flats"
-
Quality, repeatability, and necessity are unknown
Summary
SOP is in a state to be used for routine observing. Efficiency
is severely compromised by the unfinished tasks. Hangs and crashes not
only waste time, but even worse, the target tile (or sky coordinates) moves
through and even out of the allotted hour angle window. QA, documentation,
and repeatable tests of the spectroscopic system are required to ensure
survey quality data, as well as uniform and consistent data over the course
of the survey. These tasks are currently being addressed with
a target completion by the start of the summer shutdown.
Review of Observing Systems and Survey Operations
Apache Point Observatory
April 25-27, 2000
|