Sparo uses the Yerkes/Goddard data system - specifically a version that has been built for SPARO and reprogrammed for the SPARO experiment. This is called the SPARO data system. The data system consists of two parts: an A/D box out on the telescope (built at Yerkes and designed at Goddard) and a Macintosh Quadra in the control room (the 'data computer'). They are linked by optical fibers that run through the cable wrap. The SPARO data system will be mounted on the Viper telescope in a heated box that is identical to, and adjacent to, the box that holds the CMU data system. Taking data with Sparo involves turning a half-wave plate very frequently during the observations. This will be done with a serial-interface stepper controller that is controlled by the data computer (the Mac Quadra). The data computer will get feedback from an encoder at 4K that senses the half-wave plate's position. The data computer stores all the data on its disk.
Interface between Viper telescope control computer (comsoft PC), Viper control computer, and SPARO data computer is a major issue to be resoved in Summer 1998 prior to deployment to Pole.
Operating a He-3 cryostat during the Antarctic Winter requires monitoring the cryogenic health of the experiment. He-3 pressure, temperatures of He-3 stages, temperature of pumped pot, pressure in pumped pot pumping line, He-4 level, all need to be monitored (see lessons learned). Such monitoring is critical for a liquid helium system. We will have a separate, dedicated, computer system for this job, the "CryoController".
The Cryocontroller will be a 486 Industrial PC running labview. It will also be used for cycling the He-3. Because of the continuously pumped He-4 1.5 K stage, cycling the He-3 involves only manipulating heat switches and heating the charcoal pumps. We plan to do this via computer, using the labview interface.
If program decides cryostat is having a serious problem, an alarm will be set, causing some combination of: change in TTL level on an output line, warning written to data file, call sent out via TCP/IP protocol, beep sound. There could be several levels of seriousness (e.g. ice plug vs. minor change in equilibrium). (DC)
For South Pole Winter-over operation, documentation is very important. Cryocontroller documentation should be placed here via links from the above items. Source code should also be available from belmont in recoverable form.
The data system, and serial-interface indexer for the half-wave plate, will be out on the telescope. Where should the CryoController be located?
A failure of a component in the CryoController could lead to accidental dewar warm-ups and other cryogenic problems. Thus, in general, such failures will need to be repaired quickly. From this point of view, the CryoController should not be located out on the telescope where problems are difficult to debug and repair quickly. It should be in the Control room (formerly called Milvan) where direct readings of temperatures, He-4 level, etc. can be taken by bypassing the CryoController, if necessary.
Implicit in this argument is the assumption that the cable-wrap will be highly reliable. We will assume that the cable wrap is highly reliable. We do of course need to plan for a serious cable wrap failure. In this situation, we still need to keep Sparo cold while the cable-wrap gets fixed. A spare CryoController-to-Sparo cable that can be run directly up to the focal plane of a parked Viper would be one possible plan. Another would be removing Sparo to the Milvan.
A schematic view of the Sparo Computer Network appears below.
