VEGAS Status Update - last updated 21 April 2015

VEGAS Bank to Bank Variations Resolved

This problem has been resolved. See this page for more details.

Introduction

Initial ("Phase A") commissioning of VEGAS is now complete. This means that we have performed initial testing of all modes, and basic switching schemes, and VEGAS is ready for initial astronomical observations. However, we have not yet completed detailed commissioning of all modes; for example we have not yet tested deep integrations, or sub-beam nodding observations. Therefore observers should proceed with care. Real time display of VEGAS data can be viewed from Green Bank computers via vegasdisplay.gb.nrao.edu .

Most of the initial problems with VEGAS have been found and fixed. However, a few problems remain. The main outstanding one is some variation in the measured values of Tsys and line intensities from Bank to Bank. This is believed to be a problem in the digital signal processing path in the FPGAs, where at some point the dynamic range gets too small. This is being actively worked on.

A complete list of current issues is available at VegasToDoList

Operation and Trouble Shooting Guide

VEGAS "Features"

  • Due to the nature of the ADCs, there is a large single-channel "spike" in the centre channel of the spectrum. This is automatically interpolated over by the SDFITS program and the real-time display.
  • The vegasdm (total power monitor) may show a "Square-Wave" feature. This is normal.
  • The target levels for the Intermediate Frequency Rack , or iftarget levels, could potentially lead to some problems due to the larger potential bandwidth to the backend of VEGAS (this is not so much a VEGAS problem, as a problem not realized until VEGAS). We will explore the parameter space further as time allows.

VEGAS commissioning tests

We have tested the following:
  • Modes 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,24,25,29,
  • We have not yet had time to fully test modes 21,22,23,26,27,28, but since they are very similar to other modes, we do not anticipate problems.
  • Modes 14 and 19 are NOW currently available
  • From a preliminary inspection, all of the above modes appear to be operating as expected
  • A detailed commissioning memo is under preparation
  • We have tested position switched and frequency switched modes
  • We have not yet tested subreflector nodding and deep integrations

Other Notes

The online SDFITS filler cannot start filling until the end of the scan

Unlike the GBT Spectrometer, VEGAS does not flush its FITS files to disk until the scan is over. Thus the online filler cannot start until the scan is finished, and so obervers in turn cannot inspect their data until the scan is finished. This problem will be resolved later this year when data-streaming is released.

In the meantime, we urge:
  • Use the real time display (vegasdisplay.gb.nrao.edu - only accessible via GB machines)
  • Before starting your first long scan, run a short scan, process the data in GBTIDL, and confirm that everything is operating as you expect.

The Observers' Computers do not mount the "lustre" file system

Unlike for the GBT spectrometer, VEGAS data is stored on our lustre filesystem. This is not mounted by titania or ariel. If you need to fill your data manually, you need to be on one of the GBT data reduction computers that has a lustre mount point, see the list here.

Miscellaneous

Topic revision: r22 - 2015-04-21, RichardPrestage
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback