December, 2011 Test Session


The VEGAS team has been working on the first two firmware modes and the supporting software. Hardware development and design is getting pretty far along, with prototypes of all the subsystems in hand, or close to it. It is time to bring all the players together and test the hardware, and the first 2 software/firmware modes.

Test Objectives

The overall objective is to make sure the hardware, software, and firmware teams are on the same page, and the first 2 modes work properly. We have selected 2 modes to test that will exercise most of the features of the system.

Mode 1 tests

Mode 1 is a wide-band mode. It is, in the end, to provide 1.25 GHz of usable bandwidth from the spectrometer. In these first tests, it is likely that we cannot support the full wide bandwidth due to firmware constraints. We will endeavor to test at the fastest clock speed that can be supported at the time of the test.
Features to test
  • Monitor and Control system integration
  • Spectrum layout (i.e. is it flipped or not)
  • Data handling capability of the host
  • Switching signal generation and binning
  • ???

Mode 13 Tests

Mode 13 is a multi-window narrow-band mode. The signal is sampled and downconverted into 8 narrow sub-bands in the FPGA, then sent to the GPU for further spectral analysis. This mode tests the network speed and the host data handling capability, in addition to the GPU spectum generation, and the FPGA bandpass processing.
Features to test
  • Monitor and Control system integration
  • Spectrum layout (i.e. is it flipped or not, are the bands flipped around, etc.)
  • Data handling capability of the host and network stack
  • Switching signal generation and binning
  • ???

Test Hardware

Target Hardware

  • 2 ROACH-I systems
  • 4 ADC boards (NS 3 GS/S boards)
  • 2 VEGAS Analog front-end signal conditioners
  • 2 Valon Synthesizer boards
  • 10 MHz reference distribution box
  • 1 PPS reference distribution box
  • Switching signal distribution/interface box
  • 2 GPU computers
    • North
    • South
  • 2 Host computers for mode 13 tests
    • Pork
    • 43m-map
  • All the interconnections and cruft that goes with it

Test Equipment

  • Spectrum analyser
  • Signal Generator
  • Noise sources
  • digital Oscilloscope
  • Pads, cables, probes, etc.


  • ???


  • ???


Preliminary Tests

These tests will be run by NRAO GB people prior to the test week. We will ensure that the following is working:
Hardware Setup and Functional Tests
  • All roaches and computers are networked and booting properly
  • Clock levels are set properly for the frequencies to be used
  • All reference signals are present and at the right levels
  • A noise spectrum through the system looks "reasonable"
  • A noise + tone through the system looks "reasonable"
Software Setup and Functional Tests
  • All managers running and communicating
  • BOF files copied to the right directories
  • Basic mode 1 and 13 bofs load and parameters set
  • Communications with Synthesizer working
  • ???


Mode 1 testing
  • Load correct newest bof
  • Verify that data comes out as expected
    • In snap blocks
    • In buffers in the host
    • On disk
  • Noise source testing
    • Verify channel numbers with tone+ noise
    • Verify noise integrates down with time
  • On-sky testing
    • Anish to fill in.

Topic revision: r1 - 2011-11-28, JohnFord
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