next up previous
Next: Enhanced Correlator Up: Correlator Issues Previous: Options


Future Correlator

The baseline ALMA project includes development work for a 2nd-generation ``Future Correlator''. Such a Future Correlator should have higher performances than the Baseline Correlator and should take into account the current fast technological progress, e.g. in chips and interconnectivity. As stated at the correlator Preliminary Design Review held in Charlottesville in January 2000, a final commitment for the 64 antenna ALMA correlator should be made by 2003.

On the European side a Digital Hybrid Correlator design is being developed, whereas in Japan an FX design is studied. Both projects are aimed to provide a greater number of channels at the lowest spectral resolution, higher sensitivity, and higher flexibility than the Baseline Correlator. Such improvements should lead to an important increase in the scientific returns of ALMA. As a consequence, ASAC agrees that the development of the Future Correlator is an important part of the baseline ALMA project, and strongly recommends that these studies are continued. Furthermore, the ASAC encourages the European, Japanese, and North American groups to cooperate in their Future Correlator studies.

The ASAC noted with great interest that the hybrid correlator studied by the European team will, by going from full 2 bits to full 3 bits in the correlation process have a 9% higher sensitivity than the baseline correlator, which is the equivalent of adding 9% of collecting area, i.e. the equivalent of 6 additional antennas for a 64-element interferometer. Moreover, the European project is also considering a 4-bit correlation scheme.

The European team solicited ASAC input on various design options, which resulted in the following recommendations:

1.
Spectral resolution: A highest spectral resolution of 5 kHz is required, corresponding to 0.05 km/s at 30 GHz. This resolution is, e.g., necessary for the observation of lines in cold dark molecular clouds.

2.
Required number of channels: A total number of 8000 channels is sufficient for most astronomical observations. Observations using multiple sub-bands and polarizations will accordingly have less channels available per subband and/or polarization.

3.
Minimum acceptable dump time for on-the-fly observations: The specification for the minimum dump time in total power mode is 1 msec. For spectral line observations the minimum acceptable dump time may be chosen to be proportional to the number of channels, e.g 1 msec for 512 channels, but 16 msec for 8000 channels.

4.
Minimum integration period for cross-correlations: For solar (continuum) observations, minimum integration times of 1 msec are required. For spectral line observations, the fastest integration period should be made compatible with the minimum dump time. In side band separation mode, integration times of $\sim$120 msec should be possible.

5.
Are on-line WVR corrections necessary? Yes, but it will be necessary to carefully test the method before eventual implementation.

6.
Sub-arrays: The specification is that the array can be divided in at least 5 sub-arrays, which are all logically independent and can observe at a maximum of 4 different frequencies. The goal should be at least 16 independent sub-arrays, which is the number provided by the Baseline Correlator.

7.
Flexibility: Since the Baseline Correlator provides a maximum of 8 sub-bands (total, not per polarization), a reasonable goal for the Future Correlator should be at least 16 sub-bands. Such an improvement would be extremely useful for spectral line surveys of different astronomical objects.

The specifications and goals stated in the above considerations should be taken into account by both the European and the Japanese projects. In summary, the ASAC remarks that the Future Correlator developments should be guided by the goals of achieving (a) higher spectral resolution, (b) a higher number of channels in wide band modes, (c) a significant gain in sensitivity, (d) high configuration flexibility, and (e) lower power consumption.

The ASAC strongly recommends that a comparative presentation of the different correlator concepts that are presently under study (including cost estimates) should be made at a future ASAC meeting.


next up previous
Next: Enhanced Correlator Up: Correlator Issues Previous: Options
Al Wootten
2000-10-10