Pulsar bin-mode observations

When observing pulsars with the ATCA, the correlator is capable of sampling synchronously with a pulsar. In this way, during a single integration, the correlator will produce a number of bins, each of which sample at a particular range in phase of the pulsar cycle. Much of the time, when reducing a bin-mode experiment, you will want to process all the bins without regard to the bin number. In this case, to a large degree, processing a bin-mode observation differs little from a normal observation. In particular:

Obviously, however, there are times that you will want to distinguish between the bins. The bin visibility selection sub-command is the main mechanism for this. The general form is:

select=bin(n1,n2)
This selects bins n1 to n2 inclusive. If only a single bin is given, then only that bin is selected - numbering of the bins starts at 1.

The most common places where you will want to select a subset of bins is in the imaging and possibly self-calibration steps (the pulsar, during its `on' phase, may make an excellent phase self-calibration source).

Although almost all visibility-data tasks support bin-mode observations, there are a few exceptions. These give warnings when they are used on bin-mode data. These warnings should be heeded!. The most notable task which fails to properly handle bin-mode experiments is uvaver when time averaging of the data is being performed. When used to time-average, uvaver will average all the bins together. Note, however, that uvaver works correctly (i.e. treating the bins as separate) when no time averaging is involved.

Miriad manager
2016-06-21