Wide band calibration - gpcal - nfbin

Is MIRIAD being a pain? Let us know your experience.

Moderator: Mark.Wieringa

Post Reply
Mark.Wieringa
ATCA Expert
Posts: 297
Joined: Mon Feb 08, 2010 1:37 pm

Wide band calibration - gpcal - nfbin

Post by Mark.Wieringa »

Today I've checked in the first stage of wide band calibration modifications to the Miriad calibration tasks, it should appear on all servers overnight.
The only visible difference is a new parameter: nfbin in gpcal. This specifies the number of frequency bins to use when solving for complex gains and leakages.
The binned solutions from gpcal are written to a separate set of calibration tables (gainsf and leakagef). The standard continuum solutions are still done as well.
Only a few of the tasks that deal with calibration tables know about the new tables yet, but the binned solutions are applied (by interpolating in frequency) whenever the calibration is applied to the data. At this stage gpboot has been updated to take notice of the binned solutions, but mfboot not yet. Gpplt will plot the binned solution and gpcopy will copy them.

If you don't use the nfbin parameter everything should work as before, but let me know if it doesn't. If you do use it and you get unexpected results let me know too.

One thing to be careful with is data with extensive channel flagging (e.g., 16cm) - if you have many bins, some may end up with very little data and produce noisy solutions.

Cheers,

Mark
Mark.Wieringa
ATCA Expert
Posts: 297
Joined: Mon Feb 08, 2010 1:37 pm

Re: Wide band calibration - gpcal - nfbin

Post by Mark.Wieringa »

mfboot is now also 'wide band aware' and will scale the standard 'gains' and the freq binned 'gainsf' tables by the same factor.

Cheers,

Mark
Post Reply