FBSCOR FBSCOR operates on Caltech MERGE-format (u,v) data files, applying a correction for an error in the JPL/Caltech Block II VLBI Correlator hardware. This error in the fractional bit shift (fbs) was fixed on 10 June 1991, so data correlated after that date DO NOT require this correction. Details of the nature of the problem, and its correction, are available on request. The error resulted in an amplitude reduction which is proportional to baseline length and cosine (declination). The worst-case error is about 8%, though a typical (point source) sensitivity loss for a multi-station experiment is only 2--3%. Most mapping experiments will not be affected noticably by this error. Only observations designed specifically for high dynamic range (e.g., long (u,v) tracks on 10 or more stations, many 'crossing points', good coherence, and good a-priori calibration) are likely to be affected. This program can be run at any stage in the calibration and mapping process. For raw data, it should be applied BEFORE calibration, as minor inconsistencies may otherwise arise. For datasets which have been already mapped, run FBSCOR on the final AMPHI output, and re-run the last one or two cycles of hybrid mapping. Parameters The parameters names can be abbreviated to uniqueness (minimum match). End the list of parameters with a slash (/). INPUT and OUTPUT are mandatory. Other parameters have defaults which are reasonable for most cases. INPUT = "filename" : the name of the input merge-format file (e.g., INPUT = CITSCR:[SCU]FAKE.MRG). This parameter must be specified; there is no default. Only MERGE-1 format is supported. OUTPUT = "filename" : a name for the output file created by AVERAGE. This parameter must be specified; there is no default. LISTFILE = "filename" : a name for the listing file. This file gives the mean correction factor applied for each UT section of data, as well as the mean sensitivity loss, prior to correction. The filename defaults to FBSCOR.LIS. CORREXP = +1 or -1 (default +1): CORREXP = +1 performs the correction, by multiplying each amplitude and amplitude error by the appropriate correction factor. CORREXP = -1 is FOR TESTING ONLY: it is intended for use with 'perfect' data, e.g. from FAKE, for simulating the effect of the fbs error. HISTORY = "text" : a character string (up to 78 characters) in quotes, to be written in the "history" section of the output file; this is optional. (This is in addition to two history records which record the username, date, and operation performed.) FBSCOR Page 2 BANDPASS = g1, g2, g3, g4: bandpass shape, specified as amplitudes in four 0.5-MHz bins, starting at the DC edge. The size of the correction is a (weak) function of the bandpass. Values do not have to be normalized. The default set (0.8, 1.0, 1.0, 0.6) is close to the effective mean response for MkII-format data at the Block II Correlator Example $ FBSCOR INPUT = FAKE.MRG OUTPUT = FAKE_FBSCOR.MRG LISTFILE = FAKE_FBSCOR.LIS HISTORY = "*** APPLY fbs error to FAKE data, for testing ONLY***" CORREXP = -1 ! APPLY the error to 'perfect' data $ EXIT ! (default is CORREXP = +1, i.e. correct the data) History Version 1.0: 1991 Dec 11 - Original program by Glenn Berge; analysis of the fbs error and form of the correction by Robert Coker, David Fort, and Stephen Unwin. Version 1.1: 1992 Jan 17 - Change listing file (Keith Thompson). Version 1.2: 1992 Feb 6 - Change program name to FBSCOR; other minor changes (Stephen Unwin).