Bugs: dmcoords

The pointing (PNT) header keyword values in the background files are set to zero:

unix% dmlist bgevt2_c7.fits header |grep PNT
0058 RA_PNT                                  0           Real8        Pointing RA
0059 DEC_PNT                                 0           Real8        Pointing Dec
0060 ROLL_PNT                                0 [deg]     Real8        Roll (fake)

These values indicate where the optical axis was during the observation. Having zero-values will cause the CIAO tool dmcoords to produce incorrect results when run with the background file.

Workaround:

Follow the workaround in the The ACIS "Blank-Sky" Background Files thread to update the PNT header keyword values.

The WCS library that the DM uses has a problem computing coordinate transforms that involve the CAR transform.

A keyword was dropped when filtering columns from HRC evt2 files in the pre-CIAO 3.0 data model. This causes dmcoords to not have correct knowledge of the aspect solution; the tool therefore computes an incorrect zero point for the azimuth (phi) and off-axis angle (theta).

Workaround:

Use dmstat to get the DETX,DETY values:

unix% dmstat "evt2[sky=region(reg)][cols det]"

and then use dmcoords to convert these to theta and phi values.

Due to a precision issue in dmcoords, a warning may be printed even if the NOM and WCS values match. As long as the reported values are the same, this warning may be ignored:

# dmcoords (CIAO): Warning: RA_NOM DEC_NOM different from WCS

# dmcoords (CIAO): RA_NOM DEC_NOM = 224.013323 -68.725752 
                                  WCS = 224.013323 -68.725752