Problem with rspectext/wspectext

cosmologist wrote on Mar 03, 2012

Hi,

I am having a problem converting an ASCII file into a fits file.
Converting doesn't give me any error, but when I use splot to plot the spectrum I get:

ERROR: floating point overflow

And when I try to convert that file back to ASCII I get:

ERROR: floating point overflow
  "> specout)"
     line 44: onedspec$wspectext.cl
     called as: `wspectext (mode=h)'

The file I am using is the one here (without the first line):

http://vizier.cfa.harvard.edu/viz-bin/nph-Plot/Vgraph/txt?III%2f219%2f.%2fcluster%2fNGC2157.oi&-X&\*l%20[\oA]&-Y&Normalized%20Flux&--bitmap-size&600x400

Thanks!

---

Mike Fitzpatrick wrote on Mar 03, 2012

Works fine for me on a macintel system with the latest code, so maybe it is a bug that's already been fixed. What IRAF version and platform are you using, also what parameters to RSPECTEXT did you use?

---

cosmologist wrote on Mar 03, 2012

Thanks for your reply. I am using iraf on mac as well:

  NOAO/IRAFNET PC-IRAF Revision 2.14 Fri Nov 30 15:27:05 MST 2007

with these parameters:

(title  =                     ) Spectrum title
(flux   =                  yes) Flux calibrated?
(dtype  =               linear) Dispersion type
(crval1 =                INDEF) Coordinate of first pixel
(cdelt1 =                INDEF) Coordinate interval per pixel
(fd1    =                     )
(fd2    =                     )
(mode   =                   ql)

Thanks

---

cosmologist wrote on Mar 03, 2012

Am I using an older version of Iraf? :shock:

---

Mike Fitzpatrick wrote on Mar 03, 2012

The problem is in your use of INDEF for crval1/cdelt1: these value are written to the header but aren't legal FITS values, and don't make for a sensible WCS anyway. For your data table, reasonable values would be crval1=3690 and cdelt1=2, or else try using dtype=interp to interpolate the dispersion.

---

cosmologist wrote on Mar 03, 2012

Yep! That was it!! Thank you :D

---

Last post on Mar 03, 2012