imcombine arithmetic exception

jkwerk wrote on Apr 30, 2008

Hello,

I'm using IRAF version 2.12.2, for reference. I recently switched to an intel mac, and have been having difficulty with imcombine on data with which imcombine used to work well. I have tried changing virtually every parameter within imcombine (within reasonable allowed values), and am still getting an arithmetic exception every time. Below, I'm attaching an example of one of the image headers (I've tried combining two, three, up to 10 images, all with the same arithmetic exception). I have a feeling it has something to do with the image dimensions, though imcombine never complained before I switched computers so something strange is up. I'd appreciate any help you could give me! Thank you,

Jess Werk

Example image header:

ec> imhead h2cen_1047_2spec.fits l+
h2cen_1047_2spec.fits[4763,60,2][real]: h2cen1
No bad pixels, min=0., max=0. (old)
Line storage mode, physdim [4763,60,2], length of user area 2633 s.u.
Created Tue 17:01:46 25-Mar-2008, Last modified Tue 17:01:45 25-Mar-2008
Pixel file "h2cen_1047_2spec.fits" [ok]
EXTEND = F / File may contain extensions
ORIGIN = 'NOAO-IRAF FITS Image Kernel July 2003' / FITS file originator
DATE = '2008-03-25T20:01:46' / Date FITS file was generated
IRAF-TLM= '17:01:45 (25/03/2008)' / Time of last modification
DISPAXIS= 1
CTYPE1 = 'LINEAR '
CTYPE2 = 'LINEAR '
WAT0_001= 'system=world'
WAT1_001= 'wtype=linear label=Wavelength units=Angstroms'
WAT2_001= 'wtype=linear'
CRVAL1 = 3000.
CDELT1 = 0.756
CD1_1 = 0.756
CRPIX1 = 1.
DC-FLAG = 0
SLITLEN = 60 / slit length in pixels
SLITTYPE= 2 / Type of aperture
CNTRLINE= 32 / Spectrm central row
OBJECT = 'h2cen1 ' / Name of object
SLITNUM = 25 / Slit number
AIRMASS = 1.93700 /start airmass
EPOCH = 2008.03017000 /
EXPTIME = 2000.00000000 /exposure time (sec)
LC-TIME = '22:34:03' /local time
UT-TIME = '01:34:03' /
UT-END = '02:07:24' /
DATE-OBS= '2008-01-12' /
RA = ' 09:25:29.0' /
DEC = '-76:38:06.2' /
RA-D = 141.370958300 /ra-degrees
DEC-D = -76.6350556000 /dec-degrees
SITENAME= 'LCO ' /
EQUINOX = 2000.00000000 /
ST-HOURS= 4.25133 /

---

Mike Fitzpatrick wrote on Apr 30, 2008

There are two things to check here. The first is that you're not seeing a known bug, specifically:

NUMBER: 552
MODULE: imcombine
SYSTEM: V2.12.2
DATE:   Tue Feb 28 14:07:30 MST 2006
FROM:   valdes

BUG:    When the combine option is sum and weighting is selected an exception
        occurs.  The sum option was added as a later feature and a
        flag needed for weighting was not set.  There is no workaround to
        combine weighting and summing directly.  One can use "average" and
        the scale the result to the equivalent of a sum.

STATUS: Fixed for V2.12.3.

You didn't send your task parameters, but see whether you're using the sum and weighting option talked about.

Second, v2.12.2 didn't have a Mac Intel port, so I assume you're using the PPC binaries and the Rosetta emulator? This will let the binaries run but I vaguely remember having to fix the FPE handling in v2.13 because the size of some system structure had changed and so false errors were being triggered (or in some cases, ignored). Upgrading to the v2.14 release will not only let things run faster because of the native binaries, but will pick up the above fixes as well.

To rule out any issue with the image dimensions, you can use the IMSLICE task to reduce the image to 2-D. It would be interesting to know whether this fixes the problem.

Cheers,
-Mike

---

Last post on Apr 30, 2008