A couple of times recently, we've encountered a problem with the results of a sample. We're not sure if it's a software error, GPIB issue, or related to recent power fluctuations and outages that include the building housing the facility. 
The problem in question is best explained with the screenshot located at:
http://comp.uark.edu/~tmillic/curve_bad.jpg
A normal curve can be seen at:
http://comp.uark.edu/~tmillic/curve_good.jpg
Has anybody else encountered this?
Thank you,
-Tom

University of Arkansas Stable Isotope Laboratory