I just stumbled across a rather interesting short article in the July 1996 
issue of NASA Tech Briefs,  which describes a method for testing solar cell 
arrays.
It seems that measuring the capacitance of the array,  with the array dark,  
is a great deal more sensitive to damage to cells than earlier tests.  A 
defect affecting one quarter of one cell in a 40-cell string showed only 0.02 
percent difference with the older test methods,  while measuring the 
capacitance showed about a one percent change.  The temperature coefficient 
of gallium arsenide cells is given as being 5x10e-4 degrees C e-1.
Thought that might be of interest to some folks...
--- 
---------------