[evlatests] Origin of the Error

[Rick Perley]

    I have uncovered the error which led to my incorrect conclusion 
concerning phase steps throughout the Herc A observation.

    The error was that I failed to set:  DPARM(9)=1, which in AIPS-speak 
means 'don't solve for delay rate'.   Applying CLCAL -- even with 90 
hour boxcar averaging of all rate solutions -- will result in a temporal 
change of phase which somehow created the peculiar stepped phase that I 
reported on yesterday.

    So this problem comes under the heading of 'pilot error'.  But it 
also illustrates how easy it is to make a simple mistake which  can lead 
to quite disastrous consequences.