[evlatests] 400 ms tests last night

[Walter Brisken]

Last night I spent about an hour looking at 4 bands (L C X K) at 400 ms 
integration times in hope to shed more light on the 10 second dropouts.  I 
failed to remove the calibrator code so some (perhaps 1/4) of the data is 
missing.

Dropouts are clearly seen in all IFs except L band IFs B and D.  There is 
something really weird going though with frequency dependence.  In the A/C 
IF pair the amplitude drops to about 60% of the nominal value for a single 
400 ms integration before recovery, regardless of band.

On the B/D IF:
 	L-band -- dropouts not visible
 	C-band -- dropouts to 97% of nominal amplitude
 	X-band -- dropouts to 95% of nominal amplitude
 	K-band -- dropouts to 65% of nominal amplitude


This data shows "jags" in phase very clearly if that is still of interest.

-W