[evlatests] P-Band Tsys

[Rick Perley]

    ... I have found why there are no non-zero values for Tsys in my 
P-band data: 

    AIPS defaults to filling the front-end Tsys for the Cassegrain 
bands, and the back-end Tsys for PF bands.  In AIPS, one can reverse 
this with an appropriate flag being set, which I do.  The result of this 
is that I get the back-end Tsys for Cassegrain bands, and the front-end 
for PF bands.  For EVLA antennas, there is no FE Tsys. 

    It would be better if the back-end were filled by default for all 
bands, for at least the EVLA antennas...