[evlatests] Phase slopes, clarifications

[Vivek Dhawan]

This is a transition problem, and a VLBI problem but the
rate is very low, so not a serious problem. I agree it is
low priority. I set it down before I forgot, hoping to
trigger some insight rather than divert attention.

details:

Phase slope on any ONE of A, B, C, D is roughly consistent
with the round-trip phase meter delay change.

Phase slope A-B = R1-R2 = C-D = L1-L2 is non-zero sometimes;
same value on all EVLA antennas.

phase slope A-C = R1-L1 = B-D = R2-L2 = 0 always.

On EA23, A, B, C, D all jumped. A & C jumped 170 deg.
A-B and C-D were ~40 degrees. A-C = B-D = 0.
A, C set to 1341 MHz, B D set to 1666 MHz.