[evlatests] Software time Jan 11

[Barry Clark]

First item was to again check out driving VLA antennas through the CMP.
Spent most of my time cleaning up after minor code changes.  CMP came up
and worked fine from first try.  However, after a couple of hours and
many waveguide resyncs, it failed.  Hichem made a guess about a possible
cause and changed the code.  Worked well from that point, but no guarantee
that this was really the problem.

Proceeded to try controlling VLA correlator, as I had done before.  
Main difficulty was an inability to count, and therefore to convince
the correlator controller and the Modcomps that they had the same number
of antennas.  They were not happy with the Modcomp thinking 10 and the
correlator controller thinking 11.  By the time that was found and fixed
I was too tired to try anything else.  But I did get fringes with the
Executor driving the correlator.  Too weak, but I was using 50 MHz BW and
probably had wildly wrong delays.

Ken briefly tried routing the data to the PC104 port out of the correlator;
he said it wasn't complaining, but I don't know what that means.