[evlatests] AR637 1jul07: lost 2 of 10 hours

[Barry Clark]

>     1- no fringes on x-VLA (EVLA-EVLA seem OK) for first hour
>       Unfortunately this includes their only flux and pol'n calibrators!
>       (apart from one with no circular pol'n)

So this happened at all bands?  I just had an instance of it occurring
with Xsysstart, and was inclining to blame the F12 module, but that
is only used at X and Q.  Might have a look at baseline 1-2; this was
giving fringes at a very high fringe rate for me, but definite fringes
(it's a shortish baseline on the N arm).