[evlatests] More on Drop Behavior

[Ken Sowinski]

On Wed, 28 Jun 2006, Ken Sowinski wrote:
>
> When Rick finished I took a few minutes to look at this
> to verify what would happen if the executor were stopped
> or phase switching disabled.  I did this by running a
> private executro using Xyssstart and running the VLA with
> 1-2/3 second intgration interval.  When I did this there
> were no dropouts at all on antenna 18;
                                      ^^
Rick points out that I intended to write 13 for 18 here.

> I don't see how to explain that I never see the antenna 13 dropouts when 
> running Xsysstart while Rick and Walter routinely do when running their 
> test scripts.

Barry points out that the sysstart scripts update the L302s
when they feel like it, but other scripts at a very well defined
time.  This might be trying to tell us something about what
the L302 is doing.  Tests on the bench may help.  The heartbeat
supplied to the L302 in the basement is ultimately derived from
the maser and uniform, but is not synchronized to the UTC second
tick so it might not be the behave the same way as antenna.