[evlatests] Dropouts, continued ...

[Barry Clark]

Rick's file has a source reinitialization every ten minutes.  One of these
corresponds to his magic 19:42:41 IAT time.  At this time there's a lot
of initialization done.

At 20:19:43, the scrip was aborted by the operator, and another script,
starting on the same source was started.  This corresponds to the second
change in behavior he reports.

So the change in behavior appears to occur on source reinitialization.
Hoever, this doesn't appear to get us any forwarder.  A new command is
sent to the control loop DDS at this time (and only at this time), but 
that shouldn't affect the every-ten-second behavior by any mechanism I 
can think of.

There are centralized timing thingies I can think of that might produce
an effect, but it would be a phase jump, not a short timescale thing like
what we are seeing.

Anyhow, when you get set up for the lab tests, have somebody make up a 
script with source reinitializations, to get a little closer to reality.