The laser tripped around 3:20 this morning.  Suspect it is the "usual" problem.  The crystal
chiller was found off but with its dummy still in place.

    The data drops out looks like an artifact of data acquisition.

    Another observation is that the head temperatures all increase when the flow rates drop.
That is not surprising but the humidity also rises.  It might be that we have a tiny water leak
whose effect on humidity is noticeable when the temperature rises.  We have seen evidence of
tiny leaks before because of some corrosion stains in the laser base plate, however when the
laser is observed over time scales of ~30 minutes, no leak is visible.  The (possible) leak is
small enough that it does not make a noticeable difference in the water level of the chiller.