Attached are snapshops of all 3 quenches, including the magnet current, the magnet liquid
level and magnet flow rate.  Note that the latter pegs at 100 slpm.
To me it looks like the was damaged during the 2nd quench, which occurred as it
was ramping down.
Here are some observations about each quench.

Quench 1 (Oct 31, 22:40):
Level jumps to 100% at 20:15.  This is probably a result of the magnet electronics cabinet
being powered off.  A batch fill starts just as the magnet is being ramped.  The flow meter
pegs at 100, as a result of the batch fill, not the ramp.  The fill is completed at 21:40 and the
flow drops back down to its normal level while the magnet continues to ramp.  This indicates
that the magnet is superconducting.  Magnet quenches at 22:40, flow pegs as a result.  The
level jumps to 100% (this behavior is due to pressure increase during quench) and then drops
to 0%.

Quench 2 (Nov 3, 8:45)
Batch fill is completed at 20:20, followed by slow drop in flow.
Magnet is at +75 amps, and taken out of persistent mode at 8:25.  Note that flow increases
slightly (due to current in magnet leads and  persistent switch heater), but is still reasonable.
Magnet coils are still superconducting.  Magnet quenches at 60 amp (8:45) followed by liquid
level spike and drop, and flow pegs.  The batch fill should start immediately when the level
hits 20%.  I do not understand the dip in flow rate at 9:05, but it probably is not important.

Quench 3 (Nov 3, 17:45)
Liquid level is falling smoothly after a batch fill at 16:00.  At 17:10 it plateaus
briefly at 50%. (I also see that the fridge flow was decreased at this point,
explaining the plateau.)
Magnet current jumps from 0 to -12 amps at 17:25, and from -12 to -23 amps at 17:40.
Either Labview program was not running or there were communication problems.
Current then increases steadily to -26.4 amps and quenches at 17:42.
Note dramatic rate at which liquid level drops, beginning around 17:45!  Magnet is
probably resistive at this point.

Chris