Several years ago, I had a chance to connect several vacuum gages
together (all connected via T's to one hose).  When I applied a steady
vacuum, no two gages gave the same reading!  I have a feeling that this
may be the reason for all the different and conflicting readings people
are reporting.

How does one construct a water gage to test and calibrate and check
their own mechanical gage for accuracy?   A friend of mine tried to
construct a water gage, and connected it to his piano.  Somehow the
vacuum in the piano went up so high, the water was sucked up into the
mechanism!  Watch out for that!

Bruce Clark

 [ Craig Brougher and Phil Dayson have described suitable set-ups,
 [ both incorporating a water trap for safety!  -- Robbie