|
“Understanding Overpotential Testing" With
trilingual summary
Any rotating electric machine undergoes final
testing at the factory to show that it is in good working order. Part
of that involves brief application of a suitably high a-c voltage
between machine winding and frame. The NEMA standard for that "overpotential"
test is twice rated voltage plus 1000 volts, applied for one minute.
Too high a test voltage risks insulation damage, whereas too low a
voltage will not expose insulation weakness.
At successive stages of production, individual
coils or groups of coils are given in-process tests. Some
manufacturers test each coil; others use a sampling method. The
purpose is to search out insulation defects when repair is far
simpler and less costly than on a completed winding. Initially at a
high level, each succeeding test is at a lower voltage, because
repeated stress contributes to eventual fatigue failure. Such tests
are usually "momentary," lasting from 1 to 15 seconds (literature
describing in-process testing seldom defines the time involved).
Accessory circuits built into the machine will
also undergo overpotential tests. Because voltage ratings are
typically much lower, test voltage is correspondingly reduced, For
temperature sensors or other devices lacking a specified operating
voltage, NEMA standards prescribe a rating of 50 volts.
Authorities advise against repeating the
one-minute final test once a machine is in service. Any overpotential
testing then should use a voltage reduced by one-third.
Although standards mandate a-c for the final
factory test, any overpotential test is possible using d-c voltage at
1.7 times the a-c value. Reduced leakage current permits a smaller,
lighter test power supply.
From “The
Understanding Overpotential Testing" to
be published in Electrical Apparatus
October 2006
. Visit our online webstore to
order a
copy. © 2006
Barks Publications, Inc. All Rights Reserved.
|
