Page 83 - Basic PD Theory
P. 83

References


               Proof testing
               The proof test is a go-no go test that involves application of a DC test voltage that is 1.7 times the AC proof value of twice V L-L
               + 1 kV.   This test is generally used for a new winding or after major maintenance has been done on a winding and there is a
               potential for severe damage.

               Maintenance Testing
               The maintenance test is conducted at approximately 75% of the proof test voltage and may be used as more of a predictable
               testing tool.  Often, a maintenance test can be done by applying the voltage in a step, or ramp method.   The hope is by
               monitoring the current you can predict the point of failure before breakdown occurs.  Though not always possible, this
               deflection point is sometimes observed by a sudden change in the current that results in the presence of a “knee” on the current
               vs. voltage plot.  A knee is an observed phenomenon and often difficult to define mathematically, some folks say a doubling
               between voltage intervals describes the knee.  If the current avalanches, there is obviously a problem. This test is not normally
               performed on salient pole rotor windings.
                                                                Total current



                The voltage can be applied in the following methods:         Ramp test
                •  Ramp Tests - steady increase in voltage
                •  Fixed-interval step test - <3% change at 1
                   minute intervals
                •  Graded-time interval step test - remove effects                           Step test
                   of absorption current on the slope of the V/I
                   curve




                                                                              Applied Voltage

                                                        Figure 60.  The "knee" in the leakage current during a direct voltage test.
               7.  Alternating Voltage Hipot

               Like the direct voltage Hipot test, this test is a go or no-go test.  It is strictly a test to determine if     IEEE Std 432:
               the winding will survive an over-voltage stress typical of a bus fault, or is sealed from moisture   Guide for Insulation
               ingress.   It is a destructive test by design in that if a failure occurs the machine is not suitable for   Maintenance for Rotating
               return to service until the failure is repaired.  The ac hipot stresses the endwinding similarly to   Electrical Machinery (5 hp to
               that induced during normal operation.  Recommended levels are as follows:   less than 10,000 hp)
                       2 * V L-L + 1kV for acceptance of a new winding
                       1.5 * V L-L for maintenance  (some prefer 130% before maintenance and 120% after)
                                              Finds:  severe cracks


               8.  Wedge Tap Test

               If it is suspected that the coil wedges are loose then a wedge tap test should be perform to verify this.  Tap each wedge at a
               number of locations along its length with a ball peen hammer or other metal object and put a finger on the wedge near where it
               is struck.  If the wedge is tight there will be a “pinging” sound and no wedge vibration will be felt.  If the wedge is loose there will
               be a dull “thud” sound and wedge vibration will be felt.  The recommended acceptance criteria for this test is:.
                     No two adjacent wedges in the same slot should be loose




                                                                                 www.irispower.com

                                                                                  78 | P ag e
   78   79   80   81   82   83   84   85   86   87   88