CORE LOSS TESTING IN THE PRACTICAL MOTOR …

1 Copyright 1991. All Rights ReservedPage 1 CORE LOSS TESTING IN THE PRACTICAL MOTOR REPAIRENVIRONMENTABSTRACTL exseco developed the first commercial Core Loss Tester fifteen years ago. Over the years, Lexseco performedthousands of core loss tests in a MOTOR repair shop environment. Hundreds of Lexseco testers are in use aroundthe world. Until now, the only real source of guidance for performing core loss tests has been several articlespublished in the last decade. This publication will present the mathematical foundations of the toroidal transformertype core loss test and discuss the application of this test procedure to the MOTOR repair environment includingcommon core repair TO LOSSESB efore any test procedure for determining core lossesand core degradation can be explained, the nature ofsteel properties should be understood.

2 Ferrousmaterials have various electrical, magnetic, andmechanical properties that are taken intoconsideration during the design of an electricalapparatus. The object of the design is typically tomaximize output while minimizing the associatedelectrical and mechanical losses. By design, themotor is known to have a certain level of I2R copperloss, windage and friction, and stray load losseswhich accompany the specific design parameters ofthe MOTOR . Similarly, designers choose the grade ofcore steel, the desired thickness of the steel, and theprocessing techniques in an attempt to minimize thesteel losses without incurring a substantial increasein material costs.

3 These losses are divided into losses2. Eddy current lossesFigure 1 - Hysteresis Loss CurveHysteresis loss is the amount of input energyexpended to change the magnetic polarity of thesteel in conjunction with the changing polarity of thealternating current waveform. This portion of thetotal core loss is represented by the shaded portionof the hysteresis curve shown in Figure 1. Ideally,the voltage (VS) to magnetization current (IM) graphwould be represented by a single path from A to Bequal to the path from B to A. However, due to thisincrease in expended energy, the path from A to B isnot equal to the path from B to A, and a hysteresisloop is developed.

4 The area between the two pathsis identically equal to the hysteresis loss. Hysteresisloss will be dissipated in the form of currents are circulating currents resulting fromthe magnetic fields generated in the electric MOTOR . Normal eddy current flow is shown in Figure 2. Figure 2 Normal Eddy Current FlowWhen any electrical conductor is placed within amagnetic field, a current proportional to the cross-section of the conductor and the strength of the fieldis known to flow perpendicular to that field. For thisreason, steel lamination thickness is minimized toreduce the amount of eddy current the insulation between laminations breaksdown, there is an associated increase in the eddycurrent flow.

5 At the point of degradation, currentsCopyright 1991. All Rights ReservedPage 2 flow between the laminations as shown in Figure current loss is the total input power lost tothese circulating currents. Like hysteresis loss, eddycurrent losses are dissipated in the form of 3 - Interlaminar Eddy Current FlowSumming the eddy current and hysteresis losses willprovide the total core losses for any given LOSS TEST PROCEDUREST oday, there exist two types of tests for determiningthe total core loss of an electric MOTOR . The first ofthese is known as the segregated loss methoddescribed in the Institute of Electrical and ElectronicsEngineers (IEEE) Standard 112 and the CanadianStandards Association Standard 390.

6 Using linearregression analysis and various percentage loadingcharacteristics, this test method determines eachcomponent of the total losses as well as the subjectmotor efficiency. When the exact nature of corelosses in the complete MOTOR system are required,this type of test is much more effective than anyother test of its type. However, the subject motormust be in complete working order. This limits itsusage to MOTOR repairs which can be classified aspreventive maintenance repairs. Second, the testingprocedure must be performed in laboratory controlledenvironments. Subletting core loss TESTING tolaboratories specifically devoted to MOTOR TESTING isa costly and time consuming venture.

7 Conservativeestimates of the time and cost of segregated losstesting is approximately one day per MOTOR atupwards of $1000 per day. This further reduces theset of motors to non-emergency breakdowns. Obviously, for the MOTOR repair industry, thesegregated loss method for determining the condition of the core is very other method, which is extremely well suited tothe MOTOR repair industry, is commonly referred to asthe Toroidal Transformer Test or loop test as shownin Figure 4 - Toroidal Transformer TestWhile there exist several variations on the toroidaltransformer test procedure, they all share the samebasic mathematical foundations.

8 The principlesunderlying the traditional "loop test," and thecommercial Core Loss Tester, are based upon anelectromagnetic equation known as Faraday's Law. This application of Faraday's Law (1) holds that anylaminated steel core assembly, regardless of size ordesign, can be used as the core of a toroidaltransformer. Once excited to a selected backironflux density, the power lost to the core steel can bemeasured. More specifically,)dtdn( = v (1)where v is the induced voltage, n is the number ofturns in the transformer, and is the flux in thecore. Given thatc B = M (2)where BM is the flux density and c is the cross-sectional area of the toroid, we can directlyincorporate the flux density (2) into Faraday'sEquation (3).

9 Dtc) Bd(n( = vM (3)Copyright 1991. All Rights ReservedPage 3 With the loop test, the voltage (v) in Faraday'sequation is held constant at the value supplied by thefixed power source. With a known target back ironflux density (BM), this equation can be solved for thenumber of turns (n). By wrapping the calculatednumber of turns around the core, the target fluxdensity is achieved. Being both labor intensive andapplicable to only stator cores, the loop test is oftenomitted from the repair procedure of most the commercial core loss tester, the number ofturns (n) is held constant at one turn, and the voltagesource is varied.)

10 This achieves the same goal whilesimplifying the test procedure. To further enhancethe accuracy of the test flux density level, theequation can be further simplified to solve for theflux ( ) instead of the voltage (v). This solution,while proprietary to Lexseco, Inc., is both waveshapeand frequency independent, thus removing theeffects of inherent transformer harmonics from the target back iron flux density has beenreached, the excitation current, induced voltage, andthe excitation and resultant power levels aremeasured. By taking the difference between theexcitation power on the primary and the resultantpower on the secondary of the toroidal transformer,the loss to the core steel is determined.