You are on page 1of 48

AS 1359.102.

11997
IEC 34-2:1972 IEC 34-2:1972/Amd.1:1995 IEC 34-2:1972/Amd.2:1996

Australian Standard
Rotating electrical machines General requirements Part 102.1: Methods for determining losses and efficiencyGeneral
[ IEC title: Rotating electrical machines, Part 2: Methods for determining losses and efficiency of rotating electrical machinery from tests (excluding machines for traction vehicles)]

This Australian Standard was prepared by Committee EL/9, Rotating Electrical Machinery. It was approved on behalf of the Council of Standards Australia on 10 March 1997 and published on 5 July 1997.

The following interests are represented on Committee EL/9: Australian British Chamber of Commerce Australian Chamber of Commerce and Industry Australian Electrical and Electronic Manufacturers Association Bureau of Steel Manufacturers of Australia Department of Defence Electricity Supply Association of Australia Institution of Engineers Australia

Review of Australian Standards. To keep abreast of progress in industry, Australian Standards are subject to periodic review and are kept up to date by the issue of amendments or new editions as necessary. It is important therefore that Standards users ensure that they are in possession of the latest edition, and any amendments thereto. Full detail s of all Australian Standards and related publications will be found in the Standards Australi a Catalogue of Publications; this information is supplemented each month by the magazine The Australian Standard, which subscribing members receive, and which gives details of new publications, new editi ons and amendments, and of wit hdrawn Standards. Suggestions for improvements to Australian Standards, addressed to the head office of Standards Australia, are welcomed. Notification of any inaccuracy or ambiguity found in an Australian Standard should be made without delay in order that the matter may be investigated and appropriate acti on taken.

This Standard was issued in draft form for comment as DR 96091 and 96206.

www.bzfxw.com

AS 1359.102.11997

Australian Standard
Rotating electrical machines General requirements

Part 102.1: Methods for determining losses and efficiencyGeneral

Originated as part of AS 1359.331983. Revised and redesignated in part as AS 1359.102.1 1997.

PUBLISHED BY STANDARDS AUSTRALIA (STANDARDS ASSOCIATION OF AUSTRALIA) 1 THE CRESCENT, HOMEBUSH, NSW 2140
ISBN 0 7337 1178 2

ii

PREFACE
This Standard was prepared by the Standards Australia Committee EL/9, Rotating Electrical Machinery to supersede, in part, AS 1359.33 1983, General requirements for rotating electrical machines , Part 33: Methods for determining losses and efficiency . It is identical to and has been reproduced from IEC 34-2:1972, Rotating electrical machines , Part 2: Methods for determining losses and efficiency of rotating electrical machinery from tests (excluding machines for traction vehicles ), including Amd.1:1995 and Amd.2:1996 as indicated by marginal bars near the affected text. Reproduction was done by scanning the IEC text and adjusting the style of the original publication to conform with later IEC style. This Standard is a Part of the AS 1359 series listed in AS 1359.0, Part titled: Introduction and list of Parts. The objective of this Standard is to provide the rotating electrical machine industry with standard methods for determining losses and efficiency. The objective of this Revision is to clarify certain methods and to provide more details of the retardation method (from IEC 34-2 Amd.1); to amend the reference temperature used for correcting I 2 R losses (from IEC 34-2 Amd.2); and to transfer the calorimetric method to a new AS 1359.102.2. As this Standard is reproduced from an International Standard, the following applies: (a) (b) (c) (d) Its number does not appear on each page of text and its identity is shown only on the cover and title page. In the source text this Recommendation should read this Australian Standard. A full point substitutes for a comma when referring to a decimal marker. References to International Standards should be replaced by references to Australian Standards, as follows: Australian Standard 1359 Rotating electrical machines General requirements 1359.101 Part 101: Rating and performance 1359.102.2 Part 102.2: Methods for determining losses and efficiency Calorimetric method 1852 International Electrotechnical Vocabulary Rotating electrical machines Part 1: Rating and performance First supplement: Measurement of losses by the calorimetric method International Electrotechnical Vocabulary (IEV) Direct acting indicating analogue electrical measuring instruments and their accessories

Reference to International Standard 34 34-1 34-2A

50 51

The references in Paragraph A.1.4 to clauses 11 and 13 and to table II of IEC 34-2A apply respectively to Clauses 4.4, 3.7 and Table 2 of AS 1359.102.2.

www.bzfxw.com

iii

CONTENTS
Page
Clause SECTION ONE - GENERAL 1 2 3 4 Scope . . . . . . . . . . . . . . . . . . . . . . . Object . . . . . . . . . . . . . . . . . . . . . . . General . . . . . . . . . . . . . . . . . . . . . . 3.1 List of symbols . . . . . . . . . . . . Definitions . . . . . . . . . . . . . . . . . . . . 4.1 Efficiency . . . . . . . . . . . . . . . . 4.2 Total loss . . . . . . . . . . . . . . . . 4.3 Braking test . . . . . . . . . . . . . . 4.4 Calibrated driving machine test 4.5 Mechanical back-to-back test . 4.6 Electrical back-to-back test . . . 4.7 Retardation test . . . . . . . . . . . 4.8 Calorimetric test . . . . . . . . . . . 4.9 No-load test . . . . . . . . . . . . . . 4.10 Open-circuit test . . . . . . . . . . . 4.11 Sustained short-circuit test . . . 4.12 Zero power factor test . . . . . . . 5 Reference temperature . . . . . . . . ... . .. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .. .. .... . .. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... .. ..... ..... ..... . .. .. ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... . .. .. ..... ...... ... ... ...... .. .. .. . .. .. . .. ... . .. ... . .. .. .. ...... ...... ...... ..... . ..... . .. .. .. ..... . ...... ...... .. ... . 1 1 1 2 2 2 2 3 3 3 3 3 3 3 3 3 3 4

SECTION TWO - D.C. MACHINES . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . .

Losses to be included . . . . . . . . . . . . 6.1 Excitation circuit losses . . . . . . . 6.2 Constant losses . . . . . . . . . . . . 6.3 Load losses . . . . . . . . . . . . . . . 6.4 Additional load losses . . . . . . . . Determination of efficiency . . . . . . . . . 7.1 Summation of losses . . . . . . . . . 7.2 Total loss measurement . . . . . . 7.3 Direct measurement of efficiency

. . . . . . . . .

. . . . . . . . .

.. .. .. .. .. .. .. .. ..

. . . . . . . . .

. 5 . 5 . 5 . 5 . 6 . 6 . 6 . 10 . 10

SECTION THREE - POLYPHASE INDUCTION MACHINES 8 Losses to be included . . . . . . . . . . . . 8.1 Constant losses . . . . . . . . . . . . 8.2 Load losses . . . . . . . . . . . . . . . 8.3 Additional load losses . . . . . . . . Determination of efficiency . . . . . . . . . 9.1 Summation of losses . . . . . . . . . 9.2 Total loss measurement . . . . . . 9.3 Direct measurement of efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. . . . . . . . . .. .. .. .. . . . . 11 11 11 11 12 12 14 14

iv
Page
Clause SECTION FOUR - SYNCHRONOUS MACHINES 10 Losses to be included . . . . . . . . . . . . 10 1 Constant losses . . . . . . . . . . . . 10.2 Load losses . . . . . . . . . . . . . . . 10.3 Excitation circuit losses . . . . . . . 10.4 Additional load losses . . . . . . . . 11 Determination of efficiency . . . . . . . . . 11.1 Summation of losses . . . . . . . . . 11.2 Total loss measurement . . . . . . 1 11.3 Direct measurement of efficiency . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . 15 15 15 15 16 16 16 19 19

SECTION FIVE - METHODS OF TEST 12 13 14 15 General . . . . . . . . . . . . . . . . . . . . Calibrated machine test . . . . . . . . . . Zero power factor test . . . . . . . . . . . Retardation method . . . . . . . . . . . . . 15.1 General . . . . . . . . . . . . . . . . . 15.2 Composition of retardation tests 15.3 Retardation test procedure . . . 15.4 Taking of measurements . . . . . 16 Electrical back-to-back test . . . . . . . 17 Calorimetric test . . . . . . . . . . . . . . . 18 Schedule of preferred tests . . . . . . . 18.1 D.C. machines . . . . . . . . . . . . 18.2 Polyphase induction machines . 18.3 Synchronous machines . . . . . . ... ... .. . ... ... .. ... ... ... ... ... ... ... ... .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .... .... .. .. .... .... .... .... .... .... .... .... .... .. .. .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .... .. .. .... .... .... .... .... .... .... .... .... .... .... ... .. .. . .. ... .. . ........ . .. .. .. . . .. ... .. ........ ........ ........ ........ . .. .. ... ........ . . .. .. .. ... ... .. ........ 20 21 21 21 22 23 25 27 29 29 29 29 30 30

FIGURES (1 to 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 ANNEX A 2 Provisional methods for determining losses and efficiency of converter-fed cage induction machines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Copyri ght

STANDARDS AUSTRALIA

Users of Standards are reminded that copyri ght subsists in all Standards Austr alia publi cati ons and software. Except where the Copyri ght Act all ows and except where provided for below no publi cati ons or soft ware produced by Standards Austr alia may be reproduced, stored in a retr ieval system in any form or transmitt ed by any means without prior permission in writ ing from Standards Australi a. Permission may be condit ional on an appropri ate royalt y payment. Requests for permission and information on commercial software royalt ies should be directed to the head offi ce of Standards Austr alia. Standards Austr alia wil l permit up to 10 percent of the technical content pages of a Standard to be copied for use exclusively in-house by purchasers of the Standard without payment of a royalty or advice to Standards Austr alia. Standards Austr alia wil l also permit the inclusion of its copyri ght materi al in computer soft ware programs for no royalt y payment provided such programs are used exclusively in-house by the creators of the programs. Care should be taken to ensure that materi al used is from the current editi on of the Standard and that it is updated whenever the Standard is amended or revised. The number and date of the Standard should therefore be clearly identif ied. The use of materi al in print form or in computer software programs to be used commercially, with or without payment, or in commercial contracts is subject to the payment of a royalty. This policy may be vari ed by Standards Austr alia at any ti me.

www.bzfxw.com

1 AUSTRALIAN STANDARD

Rotating electrical machines General requirements Part 102.1: Methods for determining losses and efficiency General

SECTION ONE - GENERAL

Scope

This Recommendation applies to d.c. machines and to a.c. synchronous and induction machines of all sizes within the scope of IEC Publication 34-1. The principles can, however, be applied to other types of machines such as rotary converters, a.c. commutator motors and single-phase induction motors for which other methods of determining losses are generally used.

Object

This Recommendation is intended to establish methods of determining efficiencies from tests, and also to specify methods of obtaining particular losses when these are required for other purposes.

General

Tests shall be conducted on a completely sound machine with all covers fitted in the manner required for normal service, with any devices for automatic voltage regulation not a composite part of the machine itself being made inoperative, unless otherwise agreed.
1 Unless otherwise agreed, measuring instruments and their accessories, such as measuring

transformers, shunts and bridges used during the test shall have an accuracy of 0,5 or better (IEC 51), excluding three-phase wattmeters and wattmeters for low power factor, for which an accuracy class shall be 1,0 or better. Instruments shall be selected to give readings over the effective range such that a fraction of a division is a small percentage of the actual reading and can be easily estimated. On machines with adjustable brushes, the brushes shall be placed in the position corresponding to the specified rating. For measurements on no-load, the brushes may be placed on the neutral axis. Speed of rotation may be measured by a stroboscopic method, digital counter or tachometer. When measuring slip, the synchronous speed should be determined from the supply frequency during the test.
COPYRIGHT

2
When the over-all efficiency or the absorbed power is measured for a group of machines comprising two electrical machines, or a machine and a transformer, or a generator and its driving machine, or a motor and its driven machine, there is no need to indicate the individual efficiencies. If, however, these are given separately, they should be regarded as approximate. 3.1 List of symbols

A list of symbols used in the draft, with the general meanings attributed to each one, is given below:
1

C I I1 I 1r Io I or J n nN N P P1 P 1r P Fe Pf Pk Pt S s U Ue Un Ur r o or

= = = = = = = = = = = = = = = =

= = = = = = = = = = = =

retardation constant current load current at rated voltage main primary current at reduced voltage no-load current at rated voltage no-load current at reduced voltage moment of inertia speed of rotation in revolutions per minute rated speed, number of full revolutions of the shaft losses which can be directly measured power absorbed at rated voltage power absorbed by main primary winding at reduced voltage iron losses defined in accordance with 6.2 a), 8.1 a) and 10.1 a) friction and windage losses (mechanical losses defined in accordance with 6.2 b), 6.2 c), 8.1 b), 8.1 c), 10.1 b) and 10.1 c)) short-circuit losses representing the sum of the I 2 R losses in operating windings on load in accordance with 10.2 and additional load losses in accordance with 10.4 total of the losses during the retardation test angular displacement of the machine shaft slip excitation voltage across terminals of main rheostat total excitation voltage rated voltage reduced voltage for load test per unit deviation of rotational speed from rated speed load phase angle at rated voltage load phase angle at reduced voltage no-load phase angle at rated voltage no-load phase angle at reduced voltage

Definitions

For definitions of general terms used in this Recommendation, reference should be made to the International Electrotechnical Vocabulary [IEC Publication 50]. For the purpose of this Recommendation, the following definitions apply: 4.1 Efficiency The ratio of output to input expressed in the same units and usually given as a percentage. 4.2 Total loss The difference between the input and the output.
COPYRIGHT

www.bzfxw.com

3
4.3 Braking test A test in which the mechanical power output of a machine acting as a motor is determined by the measurement of the shaft torque, by means of a brake or dynamometer, together with the rotational speed. Alternatively, a test performed on a machine acting as a generator, by means of a dynamometer to determine the mechanical power input. 4.4 Calibrated driving machine test A test in which the mechanical input or output of an electrical machine is calculated from the electrical output or input of a calibrated machine mechanically coupled to the machine on test. 4.5 Mechanical back-to-back test A test in which two identical machines are mechanically coupled together, and the total losses of both machines are calculated from the difference between the electrical input to one machine and the electrical output of the other machine (see figure 1). 4.6 Electrical back-to-back test A test in which two identical machines are mechanically coupled together, and they are both connected electrically to a power system. The total losses of both machines are taken as the power input drawn from the system (see figure 2). 4.7 Retardation test A test in which the losses in a machine are deduced from the rate of deceleration of the machine when only these losses are present. 4.8 Calorimetric test A test in which the losses in a machine are deduced from the heat produced by them. The losses are calculated from the product of the amount of coolant and its temperature rise, and the heat dissipated in the surrounding media. 4.9 No-load test A test in which the machine is run as a motor providing no useful mechanical output from the shaft. 4.10 Open-circuit test A test in which a machine is run as a generator with its terminals open-circuited. 4.11 Sustained short-circuit test A test in which a machine is run as a generator with its terminals short-circuited. 4.12 Zero power factor test A no-load test on a synchronous machine which is over-excited and operates at a power factor very close to zero.

COPYRIGHT

4
5

Reference temperature

2 Unless otherwise specified, all I 2R losses shall be corrected to the temperatures given below:

Thermal class of the insulati on system

Reference temperature C

A, E 75 B 95 F 115 H 130 If the rated temperature rise or the rated temperature is specified as that of a lower thermal class than that used in the construction, the reference temperature shall be that of the lower thermal class.

COPYRIGHT

www.bzfxw.com

5 SECTION TWO - D.C. MACHINES

Losses to be included

The total losses may be taken as the sum of the following component losses: 6.1 Excitation circuit losses a) b)

I 2R losses in shunt or separately excited windings and in the excitation rheostats.


Exciter losses.

All the losses in an exciter mechanically driven from the main shaft, which forms part of the complete unit and is used solely for exciting the machine, together with losses in the rheostat in the excitation circuit of such an exciter, but with the exception of friction and windage losses. In the case of a separate excitation supply such as battery, rectifier or motor generator set, no allowance is made for the losses in the excitation source or in the connections between the source and the brushes.
NOTE - When the losses in a separate excitation system are requir ed, these should be listed separately and can be taken as the diff erence between the excitation power divided by the eff iciency of the excitation system, and the excitation power.

6.2

Constant losses a) b) Losses in active iron, and additional no-load losses in other metal parts. Losses due to friction (bearings and brushes) not including any losses in a separate lubricating system. Losses in common bearings shall be stated separately, whether or not such bearings are supplied with the machine.
NOTE - When the losses in a separate lubricating system are requir ed, these should be listed separately.

c)

The total windage loss in the machine including power absorbed in integral fans and in auxiliary machines, if any, forming an integral part of the machine. The losses in auxiliary machines such as external fans, water and oil pumps not forming an integral part of the machine, but provided exclusively for the machine in question, shall be included only by agreement.
NOTE - When the losses in a separate ventilating system are required, they should be listed separately as they are not part of the machine losses.

6.3

Load losses a)

I 2R losses in armature, and windings carrying armature current (e.g. commutating, compensating, excitation and series connected windings).
Electrical losses in brushes.

b)

COPYRIGHT

6
6.4 Additional load losses a) Losses introduced by load in active iron, and other metal parts other than the conductors. Eddy current losses in armature conductors caused by current dependent flux pulsation and commutation. Losses in the brushes caused by commutation.

b)

c)

NOTE - These losses are sometimes called additional losses, but they do not include the additional no-l oad losses in 6.2 a).

7
7.1

Determination of efficiency
Summation of losses

The efficiency can be calculated from the total losses which are assumed to be the summation of the losses obtained in the following manner: 7.1.1 Excitation losses These are: 7.1.1.1 Excitation winding I 2R losses These losses are calculated from the formula I 2 R, where R is the resistance of the shunt excitation winding (or separately excited winding), corrected to the reference temperature, and I is the excitation current. Except for case c) below, the excitation current shall be that corresponding to rated speed under rated load conditions. For case c) below, the excitation current shall be that corresponding to rated speed at no-load. If the excitation current cannot be measured during a test on load, it should be taken as: a) For shunt connected or separately excited generators with or without commutation poles; 110 % of the excitation current, corresponding to no-load at a voltage equal to the rated voltage plus ohmic drop in the armature circuit (armature, brushes and commutating windings if any, see also 7.1.3.2) at rated load current. For compensated shunt or separately excited generators: the excitation current corresponding to no-load at a voltage equal to the rated voltage plus the ohmic drop in the armature circuit (armature, brushes, commutating windings and compensating windings, see also 7.1.3.2) at rated load current. For level-compounded generators: the excitation current for the rated no-load voltage. For over-compounded and under-compounded generators, and special types of generator not covered by items a) to c): as agreed between manufacturer and purchaser. For shunt wound motors: equal to no-load excitation current corresponding to the rated voltage.

b)

c)

d)

e)

COPYRIGHT

www.bzfxw.com

7
7.1.1.2 Main rheostat losses These losses are calculated from the formula I 2R, where R is the resistance of the part of the rheostat in circuit for the rating considered, and I is the value of the excitation current defined as in 7.1.1.1 above. They are also equal to the product, IU, of the excitation current multiplied by U , the excitation voltage which must be absorbed in the rheostat. The sum of the losses, 7.1.1.1 and 7.1.1.2, is also equal to the product IU e of the excitation current I and the total excitation voltage U e .
NOTE - Where a resistance is permanently connected in series in the excitation cir cuit it should be dealt with in the same way as the main rheostat.

7.1.1.3 Exciter losses


NOTE - This applies only to the case where the exciter is mechanically dri ven from the main shaft and is used solely for exciting the main machine.

These losses include the difference between the power absorbed at the shaft by the exciter and the useful power which it provides at its terminals,* as well as the excitation losses in the exciter if this is excited from a separate source. If the exciter can be uncoupled from the main machine and tested separately, the power which it absorbs may be measured by using the calibrated-machine method. If the exciter cannot be uncoupled from the main machine, the power which it absorbs may be measured either by the method of working the main machine as a motor on no-load, or by the calibrated machine method (clause 13), or by the retardation method (clause 15), applied to the whole unit. In these three methods, the power absorbed by the exciter is obtained as the difference between the total losses of the unit measured under identical conditions, first with the exciter on-load and secondly with the exciter non-excited, the excitation being supplied by an independent source. If none of these methods is applicable, the power absorbed by the exciter is obtained by adding to the power, measured at the terminals, the different separate losses determined as under clause 6. However, mechanical friction and windage losses which are measured at the same time as those of the main machine need not be taken into account. 7.1.2 Constant losses 7.1.2.1 No-load test at rated voltage The constant losses shall be determined by running the machine under no-load conditions as a motor with rated voltage applied and with rated speed achieved by adjustment of the excitation, which shall preferably be derived from a separate source. The total electric power absorbed, less the I 2R losses in the armature and in the excitation winding or, if necessary, less the power absorbed by the exciter, gives the sum of the constant losses.

* The useful power at the terminals of the exciter is equal to the sum of the losses, 7.1.1.1 and 7.1.1.2, of the main machine. COPYRIGHT

8
7.1.2.2 Open circuit test The constant losses can be determined separately by driving the machine at its rated speed by means of a calibrated machine. The machine on test is excited (preferably from an independent source), so as to work as a generator on no-load at a voltage equal to its rated voltage, the power which it absorbs at its shaft, and which can be obtained from the electric power absorbed by the calibrated machine, giving the sum of the constant losses. By removing the excitation, the sum of the friction and windage losses is obtained in the same way. The core losses may be determined separately by subtracting the losses during this test from those measured during the previous no-load test. By lifting the brushes the brush friction loss may be determined separately by subtracting the losses during this test from those measured during the previous, unexcited test. 7.1.2.3 Retardation test In machines with large inertia, the total constant losses, as well as the separate constant losses, can be determined by the retardation method. 7.1.3 Load losses These are: 7.1.3.1 Armature circuit I 2R losses These losses are calculated from the current and the measured resistance, corrected to the reference temperature, except that where resistance measurement is impracticable due to very low resistances, calculation is permissible.
NOTE - Under this heading are included compensating windings, commutating pole windings and diverters. In the case of divert ers in parallel with a series winding, the I 2R losses should be determined using the total curr ent and the resulting resistance.

7.1.3.2 Electrical losses in brushes The sum of these losses shall be taken as the product of the armature current and a fixed voltage drop. The voltage drop allowed for all brushes of each polarity shall be 1.0 V for carbon or graphite brushes and 0.3 V for metal-carbon brushes, i.e. a total drop of 2.0 V for carbon or graphite brushes, and 0.6 V for metal-carbon brushes. 7.1.4 Additional load losses Unless otherwise specified, it is assumed that these losses vary as the square of the current, and that their total value at maximum rated current is, for:

Uncompensated machines
1% of the rated input for motors; 1% of the rated output for generators.

Compensated machines
0.5% of the rated input for motors; 0.5% of the rated output for generators.

COPYRIGHT

www.bzfxw.com

9
For constant speed machines, the rated output or input as appropriate is taken as the output or input which would be obtained at maximum rated current and maximum rated voltage. For variable speed motors where the speed change is obtained by applied voltage, the rated input is defined at each speed as being the input when the maximum rated current at any speed is associated with the applied voltage of the particular speed considered. For variable speed motors where the increase in speed is obtained by weakening the field, the rated input is defined as being the input when the rated voltage is associated with the maximum rated current. For variable speed generators where the voltage is maintained constant by varying the field, the rated output is defined as being the output which is available at the terminals at rated voltage and maximum rated current. The allowances for additional losses at the speed corresponding to the full field shall be as specified above. The allowances for additional losses at other speeds shall be calculated using the appropriate multiplying factors given in table 1.
Table 1 Multiplying factors for different speed ratios
Speed rati o 1.5:1 2 :1 3 :1 4 :1 Mult iplying factor 1.4 1.7 2.5 3.2

The speed ratio in the first column of table 1 shall be taken as the ratio of actual speed under consideration to the minimum rated speed for continuous running. For speed ratios other than those given in table 1 the appropriate multiplying factors can be ascertained by interpolation.
NOTE - The additional load loss may be obtained fr om an input-output test or from a back-t o-back test by subtr acting from the total measured losses all other known losses.

7.1.4.1 Change in core loss due to load In general, this variation is usually negligible. By special agreement, for very low voltage machines, the sum, 6.2 a) and 6.4 a), may be measured as described for the constant losses in active iron, 6.2 a) by one or other of the two methods, by operating as a motor on no-load or as a generator on no-load, but instead of making the test at the rated voltage, the test is made at the rated voltage increased or decreased by the voltage drop in the armature circuit for the current considered, depending on whether the machine is a generator or a motor.
1 7.1.4.2

Additional load losses in d.c. motors supplied by static power converters

Whenever the current ripple factor (see 2.29, IEC 34-1: 1994) of the armature current exceeds 0,1, the additional losses caused by the a.c. component of the armature current shall be considered in addition to the losses specified in 7.1.4. They shall be calculated as the eddy current losses caused by the fundamental component of the above-mentioned a.c. component. The method of calculation used shall be the subject of agreement between manufacturer and purchaser.

COPYRIGHT

10
1 7.2

Total loss measurement

7.2.1 Electrical back-to-back tests (see clause 16) When identical machines are run at essentially the same rated conditions, the losses supplied from the electrical system are assumed to be equally distributed and the efficiency is calculated as 7.3.3. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made. 7.3 Direct measurement of efficiency

7.3.1 Braking test When the machine is run at rated conditions of speed, voltage and current, the efficiency is then taken as the ratio of output to input. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made. 7.3.2 Calibrated machine test (see clause 13) When the machine is run at rated conditions of speed, voltage and current, the efficiency is taken as the ratio of output to input. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made. 7.3.3 Mechanical back-to-back test When identical machines are run at essentially the same rated conditions, the losses are assumed to be equally distributed, and the efficiency is calculated from half the total losses and the electrical input (in the case of a motor) or electrical output (in the case of a generator). The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made.

COPYRIGHT

www.bzfxw.com

11 SECTION THREE - POLYPHASE INDUCTION MACHINES

Losses to be included

The total losses may be taken as the sum of the following component losses: 8.1 Constant losses a) b) Losses in active iron, and additional no-load losses in other metal parts. Losses due to friction (bearings and brushes, if not lifted during operation) not including any losses in a separate lubricating system. Losses in common bearings shall be stated separately whether or not such bearings are supplied with the machine.
NOTE - When the losses in a separate lubri cating system are required these should be listed separately.

c)

The total windage loss in the machine, including power absorbed in integral fans, and in auxiliary machines, if any, forming an integral part of the machine. The losses in auxiliary machines such as external fans, water and oil pumps not forming an integral part of the machine, but provided exclusively for the machine in question, shall be included only by agreement.
NOTE - When the losses in a separate ventilating system are required they should be listed separately.

8.2

Load losses a) b) c)

I 2R losses in primary windings. I 2R losses in secondary windings.


Electrical losses in brushes (if any).

8.3

Additional load losses a) Losses introduced by load in active iron and other metal parts other than the conductors. Eddy current losses in primary or secondary winding conductors caused by current dependent flux pulsation.

b)

NOTE 1 - Losses, 8.3 a) and b), are sometimes called additional losses, but they do not include the additional no-l oad losses in 8.1 a). NOTE 2 - In the case of auxiliary machines such as phase advancers driven mechanically fr om the main shaft, the losses should be included in the same way as the exciter losses are included for synchronous machines. Losses in separately dri ven phase advancers or regulating equipment should be given separately for rated operating conditions of the main machine. These losses should be determined by the standard method for the types of apparatus involved.

COPYRIGHT

12
9
9.1

Determination of efficiency
Summation of losses

The efficiency can be calculated from the total losses which are assumed to be the summation of the losses obtained in the following manner: 9.1.1 Constant losses 9.1.1.1 No-load test at rated voltage The sum of the constant losses, 8.1 a), b) and c), is determined by running the machine as a motor on no-load. The machine is fed at its rated voltage and frequency. The power absorbed, decreased by the I 2 R losses in the primary winding, gives the total of the constant losses. The I 2R losses in the secondary winding may be neglected. 9.1.1.2 Calibrated machine test (see clause 13) The constant losses may be determined separately by driving the machine, disconnected from the network, at its rated speed by means of a calibrated motor (see 9.2.2). With the brushes, if any, in place, the power absorbed at the shaft of the machine, which may be deduced from the electrical power absorbed by the calibrated motor, gives the sum of the losses in 8.1 b) and 8.1 c). With the brushes, if any, lifted the sum of the bearing friction losses and the total windage losses is obtained in the same manner. The losses described in 8.1 a) may be obtained from the test described in 9.1.1.1 by subtraction. 9.1.1.3 No-load test at variable voltage The losses described in 8.1 a) and the sum of the losses described in 8.1 b) and c) may alternatively be separated by running the machine as a motor at rated frequency but at different voltages. The power absorbed, less the I 2R losses in the primary winding, is plotted against the square of the voltage. This, at low values of saturation, will give a straight line which can be extrapolated to zero voltage to give the sum of the losses, 8.1 b) and c). It should be borne in mind that at very low voltages, losses plotted on the diagram may be high because of the increased secondary winding losses with increased slip. When plotting the straight line, those values should not be taken into account. If the motor is started with a short-circuited secondary winding and the brushes are lifted (which is possible if the supply generator is started at the same time as the motor) the bearing friction and total windage losses are obtained at zero voltage by extrapolation as above.
NOTE - For wound rotor motors a synchronous no-load test can be carr ied out as for synchronous machines with d.c. excitation in two rotor phases (or three if desired).

9.1.2 Load losses 9.1.2.1 Load test The losses described in 8.2 a) are calculated from the resistance of the primary windings measured using direct current and corrected to the reference temperature, and from the current corresponding to the load at which the losses are being calculated.

COPYRIGHT

www.bzfxw.com

13
To determine the losses in 8.2 b) when an on-load test is made, the secondary winding losses are taken to be equal to the product of the slip and the total power transmitted to the secondary winding, i.e. the power absorbed, decreased by the core losses in 8.1 a) and the I 2R losses in the primary winding in 8.2 a). This method gives directly the sum of the losses in 8.2 b) and 8.2 c) for wound rotor machines, and the losses in 8.2 b) for cage machines. For this latter type of machine, this is the only applicable method as it is not possible to measure the resistance and current of the secondary winding directly. When use is made of this method, the slip may be measured by a stroboscopic method or by counting the beats of a permanentmagnet millivoltmeter connected between two rings (for motors with wound secondary windings) or the terminals of an auxiliary coil (for motors with short-circuited secondary windings) or between the ends of the shaft. 9.1.2.2 Calculated values For wound rotor motors, the losses in 8.2 b) may be calculated from the resistance measured by direct current and corrected to the reference temperature, and from the secondary current calculated from a circle diagram or equivalent circuit, account being taken of the true transformation ratio of the machine. The type of circle diagram to be used should be agreed between manufacturer and purchaser. To make an on-load test, the losses in 8.2 c) in the brushes cannot be measured directly and these losses shall be taken as the product of the current flowing in the brushes and a fixed voltage drop. The voltage drop in all brushes of the same phase shall be taken as 1.0 V for carbon or graphite brushes, and 0.3 V for metal-carbon brushes. 9.1.2.3 Load test at reduced voltage This method is also applicable to cage rotor machines. When the voltage is reduced, while keeping the rotational speed of the machine constant, the currents diminish approximately in proportion to the voltage, and the power approximately in proportion to the square of the voltage. When the voltage is down to half its rated value, the currents will then be reduced to about one half, and the power to about one quarter, of their values at the rated voltage. When a load is applied to an induction motor at a reduced voltage U r, the power absorbed P 1r , the main primary current I 1r and the slip s are measured, as well as the no-load current I or , at the same reduced voltage U r , and the no-load current Io at the rated voltage U n. The current vector I 1 of the load at rated voltage is obtained by constructing a vector diagram (figure 3) in the following manner: To the current vector I 1r multiplied by the ratio

add the vector:

COPYRIGHT

14
The resultant vector represents the current which would flow at the rated voltage U n for the following absorbed power:

By means of the values I 1 , P1 , thus determined, and with the slip s measured at reduced voltage, it is then possible to calculate the on-load losses, as indicated in 9.1.2.1. 9.1.3 Additional load losses Unless otherwise specified, it is assumed that the losses specified in 8.3 a) and 8.3 b) vary as the square of the primary current and that their total value at full load is equal to 0.5% of the rated input for motors and 0.5% of the rated output for generators.
NOTE - For some designs of small machines these losses might be higher than 0.5% of the rated input. If , for a part icular case, the value is of import ance, the loss should be determined by the dir ect method of eff iciency measurement.

1 9.2

Total loss measurement

9.2.1 Electrical back-to-back test (see clause 16) When identical machines are run at essentially the same rated conditions, the losses supplied from the electrical system are assumed to be equally distributed and the efficiency is calculated from half the total losses and the electrical input to one machine. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made.
NOTE - Where a gear box is requir ed, as in the case of induction motors, it is necessary for the loss in this to be deducted fr om the electr ical input before determining the losses in the electr ical machine.

9.3

Direct measurement of efficiency

9.3.1 Braking test When the machine is run at rated conditions of speed, voltage and current, the efficiency is taken as the ratio of output to input. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made. 9.3.2 Calibrated machine test (see clause 13) When the machine is running in accordance with clause 13 at rated conditions of speed, voltage and current, the efficiency is then taken as the ratio of output to input. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made. 9.3.3 Mechanical back-to-back test When identical machines are run at essentially the same rated conditions, the losses are assumed to be equally distributed, and the efficiency shall be calculated from half the total loss and the electrical input. The driven machine operates as an induction generator if a source of reactive power is provided, and a suitable load is connected to its terminals. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made.

COPYRIGHT

www.bzfxw.com

15 SECTION FOUR - SYNCHRONOUS MACHINES

10

Losses to be included

The total losses may be taken as the sum of the following component losses: 10.1 Constant losses a) b) Losses in active iron, and additional no-load losses in other metal parts. Losses due to friction (bearings and brushes), not including any losses in a separate lubricating system. Losses in common bearings shall be stated separately whether or not such bearings are supplied with the machine. For water-driven generators and synchronous motors for pump storage schemes, the losses in thrust bearings, and if the thrust bearings are associated with guide bearings the total losses in those bearings, shall be stated separately. The thrustload, temperature of the bearings, and type of oil and oil temperature at which the loss values are valid shall also be given.
NOTE - When the losses in a separate lubri cating system are required these should be listed separately.

c)

The total windage loss in the machine including power absorbed in integral fans, and in auxiliary machines, if any, forming an integral part of the machine. The losses in auxiliary machines such as external fans, water and oil pumps not forming an integral part of the machine, but provided exclusively for the machine in question, shall be included only by agreement.
NOTE 1 - When the losses in a separate ventilating system are required they should be listed separately. NOTE 2 - For machines indirectly cooled or directly cooled by hydrogen, see 11.5 of IEC 34-1.

10.2 Load losses a) b)

I 2R losses in primary windings. I 2R losses in starting or damping windings.

NOTE - These are significant only for single-phase machines.

10.3 Excitation circuit losses a) b)

I 2R losses in the excitation windings and in the excitation rheostats.


All the losses in an exciter mechanically driven from the main shaft which forms part of the complete unit, and is used solely for exciting the machine, together with losses in the rheostat in the excitation circuit of such an exciter, but with the exception of friction and windage losses. Losses in rotary rectifiers and in a gear, rope or belt, or similar drive between shaft and exciter should be included. All the losses in any apparatus for self-excitation and regulation receiving its input from the a.c. supply connected to the terminals of the synchronous machine.

COPYRIGHT

16
In the case of a separate excitation supply such as a battery, rectifier or motor generator set, no allowance is made for the losses in the excitation source or in the connections between the source and the brushes. c) The electrical losses in brushes.

10.4 Additional load losses a) Losses introduced by load in active iron and other metal parts other than the conductors. Eddy current losses in primary winding conductors.

b)

11 Determination of efficiency
11.1 Summation of losses The efficiency can be calculated from the total losses which are assumed to be the summation of the losses obtained in the following manner: 11.1.1 Excitation circuit losses 11.1.1.1 Excitation winding I 2R losses These losses are calculated from the formula I 2 R, taking for R the resistance of the excitation winding corrected to the reference temperature, and for I the value of the exciting current for the particular rating of the machine, measured directly during the on-load test or calculated when this test is not possible. Where such a calculation is made, the method to be used is for agreement between manufacturer and purchaser. 11.1.1.2 Main rheostat losses These losses are calculated from the formula I 2R, where R is the resistance of the part of the rheostat in circuit for the rating considered, and I is the value of the exciting current for the rating considered defined as in 11.1 .1.1. They are also equal to the product IU of the excitation current at the particular rating, and the voltage U at the terminals of the rheostat.
NOTE - Where a resistance is permanently connected in series in the excitation cir cuit it should be dealt with in the same way as the main rheostat.

11.1.1.3 Electrical losses in brushes The sum of these losses shall be taken as the product of the excitation current at the rating considered and a fixed voltage drop. The voltage drop allowed for all brushes of each polarity shall be 1.0 V for carbon or graphite brushes, and 0.3 V for metal-carbon brushes, i.e. a total drop of 2.0 V for carbon or graphite brushes, and 0.6 V for metal-carbon brushes. The sum of the losses according to 11.1 .1.1, 11.1.1.2 and 11.1 .1.3, is also equal to the product IU e , of the exciting current I and the total excitation voltage U e .

COPYRIGHT

www.bzfxw.com

17
11.1.1.4 Exciter losses
NOTE - This applies only to the case where the exciter is mechanically driven from the main shaft and is used solely for exciting the synchronous machine.

These losses include the difference between the power absorbed at the shaft of the exciter and the useful power which it provides at the terminals of the exciter*, and the excitation losses of the exciter if this machine itself is excited by a separate source. If the exciter can be uncoupled from the main machine and tested separately, the power which it absorbs may be measured by the calibrated machine method. If the exciter cannot be uncoupled from the main machine, the power which it absorbs may be measured either by the calibrated machine method or by the retardation method applied to the whole unit. In these two methods, the power absorbed by the exciter is obtained as the difference between the total losses of the unit measured under identical conditions, first with the exciter on-load and secondly with the exciter not excited, the excitation being furnished by an independent source. If none of these methods is applicable, the separate losses should be determined as described under clause 6 for d.c. machines (see 7.1.1.3, last paragraph).
NOTE - The manufacturer and purchaser should agree on the method of determining the losses in apparatus for self-excitation and regulation receiving their input from the a.c. lines connected to the terminals of the machine.

11.1.2 Constant losses 11.1.2.1 Unity power factor test at rated voltage and frequency The sum of the constant losses is generally determined by the method of running the machine as a motor on no-load. The synchronous machine is fed at its rated voltage and rated frequency, so as to work as a motor on no-load. The excitation is adjusted so that the machine absorbs the minimum a.c. current. The electrical power absorbed, decreased by the I 2 R loss in the primary windings, and, if appropriate, by the power absorbed by the exciter, gives the sum of the constant losses.
NOTE - This latt er correction may be avoided by the use of a separate source of excitation power.

11.1.2.2 Open circuit test The sum of the constant losses, 10.1 a), 10.1 b) and 10.1 c), the losses 10.1 a), and the sum of the losses, 10.1 b) and 10.1 c), may also be determined by driving the machine at its rated speed by means of a calibrated machine. The machine is excited by an independent source so as to work as a generator with open circuit at a voltage equal to its rated voltage. The power which it absorbs at its shaft, and which may be calculated from the power absorbed from the calibrated motor, gives the sum of the constant losses 10.1 a), 10.1 b) and 10.1 c). By removing the excitation, the sum of the losses, 10.1 b) and 10.1 c) is obtained in the same manner. The core losses 10.1 a) are obtained by subtraction. Given the small number of brushes used on synchronous machines, it is generally not possible to separate the brush friction losses from the sum of the other constant losses by means of a test with the brushes lifted.

* The useful power at the terminals of the exciter is equal to the sum of the losses according to 11.1.1.1, 11.1.1.2 and 11.1.1.3, of the main machine.
COPYRIGHT

18
11.1.2.3 Retardation test (see clause 15) The sum of the constant losses 10.1 a), 10.1 b) and 10.1 c), the losses 10.1 a) and the sum of the losses, 10.1 b) and 10.1 c) may be determined by using the retardation method. 11.1.2.4 Unity power factor test at variable voltage The losses 10.1 a), 10.1 b) and 10.1 c) may be separated by running the machine as a motor at rated frequency, but at different voltages as described in 9.1.1.3 of Section Three. The power factor shall be maintained at unity by adjusting the excitation current during the test. 11.1.2.5 Variable cooling gas density test For machines cooled by a gas at variable pressure, the total windage loss may be separated from the friction losses by tests at different densities of cooling gas.
NOTE - Tests at diff erent speeds are under consideration.

11.1.2.6 Calorimetric test (see clause 17) The bearing losses may be separately determined when possible by using the calorimetric method.
NOTE - The determination of losses in thrust bearings, possibly combined with guide beari ngs, in vert ical shaft machines, should only be made by agreement.

1 11.1.3

Load losses

These consist of I 2 R losses in primary windings. The I 2R losses in the primary winding are normally measured during the short-circuit test described in 11.1.4. When they are to be given separately, the losses are calculated from the rated current and the resistance of the windings corrected to the reference temperature. 11.1.4 Additional load losses Unless otherwise specified, the sum of the losses, 10.4 a) and 10.4 b) is measured by means of the short-circuit test method.

The machine to be tested, with its primary winding short-circuited, is driven at its rated speed and so excited that the current in the short-circuited primary winding is equal to the rated current. The power absorbed at the shaft, decreased by the mechanical losses, 10.1 b) and 10.1 c), and the power absorbed by the exciter, if appropriate, presents the sum of the load losses and the additional losses, 10.2 and 10.4. If the leakage reactance is abnormally high, as for a machine for high frequency, a correction shall also be made for core losses. The load losses vary in different senses as a function of the temperature. The sum of the load losses and additional losses is assumed to be independent of the temperature and no 1 correction is made to a reference temperature. Unless otherwise specified, it is assumed that the additional load losses vary as the square of the armature current.
NOTE - It is recognized that the sum of the additional losses, 10.4 a) and 10.4 b), thus determined, is generally a little higher than the losses which actually exist at rated load.

The power absorbed at the shaft of the machine during the short-circuit test may be measured by the calibrated machine method (clause 13), or by the retardation method (clause 15).

COPYRIGHT

www.bzfxw.com

19
1 11.2

Total loss measurement

11.2.1 Electrical back-to-back test (see clause 16) When identical machines are run at essentially the same rated conditions, the losses are assumed to be equally distributed and the efficiency shall be calculated as in 11.3.3. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made. 11.2.2 Zero power factor test (see clause 14) When the machine is run at rated conditions of speed, voltage and current, the total losses are equivalent to the absorbed power during the test, corrected for the difference between actual and the full-load exciting current losses. 11.3 Direct measurement of efficiency 11.3.1 Braking test When the machine is run at rated conditions of speed, voltage and current, the efficiency is taken as the ratio of output to input. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made. 11.3.2 Calibrated machine test (see clause 13) When the machine is run at rated conditions of speed, voltage and current, the efficiency is taken as the ratio of output to input. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made. 11.3.3 Mechanical back-to-back test When identical machines are run at essentially the same rated conditions, the losses are assumed to be equally distributed, and the efficiency shall be calculated from half the total losses and the electrical input. The test shall be made as nearly as possible at the temperature attained in operation at the end of the time specified in the rating. No winding temperature correction shall be made.

COPYRIGHT

20 SECTION FIVE - METHODS OF TEST

12

General

Tests can be grouped in one of the three following categories: a) Input-output measurement on a single machine. This usually involves the measurement of mechanical power into, or out of a machine. Input and output measurement on two machines connected back-to-back, e.g. two identical machines or a test machine coupled to a calibrated machine. This is done to eliminate the measurement of mechanical power into or out of the machine. Measurement of the actual loss in a machine under a particular condition. This is not usually the total loss, but comprises certain component losses. The method may, however, be used to calculate the total loss or to calculate a component loss. The choice of test to be made depends on the information required, the accuracy required, and the type and size of the machine involved. Where alternative methods are available for a particular type of machine, the preferred method is indicated (see clause 18).
1 A distinction is made between direct and indirect efficiency determination.

b)

c)

The direct determination of efficiency is made by measuring directly the power supplied by the machine and the power absorbed by it. The indirect determination of efficiency is made by measuring the losses of the machine. Those losses are added to the power supplied by the machine, thus giving the absorbed power. The indirect determination may be carried out by the following methods: (i) (ii) determination of separate losses for summation; determination of total losses.

NOTE - The methods for determining the eff iciency of machines are based on a number of assumptions; it is therefore not possible to make a compari son between the losses obtained by the dir ect method of measurement and those obtained by the measurement of the separate losses.

Unless otherwise specified, the guaranteed efficiency of a machine is that which is based on the determination of separate losses, but when there is a choice of method, the evaluation of efficiency should be based on the accuracy obtainable from the method, the efficiency and the type of machine involved.*

* In some countr ies 90% efficiency is accepted as a basis for using the indirect method whereas some other countri es prefer a lower value, e.g. 70%. COPYRIGHT

www.bzfxw.com

21
When the efficiency or total loss is derived from the measured input and output power, any inaccuracy in these measurements appears as a direct error in the efficiency (e.g. with an accuracy of power measurement not better than 1%, the efficiency can be 2% in error or the total losses can be in error by 2% of the total input power). On small machines or machines with relatively low efficiencies (say below 90%),* this method may be quite acceptable and gives a convenient form of test for such machines. On these and other machines efficiency can be obtained with high accuracy by the calculation of losses from direct measurements.

13

Calibrated machine test

The machine of which the losses are to be measured is separated from the network, uncoupled from its driving motor if necessary, and driven at its rated speed by a calibrated motor, that is by an electric motor of which the losses have been previously determined with great accuracy, such that it is possible to determine the mechanical power which it furnishes at its shaft, knowing the electric power which it absorbs and its speed of rotation. The mechanical power transmitted by the calibrated motor to the shaft of the machine under test is a measure of the losses of this latter machine for the working conditions under which the test is made. In this method, the machine tested may be on no-load, excited or not excited, with or without brushes or short-circuited, which enables categories of losses to be separated. As an alternative, the calibrated motor may be replaced by a dynamometer or by any other motor driving the machine under test through an appropriate torsionmeter, which enables the torque transmitted to the machine under test to be known, and hence the mechanical power absorbed by this latter machine. When use is made of this alternative, the speed of rotation, which comes directly into the calculation of the power, must be measured with extreme care.

14

Zero power factor test

The machine operates as a motor at no-load and at rated speed, with a power factor in the neighbourhood of zero, while the excitation current is adjusted so that the machine carries its rated primary current. The supply voltage is such that the magnetic losses have the same value as in no-load operation at rated voltage. The supply voltage is usually equal to the rated voltage unless this would give an active iron loss appreciably greater than that at full load. In principle, the reactive power should be positive, i.e. over-excited, but when this is impossible because the exciter voltage is not sufficient, the test can be made with absorption of the reactive power (i.e. under-excited).
NOTE - The accuracy of this method is dependent upon the accuracy at low power factor of the wattmeters used.

15

Retardation method

A retardation method can be used for determining the separate losses of rotating electrical machines.

* In some countr ies 90% efficiency is accepted as a basis for using the indirect method whereas some other countri es prefer a lower value, e.g. 70%. COPYRIGHT

22
1 The methods of determination of losses covered by this clause are basically intended for large

synchronous machines, but the principles used can also be applied to other machines (a.c. induction and d.c. machines, exhibiting mainly an appreciable rotational inertia) using the appropriate losses for such machines. 15.1 General The retardation method is used to determine: sum of the friction loss and windage loss (mechanical losses) in machines of all types; sum of losses in active iron and additional open-circuit losses in d.c. and synchronous machines; sum of I 2 R losses in an operating winding and additional load losses (short-circuit losses) in synchronous machines.

15.1.1 Fundamentals The total of the loss P t which retard the machine is proportional to the product of the speed to which these losses correspond and the deceleration at this speed:

When n is expressed in rev/min and P t is given in kW, then the retardation constant C is: The value of deviation shall not be greater than 0,1 and may have to be less than this depending on the characteristics of the machine. where J is given in kg/m 2. The deceleration dn/dt can be obtained either directly, using an accelerometer, or indirectly, by one of the methods given in 15.1.2, 15.1.3 and 15.1.4 below. 15.1.2 Method of the chord This requires the measurement of the time interval t 2 t 1 during which the speed of the tested machine changes from n N (1 + ) to n N (1 ), see figure 4. The ratio of speed interval 2 n N to time interval t 2 t 1 is approximately the deceleration at rated speed:

COPYRIGHT

www.bzfxw.com

23
1 15.1.3

Method of the limiting secant

This is a variant of the method of the chord and is intended to be applied in cases when the speed of rotation cannot be increased above the rated value. The instant of time when the speed of rotation is of the rated value n N is marked as t 1, and the time instants at which the speed of rotation acquires the values of (1 )n N are marked as t 2. The deviation is successively decreased, and the time derivative of the speed of rotation is the limit of the tangent of the angle made by the line passing through the points t 1 and t2 with the time axis, as approaches zero, see figure 5.

15.1.4 Method of the average speed of rotation If t 1, t2 and t3 represent the successively recorded time readings, the shaft making N complete revolutions within the time interval between any two subsequent readings, then the average values of speed during the time intervals shall be:

and the deceleration of the shaft at an intermediate moment of time t 2 The second test gives the total of mechanical losses and iron losses from the formula:

Calculated values of deceleration are plotted against the average values of speed of rotation. The value of deceleration at the rated speed of rotation is determined from the curve. 15.2 Composition of retardation tests 15.2.1 Composition of tests with known moment of inertia When the moment of inertia of a machine rotating part is known by measurement or by design, then for a d.c. machine two basic retardation tests are sufficient: the machine running unexcited and the machine running open-circuited, excited at rated voltage at rated speed. For a synchronous machine a third retardation test should be made with the armature winding being short-circuited and the excitation set to give the rated armature current. The first test gives the mechanical losses of the tested machine from the formula:

COPYRIGHT

24
1 The third test gives the sum of mechanical losses and short-circuit losses from the formula:

In the above equations are the values of speed derivative in time in the first, second and third tests respectively. The iron losses are determined as the difference of the losses measured in the second and first tests. The sum of the I 2 R losses and the additional losses in the armature circuit are determined as the difference of losses measured in the third and first tests. Separation of this sum into components, if required, is done by subtracting from it the I 2R losses in the armature circuit calculated from the armature circuit resistance corresponding to the test temperature. For this purpose the winding temperature shall be deduced by the appropriate method of temperature measurement directly after each retardation test with the armature circuit being short-circuited. 15.2.2 Composition of test with unknown moment of inertia When the moment of inertia of a machine rotating part is not known, or the machine is coupled mechanically to other rotating parts, e.g. a turbine, whose inertia is not known, then some additional tests shall be carried out to determine the retardation constant C . In the instance where there is a possibility to run the tested machine as an unloaded motor from a power supply of the proper voltage, number of phases and frequency (in the case of a.c. machines), and the power supplied to the tested machine can be measured, (equal to the sum of the mechanical losses and iron losses as the armature circuit I 2R losses are usually ignored), then the retardation constant C is determined from the formula:

If the measurement of power is difficult because of frequency oscillations of the power supply, then as an alternative the energy supplied to the tested machine may be measured with an integrating meter. For this purpose it is necessary to run the machine as a motor for some time at constant supply conditions. In the instance where there is no possibility of running the tested machine as an unloaded motor, then, in addition to the three retardation test considered in 15.2.1, one more retardation test shall be conducted. The tested machine in this case is slowed down by any losses P which can be measured and are of the same order as the expected losses P Fe, and P k . For this purpose the open-circuit or short-circuit losses of a connected transformer can be used, which are separately measured. Alternatively, if an exciter or auxiliary generator mounted on the tested machine shaft is available, its load with a ballast resistance may be used.

COPYRIGHT

www.bzfxw.com

25
1 If the tested machine is slowed down by the transformer open-circuit losses, and the short-

circuit losses according to the transformer open-circuit current are ignored, then hence

When the tested machine is slowed down by the transformer short-circuit losses, usually the iron losses corresponding to magnetic flux in the short-circuited transformer are ignored. Hence and

When the tested machine is slowed down by an exciter or auxiliary generator loaded with a ballast resistance, the retardation losses consist only of the tested machine mechanical losses P f and the measured load P (with allowance for efficiency of the load machine that can be determined by calculations). Hence:

so that

15.3 Retardation test procedure 15.3.1 State of a tested machine during retardation tests A tested machine shall be completely assembled as for normal operation. The bearings shall be run in prior to the test. The air temperature shall be adjusted wherever possible to the normal temperature at which the windage loss measurement is required by throttling the air coolant flow. The bearing temperatures shall be adjusted to the normal temperature at which the bearings operate with rated load, by adjusting the coolant flow.

COPYRIGHT

26
1 15.3.2

Tes ted machine coupled with other mechanisms

When possible, the tested machine shall be uncoupled from other rotating parts. If the machine cannot be uncoupled, all possible steps shall be taken to reduce the mechanical losses in other rotating parts, e.g. by partial dismantling or in the case of a water turbine, by removing water from the runner chamber. Means shall also be taken to eliminate the possibility of water flowing from the upstream side and from drawing water by the rotating runner from the downstream side. Rotation of the runner in the air produces windage losses which can be stated experimentally or from calculations by agreement between manufacturer and purchaser. 15.3.3 Rotation of a tested machine In some cases the tested machine can be driven by its normal prime mover, e.g. by Pelton turbine where the water supply to the runner can be cut off instantly. However, the tested machine is usually running as a motor on no-load, fed from a separate source with a wide range of variable speed in all cases the excitation shall be obtained from a separate source with a rapid and precise voltage control. The excitation from the inherent mechanically-coupled exciter is not recommended in principle, but may be permitted in those cases when the value of the deviation of speed is relatively small, e.g. it does not exceed 0,05. In all these cases the losses in exciters coupled to the shaft of the tested machine shall be taken into account. 15.3.4 Procedure performed prior to starting the tests Each test begins with the tested machine being rapidly accelerated to a speed above (1 + ) n N so that during deceleration to this speed the machine can be placed in the required condition, namely: the machine is disconnected from a supply source; in the case of retardation by only mechanical loss, the machine field is suppressed; in the case of retardation by the sum of the mechanical loss and short-circuit losses, the machine field is suppressed, the armature terminals are short-circuited and the machine is reexcited to the preset short-circuit current; in the case of retardation by the transformer losses after field suppression, the tested machine is connected to the transformer previously set to a certain state (at no load or short-circuit) and excited to the preset values of current or open-circuit voltage; in the case of retardation by the exciter load losses or auxiliary generator mounted on the machine shaft, the tested machine field is suppressed and the specified load is set simultaneously.

In all cases described above a sufficient time delay shall separate the switching off of the supply and starting the measurements to allow electromagnetic transients to decay. In the case of retardation by the sum of mechanical and iron losses or by the open-circuit losses of a supply transformer, no procedures are required after the machine is disconnected from the supply if the tested machine excitation corresponds to the preset open-circuit voltage, in the case of a synchronous machine, at rated speed and unity power factor.

COPYRIGHT

www.bzfxw.com

27
1 15.3.5

Procedures during retardation

The readings of all instruments used for each test (field current ammeter, open-circuit voltage voltmeter, short-circuit current ammeter) and of all instruments required to measure the power in additional retardation tests when the moment of inertia J is not known shall be taken at the instant when the tested machine passes through rated speed; no readings at this instant are required in the case of an unexcited retardation test. The measured values of open-circuit voltage or short-circuit current shall not differ from the preset values by more than 2%. The calculated final value of the speed derivative in time for each of the tests shall be adjusted proportionally by the ratio of the square of the preset value to the measured value. 15.3.6 Program of retardation tests The retardation tests shall be conducted as a series without interruption, whenever possible. It is recommended that the series start and finish with some retardation tests of an unexcited machine. If for any reason the test series is not conducted in a continuous manner then it is recommended that each subsequent series of tests start and finish with some unexcited retardation tests. Tests may be either repeated several times at the same preset values of open-circuit voltage or short-circuit current, e.g. at rated values, or at various values within limits of the order of 95% - 105% of the rated values. In the first case the arithmetic mean values obtained from all measurements are assumed to be the real measured value of each type of loss. In the second case the values are plotted on a curve as a function of voltage or current. Real measured values are assumed to be those occurring at the points of intersection of the preset values of voltage or current as read from the curves. Additional retardation tests, when the moment of inertia of the tested machine is not known, shall be conducted at the same values of voltage or current as those obtained with the winding open or short-circuited. If this is not possible the respective values shall be determined from curves as indicated above. 15.4 Taki ng of measurements 15.4.1 Methods of measurements The measurements taken during retardation tests are aimed at obtaining the required value of the speed derivative in time and may be performed by one of the three methods: a) accelerometric - direct measurement of deceleration with time:

b) tachometric - by determining the dependence of speed with time:

c) chronographic - by determining the dependence of angular displacement of the tested machine shaft with time:

COPYRIGHT

28
1 For all cases recording measuring instruments may be used both with continuous and with

discrete recording of measured values and time. 15.4.2 Accelerometric method The dependence of speed on time for large machines having a complex ventilation route may not be regular. As a consequence of this the instantaneous values of deceleration during retardation at the moment of passing through rated speed may be random. Therefore, true values of the speed derivative may be determined by plotting measured decelerations versus time or speed and using a suitable curve fitting or correlation technique. 15.4.3 Tac hometric method A plot of speed versus time is obtained from the results of measurements. On this plot the time instants are defined at which the speed acquired the values indicated for the chord or limiting secant method. The differences between the times at the lower and upper limits of speed are used to calculate the decelerations. If there is an exciter or any other electrical machine on the tested machine shaft, it can be used as a tachogenerator, provided that the voltage signal does not pulsate with the speed of rotation of the tested machine. The excitation shall be supplied from a stable d.c. source, such as a separate storage battery. If the voltage signal does pulsate with the speed of rotation or when there is no such tachogenerator on the tested machine, a coupled d.c. machine may be used. It can be driven from the shaft of the tested machine by a seamless belt or by other means to provide smooth rotation. Readings of the speed may be made either in the exact time intervals, specified by the respective method, in which case there is no need for special recording of time or of signals from the tested machine shaft; in this case, the readings of time shall be taken concurrently with readings of speed. There is no need to take readings with each turn of the shaft; usually 30 to 40 readings during the whole test are quite sufficient. With the availability of high-accuracy measuring instruments, the measurement of speed of rotation may be substituted by measurement of the instantaneous values of speed or of the period of the voltage of the tested machine or of any other a.c. machine situated on its shaft; it is not necessary that the number of pole pairs of both machines is equal. 15.4.4 Chronographic method The time-counters used may be either visual indicators with continuous (non-stepwise) motion of the pointer, or digital indicators with printers (electrical or mechanical). Time readings shall be taken according to the signals obtained from the tested machine shaft either with each complete revolution of the shaft or for a known number of revolutions.
NOTE - If when using the tachometri c method the speed of rotation is determined by signals fr om the tested machine shaft, then the time readings may be used both for tachometr ic and chronographic methods, thus providing a mutual check.

In some cases, when the unit has smooth deceleration characteristics, sufficient accuracy can be obtained by measuring the time for retardation between two speeds with the same difference to the rated speed

COPYRIGHT

www.bzfxw.com

29
1 The stator voltage frequency provides the best means of determining the speed of a

synchronous machine. 15.4.5 Measurement of losses in bearings The losses in bearings and thrust bearings can be subtracted from the total sum of the mechanical losses, if required. These may be determined by the calorimetric method in accordance with IEC 34-2A. If the tested machine uses direct-flow cooling of the bearings, these losses are distributed between the tested machine and any other coupled to it mechanically, such as turbine, in proportion to the masses of their rotating parts. If there is no direct-flow cooling, the distribution of bearing losses shall be determined from empirical formulae by agreement between manufacturer and purchaser.

16

Electrical back-to-back test

This method is applicable when two identical machines are available. The machines are coupled mechanically and electrically so as to operate at rated speed, one as a motor and the other as a generator. The actual temperature at which the measurements are carried out should be as close as possible to the working temperature and no further correction should be made. The losses of the assembled machines are supplied either by a network to which they are connected, or by a calibrated driving motor, or by a booster, or else by a combination of these various means. The average value of the armature currents is adjusted to the rated value, the average of the voltage of the two armatures is above or below the rated voltage by an amount equal to the voltage drop, depending on whether the d.c. machines are intended to be used respectively as generators or as motors. Where two induction machines are electrically connected, they should be mechanically coupled with a speed adjusting device, such as a gear box, to ensure the correct circulation of power. The magnitude of power circulated depends upon the difference in speed. The electrical system supplying the losses to the two machines will be required to provide magnetizing kvar to both machines. When two synchronous machines are electrically connected, they should be mechanically coupled with a correct angular phase relationship. The magnitude of the power circulated depends upon the difference in phase angle between them.

17

Calorimetric test

Measurement of losses by calorimetric methods shall be performed in accordance with IEC 34-2A.

18

Schedule of preferred tests

18.1 D.C. machines The preferred test for d.c. machines is in accordance with 7.1 and the preferred method of calculating the efficiency is in accordance with 7.1.2.

COPYRIGHT

30
18.2 Polyphase induction machines The preferred test for polyphase induction machines is in accordance with 9.1 and the preferred method of determining the constant losses is in accordance with 9.1.1.1. 18.3 Synchronous machines The preferred test for synchronous machines is in accordance with 11.1 and the preferred method of determining the constant losses is in accordance with 11.1 .2.1.

Figure 1 - Mechanical back-to-back test.

Figure 2 - Electrical back-to-back test.

COPYRIGHT

www.bzfxw.com

31

Figure 3 - Vector diagram for obtaining vector of load current I 1 at rated voltage.

COPYRIGHT

32
1

Figure 4 - Method of the chord

Figure 5 - Method of the limiting secant

COPYRIGHT

www.bzfxw.com

33

Annex A (Informative) Provisional methods for determining losses and efficiency of converter-fed cage induction machines
INTRODUCTION

This annex applies to cage induction machines with rated frequencies up to 120 Hz supplied by converters which have an intermediate circuit and are of the following types: I-converters and U-converters, typically Pulse Width Modulated (PWM). The methods to determine losses and efficiency given in section 3 are partly no longer applicable and this annex indicates the test modifications that are necessary.
NOTE - The six-step converter is a special case of the pulsed converter.

In general, when fed from a converter, the motor losses are higher than during operation on a sinusoidal system. These additional losses depend on the harmonic spectrum of the impressed supply quantity (either current or voltage). Their magnitude is influenced by circuitry and control method of the converter. Consequently a simple factor to cover these additional losses cannot be found. The determination of losses and efficiency will therefore preferably use procedures where the motor is operated together with the same converter with which it is going into service. It is also understood that suitable methods shall not require the knowledge of design data of the motor, such as the rotor bar geometry.

A.1
A.1.1

Determination of losses and efficiency of converter-fed motors


Components of the additional losses

In cage induction motors additional losses1 ) are produced due to the harmonics in either current or voltage; they are made up of the following components: a) additional I 2R losses in primary windings; b) additional I 2R losses in secondary windings; c) additional losses in active iron.
NOTE - The physical effects giving rise to the additional losses are treated in Chapter 5 of IEC 34-17, 1992: Guide for the application of cage induction motors when fed from convert ers.

1)

These additional losses are due to harmonics of the supply and do not contain the additional losses described in 8.1 a) and 8.3 which refer to sinusoidal supply of fundamental frequency only. COPYRIGHT

34
2

A.1.2

Efficiency determination by input-output measurement

The motor input-output measurement as indicated in clause 12 is a preferred method since all additional losses are incorporated in the result (see clause A.3); however, the measuring equipment must have sufficient accuracy for measuring power, torque and speed as well as an appropriate frequency range. Therefore, additional requirements for measuring instruments and accessories beyond the contents of clause 3 have to be specified (see clause A.2). To keep within a required relative tolerance of the resulting motor efficiency, the maximum relative error ( P /P in ) max of the power measurement has to be decreased with increasing efficiency, as shown in figure A.1.

Figure A.1 - Maximum permissible relative error (P/ P in) max of input as well as output measurement
2)

There is also the possibility to determine the overall efficiency of the complete system consisting of converter and motor by input-output measurement, applicable on agreement between manufacturer and purchaser. In this case the motor efficiency cannot be determined separately. A.1.3 Efficiency determination by summation of losses

A number of presumptions made in 9.1 are no longer valid for motors fed from converters. In the no-load test, the I 2 R losses in the secondary winding (9.1.1.1) may not be neglected. Therefore the iron losses cannot be separated. The no-load test at variable voltage (i.e. at variable flux) according to 9.1.1.3 cannot be carried through with many commercial converters, due to the limited range of adjustment; consequently there will be no possibility to separate the friction and windage losses (8.1 b) and c)) from the other losses by a no-load test. Concerning the load test, the statement in 9.1.2.1 that the secondary winding losses are taken to be equal to the product of the slip and the total power transmitted to the secondary winding is only valid for a machine operated with a sinusoidal current of fundamental frequency. Moreover, to calculate the I 2R losses of the primary winding by means of the resistance measured using direct current (9.1.2.1) will produce an error due to eddy currents.

2)

= tolerance as described in IEC 34-1, table VIII , items 1 and 2. The curves are based on a simplified error consideration, assuming err ors P of equal magnitude in P in and P out . Figure A.1 is a graph of the equation (P /P in) max = . (1 ) / (1 + ). COPYRIGHT

www.bzfxw.com

35
2

Hence when using the method of summation of losses certain assumptions have to be made (see clause A.4). A.1.4 Efficiency determination by the calorimetric method

The calorimetric method is especially useful for application to converter-fed motors since the losses are measured independently of the waveforms of voltages and currents. The calorimetric calibration method according to clause 3 of IEC 34-2A has been found of advantage since it does not require measurement of the mass rate of flow; hence the density of the cooling medium, being functions of humidity and temperature, need not be known. Moreover, the variation of specific heat capacity can usually be disregarded. In a set-up according to Figure A.2 the power absorbed in the dissipation resistor can be measured without difficulty, so that the motor losses may be calculated from the proportion:

where

Pv Pd T 1, T 2, T 3

represents the motor losses; represents the power absorbed in the dissipation resistor; represents the measured temperatures at the points indicated in Figure A.2.

The measuring accuracy depends mainly on the magnitude of temperature rise values ( T 2 T 1 ) and (T 3 T 2 ). The measurement has to be made in accordance with clause 13 of IEC 34-2A, to enable an accuracy of measurement as indicated in clause 15 and table II.

Figure A.2 -

Schematic diagram of a test set-up for the calorimetric calibration method

A.1.5 Efficiency determination by summation of losses from tests on a sinusoidal system with lumped increments to take care of the additional losses Often standard design motors and converters are coupled together only on the site of service. Especially in these cases a method to determine efficiency by adding a lumped increment to the known losses on sinusoidal supply would be welcome, but, as mentioned before, there is no chance to define suitable values covering the wide variety of converter circuits and control methods. At the present state, experimental data collected for a certain range of output and circuitry allow only a limited statement on lumped increments (see clause A.5).
COPYRIGHT

36
2

A.1.6

Recommended methods

Depending on the rating of the motors, application of the following methods is recommended for machines of rated frequency 50 Hz or 60 Hz: a) Input-output measurement (clause A.3) when fed from I- or U-converter for motors 50 kW. The method may also be applied to motors of higher rated output by agreement between the purchaser and the manufacturer.3) b) Summation of losses when fed from I-converter (A.4.1) or U-converter (A.4.2) for motors >50 kW. Summation of losses with sinusoidal supply (9.1) and no-load test when fed from U-converter (clause A.4.3) for motors tested in a test-shop (irrespective of rating). The calorimetric method (A.1.4) applicable to all ratings with I- and U-converter.

c)

d)

A.2

Requirements for the measuring instruments

Instruments for r.m.s. current and voltage and for active power are necessary. In the input-output measurement method, the latter determines, together with the equipment to measure torque and speed, the accuracy of the results. Regarding the contribution of the harmonics to the losses, care has to be taken to select measuring equipment capable of operating in the range of relevant frequencies with sufficient accuracy. The following is required of the frequency range4) fr of the measuring equipment with inclusion of instrument transformers, transducers and shunt resistors:

f r = 10 f 1 for six-step converters; f r = 6 f p for PWM converters, with a maximum of 100 kHz;
where

f1 fp

is the maximum rated frequency; is the maximum pulse frequency (carrier frequency).

For six-step converters, these requirements can be met by conventional electrodynamic instruments. For PWM converters it is necessary to employ equipment with a broader frequency range. These will preferably be electronic instruments with AD-converters and digital data processing.

3)

If by agreement the method is applied to motors of higher rating, it has to be accepted that the error of the determined efficiency can exceed the tolerance values given in IEC 34-1, Table VIll. For conventional indicating measuri ng instr uments (see IEC 51) the accuracy is specified for the nominal fr equency (e.g. 0,2 % for 40 .. . 60 Hz), while an additional error of the class accuracy is tolerated at an upper specified frequency (e.g. 0,4% at 1000 Hz). For electr onic measuring instruments, usually a frequency range is indicated which means the upper specified fr equency. The accuracy is given both for 50 Hz or 60 Hz and for the upper specified fr equency. In the following this will be called the frequency range of an instrument. COPYRIGHT

4)

www.bzfxw.com

37
2
NOTES 1 With high values of pulse fr equency the two-watt meter (Aron) method should not be used since, due to capacitive currents, the sum of input curr ents may differ fr om zero. Hence, one power measuri ng instr ument per phase shall be applied. 2 It is considered that the following accuracies are within reach with appropriate measuring equipment: power with 0,5%, torque with 0,4% and speed with 0,1%. 3 The converter output harmonics and their dominant order numbers depend on the modulation method. Basic considerations are given in clause A.6.

A.3

Total loss measurement

To determine the efficiency of motors when fed from converters, the braking test ((9.3.1) see amendment 1) or the calibrated machine test ((9.3.2) see amendment 1) may be applied. These methods are also applicable to operating conditions with other than rated frequency, including field-weakening. This direct method is restricted to machines not exceeding the limit of rated power referred to in A.1.6.

A.4

Summation of losses

The summation of losses can be applied by means of a modified no-load test (9.1.1.1) and a modified load test (9.1.2.1). Both are carried through at rated frequency and voltage, the voltage being adjusted according to the features of the converter, e.g. by means of an inherent characteristic curve or by field-oriented control. Different modifications have to be observed for I-converters and U-converters. A.4.1 Motor supplied by I-converter with block-wave output.5)

It is assumed that the waveform of the current does not change between no-load and full load, so that the relative harmonic content is independent of the load. It is also assumed that the additional iron losses are predominantly losses due to magnetic reversal of leakage fluxes. Under these assumptions the additional losses due to harmonics depend mainly on the current and vary with the square of the primary r.m.s. current. In theory this allows determination of the additional losses under load from the difference of constant losses measured in no-load tests on both sinusoidal and converter supply. In practice, however, considerable errors would have to be expected from such a method. A.4.1.1 No-load test when fed by converter The difference between the measured input and the primary winding losses as calculated from the d.c. resistance contains the constant losses (according to 8.1, taking only the fundamental into account) and the following components: additional primary winding losses due to increased resistance because of eddy currents, additional iron losses, secondary winding losses due to harmonics.

5)

Specifications for supply by I- converters with pulse-width-modulation are not contained in this document. COPYRIGHT

38
2

All these components will appear within the sum of the constant losses when applying the noload tests (9.1.1.1). A.4.1.2 Load test when fed by converter When applying the method of 9.1.2.1 with readings from the load test with converter, the calculated secondary winding losses will be too small. In order to compensate for the difference of harmonic secondary winding losses under load and no-load, and for eddy current losses of the harmonics in the primary winding, 0,5% of the input power shall be added to the losses at full load (i.e. additionally to the lumped additional losses of 0,5% prescribed by 9.1.3), unless otherwise agreed, and assuming that the losses vary as the square of the primary current. A.4.2 Motor supplied by U-converter
NOTE - The following applies both to six-step and PWM convert ers.

It may be assumed that the absolute values of the harmonic currents are independent of the load. The efficiency will be obtained by applying a no-load test and a load test both with converter. The additional losses are contained in the sum of the constant losses as determined from the no-load test according to 9.1.1.1. Consequently the I 2R losses in the secondary winding to be taken into account in the summation of losses are those due to the fundamental slip frequency currents only. They are determined from the load test in a way different from 9.1.2.1; they are taken to be equal to the product of the slip and the fundamental power transmitted to the rotor, i.e. the power absorbed, decreased by the I 2 R losses in the primary winding and the sum of the constant losses except the friction and windage losses. The latter can be determined from a retardation test. The lumped value of the additional load losses due to the fundamental according to 9.1.3 has to be applied. Figure A.3 gives a diagram of the harmonic power and corresponding additional losses. Separation of the additional losses is only possible when the constant losses with sinusoidal supply are known.

COPYRIGHT

www.bzfxw.com

39
2

Figure A.3 - Power and losses diagram 6) : a) with a sinusoidal supply, and b) harmonic power and corresponding additional losses due to non-sinusoidal supply

A.4.3 Summation of losses by a load test with sinusoidal supply and a no-load test with converter supply When a sinusoidal supply is available, the load losses shall be determined from the load test as specified in 9.1.2.1. The no-load test is performed according to 4.9. The additional load losses as specified in 9.1.3 have to be applied.

A.5 Method with assumed increments to the losses determined on sinusoidal supply
Experience available from three-phase cage induction motors tested both with sinusoidal waveform and with converter supply has been evaluated to obtain the additional losses, expressed as percentage of the input, whilst the output was the same in both cases. For machines between 30 kW and 1015 kW rated output, operated with six-step I-converter and 50 Hz or 60 Hz rated frequency, the additional losses due to converter supply are between 0,6% and 1,25% of the input. A suitable assumed increment for motors above 30 kW rated output is 1%.

6)

Symbols in Pin, f = P = Pout = Pin, h = Pout , h =

figure A.3: fundamental input power; power transmitt ed to the secondary winding; output; harmonic input power; harmonic output power. COPYRIGHT

40
2

For machines with U-converters the results depend on pulse-frequency, pulsegenerating scheme and modulation index. The experimental results from PWM converters with sinusoidal reference wave showed additional losses from almost negligible values up to 3% of the input. With six-step U-converters a suitable assumed increment is 1,5%.

A.6
A.6.1

Basic considerations concerning converters


Typi cal converter output waveforms

Figure A.4 -

Typical voltage and current waveform: a) six-step I-converter drive; b) six-step U-converter drive; c) PWM U-converter drive

A.6.2

PWM-converter harmonics

The harmonic spectrum of the converter output voltages and currents depends on the modulation method. Analytical expressions can be found for synchronous pulse generation by interaction of a carrier-wave and a reference-wave, with a constant ratio of their frequencies. Examples are U-converters with pulse-width modulation. The spectrum of voltage harmonics can then be described in terms of fundamental frequency f 1, pulse frequency f p and modulation index M.

COPYRIGHT

www.bzfxw.com

41
2

Let f 1 be the converter fundamental frequency, then the harmonics will be of fn = n . f 1; n being an integer for synchronous modulation methods. With three-phase bridge six-step converters the harmonics are of the order n = 5, 7, 11, 13..., the harmonic content decreasing with increasing n. The pulse frequency f p is equal to the number of turns-on per second of the valves in one main branch of the converter; for PWM modulation it is identical with the sampling frequency. The quantity p = f p/f 1, called the carrier ratio, is selected to be a multiple of 3 to give a balanced three-phase output. Sine- and square-wave modulation produce a harmonic spectrum at the output with harmonics of the order numbers n = p m g (m g = 2, 4, ...) and n = 2p m u (m u = 1, 3, ...). The modulation index M is defined as the ratio of the reference wave amplitude and the carrier wave amplitude. Over a wide range of M a harmonic near 2 f p is dominant in the motor terminal voltage spectrum. Given the voltage harmonics at the motor input, the current harmonics will depend on the motor impedance. Asynchronous pulse-pattern generation schemes can result in fractional harmonic order numbers. Motor current harmonics can further be decreased by inserting filters between converter output and motor input. In the case of I-converters operated with pulse-width modulation, a filter (with capacitors) is essential for its performance.

COPYRIGHT

This page has been left intentionally blank.

www.bzfxw.com