इंटरनेट मानक

Disclosure to Promote the Right To Information

Whereas the Parliament of India has set out to provide a practical regime of right to
information for citizens to secure access to information under the control of public authorities,
in order to promote transparency and accountability in the working of every public authority,
and whereas the attached publication of the Bureau of Indian Standards is of particular interest
to the public, particularly disadvantaged communities and those engaged in the pursuit of
education and knowledge, the attached public safety standard is made available to promote the
timely dissemination of this information in an accurate manner to the public.

“जान1 का अ+धकार, जी1 का अ+धकार” “प0रा1 को छोड न' 5 तरफ”

Mazdoor Kisan Shakti Sangathan Jawaharlal Nehru
“The Right to Information, The Right to Live” “Step Out From the Old to the New”

IS 649 (1997): Methods for testing steel sheets for

magnetic circuits of power electrical apparatus [MTD 4:
Wrought Steel Products]

“!ान $ एक न' भारत का +नम-ण”

Satyanarayan Gangaram Pitroda
“Invent a New India Using Knowledge”

“!ान एक ऐसा खजाना > जो कभी च0राया नहB जा सकता ह”

है”
ह
Bhartṛhari—Nītiśatakam
“Knowledge is such a treasure which cannot be stolen”
 IS 649 : 1997
(Reaffirmed 2001)

METHODS OF TESTING STEEL SHEETS

FOR MAGNETIC CIRCUITS OF POWER
ELECTRICAL APPARATUS
( Secod Revision, )

ICS 77.1403); 77.IJO.W; 2O.O-K~.10

BUREAU OF INDIAN STANDARDS

h4ANAK BIHAVAN, 9 BAHADUR SHAIH ZAFAR h4ARC;
NEW DELHI 1 10002

l’rice Croup 11
Wrought Steel Products Sectional Committee MTD 4

FOREWORD

This Indian Standard (Second Revision) was adopted by the Bureau of Indian Standards, after the draft
finalized by the Wrought Steel Products Sectional Committee had been approved by the Metallurgical
Engineering Division Council.

This standard was first issued in 1955 and subsequently revised in 1963. While reviewing the standard in
the light of experience gained during these years, the Committee decided to revise it to bring it in line
with the present practices being follwed internationally.

The main modification in this revision relates to incorporation of test methods for insulation resistance
test, resistivity, ductility, density, size and shape measurement and tests on insulation coating. This
modification has become necessary in view of the fact that these tests have been incorporated in all major
standards of the world. All the tests are given in details so that many laboratories in this country can get
guidance in installing equipment to carry out these tests.

In the preparation of this standard, assistance has been derived from the following overseas standards:

IEC404-2 : 197X Magncticmaterials, Part 2: Methods ofmeasuring ofmagnetic, electrical and physical
properties of magnetic sheet and strip, issued by Internalional Elcctrotechnical Commission (IEC).

DIN 50642 June 1975 Testing of metallic materials - Testing of shape variation of electrical steel
sheet and strip and determination of internal stresses, issued by DIN Germany.

JIS C-2550-1986 Japanese industrial standards - Methods of tests for magnetic steel sheet and strip,
issued by Japanese Standards Association.

BS 6404 : Part 2 : 1985 Magnetic materials, Part 2: Methods of measurement of magnetic electrical
and physical properties of magnetic sheet and strip, issued by British Standards Institution.

BS 6404 : Section 84 : 1986 Specification for cold rolled non-oriented magnetic steel sheet and strip
delivered in the finally annealed state, issued by British Standards Institution.

1991 Annual Book of ASTM Standards Section 3, Volume 03.04 Magnetic propcrtics, metallic
materials for thermostats, electrical resistance and heating contacts, issued by American Society for
Testing and Materials.

In reporting the result of a test made in accordance with this standard, if the final value, obscrvcd or
calculated, is to be rounded off, it shall be done in accordance with IS 2 : 1960 ‘Rules for rounding off
numerical values (revised)‘.
 IS 649 : 1997

Indian Standard

METHODS OF TESTING STEEL SHEETS

FOR MAGNETIC CIRCUITS OF POWER
ELECTRICAL APPARATUS
( Second Revision )
1 SCOPE 3.4 Core Plate -A generic term for any insulating
material, formed metallurgically or applied exter-
1.1 This standard prescribes methods of test for
nally as a thin surface coating, on a sheet or strip
determining the requirements of magnetic steel
stock used in the construction of laminated and
sheets and strips used for the construction of
tape wound cores.
magnetic circuits of power electrical apparatus.
3.5 Density - The ratio of mass to volume of
1.2 It covers the methods for the measurement of
material. The cgs unit is g/cm3.
magnetic, electrical and physical properties and
insulation coating test of magnetic steel sheets and 3.6 Eddy Current -An electriccurrent devclopcd
strips. in a material due to induced voltages dcvclopcd in
the material.
2 REFERENCES
3.7 Electrical Steel, Grain Oriented - A flat
The following Indian Standards arc necessary
rolled silicon-iron alloy usually containing
adjuncts to this standard:
approximately 3 pcrccnt silicon, having enhanced
IS No. Title magnetic properties in the direction of rolling and
648 : 1994 Non-oriented electrical steel normally used in transformer cores.
sheers and strips for magnetic 3.8 Electrical Steel, Non-oriented - A flat rolled
circuits Cfourth revision) clcctrical steel which has approximately the same
3024 : 1996 Grain oriented electrical steel magnetic properties in all directions.
sheets and strip (first revision)
3.9 Frequency, Cyclic, f - The number of hertz
13795 Glossary of terms relating to
(cycle/second) of a periodic quantity.
(Part 1) : 1993 special alloys: Part 1 Magneic
materials 3.10 Hertz, JIz -The unit of cyclic frcquency,J
3 TERhlINOLOGY 3.11 Ilysteresis Loss, Rotational -The hysteresis
3.1 For the purpose of this standard, the defini- loss that occurs in a body when subjected to a
tions given in IS 13795 (Part 1) : 1993 shall apply, constant magnetizing force, the direction of which
in addition to the following. rotates with respect to the body, tither in con-
tinuously cyclic, or in a repeated oscillatory man-
3.2 Apparent Power, Pa - The product (volt-
ner.
amperes) of the rms exciting current and applied
rms terminal voltage in an electric circuit contain- 3.12 Insulation Resistance - The apparent
ing inductive impedance. The components of this resistance between adjacent contacting lamina-
impedance due to the winding will be linear, while tions, calculated as a ratio of the applied voltage to
the components due to the magnetic core will be conduction current. This parameter is normally a
non-linear. The unit of apparent power is volt- function of the applied force and voltage.
ampere VA.
3.13 Magnetostriction - The change in dimcn-
3.3 Apparent Power, Specific, P:,(R,Q - The value sions of a body resulting from magnetization.
of the apparent power divided by the active mass of
the specimen, that is, volt-ampcrcs per unit mass. 3.14 Stacking Factor (Lamination Pactor, Sl>ace
The values of voltage and current are those Factor), S -A numeric, less than unity and usually
developed at a maximum value of cyclically varying expressed as a percentage, which is defined as the
induction B and specified frcqucncy,f. ratio of the uniform solid height h of the magnetic

I
IS 649 : 1997

material in a laminated core to the actual height SECTION 1 MAGNETIC TESTING

h’(core build-up) when measured under a specified
pressure S is thus equal to the ratio of the volume 5 TESTING CONDITIONS
of magnetic material in a uniform laminated core 5.1 Temperature
to the overall geometric volume in the core.
4 TEST ITEMS The magnetic tests shall be carried out at 27kS’C.

4.1 The test shall be made for the following items 5.2 Magnetizing Condition
in conformity with the provisions of Section 1 to
Section 11. 5.2.1 After demagnetizing the specimen, mag-
netization shall be conducted by applying mag-
4.2 Magnetic Tests netizing force to it, so that both the positive and
4.2.1 a.c. magnetization characteristic tests, iron negative maximum magnetic flux density induced in
loss tests and apparent power tests at commercial the specimen become identical. On a.c. test the
frequency. discrepancy between the form factors of secondary
induced voltage and sine wave shall be within
4.2.2 d.c. magnetization characteristic tests.
+.5 percent.
4.3 Electrical Tests
6 TEST SPECIMENS
4.3.1 Insulation Resistance Test
4.3.2 Resistivity Test 6.1 The practice to be followed for a test lot and
selection and preparation of test specimens is as
4.3.3 Determination of Den&
follows.
4.4 Physical Tests
6.1.1 Test Lot
4.4.1 Stacking Factor
A test lot may be composed of coils or cut lengths.
4.4.2 Ductility Test
A test lot of coil product may consist of one or more
4.4.3 Internal Stress coils having essentially the same treatment and
4.5 Size and Shape Measurement composition.
4.5.1 Size Measurement 6.1.2 Selection and Preparation of Test Specimen
4.5.2 Thickness 6.1.2.1 The Epstein test sample shall be the stand-
4.5.3 Width and Length ard specimen for determinations of the magnetic
properties of flat rolled electrical steels, except
4.5.4 Cutting Burr Measurement
when otherwise established by mutual agreement
4.5.5 Out of Square between the manufacturer and the purchaser.
4.5.6 Flatness Measurement ( Wave Factor) 6.1.2.2 The standard Epstein test specimen shall
4.5.7 Bowing or Residual Curvature be composed of test strips preferably cut from test
4.5.8 Edge Camber sheets in a manner shown in Fig. l(a) or l(b). One
half of the strips are cut parallel and the other half
4.6 Tests on Insulation Coating cut perpendicular to the direction of rolling.
tSCRAP SCRAPA

(a) ~SCRAP

%.CRAPr

FIG.~ SUGGESTEDDISTRIBUTIONOFSTRIPTOBECUTFROMSHEETSFORMAGNETICTESTS

2
 IS 649 : 1997

6.1.2.3 When less than the total number of strips 6.1.2.12 The total weight of the sheets shall be not
obtained from the sampled area are needed for the less than 400 g and it should be determined within
test specimen, the excess strips should be discarded +l g.
equally from all locations in the sampled areas. For
6.1.2.13 When the cross sectional area of sheet
instance, if approximately one fourth of the total
material test specimens is required, it shall be
strips obtained in excess, every fourth strip should
calculated from the measurements of weight and
be discarded.
length using a density value in accordance with
6.1.2.4 The Epstein test specimen shall consist of 6.1.2.14.
strips sheared or punched in a width of 30 mm and 6.1.2.14 Density
not less than 280 mm long. For ease of assembling
the specimen in the test frame, it is desired to use Unless otherwise specified by the manufacturer,
strips slightly longer than 280 mm and a length of the following densities may be assumed for cal-
305 mm is recommended. culation purposes.

6.1.2.5 The test strips shall be as nearly rectangular Silicon Content Assumed Density
as possible and shall conform to the specified Percent g/cm3
dimensions within 20.8 mm. up to 0.5 7.85
Over 0.5 to 2.0 7.75
6.1.2.6 The test strips shall becut with sharp shears
Over 2.0 to 3.5 7.65
or dies to avoid excessive burring or distortion.
Over 3.5 to 5.0 7.55
6.1.2.7 For grain oriented steel sheets, the strips Method of determination of the density of magnetic
shall be cut parallel to the direction of rolling sheet shall be as per Section 6.
according to Fig.lc. The samples of oriented
material before testing shall be stress relief an- SECTION 2 STANDARD TEST METHOD
nealed after cutting at a temperature of 8OOstr2O”C FOR ALTERNATING-CURRENT MAGNETIC
in a non-oxidizing, carbon free atmosphere. They PROPERTIES OF MATERIALS AT POWER
shall be held at full temperature for a minimum FREQUENCIES USING WATTMETER,
period of 15 minutes and cooled in the furnace to AMMETER, VOLTMETER METHOD AND
below 100°C before removal. 25-cm EPSTEIN TEST FRAME
6.1.2.8 For non-oriented steel sheets, the strips 7 SCOPE
shall be cut as per 6.1.2.2 and shall be tested without
7.1 This test method covers tests for the magnetic
any heat treatment.The test may also be carried out
after ageing at a temperature of 225°C for 24 hours, properties of basic flat-rolled magnetic materials at
if agreed upon between the manufacturer and the power frequencies (25 to 400 Hz) using a 25-cm
purchaser. Epstein test frame and the 25-cm double- lap-
jointed core with corner setting. It covers the deter-
6.1.2.9 From material in coil form, prepare the test mination of core loss, volt-amperes, rms and peak
strips from test sheets cut from one or both ends of exciting current, and a.c. permeability and related
the coil. properties of flat-rolled magnetic materials under
a.c. magnetization.
6.1.2.10 From material in cut length form, two or
more test sheets shall be taken from the test lot. 7.2 This test method provides a test for core loss
and exciting current at moderate and high induc-
6.1.2.11 The minimum number of test pieces to be
tions up to 15 kG (1.5 T) on non-oriented electrical
cut in the case of the standard thickness of sheet steels and up to 18 kG (1.8 T) on grain oriented
shall be as under:
electrical steels.
Thickness of Sheet No. of Test Pieces
7.3 The frequency range of this method is normally
mm A4in
that of the commercial power frequencies 50 to
1.00 12 60 Hz.
0.65 16 7.4 This test method also provides procedures for
0.50 calculating a.c. impedance permeability from
0.35 measured values of rms exciting current and for a.c.
20
0.27 peak permeability from measured peak values of
total exciting currents at magnetizing forces up to
NOTE- In no case shall the specimen consist of less than
about 150 0, (12 000 A/m).
twelve strips and shall be a multiple of four.

3
IS 649 : 1997

7.5 The specimen for this test shall be selected and capable of carrying very much higher currents than
prepared for testing in accordance with provisions normally are encountered, in order to limit primary
of 6 of Section 1. circuit resistancevalues that will not cause appreci-
able distortion of flux wave form in the specimen
8 BASIC CIRCUIT when relatively high non sinusoidal currents are
being drawn. A primary circuit current rating of
8.1 Figure 2 shows the essential apparatus and 30 A is usually adequate for this purpose. Although
basic circuit connections for this test. Terminals 1 the current drain in the secondary circuit is quite
and 2 are connected to a source of adjustable a.c. small, the switches and wiring of these circuits
voltage of sinusoidal waveform and sufficient should be rated for at least 10 A to ensure that the
power rating to energize the primary circuit lead resistance is so small that the voltage available
without appreciable voltage drop in the source im- at terminals of all instruments is imperceptibly
pedance. The primary circuit switches Sr, S2 and SJ lower than the voltage at the secondary terminals
as well as all primary circuit wiring should be of the Epstein test frame.

*
Y I
b

r
s6
s
ul
TEST
FRAME
0
% Vr‘ms

FIG. 2 BASC CIRCUIT DIAGRAM FOR WATTMETIZRMETHOD

9 APPARATUS 9.1.1.1 The test frame shall consist of four

solenoids (each having two windings) surrounding
9.1 The apparatus shall consist of as many of the
the four sides of the square magnetic circuit, and a
following component parts as are required to per-
mutual inductor to compensate for air-flux within
form the desired measurement functions.
the solenoids. The solenoids shall be wound on
9.1.1 Epstein Test Frame (Fig. 3) nonmagnetic, nonconducting forms of rectangular
cross section appropriate to the specimen mass to
be used. The solenoid shall be mounted so as to be
accurately in the same horizontal plane, and with
the centre line of solenoids on opposite sides of the
square, 25OkO.3 mm apart. The compensating
mutual inductor may be located in the centre of the
space enclosed by the four solenoids if the axis of
the inductor is made to be perpendicular to the
plane of the solenoid windings.
9.1.1.2 The inner or potential winding on each
solenoid shall consist of one fourth of the total
m =OGm
number of secondary turns evenly wound in one
layer over a winding length of 191 mm or longer of
each solenoid. The potential windings of the four
solenoids shall be connected in series so their vol-
tages will add. The outer or magnetizing winding
shall consist of one fourth of the total number of
primary turns evenly wound over the winding
FIG. 3 2.5cm EPSTEIN FRAME length of each solenoid. These individual solenoid
 IS 649 : 1997

windings, too, shall be connected in series so their error in exciting current at 15 kG (1.5 T> caused by voltage
error may be 10 to 20 times as large as the voltage error. An
magnetizing forces will add. The primary winding effort should be made to maintain the calibration at 0.25
may comprise up to three layers using two or more percent (or better) of the true voltage reading at all scale
wires in parallel. points from half-scale to full-scale deflection. Voltage scales
should be such that the instrument is not used at less than
9.1.1.3 Primary and secondary turns shall be half-scale deflection. Care should also be taken to avoid
wound in the same direction, with the starting end errors due to temperature and frequency effects in the instru-
ment.
of each winding being at the same corner junction
of one of the four solenoids. This enables the poten- 9.1.3 RMS Viltnteter, Vm
tial between adjacent primary and secondary turns
to be a minimum through out the length of the A true rms-indicating voltmeter shall be provided
winding, thereby reducing errors due to electro- for evaluating the form factor of the voltage in-
static phenomena. duced in the secondary winding and for evaluating
the instrument losses. The accuracy of the rms
9.1.1.4 The solenoid windings on the test frame voltmeter shall be the same as that specified for the
may be of any number of turns suited to the flux voltmeter. The resistance of the rms voltmeter
instrumentation, mass of specimen, and test fre- shall not be less than 500 ohms per volt of full-scale
quency. Windings with a total of 700 turns are indication.
recommended for tests in the frequency range of 25
through 400 Hz. 9.1.4 Wattrneter

9.1.1.5 The mutual inductance of the air-flux 9.1.4.1 Electrodynanzonleter wattnleter, W

compensating inductor shall be adjusted to be the A reflecting type electrodynamomcter wattmeter is
same as that between the test-frame windings and desirable for all specimen masses and necessary for
within one turn of the compensator secondary. Its specimens lighter in mass than about 6 g/mm of
windings shall be connected in series with the strip length (see Fig. lb). For specimens weighing
corresponding test-frame windings so that the volt- more than this, a direct indicating low-power-factor
age induced in the secondary winding of the induc- electrodynamometer wattmeter of highest avail-
tor by the primary current will completely oppose able sensitivity may be used. For this later type of
or cancel the total voltage induced in the secondary instrument a 5 percent power-factor type is
winding of the test-frame when no sample is in desirable so that readings will not have to be taken
place in the solenoids. at less than 25 percent of full-scale indication. The
rated accuracy of measurement of the wattmeter, at
9.1.2 Flux Voltmeter, Vf
the frequency of test and for unity power-factor
A full wave true average responding typevoltmeter, loads, shall not be poorer than 0.25 percent of
with scale reading in average volts multiplied by full-scale deflection. For general testing, resistance
1.111, so that its indications will be identical with of the potential circuit of this instrument should
those of true rms meter on a pure sinusoidal volt- not be less than lOOohms per volt full-scale for each
age shall be provided for evaluating the peakvalue voltage range, and the inductance of the potential
of the test induction. To produce the estimated circuit should be such that the inductive resistance
accuracy of the test under this method the full scale at the test frequency will not exceed 1 ohm per 1000
meter error shall not exceed 0.25 percent (see ohms of resistance of this circuit unless the poten-
Note). Meters of 0.5 percent or more error may be tial circuit is compensated for its reactance. If tests
used at reduced accuracy. The resistance of the flux arc to be made at 15 kG (1.5 T) on ‘half-and-half
voltmeter shall not be less than 1000 ohms per volt grain specimen, the resistance of the wattmeter
of full-scale indicationand should have a resistance potential coil circuit must not be less than 5 000
high enough (3 000 to 10 000 ohms per volt) to ohms for each ohm of inductive reactance in the
avoid calibration and linearity errors but not potential circuit unless the instrument is adequate-
high enough to introduce electrostatic errors if ly compensated for errors due to its reactance (see
used with the mutual inductors as a peak ammeter Note 1). For tests at inductions up to 10 kG (1.0 T)
(see 9.1.6). A variable resistance, standard-ratio the current coils of the wattmeter should not have
transformer or other variable scale multiplying a resistance or reactance which exceeds 1 ohm. For
device may be employed to permit the flux tests at 15 kG (1.5 T) the current coil resistance or
voltmeter to be adjusted to indicate directly in units reactance should not exceed 0.25 ohms (see Note
of flux density if the combination cl” basic instru- 2). For power frequency testing of high loss
ment and scale multiplying device conforms to the material at any induction or for general testing at
specification stated above. high inductions it is desirable to have the current
NOTE - Inaccuracies in setting the test voltage produced coil resistance and reactance as low as 0.1 ohm. The
errors approximately two times as large in the core loss. The current rating of the wattmeter coils for low induc-

5
IS 649 : 1997

tion testing or at moderate inductions for oriented an air-core mutual inductor the peak value of its
material may be 1 or 2 ampere. Generally when flux (and hence the indications of the fluxvoltmeter
testing at 15 kG or higher a current-coil rating of 5 connected to its secondary winding) will be propor-
ampere or more is required (see Note 3). tional to the peak value of its primary current. A
mutual inductor used for this purpose must have
NOTES
reasonably low primary impedance so that its inser-
1 Failure to observe these limitations may necessitate cor- tion will not materially affect the primary circuit
rection for phase-angle errors in the indications of the watt-
meter. A variable resistance or other suitable variable-scale conditions, and yet must have sufficiently high
multiplying device may be employed to permit the wattmeter mutual inductance to provide a satisfactorily high
to indicate directly in watts per unit mass if the combination voltage to the flux voltmeter for primary currents
of the basic instrument and multiplier conforms to the corresponding to the desired range in the magnet-
specifications stated above.
izing force Hp. The secondary impedance of the
2 This is necessary to avoid excessive distortion of flux mutual inductor is important in relation to the
waveform in the test specimen due to nonlinear impedance
current in this circuit and should be low if any
voltage drops in series with the primary winding of the
Epstein frame. significant secondary current is drawn by a low
impedance flux voltmeter. In any case, the addition
3 This may be necessary to avoid objectionable or destruc-
tive temperature rise in thecurrent coils. For general testing of the fluxvoltmeter should not change the “mutual
at very high inductions the wattmeter current coils should inductor secondary terminal” voltage by more than
have a rating of 10 ampere or more. 0.25 percent. The voltage waveforms are extremely
peaked under normal test conditions and the flux
9.1.4.2 Wattmeter other than electrodynan~orneters
voltmeter should be capable of handling the high
It is anticipated that new developments in crest factor. Under sinusoidal calibration proce-
instrumentation will provide electronic, thermal, dures as indicated, the crest factor capabilities of
or other types of wattmeters which may be useful at the meter are not checked, and thus care should be
very lower power factors while retaining sufficient exercised that the crest capabilities are adequate. It
accuracy of measurement for use under provisions is important that the mutual inductor to be located
of this method. When any such wattmeter has in the test equipment in such a position that its
satisfactorily demonstrated its ability to meet the windings will not be linked by a.c. leakage flux from
requirements of 9.1.4.1 and to maintain required other apparatus. Care should be taken to avoid
accuracy levels of 11 it shall be acceptable for use locating it so close to any magnetic material or any
with this method and may replace the conducting material that its calibration and
clectrodynamometer instrument. linearity might be affected.
9.1.5 RMS Ammeter, A 9.t.6.2 Electronic peak-to-peak ammeter

A true rms-indicating ammeter is needed if meas- Even at commercial power frequencies there can be
urements of exciting current are to be made. A appreciable error in the measurement of H if wind-
nominal accuracy of 1.0 percent of full-scale or ing capacitances and inductances and flux
better is required for this instrument. The instru- voltmeter errors begin to become important at
ment must have very low internal impedance to some of the higher harmonic frequencies oc-
avoid contributing to the distortion of the flux casioned by the extremely nonsinusoidal character
waveform. of the voltage waveform induced in the secondary
of the mutual inductor by the nonsinusoidal excit-
9.1.6 Devices for Peak-Cwrent Measurement ing-current waveform. In such cases the crest or
9.1.6.1 Mutual-inductorpeak ammeter peak-current measurements may be made with an
electronic voltmeter whose indications are propor-
A means of determining the peak value of the tional to the peak-to-peakvalue of thevoltage drop
exciting current is required if evaluation of the a.c. that results when the exciting current flows through
magnetizing force is to be made by the peak-current a low value of standard resistance connected in
method. An air-core mutual inductor used in con- series with the primary winding of the test-frame.
junction with a flux voltmeter comprise the This electronic peak-to-peak reading voltmeter
apparatus most frequently used for this measure- should have a nominal full-scale accuracy (see
ment at 50 or 60 Hz. Use of this device is based on Note) of at least 3 percent at the test frequency and
the same theoretical considerations that dictate the be able to accommodate voltages with a crest factor
use of the flux voltmeter on the secondary of the of up to approximately 5. Care must be exercised
test frame to measure the peak inductions; namely, that the standard resistor (usually in the range 0.1
that when a fluxvoltmeter is connected to a test coil to 1.0 ohm) carrying the exciting current has adc-
the flux voltmeter indications are proportional to quate current carrying capacity and is accurate to at
the peakvalue of flux linking the coil. In the case of least 0.1 percent in value. It must have negligible

6
 IS 649 : 1997

temperature and frequency characteristics under

the conditions applying in this method. If desired,
the value of the resistor may be such that the peak-
reading voltmeter indicates directly in terms of H,
provided that the resistor otherwise confirms to the
limitations stated above. Normally this resistor will
replace the mutual inductor in the circuit of Fig. 2
and the shorting switch 5’3is used to remove this extra
resistance from the primary circuit when not in use.
NOTE -Because electronicvoltmeters are more subject to
change of calibration than conventional instruments, they
should be used only where means is also provided for
frequent and convenient checking of their calibrations to
ensure maintenance of the accuracy requirements.
FIG. 4 DOUBLE-LAPPED JOINTS
9.1.7 Power Supply
An a.c. power supply capable of satisfying the mag- 10.4 Demagnetization
netizing condition as given in LO.4 shall be used. When measurements of any magnetic property are
The stability of voltage and frequency while testing to be made at inductions below 10 kG (1.0 T), first
shall be within kO.2 percent. demagnetize the specimen. Accomplish this by
10 PROCEDURE initially applying a voltage sufficient to magnetize
the specimen to an induction above the knee of its
10.1 Prior to testing, check the specimen strip for magnetization curve (where the exciting current
length to see that they conform to the desired increases sharply for small increase in induction)
length to within 0.8 mm. Also check the specimen and then decrease the voltage slowly and smoothly
to see that no dented, twisted, or distorted strip has (or in small steps) to a very low induction. After this
been included. Strips having readily noticeable demagnetization, make tests immediately (to
shearing burrs also may be unsuitable for testing. obtain a test value within 2 or 3 minutes) for the
Weigh the specimen on a scale or balance capable desired test points. Make tests at several values of
of determining the mass within an accuracy of 0.1 induction in order of increasing induction values.
percent. Demagnetization may often be omitted for the test
10.2 Divide the test specimen strips into four frequency of 50 Hz.
groups containing equal numbers of strips having 10.5 With the required apparatus connected as in
very closely the same mass, for testing. Insert the Fig. 2 and with terminals 1 and 2 connected to the
strips (always a multiple of four in number) into the power source, then with switches S2, Sa and S4
test-frame solenoids one at a time, starting with one closed, S5 closed to the test frame size, and St and
strip in each of two opposite solenoids and then & open, adjust the voltage of the power supply to
inserting a strip into each of the other two solenoids a point where the fluxvoltmeter indicates the value
so that these later strips completely overlap the of voltage calculated to give the desired test induc-
former tap at the four corners (see Fig. 4). This tion in accordance with the equation of 11.1. Be-
completes one layer of strips constituting a com- cause the action of the air-flux compensator causes
plete flux path with four overlapped joints. Build a voltage equal to that which would be induced in
up successive layers in the same fashion until the the secondary winding by the air-flux to be sub-
specimen is completely assembled. With specimens tracted from that induced by the total flux in the
cut half with and half across grain, arrange all the secondary, the induction calculated from this
parallel or “with grain” strips in two opposite voltage will be the intrinsic induction&=@ -
solenoids and all the cross-or transverse-grain ,uoHp). In most cases the values of intrinsic induc-
strips in the other two opposite solenoids. tion, Bi, arc not sufficiently different from B to
10.3 If the specimen strips are reasonably flat and require that any distinction be made. Where ,L~*H~
have a reasonable area of contact at the corners, a is no longer insignificantly small compared to Bi, as
sufficiently low reluctance is usually obtained at very high inductions, determine thevalue ofB by
without resorting the pressure on the joints. When adding to Bi either a measured value of;coHp or a
the joints are unavoidably poor, the use of light nominal value known to be reasonably typical of the
pressure on the joints, with the use of nonmagnetic class of material being tested.
corner weights of about 200 g, is permissible al- 10.6 Core Loss
though it may introduce some additional stresses in
strain-sensitive materials. When the voltage indicated by the flux voltmeter
has been adjusted to the desired value, read the

7
 IS 649 : 1997

wattmeter. Some users, particularly those having 10.9 RMS Exciting Current
wattmeters compensated for their own losses (or
Measure therms exciting current when required, by
burden), will desire to open switch SS to eliminate
having Si and Ss closed; S2 S4, and S6 open; and Ss
the flux voltmeter burden from the wattmeter in-
closed to the test-frame side; then with the ammeter
dication others will likely choose to have Ss and Sg
on a suitable scale range, adjust the voltage to the
closed when measuring the losses, so that all instru-
correct fluxvoltmeter for the desired test induction.
ments may be read at the same time. In the later
When the setting of voltage-is correct, open Ss and
case the combined resistance load of the flux
read the ammeter with no current drain in the
voltmeter, rms voltmeter, and potential circuit of
secondary circuit. IfSs is kept closed to monitor the
the wattmeter will constitute the total instrument
induction during the current reading the current
burden on the wattmeter. Exercise care so that the
drain of the flux voltmeter will be included in the
combined current drain of the instruments does not
ammeter indication. If exciting current is to be
cause an appreciably large voltage drop in the
reported in terms of ampere-turns per unit path
secondary circuit resistance of the test-frame. In
length, volt-amperes per unit mass, or permeability
such a case the true induction in the specimen may
from impedance, calculate the values of these
be appreciably higher than is apparent from the
parameters from the equation given in 11.4.
voltage measured at the secondary terminals of the
test-frame. In any event, power due to any current 10.10 Permeability
drain in the secondary circuit at the time of reading
When permeability from peak current is required,
the wattmeter must be known so it can be sub-
determine the peak value of the exciting current for
tracted from the wattmeter indications to obtain
a given induction by having Ss open to insert the
the net watts due to core loss.
primary of the mutual inductor, St and S2 closed to
10.7 Obtain the specific coreloss of the specimen protect the ammeter and wattmeter from the pos-
in watts per unit mass at a specified frequency by sibility Of eXCXSSiVC CUrrentS. S4 and S6 open t0
dividing the net watts by that portion of the mass of minimize secondary loading, and Ss at first closed
the specimen constituting the active magnetic flux to the test-frame side. Then adjust the voltage to
path (which is less than the mean geometric path the current value for the desired induction, at this
length) in the specimen. Equations and instruc- point through Ss to the mutual inductor side to
tions for computing the active mass of thespecimen observe the corresponding value of flux volts (En,,)
and the specific coreloss are given in 11.2. at the secondary of the mutual inductor. For use in
10.8 Measure the rms value of the secondary volt- this manner at full accuracy the fluxvoltmeter must
age by having both Ss and S6 closed, and the voltage be such that the restrictions of 9.1.6.1 are met.
. adjusted to indicate the correct value of flux volts. Equations involved in the determination of peak
On truly sinusoidal voltage both voltmeters will current and HP by the mutual inductor method are
indicate the same voltage, showing that the form given in 11.6.
factor of the induced voltage is 1.111. When the
10.10.1 Various types of a.c. permeability may be
voltmeters give different readings, the ratio of the
determined from measurements described in these
rms value to that indicated by the flux voltmeter
methods. It should be understood that these a.c.
reveals the ratio by which the form factor of the
permeabilities are in reality mathematical defini-
induced voltage deviates from the desired value of
tions each based on different specified assump-
1.111. Determining the induction from the readings
tions. Therefore their individual values may differ
of flux voltmeter assures that the correct value of
considerably from each other and from the normal
peak induction is achieved in the specimen, and
d.c. permeability, ,LL
hence that the hysteresis component of thecoreloss
is correct even if the wave form is not strictly 10.11 If the peak reading voltmeter and standard
sinusoidal; but the eddy-current component of the resistors are used instead of the mutual inductor
coreloss, being due to current resulting from a non- and flux voltmeter for determining peak current,
sinusoidal voltage induced in the cross section of follow the same procedure as for the mutual induc-
the strip, will be in error depending on the deviation tor method except use switch Ss only on the test-
of the induced voltage from the desired sinusoidal frame side because a separate meter indicates the
wave shape. This error in the eddy-current com- peak current value. Equations involved in the
ponent of loss can be readily corrected by calcula- determination of peak current and HP by the peak-
tions based on the observed form factor and the reading voltmeter method are given in 11.6.
approximate percentage of eddy-current loss for
11 CALCULATIONS
the grade of material being tested if the correction
is reasonably small. The equations involved in 11.1 Calculate the value of the flux voltage, Ef in
determining this correction is given in 11.3. volts, at the desired test induction in the specimen

8
 IS 649 : 1997

(when corrected for flux due to H in the material ntl = 0.94 m/41 = 0.235 m/l
and in the air space encircled by the test winding where
through the use of the required air-flux compen- HZ = the total specimen mass in kg, and
sator) in accordancewith Ihe following basic equation. I = actual strip length in m.
Ef = 4.443 Bi ANzf
The equation giving the specific core loss in watt
where per kilogram of Epstein specimens then becomes;
Bi = maximum intrinsic flux density, in tesla (T);
P@ ; f) = (w - E2/R)l/0.235 M!
A = effective cross-sectional area of the test
specimen in square metres (m2); 11.3 Form Factor Correction
iV:! = number of turns in secondary winding, and The percent error in form factor is given by the
f = frequency, in cycles per second (Hz). following equation:
In the case of Epstein specimens, where the total F = (100 E/Ef) - 100
number of strips is divided into four equal groups assuming (see Note 1) that:
comprising the magnetic circuit, the mass of the
specimen in each of the four legs of the magnetic Observed P,(B ; 0 = [(corrected PC@;r~)/lOO]h +
circuit becomes m/4, and the effective cross-sec- (corrected Pc(B ; f))Ke/lOO
tion, A in square metres, of each leg is: The corrected coreloss, which shall bc computed
A = m/41P when F is greater (see Note 2) than + 1 percent is:
where Corrected Pq~;q = (observed Pc@;t)) lOO/(h+Ke)
nz = total mass of specimen strip in kg, where
I = length of specimen strips in m (usually
Observed Pc(B;f) = specific core loss calculated
0.28 or 0.305m), and
= by the equations in 11.2,
P standard assumed density of specimen
h = percentage hysteresis loss at induction B,
material in kg/m3.
e = percentage eddy-current loss at
Then, in the Epstein frame: induction B, and
Ef = (l.IllBi mN2f)/lP K = (E/E#
Obviously, h= 100-c” if residual losses are con-
For testing of Epstein specimen at 50 Hz and a
sidered negligible.
secondary winding of 700 turns the equation
simplifies to: The value of h and e in the above equation are not
critical when waveform distortion is low. Typical
Ef = (3.888 X 104Bi,)/ZP values of eddy current loss at 50 Hz for the common
11.2 Core Loss Calculation classes of materials, strip thicknesses and specimen
form are shown in Table 1.
To obtain the specific core lo ,s of the specimen in
watts per unit mass, it is necessary to subtract all NOTI
secondary circuit power included in the wattmeter 1 In determining the form factor error it is assumed that the
indication before dividing by the active mass of the hysteresis component of core loss will be independent of the
specimen, so that for a specific induction and Corm factor if the maximum value of induction is at correct
value (as it will be if a flux voltmeter is used to establish the
frequency the specific core loss in watts per kg is as value of the induction) but that the eddy-current component
follows: ofcorcloss, being thefunctionofthermsvalueofthevoltage,
will be in error for nonsinusoidal voltages. While it is strictly
Pc(B ; f) Or PC/,,,= ( w-E2/R)hlI
true that frequency or form factor separations do not yield
where true values for the hysteresis and eddy-current components.
Yet they do separate the core loss into two components, one
w= watts indicated by the wattmeter,
which is assumed to vary as the second power of the form
E = rms volts for the secondary circuit, factor and the other which is assumed to be unaffected by
R = parallel resistance ofwattmeter potential form factor variations. Regardless of the academic difficul-
circuit and all other secondary loads in ties associated with the characterizing these components as
hysteresis and eddy-current loss, it is observed that the equa-
ohms, and
tion for correcting core loss or waveform distortion of voltage
l?ll = active mass in kg. based on the percentages of first-power second-power of
In the 25-cm Epstein frame it is assumed that frequency components of the core loss does accomplish the
desired corrections under all practical conditions if the form
0.94 m is the effective magnetic path with specimen factoris accuratelydetermmed and thcdistortion not excessive.
strips 0.28 m or longer. For the purpose ofcomput-
2 It is recommended that tests made undcrconditionswherc
ing coreloss the active mass ntl of the specimen
the percent error in form factor, F, is greater than 10 percent
(Icss than the total mass) is assumed to be as be considered as likely to be in error by an excessive amount,
follows: and that such conditions be a\:- !cd.
IS 649 : 1997

Table 1 Eddy-Current Loss

(ClnLlse 11.3)

Material Specimen Assumed Fdcly-Current Loss, Percent

at SOIIzfor Strips Thicknesses
A
f \
0.27 0.30 0.35 0.50 0.65
mm mm mm mm mm

(1) (2) (3) (4) (5) (6) (7)

Nonoriented Half Longitudnal - - 20 30 40
and Half Transverse
Nonoriented Longitudinal - - 25 35 45

Oriented Longitudinal 50 50 60 - -

11.4 RMS Exciting Current

RMS exciting current is often normalized for 2 Hz is computed from the nnsvalue of the complex cxci ting
current by assuming a crest factor of n. Thus it is based on
circuit parameters by converting to the following
a sinusoidal current having a rmsvalue equal to the rmsvalue
forms: of the complex current.
RMS exciting force, NIIIII = NrI/O.94
= 1.064N11, A/m or 11.53 For control in the production of magnetic
Apparent a.c. magnetizing force, H, = v’TNd/Z1 materials, it is often desirable to determine an a.c.
= 1.504NtZ,A/m permeability value that is more directly comparable
where to the d.c. permeability of the specimen. This is
Nr = number of turns in primary winding, . accomplished by evaluating H, from the measure
I = rms value of exciting current, and peak value of the exciting current at some value HP
Hz = apparent a.c. magnetizing force A/m. sufficiently above the knee of the magnetization
curve that the component of the exciting current is
For the 700-turn Epstein frame, appreciably greater than the core-loss component.
Apparent a.c. magnetizing force, Hz = 1053.I,A/m Such a test point for many commercial materials is
Specific exciting power, PC,; f) = E&zt,exciting rms an HP value of 796 A/m. Permeability determined
VA/kg. in this way is characterized by the symbol,+, and is
where computed as follows (see Note 1):
nrr = active specimen mass in kg, and ,uP = BJHr
I = rms in amperes.
where
11.5 Permeability HP = peak exciting magnetizing force
evaluated from measurements of peak
11.5.1 For various types of applications, certain
current made either with the per-
types of a.c. permeability data (in H/m) are more
meability-inductor or peak-reading-
useful than others.
voltmeter methods [see 9.1.6.1 and
11.5.2 One type of a.c. permeability directly re- 9.1.6.21 and in accordance with the cqua-
lated to the rms exciting current (or rms excitation) tion in 10.6.
or a.c. impedance is characterized by the symbol
11.6 HP from Peak Exciting Current
,u, and is computed as follows (see Note 1):
= BiIHz = 0.665 BJNlI
,LJ.~ To evaluate peak exciting current [or peak exciting
= (0.950 x lo-‘)BJlfor the700 turn-frame magnetizing force, HP) by the mutual-inductor
method, the relationship between secondary flux
where volts and peak value of primary current for the
B; = Teslas, and mutual inductor must be established at the desired
I = rms amperes (see Note 2). frequency. This must be done by passing a
NOTES sinusoidal current through the primary of the
1 For simplification and convenience in the calculation of
mutual inductor and reading the resulting “opcn-
a.c. permeabilities thevalue of& is used to replace& in the circuit” secondary flux volts, Er,,, with a flux
permeability equation. This entails no loss of accuracy until voltmeter of very high impedance. Then a calibra-
,&fP becomes appreciable in magnitude when compared to tion constant can be established such that, for the
thevalue of& If greater accuracy is essential,&, or (Bi +
frequency of calibration:
,&I,,) should be used to replace & in these equations.The
magnetic constantpo is equal to 4x X 10-7H/m. K,,,, flux VlpeakA = Edl, = EdId

10
 IS 649 : 1997

where nonoriented material is estimated at +3 percent.

II) = pcakvalue, and The precision of exciting current measurement is
I = rms value of the calibrating sinusoidal estimated at 25 percent at commercial power
current. frequencies.
Then, at the calibration frequency: 13 AGEING TEST
H,,AIm = N11,lll = NIEc,,,,/O.~~K~ 13.1 The samples shall be subjected to accelerated
= 1.064N1Ef,,,/K,” = 744.7(&/K,,,) ageing test for 24 hours at a temperature of 225°C.
for the 700-turn test frame. Often theactual mutual 13.2 Subject to agreement, the ageing test shall be
inductance of the inductor is known from a.c. performed for 600 hours at a tempcraturc of 100°C.
bridge or d.c. ballistic measurcmcnts. In that case, Intermediate measurements can be made after 200
it can bc shown that: and 400 hours in order to ensure that the ageing has
Efn,, flux V = 4.443jZ,,,I, finished at the end of 600 hours.

where SECTION 3 STANDARD TEST METHOD

L,,, = mutual inductance in Henry, and FOR DIRECT-CURRENT MAGNETIC
H,,, A/m = NtErm/4.443jZm11 = 0.2394N1Er~,i’, PROPERTIES OF MATERIAL USING
If L,,, is made exactly 33.52 mH, then for test at 50 THE BALLISTIC METHOD
Hz in the 700-turn Epstein frame: 14 SCOPE
HP = 100Erm
14.1 This test method covers d.c. ballistic testing
11.6.1 The magnetizing force from the peak value for the determination of basic magnetic properties
of the exciting current may also be determined by of materials in the form of double lapped testing
the peak-reading voltmeter method that involves cores. It includes tests for normal induction and
the measurement of the peak value of voltage drop hysteresis taken under conditions of steep wave
across a small resistor carrying the exciting current. front reversals of the direct-current magnetizing
The relationship between H,), the peak voltage force.
across the standard resistor, R,, and peak exciting
14.2 This test method shall be used in conjunction
current Z,,is:
with Section 1.
E,, = RJp or Ic,A = Ey.,,/2Ro
14.3 This test method is suitable for a testing range
whcrc from very low magnetizing forces up to 200 Oe or
E,,.,, = peak to peak value in volts of the sym- more (15.9 kA/m or more). The lower limit is dcter-
metrical voltage drop across R,. mined by integrator sensitivity and the upper limit
by heat gcncration in the magnetizing winding.
The magnetizing force from peak exciting current Special techniques and short duration testing may
is then: extend the upper limit of magnetizing force.
H&/m = NlE,,.,,/2Rdl = 0.5319N1E,.,,lR~ 14.4 Testing under this test method is inherently
If t hc standard resistor is made exactly0.3723 ohms, more accurate than other methods. When specified
then for tests at all frequencies in the 700-turn dimensional or shape requirements are observed,
Epstein frame, the measurements are a good approximation to
absolute properties. Test accuracy available is
H,, = 1 OOO.E,,.,
primarily limited by the accuracy of instrument-
In 11.6, the pearmcability from peak exciting cur- ation.
rent will be (see Note 1 of 11.5.2);
14.5 This test method permits a choice of test
specimen to permit measurement of properties in
12 l’RECISION AND BIAS any desired direction relative to the direction of
crystallographic orientation without interference
12.1 This is a basic method and although its true from external yoke system.
bias is not known it is assumed to be the same as its
precision. 14.6 The acceptable minimum width of strip used
in such test specimens is also sensitive to the
12.2 Precision of core loss test at 50 Hz is estimated material under test.
to be within 1 percent for nonoriented materials
core loss at 10 kG (1.0 T) and for oriented material 15 APPARATUS
at 1.5 kG(1.5 T). Precision at 15 kG(1.5 T) on 15.1 Epstein Test Frame-Same as 9.1.1.

11
IS 649 : 1997

15.2 d.c. Power Supply within +O.l pcrccnt) in either polarity when nor-
mal induction testing is conducted, current rcvcr-
The preferred source of d.c. current is high quality
sals can be conducted with no overshoot or
linear power supply of either unipolar or bipolar
oscillation and the magnetizing current is truly zero
operation. The power supply must exhibit high
for the zero current programming signal.
stability and very low ripple in order to achieve the
most accurate results. Programmable bipolar 15.7 Hysteresis Switch, S2 (When Required)
operational amplifier power supplies are satisfac- This switch should conform to requircmcnts given
tory for this type of testing. Other stable source of in 15.6.
d.c. current such as storage batteries is permitted.
15.8 Integrator, F
15.3 Main-Current-Control Rheostat, RI
Due to their superior accuracy, stability and ease of
When nonprogrammeable. sources of d.c. current operation, electronic charge integrators are the
such as storage batteries are used, rheostats must preferred means of measuring magnetic flux. In-
be used to control the current. These rheostats tegrators utilizing either operational amplifier and
must have sufficient power rating and heat- capacitor feedback (analog integrator) or pulse
dissipating capability to handle the largest test cur- counting are permitted. The accuracy of the in-
rent without undesirable changes in resistance and, tegrator must be better than 1 percent full scale. If
therefore, magnetizing current during conduct of analog display meters are used to read the value of
the test. flux, the measurement should bc made on the upper
15.4 Hysteresis-Current-Control Rheostat, R2 two-thirds of the scale. Analog integrators must
have drift adjust circuitry and the drift should not
The hysteresis-current-control rheostat, when exceed 100 Maxwell-turns (10e6 Wb turns) per
required, must have the same power rating and minute on the most sensitive range. It is also
resistance as the main-current-control rheostat. desirable that the integrator have appropriate scal-
15.5 Ammeter, A ing circuitry to permit direct reading of either flux
(p) or flux density (B). Ballistic galvanometers or
Measurement of the magnetizing current can bc moving coil fluxmetcrs arc allowed provided the 1
accomplished with either a d.c. ammeter or a percent full scale accuracy rcquiremcnt is met.
combination of a precision shunt resistor and d.c.
voltmctcr. The meters and shunt resistor, if used, 15.8.1 By agreement be1 ‘?en the parties, a ballis-
must have an accuracy of atleast 0.25 percent. To tic galvanometcr or charge integrator may replace
improve test accuracy multirange digital ammeters the flux meter.
orvoltmeters are preferred. Autorangingcapability 15.8.2 When a Ballistic Galvanometer is used, this
is desirable for convenience but is not essential for should be of periodic time not less than 7 seconds.
this test method. If analog meters arc used, the
ranges must be such that all test readings are made 16 CALIBRATION
in the upper two-thirds of the scale. 16.1 Calibration of Integrator
15.6 Reversing Switch, S1 Practical operating experience has shown that
Due to the low resistance nature of the magnetizing provided a proper warm up period is allowed,
circuit, it is imperative that high quality switches be electronic indicators require infrequent calibra-
used. Changes in switch resistance upon reversal tion and unlike ballisticgalvanometers, calibration
will cause deviation from the cyclically magnetized is not an integral part of this test method. When
condition which if excessive will impair test calibration is required it can be accomplished with
accuracy and precision. Experience has shown that either a mutual inductor or a volt-second source.
mercury switches are the best suited for this ap- Due to their traceability to the fundamental units
plication. Knife blade switches or mechanical or of voltage and time, volt-second sources are the
electrically operated contactors can also be used preferred means of calibration. The accuracy of
provided the requirement for uniform and equal either the mutual inductor or volt-second source
contact resistance can be maintained. Due to the must be better than the rated full scale accuracy of
presence of leakage currents in the open condition, the integrator.
solid state relays are not permitted. The difficult& 16.2 Calibration of Ballistic Galvanometer
inherent in the use of main current reversing
switches can be minimized by use of linear power The galvanometer scale and B-circuit may be
supplies capable of accepting a remote program- calibrated using current reversals in the mutual
ming signal. Such power supplies are permitted inductor. The following equatioli shall be used to
provided that the magnetizing current is equal (to determine the calibration values:

12
 IS 649 : 1997

I, = BNA/(L, x 10’) 17.3 The strip shall be weighed accurately before

where assembling and their mean cross section calculated
Ic = current required for reversal in the from the formula:
primary of the mutual inductor L, to A = mlpl
calibrate the B-circuit for a desired
where
deflection in A,
A =
mean cross section of ‘the test strips
B = flux density in the test specimen at
calibrated deflection in G, in m*,
m= total mass of the strips in g,
N= number of turns in B-sensing coil,
A = cross sectional area of test specimen in P density in g/m3, and
1 1 mean length of the test strips in m.
cm*, and
Lm = value of calibrating mutual induction in 18 PROCEDURE
units of mH.
The equation can also be written as. 18.1 In Fig. 5 the d.c. power source supplies mag-
netizing current measured by ammeter A. Rheos-
Ic = ONI(L, x 10’)
tats Rl and R2 and switches SI and S:! determine the
where p = BA or total magnetic flux, Mx magnitude and direction of the current as required
16.2.1 Using the above equation substitute in the by the various operations. In general, three types of
value of flux density B, which corresponds to the switching operation are required in ballistic testing.
desired calibration flux density and the values of the One is reversal magnetizing-current direction
specimen area turns and mutual inductance. This without change of magnitude as required for estab-
gives the value of current which must be reversed lishing a cyclically magnetized condition and in
in the mutual inductor. Set this value of current normal induction tests. This is accomplished by
through the mutual inductor and observe, the throwing switch St from one side to other. Asecond
galvanometer deflection on current reversal. The is reduction of magnitude of magnetizing current
value of the calibrating resistor is then adjusted to without change of direction. This operation is re-
make the galvanometer deflection on current quired to measure points on the hysteresis loop in
reversal swing from zero to the desired scale the first quadrant. This is done by opening switch
deflection for the calibrated deflection point. S2.,The third operation combines reversal of mag-
Usually the scale is calibrated to make the deflec- netizing current direction with a reduction in mag-
tions on reversal equal to the B value of calibration nitude. This operation is required to measure
or some simple multiplier of it. points on the hysteresis loop in thesecond and third
16.3 For basic material evaluation the gal- quadrants. Obtain this reversal and reduction by
vanometer shall be calibrated with sufficient num- simultaneously throwing switch Si from one side to
ber of current values to provide a calibration curve the other and opening switch S2. Use care to be sure
which is accurate to 0.1 percent of full scale or 0.2 S2 is opened beforeS is reclosed for reversal. When
percent of smallest scale division. When desired determining the hysteresis loop, switches Si and S:!
because of non linearity or other reasons the test must be operated to traverse the loop in the same
deflection points may be calibrated independently direction between successive measurements so as
without completing a full scale calibration. to preserve the cyclically magnetized state of the
test specimen.
17 PREPARATION AND ASSEMBLY OF TEST
SPECIMENS 18.2 Demagnetize the test specimen immediately
17.1 The test specimen shall be cut from the sheet prior to testing. To demagnetize with direct cur-
selected at random from the batch of 5 tonnes or rent, first establish a magnetizing force sufficiently
part thereof to be tested. Aspecimen shall have the large to cause the flux density in the specimen to
same width as the Epstein test specimen. For non- reach a value greater than the knee of the normal
oriented sheets, one half of pieces shall be cut induction or magnetization curve. Then slowly
parallel to and the other half perpendicular to the reduce the magnetizing current to zero while simul-
direction of rolling as given in Fig. la and lb. For taneously operating the reversing switch at one half
oriented sheets the pieces shall be cut parallel to second or longer intervals. An auxiliary circuit
the direction of rolling as shown in Fig. lc. The total using a time delay relay to effect switch reversal will
number of test pieces taken for this test shall not be make this operation more reproducible and less
less than 12. tedious. When the test specimen consists of thin
strip (less than 0.001 m thick) alternating current
17.2 All the strips shall be cut from a single sheet
and shall be distributed symmetrically over the demagnetization using 50 Hz or lower frequency
entire area of the sheet as far as practicable. and autotransformers can be used.

13
IS 649 : 1997

1 FLUXME’IER

FIG.~CIRCUITFORD.C. TESTING

18.3 To obtain the flux density (B) corresponding difference in flux density between the point of
to a specific magnetizing force (H), establish the maximum magnetization current and the point
proper magnetizing current using equation 1, cycle corresponding to the hysteresis loop measurement
the reversing switch several times to establish the current.
cyclically magnetized condition, zero the integrator
19 CALCULATION
and execute the proper switching procedure as
found in 18.1. The value of the flux or flux density 19.1 The mean magnetizing force applied to the
can then be computed from the integrator reading. test specimen by the current through the magnet-
Additional test points on the normal induction izing coil is determined from the equation:
curve can be obtained without demagnetization if H = NI/fl ... (1)
they are obtained in ascending order of B or H.
Otherwise it is necessary to demagnetize prior to where
additional testing. H = magnetizing force, A/m
N = number of turns in magnetizing coil Nr,
18.4 To obtain the magnetizing force correspond-
Z = current through the magnetizing coil in
ing to a specific flux density, a procedure similar A, and
to 18.3 is used with the exception that the magnetiz-
It = mean magnetic path length in m.
ing current, and therefore magnetizing force must
be found by trial and error. If the specified flux 19.2 The Epstein test frame coils are built consid-
density is exceeded, demagnetization is usually re- erably larger than the test specimen cross sectional
quired before proceeding further unless operating area. To avoid the need for manual air-flux
at very low flux densities. correction a compensating mutual inductor is built
into the test-frame. This means that the flux density
18.5 Electronic integrators do not determine flux measurements are intrinsic flux density, Bi, meas-
densities directly, rather the change in flux linkages urements. To obtain normal flux density, B, the
(N2_0) is measured. This result is converted to following equation must be used:
changes in flux density by division by the specimen
B=Bi+p,Jl ... (2)
cross-sectional area A, and number of secondary
turns, N2. To determine the actual value of flux
where
density the starting or reference points must be
B =normal flux density of test sample in T,
known. In the case of normal induction or mag-
Bi =intrinsic flux density of test sample in T,
netization curve measurements, it is customary to
H =magnetizing force, A/m, and
zero the integrator and measure the change in flux
/lo =magnetic constant of free space (in SI
density for a fully reversed change in magnetizing
system p, = 4 JGx 10m7H/m).
current. In this instance, the true value of flux den-
19.3 Permeability is calculated as follows:
sity is one half of the total change in flux density.
For hysteresis loop determination, the integrator is ,U = B/H = (WH) + ~0 ... (3)
zeroed at the point of maximum magnetization.
The resulting change in flux density is equal to the where ,u = normal permeability, H/m.

14
 IS 649 : 1997

20 REPORT lamination resistance, interlaminar resistance,

which refers to the average resistivity of two or
20.1 When normal induction (flux density) values
more adjacent insulative surface in contact with
or hysteresis-loop points have been measured for
each other.
the purpose of reporting basic material properties
the following shall be reported along with the test 22.3 The apparatus is popularly known as Franklin
data. Tester.
20.2 Heat treatment or other processing applied to 23 SUMMARY OF TEST METHOD
the test specimen prior to testing.
Ten metallic contacts of fixed area are applied to
20.3 When permeability is reported, the cor- one of the surfaces of the specimen and electrical
responding values of B or H must be reported. contact is made with the base metal by two drills.
The effectiveness of the surface insulation is then
20.4 When hysteresis-loop properties are
indicated by a measurement of average electrical
reported, the values of peak magnetizing force or
current flowing between the contacts and the base
peak flux density used shall be reported.
metal under specified applied voltage. This
20.5 When saturation or other flux density values measurement can be used directly as an indicator
are reported, the value of magnetizing force must of insulation quality or may be converted to an
be reported. apparent surface insulation resistivity value.
21 PRECISION AND BIAS 24 SIGNIFICANCE AND USE
21.1 The accuracy of determining magnetizing 24.1 This test method is particularly suitable for
force H is usually dependent on the accuracy of quality control in the application of insulating
current measurement, and ability to maintain iden- coatings.
tical current after reversal and in the accuracy of
24.2 Insulating quality of a coating is measured by
determining magnetic path length. For the Epstein
a current that ranges from zero for a perfect
frame, due to corner joints, there is some uncertain-
insulator to 1.00 A for a perfect conductor.
ty as to the true path length, the determination of
H will be within 5~2.0 percent. 24.3 Single readings should not be considered
significant since the nature of the test device and
21.2 The accuracy of determining flux density, B, is
specimen are such that successive measurements of
usually dependent on the quality of integrator
a specimen often yield different values.
calibration, on the uniformity of material and
accuracy of determining the cross sectional area of 25 APPARATUS
the test specimen. When the best instrumentation
25.1 The apparatus, as shown in Fig. 6 and 7 shall
and calibrations are used, the fluxdensity, B, will be
consist of the following.
determined within + 1 percent.
25.2 Test Head
21.3 When permeability is calculated the errors
associated with both B and Hare included. For the The test head shall consist of a mounting block on
Epstein test-frame the material permeability which parts are assembled.
determinations should be within +3 percent.
25.3 Two parallel longitudinal rows of fivever-
tically mounted steel rods free to move axially
SECTION 4 STANDARD TEST METHOD
against surrounding spiral springs or other means
FOR SURFACE INSULATION RESISTIVITY
to apply equal pressure.
OF SINGLE-STRIP SPECIMENS

22 SCOPE 25.4 Brass, stainless steel or other suitable metallic

contact button on each rod, but insulated from it.
22.1 This test method covers a means of testing the
Articulation of tips improves contact by compen-
surface insulation-resistivity of single strips or sating for minor misalignments. Avoid soft metals,
punchings of flat rolled electrical steel under
poor conductors or metals subject to oxidation or
predetermined conditions of voltage, pressure and attack by solvents used in cleaning. Due to low-
temperature. voltage circuitry (0.5 V) all contacting surfaces
22.2 The term surface insulation resistivity used in must be kept clean. Full area contact of tips to core
this method refers to the effective resistivity of a plate is needed to avoid decreases in Franklin
single insulative layer tested between applied bare amperage. Check with known samples or standard
metal contacts and the base metal of the insulated test lots. The total contact area of the ten contact
test specimen. It is not the same as the terms inter- buttons shall be 6.45 cm2.

1.5
IS 649 : 1997

TEST SPECIMEN

Fx.6 A~PARATIJSOFSURFACEINSULATIONRESISTIVITYMEASUREMENT

BAT TERY=L
(1*5.10 2V)

FILM

FIG. 7 CONNBCTIONSFOR CONTACTSAND RESISTORS

25.5 A 5 ohms (+ 0.1 percent) resistor connected press and hot plate must provide a smooth, flat and
to each contact button. Contacts with their in- rigid support for the test specimen.
dividual resistors shall be connected in parallel as
25.8 IIot Plate
shown in Fig. 7.
25.6 Electrical contact with the base metal is made The hot plate shall be such that the test specimen
through two 3 mm diameter twist drills (preferably can be heated to the temperature of test with auto-
carbide tipped) or hardened and pointed rods. matic control to maintain the test temperature.
These are vertically mounted and spring loaded in
25.9 Test Head Power Supply
spiral slotted sleeves to impart a twist while
piercing the coating. The instrument may be operated from batteries or
25.7 Hydraulic Press from a voltage regulated d.c. power supply. For
battery operation, either storage or dry cell types
The hydraulic press shall have a capacity of 10 000 may be used with appropriate control rheostats for
N, with mountings to accommodate the test head, setting voltages during tests. Cdmmercial power
test specimens or punchings and a hot plate. The supplies are available for use with this equipment.

16
..,,,. .... .. .
r
,,!

--+

1.s 649:199:

po~ili~.c ixmntiu!. ConncLL the w)lm]ctc~ to show

V()[MLW :{cr~ss the tCSL head. W%cn COnnccted in
this nmnncr the vdtmclcr kxt cxrrrcnt wiil appear
as an error current in thcammeter readingwtd shall
bc SUb&aCICYJfr~m the ammeter readings. on short
circuit the correct load current k LOO .4 plus the
vol[mcic.r current. To eiirnin~te this correction it is
pcrmissibktooffscx Ihc amrnctcr 10 a zero position
when the test head is up (m}! making contact) 10
r-cturn the ammeter to a xv-o reading ~~i!h 0.5 V
applied Mtwccn the head and the test NM plate.
~7,~ _f”()cllsure ~~rrect contact button condition,
I?UIkc Mshort circuit test occasionally by lesting a
l?Jrc metal surface. When the short circuit current
is kss than 0.99 A, clean the contacts. The use of
solvcnl.s for ckaning is preferred {o abrasives bc-
cause ~he Iatcr can result in rounded tips with
tutucc.d contact areas.
27,3 The rccommmkt standard prcwrre for ]?ur-
pose of comparative tests shall be 2.1 MPti. Other
prussurcs, dcpendingupon the applications. maybe
agrccxi upon by the manufacturer and the pur-
chaser. lf more than otw ICS!pressure is 10 bc used.
applv the preSsurcs in ascending orxlcr. DIIring test-
ing. ~pply the pressures only once, t-mLan applhxl
prcsstrre may bc incrcascd to a higher vaiuc.
27A If both sides of the specimen art coale(f (10not
usc [he same area to test both sides.
2?,5 The recommcnckd st.anrlard test
tcnlperdt urcs orc roorrr tcmpcrafurc or 150C’C.
(-MM [cmpcraturcs and lhc sequence of tcmpera-
lurc~ dcpmtrling upon the &pplicati{)n may be
ayru.cd upon by the rnanufacturct - and ihc pur-
ch:l.~cr.When wsts are mxk al elevated tcrnpcra-
1tJrcs aliow sufficient t irne (usually 30 sccorxls) [0
h“~aI the spccimcn to the specified temi]erat ure.

27.6 Piacc the specimen on the plate bcrwath the

tt’st ticad and prsshhxi it so Ihiit al: conuicts are
within Ihc test area whcrr the test head is brought
in con [act with the spccimcn. Apply the spccificd
pressure. Adjust ~he whage to 0.50 V and read the
ammeter.

2X C,41.CUATIONS

2S.1 The average of electrical current mc.asure-

mcnts is usually acceptable for m’aiualing surface
insuifition. Average the current readings for each
surface for a minirnurn of fi~’e specimens. The
reported va!ue for a test k)t shall hc (he average of
bolh surfaces,

28.2 In the event electrical resistivity value is

desired, the average unit KCSiStanCCp~r lamillatk!~l
I,I,WO surfaces) ma~ bc calculated i<)two significant
jIgl)rcs as f~l]ows:-
IS 649 : 1997

Ri = (6.4.5/I)- 6.45 ... (1) 33 TEST SPECIMEN

Strips with a minimum length of 250 mm and a
where
maximum width of 30 mm, as for example
Ri = average surface resistivity in ohms/cm’
Epstein test specimen strips shall be used for the
per lamination (two surfaces), and
measurement. The oxide or other insulating coat-
I = ammeter reading A.
ing need not be removed except from places where
29 PRECISION electrical contacts must be made. The current
contacts should be in the form of transverse clamps
Even with the best practices in design, instrumen- covering at least 80 percent and preferably the
tation, maintenance and operation, the entire width of the specimen. The potential con-
repeatability and reproducibility of the test method tacts can be either knife edge or point contacts.
are greatly influenced by the nature of the surfaces
of the test specimens. Hence it is not considered 34 PROCEDURE
possible to state meaningful values for repeatability 34.1 Measure the electrical resistance of the test
and reproducibility that are universally applicable. specimen using a Kelvin type resistance bridge or
potentiometer-ammeter system having separate
SECTION 5 DETERMINATION OF current and potential leads.
RESISTIVITY OF MAGNETIC SHEET AND STRIP
34.2 The distance between each potential lead
30 SCOPE contact and the corresponding current lead contact
This test method covers determination of the resis- shall be at least twice the width of the test specimen
tivity of magnetic sheet and strip. with the two potential contacts lying between the
current contacts. The distance between the
31 SUMMARY OF METIIOD potential contacts shall be not less than 120 mm.
The electrical resistance of 250 mm long test 34.3 The dimension of each potential contact in
spccimcn is measured with a Kelvin type resistance the direction of the length of the specimen shall be
bridge or potentiometer-ammeter method. The not more than 0.5 percent of the distance between
resistivity is then calculated from the resistance the potential contacts.
measurement and the dimensions of the specimen,
and is known as the electrical resistivity of the 34.4 The contacts to the specimen shall be located
material. This value is equal to the resistance centrally with respect to the specimen’s width
between opposite faces of a cube of unit dimension and the current contacts shall COVCI
dimensions. more than 80 percent of the width. A rcliablc
contact shall be made with the specimen b!
32 APPARATUS both the current and potential leads.
32.1 Kelvin type resistance bridge or a d.c. potcn- 34.5 Specimen temperature during test should bc
tiomcter and d-c ammeter providing resistance about 27 + 5°C.
measurements to an accuracy within 0.5 percent of
the accepted true value. 34.6 To climinatc errors due to contact potential,
take two readings, one direct and one with the
32.2 If a potentiometer is used, a suitable d.c.
current rcvcrsed, in close succession,
source and ammeter are required to establish and
measure the total current in the specimen, which 34.7 The clcctrical current in the test specimen
should be limited to avoid excessive heating. The must be limited to avoid over heating but must be
required R is then the ratio of the measured potcn- adequate to provide sufficient sensitivity toshowan
tial drop to the measured current. When the potcn- out-of-balance condition when the resistance read-
tiometer is balanced, no current flows in the ing is changed to 0.5 percent of the value recorded.
potential leads so that any contact resistance at the If the current is too low, sensitivity is low also, and
potential point is of no consequence. a balance can be shown for a broad range of
32.3 The Kelvin bridge is calibrated to read dircct- resistance.
ly the resistance between the potential points
35 CALCULATION
without knowledge of the currents in the specimen.
Contact resistance at a potential points and the 35.1 Determine the average cross sectional area of
resistance of the four leads to the specimen are not the test specimen from the weight, length and
a part of the required R and are usually negligible density as follows:
portions of the corresponding components of the
A = (nz//)P ... (1)
bridge system.

18
 IS 649 : 1997

where specimens or determinations of weight percentages

A = cross sectional area of test specimen of silicon and aluminium contents provide reliable
in m2, means of establishing densities for commercial iron
112 = mass of test specimen in kg, base flat rolled products.
1 = length of test specimen in m, and
39 TEST SPECIMEN
P = density of test specimen in kg/m3.
NOTE - Equation (1) assumes a negligible mass of any Test specimen strips of suitable dimensions, for
coating material. example Epstein strips may be used for the
measurements.
35.2 Calculate the resistivity from the measured
value of electrical resistance and the cross sectional 40 IMMERSION METHOD
area determined by equation (I ) as follows:
40.1 In arbitration test, the conventional immer-
y = RAII:! ... (2) sion method must be applied. This method is ap-
propriate only when test specimens without any
where coating are used. The test specimens should be
degreased prior to use in the test.
P = electrical resistivity of the material in
ohm. m, 40.2 By means of hydrostatic balance the test
R = resistance of electrical path in ohm, specimen is weighed before and after immersing it
in water of known densityp,,,. All air bubbles should
A z cross sectional area of electrical path in be removed from the surface of the test specimen.
m2, and Then the densitypw of the test specimen is given
12 = length of electrical path between by:.
potential contacts on the test specimen in Pm = ml(m - mi) .pw ... (1)
m.
35.3 The resistivity units in ohm metre shown in where
equation (2) can be converted to micro ohm cm by P m= density of test specimen in kg/m3,
multiplying the ohm metre figure by 1O’micro ohm pw = density of water in kg/m3,
cm per ohm metre . For example, if the resistivity I72 = mass of test specimen in kg ,and
of a 1 percent silicon plus aluminium steel is Wli = apparent mass of test specimen im-
0.25 x 10e6 ohm metre or 0.25 x lo-” x lo8 micro mersed in water in kg.
ohm cm per ohm men-e is equal to 25 micro ohm cm. In order to obtain comparable results, it is
36 ACCURACY necessary to test at least five test specimens and to
Accuracy of the method for measuring resistivity of average the results.
steels of usual surface finish is estimated to be 41 INDUSTRIAL METHODS
within +2 percent.
41.1 For industrial tests the following methods are
SECTION 6 DETERMINATION OF THE
recommended.
DENSITY OF THE MAGNETIC SHEET 41.1.1 Electricnl Method
37 SCOPE According to this method the density of magnetic
Density values normally used in the calculation of sheet can be determined by measuring the electrical
the magnetic cross-section of the Epstein resistance of the test specimen.
specimens in accordance with the equation 1 (see 41.1.2 The electrical method is applicable only to
Section 5) are those values defined as the conven- silicon bearing sheet, the total silicon and
tional density in product standards. aluminium content of which lies between 1 and 5
38 METHODS TO BE USED percent of mass. Aluminium, if present, shall not be
higher than 0.4 percent. This later value is also the
38.1 The immersion method is a fundamental limit for the total of other common constituents.
method of determining density but, in application, This method recognizes the fact that both density
requires that the surfaces of the test specimen be and resistivity are functions of the aluminium and
essentially free of oxide films or applied coatings. silicon contents.
However, in case of arbitration, only this method
will be valid. 41.1.3 The resistance of a test specimen strip is
given by:
38.2 For industrial purposes, measurements of
R, = Pl,IA ... (2)
clcctrical resistance on coated or uncoated
is 619: 1997

The test specimen strip is wcighcd and the lotal I = total length of test specimen strip in m,
length measured :o within 20.2 percent or better. and
The clcctrical resistance is measured according to I, = dislancc bctwccn the potential contacts
the procedure described in Section 5. The mass i\ in m (see Section 5)
given by:
By combining equations 2 and 3, the cross scclion
nt = Pm IA ... (3) A and therefore the thickness of lest specimen are
eliminalcd.
where
pm . p = R~,t/l.l~ (9
R, = measured electrical resistance of test
specimen strip in ohms,
Density can be read from the curve ofp”, against
P = rcsistivity of the test material in ohm m,
pn, . p as shown in Fig. 8.
P 111= density of the test material in kg/m3,
n1 = mass of test specimen in kg, In order to obtain comparable results, it is ncces-
A = cross section of test specimen strips sary to test at least five test specimens and to
in m2. avcragc results.

cl*002 0903 o-ooc O-005

pm.qcn kg/ml ___c_

FIG. 8 DENSITYp,,,AGNNST PRODUCTOm . p

41.1.4 Chemical Method SECTION 7 STANDARD TEST METHOD

FOR STACKING FACTOR OF MAGNETIC
For iron based alloys the relationship found in
MATERIALS
practice between density and silicon content or
silicon and aluminium content is given by equation: 43 SCOPE
pm = 7.865 - 65 (Psi + 1.7 PAI) (5) This test method covers measurement of the stack-
ing factor also known as lamination factor of a
whcrc
spccimcn composed of strips cut from magnetic
P “1 = density in kg per m3, material.
Psi = proportion of silicon (in percent, mass),
and 44 SUMMARY OF TEST METIIOD
PAI = proportion of aluminium (in percent,
mass). The laminated test specimen is subject to pressure
in a compression device and the resulting volume is
42 REI’ItODUCIRILITY then determined from the measured specimen
The reproducibility of the results obtained from height, width and length. An equivalent solid
each method is characterized by a standard volume is calculated from the specimen mass and
deviation of the order of 0.2 to 0.3 percent. the true density of the specimen material. The ratio

20
 IS 649 : 1997

of the calculated (equivalent solid) volume to the 47.2 Stack the strips evenly and place them
measured volume is the stacking factor. symmetrically between the two flat plates in the
compression testing machine.
45 APPARATUS
47.3 Apply pressure so it is distributed uniformly
45.1 Testing machine across the test specimen. The recommended
A compression testing machine or other standard minimum test pressure shall be 3.5
compression device capable of exerting the kgf/cm2 and is gradually applied.
specified pressure. 47.4 Calculate the average separation of the back-
45.2 Metal plates ing plates or stack height at the required pressure
from measurements of plates separation.
Two flat smooth rigid metal plates with square
47.5 When using a compression testing machine
edges and ends are required. They shall be of
make four measurements of the separation, one at
sufficient stiffness to ensure practically uniform
each corner of the backing plates.
pressure in the sample. Each plate shall be 215 mm
long and have a minimum width of 50 mm so that 47.6 When using a compression device designed
the area of strips under pressure when testing 30 specifically for determining this factor only two
mm wide specimens will be 6 450 mm2. measurements, taken on the longitudinal axis of the
strips at each end of the plates will be satisfactory.
45.3 Measuring Device
48 CALCULATIONS
The measuring device shall be capable of measuring
the height of the stack placed symmetrically with 48.1 Calculate the percentage stacking factor as
respect to the compression head on either side of follows:
the stack; Use of dial gauges capable of measuring S = mlwPt x 100
with an accuracy of one-hundredth of a milimetre
is recommended. Vernier caliper may also be used. where
s = stacking factor percent,
46 SAMPLING AND TEST SPECIMEN m= mass of test specimens in kg,
46.1 The test strip shall be selected as rep- width of test specimens in m,
r; 1 length of test specimens in m,
resentative of surface condition. Core loss test
specimens (Epstein test) not less than 16 pieces are P = density of test specimens in kg/m3, and
normally used for this purpose. t = measured average separation plate faces
in m.
46.2 The test specimens must be composed of
strips taken from one lot of steel. It must be 48.2 Length and width dimensions should be
representative of surface condition gauge and other known to at least 0.25 percent and preferably to 0.1
variables. The shearing burrs or loose particles percent.
shall be carefully removed from the pieces before
the test. SECTION 8 STANDARD TEST METHOD
FOR DETERMINATION OF DUCTILITY
46.3 The test specimen shall preferably consists of
the number of strips prescribed in Table 2. 49 DETERMINATION OF DUCTILITY OF
NON-ORIENTED ELECTRICAL STEEL
Table 2 Number of Test Pieces for SHEETS
Stacking Factor This test method covers determination of ductility
utilizing Epstein test strips and a bending device for
bending the strip over a predetermined radius. It is
Thickness, mm Number of Pieces
intended for non-oriented electrical sheet or strip
(1) ‘2”8’
0.27 covered under IS 648 : 1994.
0.30 28
0.35 24 50 SUMMARY OF METHOD
0.50 16
0.65 and above 16
A test strip is placed in the special test apparatus
designed to clamp one end of the specimen securely
46.4 Each strip shall have a minimum length of 305 while the other end is free to move, but held in
mm and width of 30+0.08 mm. tension by a spring. The specimen is repeatedly bent
47 PROCEDURE through 180“ reversals until a crack appears at the
bend or until sudden failure occurs by complete
47.1 Weigh the test specimens carefully with an rupture. The number of reversals until failure is
accuracy of rt 1 g. taken as a measure of the brittleness.

21
IS 649 : 1997

51 APPARATUS

51.1 The apparatus consists of a set of stationary

jaws and a movable arm to which is attached
another set of jaws and spring.
51.2 The stationary jaws shall have working edges
with radii of approximately 5 mm over which the
test specimen is bent. Stationary jaws shall be of the
quick clamping type.
51.3 The jaws attached to the movable arm shall
allow the specimen to move freely during bending.
51.4 The spring clamped to the free end of the
specimen shall provide sufficient tension in the
specimen to localize the bend.
51.5 Design of the movable arm shall permit a
rotation of approximately 18OO. F1c.9 APPARATUSFORBEND TEST
52 SAMPLING AND TEST SPECIMEN
52.1 The test specimen may be cut from samples
used for core loss or other tests.
52.2 The test specimens shall be about 30 mm in
width and not less than 152 mm in length.
52.3 The number of test specimens representing
each test lot shall not be less than ten.
52.4 The long axis of at least five test specimens FIG.10 DIAGRAMILLUSTRATINGMETHOD
shall be in the direction of rolling and at least five OFMAKINGBENDS
at the right angles to the direction of rolling.
55 DETERMINATION OF DUCTILITY OF
52.5 Edges of the test specimens shall be practical- ORIENTED ELECTRICAL STEEL SHEET
ly free of burrs; filing or machining to remove burrs
This method covers determination of the ductility
is permissible.
of grain oriented electrical steel covered under
52.6 When width of the material prevents cutting IS 3024 : 1996 by use of an apparatus known
specimens at right angles to the direction of rolling, variously as a thinner’s brake, hand folder, or an
all specimens shall be cut in the direction of rolling, apron brake.
and this shall be reported with the test results.
56. SUMMARY OF METJIOD
53 PROCEDURE A test specimen representing the full width of grain
53.1 Clamp the specimen tightly in the stationary oriented steel to be tested is bend through an angle
jaws in the bend test machine, Fig. 9, and place in of about 160 degree in a thinner’s brake. The
tension by stretching the spring during clamping. numbers of breaks, or fractures, occuring along the
bend determines the ductility class rating.
53.2 Bend the specimen through 9O’by use of the
movable arm and jaws, Fig. 10; then bend it through 57 SIGNIFICANCE AND USE
180’ in the reverse direction. Again bend the 57.1 This is a specialized bend test for grain
specimen through 180” in the first direction and oriented electrical steel
continually through 180’ reversals until a crack 57.2 This test is applicable to grain oriented silicon
appears at the bent or until sudden failure occurs steel in commercial thickness and widths up to 910
by complete rupture. mm.
53.3 Each full HO0 bend, including the first 90° 5X APPARATUS
bend, shall be counted as one bend in determining 511.1 The machine required to perform this test is
the number of bends withstood by the specimen. known as a thinner’s brake, hand folder, or an apron
54 CALCULATIONS brake
Express the brittleness of the test lot as the average 58.2 The brake shall be at least 910 mm wide with
of the number of bends withstood by the test an opening capactiy of approximately 0.90 mm.
specimens from that test lot. 58.2.1 The nose bar shall have a O.SOmm radius.

22
 IS 649 : 1997

58.2.2 The bending bar must be movable through ing table which is based on giving a larger numerical
an angle of at least 160’. class rating corresponding to the larger number of
58.2.3 The movable table or apron shall move brcaks and poorer ducti1itY.
around the end of the nose bar- at a distance of Clcrss Rnting Condition of Bend
approximately 6.5 mm. Class 1 Not more than one break with length
59 TEST SPECIMEN not exceeding 8 mm
59.1 Two specimens are required and they shall be Class 2 Not more than two breaks with total
sclectcd from the same general location as that of length not exceeding 1.5mm
the magnetic test specimen. Class 3 Three to eight breaks, all sizes
59.2 The specimens shall be cut transversely to the Class 4 Nine to fifteen breaks, all sizes
rolling direction and have a length equal to the Class 5 More than fifteen breaks, all sizes
sheet or strip width and a minimum dimension of
76 mm in the direction of rolling. 61.2 The class ratings shown are based on strip
widths of 610 to 910 mm.
59.3 The specimens must be free of rust, ripples
and scratches. 61.3 When evaluating steel strip narrower than
60 PROCEDURE 610mm, the number of breaks should be multiplied
by the ratio of 610 over the strip width in mm. This
60.1 The test specimen shall be at room tcmpcra- converted number of breaks will then determine
turc at the start of the test. the class rating.
60.2 Insert the test specimen into the brake and
61.4 The class rating assigned to a test lot shall be
clamp with the direction of rolling perpendicular to
the higher numerical class number of the two
the nose bar and 12 to 38 mm under the nose bar,
specimens.
thcrcby allowing the balance of the specimen width
to rest against the bending bar.
SECTION 9 METHOD OF MEASUREMENT
60.3 Bend the specimen around the nose bar at a FOR DETERMINATION OF INTERNAL
uniform rate by rotating the bending bar through STRESS
an angle of 160’.
62 SCOPE
60.4 Remove the specimen from the machine and
without straightening the bend, examine the out- 62.1 This test method specifies determination of
side face of the bend from surface breaks without internal stress.
magnification. 62.2 Definition
60.5 Count the breaks and measure for length. The variation ‘c’ from the shearing line is the
61 INTERPRETATION OF RESULTS greatest distance between the related edges of a
61.1 Classify each test specimen according to the sheet cut longitudinally. It is a measure of the inter-
length of number of breaks as shown in the follow- nal stresses (see Fig. 11).

t
ROLLING
DIRECTION

(a) Convex Gap (b) Concave Gap

FIG. 11 DEVIATIONFROMTHESHEARINGLINEDUETO INTERNALSTRESSES

23
IS 649 : 1997

62.3 Determination of the Variation from the 63.2 Test Specimens

Shearing Line
A specimen of a definite length specified is cut ihe test specimen shall consist of asheet or a length
through, parallel to the direction of rolling. The of strip of 2 metres.
two parts, neither of which may be reversed, are
weighed in such a way that they lie flat. The two cut 64 SIZE MEASUREMENT
edges are then again placed together until the smal-
lest possible gap is present. The greatest remaining 64.1 Thickness
distance between them is measured. The specimen
must be at least 1000 mm long. Measure the thickness as shown in Fig. 12 at any
point situated not less than 25 mm from the edges
62.4 Test Report
using a contact micrometer with an accuracy of0.01
Variation ‘c’ from the shearing line in mm, to the mm. For materials of width less than SC mm,
nearest 0.2 mm. measure the thickness on the longitudinal axis of
the sheet.
SECTION 10 METHODS OF
MEASUREMENT FOR DIMENSIONS 64.1.1 Deviation of thickness in transverse direc-
AND TOLERANCES tion is defined as the difference of the maximum
value and the minimum value in the thickness
63 SCOPE
(excluding the portions from edges of steel strip)
63.1 This test method deals with various tests mcasurcd in transverse direction.
performed to measure size and shape variation,
burr, and out of square.

FIG.12 MEASURINC~POIN,~SOFT~~ICKNHSS
64.2 Width and Length deviation from the ordered length shall bc as per
1s 648 : 1994.
64.2.1 Width
64.3 Out of Square
Measure the width perpendicular to the lon-
gitudinal axis of the sheet using a rule or tape This tolerance applies to cut lengths only and is the
measure. greatest deviation on an edge from a straight line at
right angles to a side and touching one corner, the
64.2.2 Length
measurement being taken as shown in Fig. 13. It
The dimensions of cut length shall be as close as can also bc measured as one-half the difference
practicable to the ordered length. The maximum between diagonals of cut length sheet.

r
OUT OF
SQUARE

FIG.13 OUTOFSQUARETOLERANCEFORCUTLENGTII

24
 IS 649 : 1997

64.4 Cutting Burr Measurement gauge stops, measure the burr height on the step,
where the burr has come in contact.
64.4.1 Method 1 64.4.2 Method 2

Figure 14 shows a stepped checking gauge for Place an external micrometer on a cut part, take the
measurement of cutting burr on the edge of the reading (h), when the ratchet sound once next
sheet. Method of measurement to be as follows. remove the micrometer from the cut part and place
on a portion nearby and taken the reading (h).
Place step gauge on the plain surface of the sheet Take the difference (h) of above two reading as the
and slide slowly towards cut edge, when the step cutting burr [see Fig. 151

h =hl- h2

FIG.14 BURRHEIGHT (h)

a DIRECTION OF MOVEMENT OF GAUGE

v-l-

/ 9 STEPS LSURFACE ‘A’RESTING

EACH WITi AN ON SURFACE OF SHEET
INCREMENT OF 0.Olmm
STARTING FROM 0.02mm
ALL SURFACES HRD & GRD

FIG. 1.5 STEPPEDCHECKINGGAUGEFORBLJRRONEDGE

65 TESTING OF SHAPE VARIATION 651.5 In the case of waviness which does not cx-
65.1 Deviation from Flatness (Wave Factor) tend over the entire width of the sheet, the depth of
thewavc is a further characteristicvalue. The depth
65.1.1 Definition t of the wave is deemed to be the greatest distance,
Variation from flatness in electrical steel sheet measured with a steel rule 20 mm wide and 0.5 mm
takes the form of waviness. Waviness can occur at thick, between the edges of the sheet and the point
the edge and in the middle of the sheet and over the at which it touches thesteel rule (see Fig.16 and 17).
entire width of the sheet.
65.1.6 Detemination of Waviness
651.2 Waviness W is the ratio of the height h of a
wave to the length I of the wave. The sheet (or length of strip) to be tested is placed
on a surface plate sufficiently large so that it does
65.1.3 The height h of the wave is the greatest not hang eve; the edges; then lift up one edge until
distance between the underside of the sheet and a it is approximately vertical and allow it to fall back
flat surface. freely. Measure the maximum height of the maxi-
65.1.4 The length 1 of the wave is the distance over mum wave (/L) and the length of the wave (I). The
percentage wave factor is equal to 100 /i/Z.
which the sheet rises above the surface plate.

25
IS 649 : 1997

FIG. 16 CHARACTERISTICVALUESFORTHEDETIZRMINATIONOFWAVINESS

F1c.17 EXAMPLEOFWAVEFORMATIONATEDGE

6.5.1.7 In the case of waves which do not extend 652.2 Determination of Bowing
across the width of the sheet, the depth of the waves
is also measured. A scale, 20 mm wide, 0.5 mm thick A sheet atleast 280 mm long and at least 20 mm
wide is clamped over a distance of 30 mm in such a
and of adequate length is slid under the sheet until
resistance is felt by means of repeated probing, the way that 250 mm thereof is freely suspended. The
distance between the lower edge of the sheet and
greatest distance to the outside edge of the sheet is
the vertical wall is measured.
to be ascertained and read off.
65.2.3 Test Report
65.1.8 Test Report
Length 2 of wave in mm, to the nearest 10 mm. Bowing ‘u’ in mm to the nearest 1 mm.
Height h of wave in mm, to the nearest lmm. Wavi- 65.3 Edge Camber or Straightness
ness Win percent to the nearest 0.1 percent. Depth
t ofwave in mm, to the nearest 5 mm.
65.3.1 Definition

65.2 Bowing The measure of straightness is the greatest distance

‘e’between the concave longitudinal edge of a sheet
65.2.LDefini/ion and a straight edge, of a definite length specified,
Bowing is the residual curvature, in the direction of laid against it. The straight edge must be at least
1000 mm long (see Fig. 19).
rolling, of strip unwound from the coil. A measure
of bowing is the greatest distance ‘u’ between a strip 65.3.2 Determination of Straightness
of sheet at least 20 mm wide and with a free length
of 250 mm and a vertical surface when the strip is The sheet is placed flat on a surface plate. After
pressed against the vertical surface over a distance applying a straight edge as per 65.3.1 to the con-
of at least 30 mm as shown in Fig. 18. It is desirable cave iongitudinal edge, the greatest distance be-
to use an Epstein test strip. tween the straight edge and the sheet is measured.

26
 IS 649 : 1997

65.3.3 Test Report

Straightness ‘e’ in mm to the nearest 0.2 mm.

CLAMPING SYSTEM7

ED

FIG. 18 MEASURINGDEVICEFOR BOWING

FIG. 19 VERIFICATIONOF THE EDGE CAMBER

SECTION 11 TESTS ON INSULATION cooling to the room temperature measure the

COATING insulation resistance.

66 SCOPE
68 RESISTANCE TO SOLVENTS
68.1 Xylene
These test methods deal with various tests per-
formed on insulation coating such as thermal effect To check the adherence of coating.
on coating, resistance to solvents, resistance to 68.1.1 Sample Size
freon, resistance to heat corrosion resistance,
adherence test and flexibility test. Suitable to test facilities. Minimum width of sheet
should be 30 mm.
67 THERMAL EFFECT ON COATING
68.1.2 Method of Test
Twelve specimen of the coated strip shall be Two clean glass beakers should be taken with equal
clamped together under a pressure of 1 N/mm2 quantity of xylene. The samples should be im-
approximately and heated in laboratory oven at a mersed in one of the beakers. Both the beakers
temperature 150°C for a period of 7 days. After should be heated to boiling for six hours. The

27
IS 649 : 1997

samples should be weighed before and after inser- 68.3.4.3 Coating should be checked to see if it has
tion in xylene. become loose.
68.1.3 Test Report 68.4 Resistance to Freon
68.1.3.1 Xylene in beaker with sample should be This test is valid only for inorganic and inorganic
compared with the xylene in other beaker to check with little organic based insulation coatings.
any change in colour.
68.4.1 Electrical steel sheet for the motor cores of
68.1.3,2 There should be no change in weight of completely sealed compressors must have high
sample which is immersed in xylene. resistance of insulation coating to freon attacks,
because the stators and rotors come in direct
68.1.3.3 Coating should be checked to see if it has
contact with the refrigerant and refrigerator oil.
become loose.
68.4.2 Sample Size
68.2 Trichloroethylene
Suitable to test facilities.
68.2.1 To check resistance of coating to degreasing
agent. 68.4.3 Method of Test
68.2.2 Sample Size Freon 22 and refrigerator oil to be taken in 1 : 3
proportion in a sealed chamber. Two clean glass
Suitable to test facilities. Minimum width of sheet
beakers should be taken with equal quantity of
should be 30 mm.
refrigerant oil of known viscosity. The samples
68.2.3 Method of Test should be immersed in one of the beakers both the
beakers should be kept in sealed chamber for 100
Procedure same as mentioned in 68.1.2. Only
hours at 105°C.
xylene is replaced by trichloroethylene.
68.4.4 Test Report
68.2.4 Test Report
68.4.4.1 Two beakers containing refrigerant oil
68.2.4.1 Trichloro ethylene in beaker with sample
should be compared for any change in colour.
should be compared with trichloroethylene in
other beaker to check any change in colour. 68.4.4.2 There should be no change in weight of
sample which is immersed in refrigerant oil.
68.2.4.2 There should be no change in weight of
sample which is immersed in trichloroethylene. 68.4.4.3 Coating should not become loose.
68.2.4.3 The coating is to be checked to see if it has 68.4.4.4 Viscosity should not change after the test.
become loose.
68.5 Resistance to Heat
68.3 Transformer Oil
This test is valid for inorganic and inorganic with
68.3.1 To check resistance of coating to little organic based insulation coatings.
transformer oil.
68.5.1 Sample Size
68.3.2 Sample Size
Suitable for test facilities.
Suitable to test facilities. Minimum width of sheet
should be 30 mm. 68.5.2 Method of Test
68.3.3 Method of Test The samples should be annealed in non-oxidizing
atmosphere at a temperature of 800°C for two
Two clean glass beakers should be taken with equal
hours.
quantity of transformer oil. The samples should be
immersed in one of the beakers. Both the beakers 68.53 Test Report
should be heated to 100 - 12O’C for eight hours.
68.5.3.1 The coating should not flake-off or burn-
The samples should be weighed before and after
off.
insertion in transformer oil.
68.6 Corrosion Resistance
68.3.4 Test Report
68.6.1 Sample Size
68.3.4.1 Transformer oil in beaker with sample
should be compared with the transformer oil in Suitable for test facilities.
other beaker to check any change in colour. 68.6.2 Method of Test
68.3.4.2 There should be no change in weight of The test is to be carried out in a humidity chamber
sample which is immersed in transformer oil. for 96 hours.

28
 IS 649 : 1997

68.6.3 Test Report 68.8 Flexibility Test

68.6.3.1 The samples should be checked after every
24 hours for rust formation, if any. Insulation coating should not peel offduring bend-
ing through 180° on 12 mm diameter mandrel.
68.7 Adherence Test
The surface coating shall be sufficiently adherent so 68.9 Coating Thickness
that it does not detach during further coating or
chip off at edges during shearing or punching. In The coating thickness of the steel sheet is to be
the reverse bending test with a bending radius of measured using an instumcnt based on magnetic
5 mm, the surface coating shall not be detached induction principle.
after bending through 90’.

29
Bureau of Indian Standards

BIS is a statutory institution established under the Bureau @In&n Standards Act, 1986 to promote
harmonious development of the activities of standardization, marking and quality certification of goods
and attending to connected matters in the country.

Copyright

BIS has the copyright of all its publications. No part of these publications may be reproduced in any form
without the prior permission in writing of BIS. This does not preclude the free use, in the course of
implementing the standard, of necessary details, such as symbols and sizes, type or grade designations.
Enquiries relating to copyright be addressed to the Director (Publications), BIS.

Review of Indian Standards

Amendments are issued to standards as the need arises on the basis of comments. Standards arc also
reviewed periodically; a standard along with amendments is reaffirmed when such review indicates that
no changes are needed; if the review indicates that changes arc needed, it is taken up for revision. Users
of Indian Standards should ascertain that they are in possession of the latest amendments or edition by
referring to the latest issue of ‘BIS Handbook’ and ‘Standards: Monthly Additions’.

This Indian Standard has been developed from Dot : No. MTD 4 ( 4048 ).

Amendments Issued Since Publication

Amend No. Date of Issue Text Affected

BUREAU OF INDIAN STANDARDS

Headquarters:

Manak Bhavan, 9 Bahadur Shah Zafar Marg, New Delhi 110002 Telegrams : Manaksanstha
Telephones : 323 0131,323 33 75,323 94 02 (Common to all offices)

Regional Offices : Telephone

Central : Manak Bhavan, 9 Bahadur Shah Zafar Marg { 323 76

38 41
17
NEW DELHI 110002

Eastern : l/14 C. I.T. Scheme VII M, V. I. P. Road, Maniktola { 337 84

86 26,337
99,337 85
9120
61
CALCUTTA 700054

Northern : SC0 335-336, Sector 34-A, CHANDIGARH 160022

{ 60 20
38 43
25

Southern : C. I. T. Campus, IV Cross Road, CHENNAI 600113

{ 235 02
15 19,235
16,235 04
23 42
15

Western : Manakalaya, E9 MIDC, Marol, Andheri (East) 832 92 95,832 78 58

MUMBAI 400093 832 78 91,832 78 92

Branches : AHMADABAD. BANGALORE. BHOPAL. BHUBANESHWAR. COIMBATORE.

FARIDABAD. GHAZIABAD. GUWAHATI. HYDERABAD. JAIPUR. KANPUR.
LUCKNOW. NAGPUR. PATNA. PUNE. THIRUVANANTHAPURAM.

Printed at Printograph, New Delhi (INDIA).

 AMENDMENT
NO. 1 JULY 2001
TO —. .—
-
IS 649:1997 METHODS OF TESTING STEEL SHEETS
FOR MAGNETIC CIRCUITS OF POWER ELECTRICAL
APPARATUS
( Second Rev&wn )

(Page 13, clause 16.2 ) — Substitute the following for existing equation:
/c= $N/(Lm x 105 )
(Page 19, clause 40.2 )
a) L.ine3 — Substitute ‘pW’for ‘pm’.
b) Line 5 — Substitute ‘pm’for ‘pW’.

(MTD4)

Reprography
IJnk BIS, New Delhi, [dim
 AMENDMENT NO. 2 MAY 2011

TO

IS 649 : 1997 METHODS OF TESTING STEEL SHEETS

FOR MAGNETIC CIRCUITS OF POWER
ELECTRICAL APPARATUS
(Second Revision)

(Page 2, Section 1) — Magnetic Testing as given in this section is for Conventional Grain
Oriented Steel (CGO), High Permeability Grain Oriented (HPGO), and Non-oriented Electrical
Steel. In case of some HPGO grades like Domain refined not amenable for heat treatment use the
test method as given in IEC 60404-3:1992 + A 1 : 2002 Magnetic materials, Part 3: Methods of
measurement of the magnetic properties of magnetic sheet and strip by means of a single sheet tester.

(Page 22, clause 55) — The clause specifies method for determination of the ductility of grain
oriented electrical steel as covered under IS 3024 : 2006. The alternate method is given in Annex A.
Either of the method can be used as per availability of apparatus.

ANNEX A
(Clause 55)

ALTERNATE METHOD 2 FOR DETERMINATION OF THE DUCTILITY OF

GRAIN ORIENTED ELECTRICAL STEEL

A-1 SUMMARY OF METHOD

A test strip is placed in the special test apparatus designed to clamp one end of the specimen securely
while the other end is free to move, but held in tension by spring. The specimen is repeatedly bent
through 90o reversals until a crack appears at the bend or until sudden failure occurs by complete
rupture. The number of reversals until failure is taken as a measure of the brittleness.

A-2 APPARATUS

Same as given in clause 51.

A-3 SAMPLING AND TEST SPECIMEN

A-3.1 The test specimen may be cut from samples used for core loss or other tests.

A.3.2 The test specimens shall be about 30 mm in width and not less than 152 mm in length.

A-3.3 The number of test specimens representing each test lot shall not be less than five.

A-3.4 The long axis of test specimens shall be in the direction of rolling.

A-3.5 Edges of the test specimens shall be practically free from burrs; filing or machining to remove
burrs is permissible.

A-4 PROCEDURE

A-4.1 Clamp the specimen tightly in the stationary jaws in the bend test machine (see Fig. 9). Place in
tension by stretching in the spring during clamping.

A-4.2 Bend the specimen through 90o use of the movable arm and jaws (see Fig.10) alternatively to
each side of its initial position. The test shall be stopped on appearance in bare metal of the first crack
visible to the naked eye, and last bend shall not be counted.

1
A-4.3 A bend of 90o from initial position with return to initial position counts one bend.

(MTD 4)
Reprography Unit, BIS, New Delhi, India
1

2
 AMENDMENT NO. 3 NOVEMBER 2012
IS 649 : 1997 METHOD OF TESTING STEEL SHEETS FOR
MAGNETIC CIRCUITS
OF POWER ELECTRICAL APPARATUS
(Second Revision)

[Page 2 Section 1 (see also Amendment No. 2)] — Insert the following
new clause at the end:

‘6.1.2.15 Magnetic testing as given this section is for conventional grain

oriented steel (CGO), high permeability grain oriented (HPGO) and
non-oriented electrical steel. In case of some HPGO grades like domain
refined not amenable for heat treatment use the test method as given in
IEC 60404-3 : 1992 + A 1 : 2002 ‘Magnetic materials Part 3: Methods of
measurement of the magnetic properties of magnetic sheet and strip by means
of a single sheet tester’.

NOTE — As an alternative to IEC 60404-3; the methods for single sheet test as given in
ASTM 804 or JIS 2556 may be followed, subject to agreement between the purchaser and the
manufacturer.’

(Page 21, Section 8, clause 49) — Substitute the following for the existing
clause:

‘49 DETERMINATION OF DUCTILITY OF ELECTRICAL STEEL

SHEETS

This test method covers determination of ductility utilizing Epstein test strips
and a bending device for bending the strip over a pre-determined radius. It is
intended for grain oriented and non-oriented electrical sheet or strip.’

(MTD 4)
Reprography Unit, BIS, New Delhi, India