CA1123716A - Method of producing low loss pressed magnetic cores from microlaminations - Google Patents

Abstract

47,099 METHOD OF PRODUCING LOW LOSS PRESSED
MAGNETIC CORES FROM MICROLAMINATIONS

ABSTRACT OF THE DISCLOSURE
A method for making pressed magnetic components having a low core loss characterized by the steps of com-pacting a plurality of substantially rectangular particles of iron alloy having a silicon content of from about 2.5%
to about 3.5% and a carbon content of up to 0.01% into a compact of predetermined configuration, and stress relief annealing the compact in a deoxidizing atmosphere so as to provide a magnetizable compact having low core loss.

Description

CROSS-REFERENCE TO RELATED PATENTS
This invention is related to U. S. Patent No.
4,158,581 issued June 19, 1979, by R. F. Krause and Norman Pavllk; U. S. Patent No. 4,158,582 issued June 19, 1979, by R. F. Krause; U. S. Patent No. 4,158,561 is~ued June 19, 1979, by N. Pavlik and John Sefko, U. S. Patent No.
4,158,580 issued June 19, 1979, by Wllliam Reynolds and N. Pavlik.
BACKGROUND OF THE INVENTION
Field of the Invention- ~
This invention relates to a method of making a magnetic core having a significantly low core loss, and more particularly, it pertains to a method for using high 3ilicon iron alloy particles which when insulated, compacted and annealed provide a magnetic core exhibiting a very low core loss.

~ Z ~ 6 Description of the Prior Art:
The term "mlcrolamination" which has been defined in U.S. Patent Nos. 3,848,331 and 3,948,690, and in general terms relates to a small rectangular particle of low carbon steel, when processed in a specific manner, is capable of being formed into a magnetic core or compact possessing soft magnetic characteristics which are useful in a ~yriad of applications, for example a light ballast, and as shown in U.S. Patent No. 3,235,675. A typical low carbon steel comprises about 0.10% carbon, less than 0.04% sulfur, less than 0.60%
manganese, and about 0.10~ silicon. Generally, the processing of microlaminations includes a decarburizing, deaxidizing, and stress relief anneal, the application of an insulating medium, and compaction of the microlaminations to a usable magnetizable compact form. The method of compaction is either uniaxial or isostatic in nature. The resultant compact exhibits a core loss of nominally 5.25 to 7 watts per pound at an induction of 14 kG. When the compact is subsequently annealed to relieve stresses resulting from the prior compaction step, the core loss increases dramatically by a factor of from 2 to 10. For example, a ring core pressed at 125,000 psi from 0.060 x 0.010 x 0.006 inch particles of low carbon steel has a core loss of approxi-mately 6.6 W/lb at 14 kG. After a 10 minute stress relief anneal at 700C in dry hydrogen, the core loss is increased to approximately 33.2 W/lb at 14 kG. Manifestly, the im-pairment of the loss characteristics is intolerable.
_UMMARY OF THE INVENTION
It has been found in accordance with this invention that the lower core loss properties are obtainable by a method of making pressed magnetic core components for use in electrical apparatus, comprising the steps of providing ,. ~3 . ~ , l~ Z 3~ ~ 47,099 microlaminationæ from an iron base alloy sheet having a slllcon content of from about 2.5% to about 3.5% and a carbon content of up to 0.010%, annealing the microlamlna-tions in a deoxidizing atmosphere to avoid formation of a heavy oxide, coating the microlaminations with a dielectric material, assembling the coated microlaminations within a container of a predetermined configuration, compacting the microlam~nations under pressure into a magnetizable compact, and annealing the compact for about 4 hours at about 9250C
in a deoxidizing atmosphere to relieve compaction stresses.
The advantage of the method of th~s invention is that the effect of a core annealing treatment is ~nexpectedly different from that found for low carbon steel microlamina-tions, and that signi~icantly lower core losses are obtained when an iron alloy material having a silicon content of from about 2,5% to about 3.5% and a carbon content of less than 0.01% is used.
BRIEF DESCRIPTION OF THE DRAWINGS
, ~
Figure 1 is a graph illustrating the effect of stearate additions on density of uniaxlally pressed iron-silicon scrap compacts;
Fig. 2 is a graph illustrating the effect of packing factor on 10 kG permeability of iron-silicon scrap compacts after 925C anneal; and Fig. 3 i8 a graph illustrating the frequency dependence of core loss on low carbon steel microlaminations and iron-silicon microlaminations.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
A very high usage of an iron-base alloy containing about 2.5% and about 3.5% silicon with less than 0.01%
carbon is found in the electrical apparatus industry. This X

~ 47,099 materlal is processed in such a manner that there exists a predetermined crystallographic orientation in the material as commercially processed. This orientation ls usually of the (110) [001] character as described by Miller Indicies, that is, the (110) plane is aligned in the ~001] direction, which is the rolling direction. Thus there are three differ- -fk~ f~oro ent axes of~g~T~L~, each of different magnitude insofar as the ease of magnitization is concerned. By utilizing these geometry-dependent magnetic characteristics, the design engineers can employ this oriented material to obtain a product having the optimum electrical character-istics at the lowest cost. Consequently, the fact that the core material is oriented is essentlal to the foregoing considerations.
The oriented material as above described is further characterized by being in the secondary recrystallized eondition in which the enormous grain growth marks the grain size readily discernible with the naked eye. Thus, the larger the grain sizeJthe better the magnetic characteristies, especially core loss since the eddy current component of the core loss decreases with increasing grain size. Typically a single grain may measure 1 inch in the rolling direction by the thickness of the rolled product. Thus metallurgists strive to produce the largest grain size possible.
In contrast thereto the present invention is directed to the use of microlaminations which have dimen-sions which are only a fraction of the size of a single grain of the oriented material described previously. Most advantageously, the microlaminations to which the process of the present lnvention i~ applicable take the form of "saw ~ Z 37 ~ 47,099 chips~ which are generated during the manufacture of wound cores for distribution transformers. Thus the technology employed for making cores for transformers employing oriented material and the technology for forming magnetic cores from microlaminations from the same materials is not synonymous.
Moreover, due to the size and method of making the magnetic cores from microlaminations, the possibility of gainfully employing the orientation characteristics is indeed remote.
According to the present invention the new method is carried out in the following preferred sequential manner:
1 Preparation of microlaminations from stock of the requisite chemical composition;

2. Annealing the microlaminations in a deoxidizing atmosphere;

3. Coating the microlaminations with an elec-trically insulating material;

4. Assembling the microlaminations within a container of predetermined configuration;
50 Compacting the microlaminations into a mag-netizable compact; and 6. Annealing the compact in a deoxidizing atmos-phere to relieve stresses.
In accordance with this invention the foregoing method is applicable to so-called "silicon steels" and preferably "oriented silicon steels". me silicon steels operable for this invention comprise iron alloys having a silicon content of from about 2.5% to about 3.5~, a carbon content of up to 0.01%, manganese from about 0.01% to 0.2%, and sulfur in the amount of 0.005% to 0.05%. A more workable ~ .

~3 ~3 1 ~6 47,099 sillcon content is from about 2.75% to 3.3%, and a preferred sillcon content is about 3.2%.
The first step comprises the provlsion or cuttlng of mlcrolaminatlon particles involved in the subsequent steps of this method. Though a relatlvely wlde range of particle sizes and thicknesses appears to be satisfactory, lt is preferred to have the microlaminations formed with the length ranglng between about 0.05 inch and about 0.150 inch, a width of between 0.01 inch and 0.02 inch, and a thickness of between 0.005 and about 0.014 inch. The mlcrolaminations are formed with a high speed rotary cutter operating on feedstock sheets in the same manner that "glitter" manufac-tured for use in Chrlstmas tree ornaments or helmets worn by motorcycle operators. However, as is pointed out, saw chips normally scrapped and produced during the fabrication of wound cores assembled on a mandrel and then sawed open are also sultable.
The second step comprlsing heating the micro-lamination partlcles involves relieving of the stresses in the steel lmparted during the formation of the microlamina-tlon. Essentlally, the heating ls a stress relief anneal in a non-oxidizing atmosphere so that the core structure of a hlgh density or packlng factor can be obtained that will ultlmately exhibit excellent magnetic characteristics.
Thus, it i8 preferred to anneal the microlaminations at a temperature within the range between about 700C and 1000C.
Thls temperature range is sufficient for relieving the stresses in the microlaminations induced during the first step of cuttlng the partlcles from feedstock. Although this step is recommended ln order to achieve the optimum magnetic ~ ~ 3 d ~ 47,099 characteristics it is not an essentlal step ln the process.
It wlll however recover optlmum ductillty so that greater density and packing factors are obtained.
The third step comprises coating of the micro-laminations with an electrlcally insulating material. The microlamlnations are preferably insulated from each other in order to provide the required core loss characteristics within the finished core. Magnesium methylate ls a preferred medium for providing an insulating coating on the lamina-B lo tions~because such coating is very thin and is sufficiently flexible to withstand the molding pressures. Though other insulating coatings may be employed, this coating provides sufficient lnterlaminar resistance to maintain the required core loss as well as other magnetic characteristics.
The fourth step consists of assembling the micro-lamlnatlons withln a container of a suitable configuration adapted for a predetermined form.
The fifth step comprises compacting the micro-laminations into a magnetizable compact at pressures of from 50,000 to 125,000 psi.
The final step comprises annealing the compact to primarily relieve the stresses incurred in the microlamina-tions during the preceding compacting step.
Particles of "oriented silicon steel" were investi-gated one of which was produced as rectangular particles by shearing and chopping of sheet stock, and the other comprised saw chip scrap generated as a result of core cutting operation during the manufacturing operation. Both types of particles had essentially the same chemical composition.
A reduction in core loss was observed when pressed 37~6 47 ogg saw chip scrap was annealed. A quantity of scrap was ball milled to separate the individual particles and remove the greater part of the saw kerf, stress relief annealed, coated with magnesium methylate, and uniaxially pressed at 125,000 psi. The 10 kG core loss was 4.3 W/lb. The compact was then annealed for 4 hours at 925C in a dry hydrogen atmos-phere and retested. The 10kG core loss was reduced to 1.7 W/lb.
Sim~larly, microlaminations were produced having dimensions of 0.060 x 0.010 x 0~011 inch, and obtained from a 4 inch wide strip of oriented iron-silicon stock material.
The particles were stress relief annealed for 1 hour at 800C in a dry hydrogen atmosphere, coated with magnesium methylate, and uniaxially pressed into a test ring core at a pressure of 125,000 psi. The 10 kG core 105s 0~ the as-pressed core was about 3.1 W/lb. The core was subsequently annealed for 4 hours at 925C in a dry hydrogen atmosphere after which the 10 kG core loss was reduced to approximately 1 W/lb.
The foregoing results made in the pressed and annealed cores produced from iron-silicon alloys are opposite to what had been previously observed when standard low carbon steel microlamination compacts were annealed. More particularly, the following examples are illustrative of the invention:
EX~LE 1 Several hundred pounds of oriented iron-silicon saw chips, which result from the core cutting operation, were received and subsequently degreased. Large strips of the scrap (greater than 1 inch in length) and miscellaneous . . ~ .

~ J~6 47,099 non-magnetlc debris were removed from the batch and four 5-pound (2300 grams) sample lots were taken. Processlng of the four lots conslsted of a combination of ball-milllng for two hours to break up the nested saw chips and remove the rough edges followed by stress relief annealing (1 hour at 800C in dry hydrogen and furnace cooled). After processing, all four lots were coated with magnesium methylate to insu-late the individual particles. Twenty-four magnetlc test rings, 6 from each lot, were then unlaxially pressed at a pressure of 125,000 psi (862 MPa). The dle wall was cleaned and coated with a zinc stearate lubricant before each pressing.
The test rings, which measured 25.4 mm ID, 4.45 mm OD, and approximately 12 mm in height, were then annealed at various temperatures for four hours in a dry hydrogen atmos-phere and furnace cooled. More specifically, one ring from each sample lot was annealed at a temperature of 870C, 925C, 980C, 1040C, and 1095C. The remalnlng rlng from each lot was left in the as-pressed conditlon. The packing factor was determlned for all rlngs, and the ac and dc magnetlc characterlstlcs, core loss (Pc/lO kG), permeablllty (~/10 kG), coercive force (Hc), and remanence (Br) were measured.
EXAMPLE II
Previous work has shown that the magnetic charac-teristics, particularly permeabllity, are strongly dependent upon the density (packing factor or P.F.) of the microlamina-tion compact due to magnetostatic interactions between the particles as well as to the effect of compact density on the apparent saturation magnetization of the sample. Therefore, an attempt was made to improve the density of the Hipersil _g_ ~ ~f ~ L17, o g g scrap compacts by adding particle lubricants to the loose chips prior to pressing. Microlamlnation scrap particles, which had previously been ball~milled, stress-rellef annealed and coated, were blended with various fractions (O, 1/8, V4, l/2, 3/4, and 1 percent by weight) of zlnc stearate lubricant and unlaxially pressed at pressures of 80,000, 100,000, and 125,000 psi (552, 689, and 862 MPa). After pressing, the rings were baked out for one hour at 425c to drive off the stearate, then annealed for 1 hour at 925c in dry hydrogen, furnace cooled, wound, and tested.
EXAMPLE III
-Sheet samples of fully-processed ll-mil (0.28 mm) thick oriented silicon steel were cut into strips 4 in. (100 mm) wide and approximately 4 ft. (91 cm) in length, slit, and chopped into microlamination particles of dimensions 0.060 x 0.010 x 0.011 inch. A 90-gram sample of this material was stress relief annealed for 1 hour at 800c in dry hydrogen, furnace cooled, coated with magnesium methylate, and uniaxially pressed into a test ring at 125,000 psi (862 MPa). The ring was annealed, (4 hours, 925c in dry hydrogen, furnace cooled). The dependence of core loss on magnetizing frequency of the annealed ring was investigated by measuring the ac magnetic characteristics at frequencies of 30, 40, 60, 200, 400, 100, and 2000 Hz.
RESULTS AND DISCUSSION
Oriented Silicon Steel Scrap The effect of pre-annealing, ball-milling, and annealing treatments on the dc and 10 kG ac magnetic char-acteristics of the pressed oriented silicon steel scrap compacts is shown in Table I. The magnetic characteristics r~ 7 ~ O g g of the as-pressed test samples compacted from unannealed scrap chips are significantly poorer than those of the stress relief annealed and pressed particles. For example, the core loss of the as-pressed rings compacted from unannealed chips is approximately double the loss of the rings compacted from annealed particles. This large difference in core loss is due to the highly strain-hardened condition of the un-annealed chips. Although strain hardening, due to the saw cutting operation, will decrease the density of the compact at any fixed pressing pressure and thus increase the core loss, the intrinsic magnetic characteristics of the scrap chips is usually impaired to a great degree by strain harden-ing. Thus, these two interrelated effects result in the poor magnetic quality of the compacts pressed from unannealed saw chips.

>~r~ 47,ogg TABLE I
Effect of Processing and Annealing On the Magnetic Properties of Hipersil Scrap Rlng Cores (125,000 p6i Pressing Pressure) Process jn~ Treatmer t Prior T Compa,c,tion Post P.F. Pc/lO kG ~/10 kG c ~ r Treat~ent (~) _ L_ (W/lb~ (e? ~ (kG
Ball Mill Plus Stress Relief Anneal . .. _ I . _ _ i __ _ None 90 4.3 145 3.8 1.6 4 hrs- 870 C ¦ B9 2.0 187 0.6 0.7 4 hrs- 925 C 1 89 1.7 187 0.6 0.6 4 hrs- 980 C 1 89 3.4 174 0.6 0.3 4 hrs-1040 C 89 2.8 170 0.7 0.5 4 hrs-1095 C ¦ 89 ¦ 3.1 ¦1720.7 0.9 Stress Relief Anneal Only None 89 I 4.1 I125¦¦ 4 .1¦ 1. 6 4 hrs- B70 C 89 1.9 181¦l 0.6¦¦ 0.5 4 hrs- 925 C 89 1.9 172~ 0.7 ~ 0.7 4 hrs- 980 C 89 3.6 16B¦¦ 0.4¦¦ 0.9 4 hrs-1040 C 89 3.0 165¦¦ 0-~ ~ O.S
4 hrs-1095 C ¦ 89 ¦ 3.7 1159~ 0.6¦ 0.7 Ball Mill Only None ~ 85 ~ 8.7 ~99~ 7.2~ 2.2 4 hrs- 870 C ~ 83 h 2.1 11 116 li 0.9 ¦¦ 0.5 4 hrs- 925 C ~ 84 ~ 2.0 ~ 120 11 0.4 ll 0.7 4 hrs- 980 C ¦¦ 88 ¦l 3.2 ~ 107 1l O-B ¦¦ 0-5 4 hrs-1040 C ll B4 ¦¦ 3.1 11 105 11 0-4 ~¦ 0.5 4 hrs-1095 C ~ 84 11 3.7 ¦¦ 105 ¦¦ 0.6 ~ 0.6 No Pre-Processing 'I 11 None ~ 84 ¦¦ 8 .8¦¦ 92 ¦¦ 7-0 ~ 1.3 4 hrs- 870 C 11 84 ~ 2.2 ~ 112 9 0.6 ¦¦ 0.5 4 hrs- 925 C ~ 85 11 2.0 ~ 103 ~ 0.6 ~ 0.5 4 hrs- 980 C ~ 84 ~ 3.5 ~ 96 ~ 0.4 ~ 0.7 4 hrs-1040 C ~ 84 ll 3.6 ~ 94 ~ 0.4 ~ 0.5 4 hrs-1095 C ¦ 84 ¦ 3.9 ~ 98 ~ 0.6 ~0.6 P.F. - Packing Factor (percent of theoretical density).
Hc and Br determined from an applied field of 100 Oe.

~ t.t~,~3~ 7,099 The operation of ball-milling to break up the nexted saw chips appears to have little effect on the overall magnetic quality of the pressed rlng cores (Table I) although subsequent work has shown that ball-milling of the scrap chips significantly increases the loose packing density of the chips (an increase of approximately 30%) and insures a more uniform die fill and more consistent pressed parts.
Therefore, ball-milling or similar processing, though not essential, is a recommended step for the optimization of the magnetic quality of a pressed compact.
The effect of annealing of the pressed scrap compacts is also illustrated in Table I. The magnetic characteristics, particularly core loss, are dramatically improved due to the relief of reintroduced residual stresses in the sample. This improvement is completely opposite to what is observed when standard low carbon steel microlamina-tions are annealed after compaction. When standard micro-lamination compacts, coated with magnesium methylate insula-tion, are annealed for any length of time and furnace cooled, the core loss increases 2-lO times above the loss measured on the test ring prior to the anneal. This increase in loss is is postulated to be due to a breakdown of the insulation at points of particle contact in the compact. As a result, although hysteresis is reduced due to the reduction of residual stresses, increased eddy currents are generated in the sample which cause a significant increase in the total observed core loss. This increase in eddy current loss is significantly greater than the reduction of hysteresis loss and thus results in a significant increase in the total losses of the sample.

~ ~J 3 ~ 47,099 The variation of core loss and permeabllity with post-annealing temperature (Table I) indicates that an apparent optimum in magnetic properties is developed after the 925C anneal which optlmum is apparently independent of prior processing. A comparison of the magnetic character-istics of the annealed rings indicates that the stress relief anneal of the particles prior to the pressing operation is necessary if optimum properties, particularly permeability, are to be achieved. This is related to the higher density of the pre-annealed compacts and the effect of thls density on the measured magnetic characteristics of the pressed rings (Fig. 1).
The effect of pressing pressure and the addition of a zinc stearate particle lubricant on the density of pressed Hipersil scrap ring cores is shown in Fig. 1, and the variation in magnetic quality presented in Table II.
Fig. 1 shows that the addition of a very small percentage of zinc stearate significantly alters the density of the pressed compacts, with the greatest benefit occurring with the addition of 1/8 to 1/4% lubricant. The relative improve-ment of the density with lubrication decreases with increasing pressing pressure. The addition of 1/8% stearate to a compact pressed at 80,000 psi (552 MPa) increases the packing factor from 82 to 86.5% or a change of 4.5 percentage points, whereas 1/8% stearate addition to a compact pressed at 125,000 psi (862 MPa) resulted in an increase in the density of only 2.3 percentage points. As the percentage of stearate is increased above the 1/8 to 1/4% optimum, the compact density begins to decrease because the stearate begins to consume an ever lncreasing volume in the compact, thus decreasing the overall sample density.

~ J~ 47,099 TABLE II
Effect of Zinc Stearate Additions On the Magnetic Characteristics of Hipersil Scrap Chipst tPost-Annealed 1 hr at 925 C) Zlnc Stearate P.F. PC/10 kG ~/10 kG ¦- Hc (%~ _ _(X) (W/lb) __ _ ~ O=e) Pressed At 80,000 psi I_ 0 82.01.74 1 90 1 0.8 0.4 1/8 1 86.61.60 1 132 1 0.8 0.4 1/4 l 86.31.56 1 128 1 0.7 0.5 1/2 1 85.51.56 1 123 0.6 0.4 3/4 1 85.21.56 1 122 0.7 0.5 1 83.81.53 109 0.7 0.4 Pressed At 100,000 psi _ _ . . ..... . _ .. _ . _ _ 0 1 86.2 1 1.64 125 ll 0.8 0.5 1/8 1 88.8 ~ 3 155 ~l 0.7 0.6 1/4 88.9 1 1.52 146 1 0.5 0.6 1/2 88.1 ~ 1.50 138 1 0.8 0.5 3/4 86.8 ! 1. 55 129 1 0.6 0.5 1 86.1 1 1.59 108 1 0.7 0.4 . _ _ ~! _ _ Pressed At 125,000 psi il 88.9 jl 1.51 ll 157 l, 0.9 0.6 1/8 ll 91.0 ,, 1.50 184 1'1 0.9 0.6 1/4 !¦ 91.3 ;l 1.46 l 171 j, 0.9 0.6 1/2 l 89.3 !l 1.46 1 146 1l 0.7 0.4 3/4 1 88.2 ll 1.47 134 '1 0.8 0.4 1 , 88.l !1 1 47 ~ 134 il 0.8 1 0.4 _ =_ ,~

tPrior to blending with stearate the scrap was ballmilled, stress-relief annealed, and coated.

~ y~ 47,099 The magnetic characteristics of the compacts pressed with the addition of the zinc stearate and subse-quently stress relief annealed for 1 hour at 925C are shown in Table II. The core loss, as well as the permeability, is improved as the density increases, regardless of whether the lncrease in density was achieved by increasing the pressing pressure or by the addition of a particle lubricant. Although the relationship between the density and the core loss is not particularly significant, the change in permeability with increased density is. This relationship is shown in Fig. 2.
Oriented Silicon Steel Microlaminations The magnetic characteristics of oriented silicon steel microlaminations produced by slitting and chopping which were thereafter annealed, coated, pressed lnto a test ring core, and annealed for four hours at 925C, are signi-ficantly better than those of the cores pressed from the oriented silicon steel saw chip scrap which were processed in a similar manner. The 10 kG loss and permeability of the 20 microlamination compact were 1.0 W/lb (0.45 W/kg) and 211 respectively, compared to 1.7 W/lb (0.77 W/kg) and 187 measured on the best scrap test ring (Table I).
A comparison of the loss characteristics of the silicon steel microlamination test ring with the loss behavior of a standard low carbon steel microlamination compact is shown in Fig. 3. The 60 Hz loss characteristics indicate the loss behavior of the silicon compact is significantly better, at all levels of induction, than the standard low carbon steel microlamination sample. Also, the frequency dependence of the losses for both samples, Fig. 3, indicates ~ 9~ 47,099 the sillcon steel sample is slgni.flcantly better over the entire range of frequencies and lnductlons lnvestlgated.
However, it should be noted from the slopes of the loss-frequency curves, that at frequencies substantially higher than those investigated, the core loss of the standard low carbon steel microlaminatlon compact will be superior to the silicon steel sample, due to a lower eddy current loss component in the low carbon steel compact.

Claims (12)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A method of making pressed magnetic core components having improved magnetic characteristics ( a low core loss property) for use in electrical apparatus, comprising the steps of (a) providing silicon-iron alloy sheet stock having a thin coating of silicate glass on two opposite sides thereof and having a silicon content of from about 2.5% to 3.5%, a carbon content of up to 0.01%, manganese of from about 0.01%
to 0.2%, and sulfur in the amount of from 0.005% to 0.25%, (b) forming microlaminations from the silicon-iron alloy sheet stock, which microlaminations comprise a coating of silicate glass on two opposite sides thereof, (c) coating the microlaminations with a dielectric material, (d) assembling the microlaminates within a mold of predetermined configuration, (e) compacting the microlaminates into a magnetizable compact, and (f) annealing the compacted microlaminates to relieve stresses.

2. The method of claim 1 in which the silicon content is from 2.75% to 3.3%.

3. The method of claim 2 in which the silicon content is about 3.2%.

4. The method of claim 1 in which the alloy stock is an oriented silicon steel.

5. The method of claim 1 in which the dielectric material is magnesium methylate.

6. The method of claim 1 in which the microlamina-tions are annealed at step (f) at a temperature between about 900°C and about 950°C.

7. The method of claim 6 in which the temperature is between about 910°C and about 940°C at step (d).

8. The method of claim 7 in which the anneal temperature is about 925°C.

9. The method of claim 8 in which the anneal time is about 4 hours.

10. A method of producing pressed magnetic core components having improved magnetic characteristics (a low core loss property) for use in electrical apparatus, comprising the steps of (a) providing silicon-iron alloy sheet stock having a thin coating of silicate glass on two opposite sides thereof, (b) forming particles of predetermined size and shape from said silicon iron alloy having a silicon content of from about 2.5% to about 3.5%, a carbon content of up to 0.01%, manganese of from about 0.01% to 0.2%, and sulfur in the amount of from 0.005% to 0.25%, (c) annealing the particles in a deoxidizing atmosphere, (d) coating the particles with a dielectric material, (e) assembling the particles within a container of predetermined configurations, (f) compacting the particles into a magnetizable compact, and (g) annealing the compact to relieve stresses.

11. The method of claim 10 in which dielectric material comprises magnesium methylate.

12. The method of claim 11 in which the annealing step (f) occurs for about 4 hours at about 925°C.