US2890143A - Annealing process for minimizing core loss in hot-rolled electric sheet segments - Google Patents

Description

PROCESS FDR MINMIZING C RE Jakob BingeLBerlin, Germany No Drawing. Application January 11*, 1956 Sel'ialNm 558,604

' 1 "Claim. (Cl. 148-112) l 0... LOSS IN HOT-ROLLED ELECTRIC SHEET SEG- United States Patent C) ,gsilicon contents. 1.5 %-3.5'% Si).,.magnetic and mecl 'apical -properties to be retained, which are not simultanequslypresentin high silicon sheetsw Very: lowlosses are achieved with sheets, annealed in this way, and enhanced magnetization, Without the necessity of'starting with a. hard and brittlematerial withv reduced magnetization. The process does not demand the use of high grade lelectric furnace smelted' steel; sheets of cheap Siemens Martin steel are used. No use is made of grain orientation by cold rolling and. thus low core losses and high magnctizationtinall directions of the sheet are achieved, for which reason the invention is particularly suitable for the. treatment of the core material in generators and motors. As the new process does not depend on any metallurgical purification of the core material by quick flowing hydrogen withthe objectof decarburisation etc.,

it can be carried. out in: a static or creeping protective gas, that is in the simplest way under the purest conditions. a invention-isbased on new discoveries in the, field of metal physics.

I Jynamo sheets with'silicon contentsabove 2% are at present stage of'technics annealed at about 800 C., for one thing to eliminate the distortion hardenings arising during processing, for another to convert the released carbon into graphite. The simplest method for this box annealing isv in air. Box annealing at higher temperatures is deprecated at the present stage of'technics, for at 1000 C. a deteriorationin thelossfigures and mechanical properties becomes noticeable, which is known as over -ozoneattire...

With the present invention electric sheet segments, part c l y th se nt o yn m e ec r machin can under 'certain conditions be annealed at much higher temperatures (1100'-'-1250 C.) with the object of reducing core loss and enhancing magnetization.

It is, of course, known that transformer sheets are annealed at above 1100 C. in a high vacuum, though improvements of quality are only achieved with a vacuum superior to 10- Torr. in consequence of a crystal purification through pumping out the released and reaction gases from the sheets. This method of increasing the initial permeability is unsuitable for dynamo sheets, for in the vacuum during the high-temperature annealing a brittleness arises in the sheets insupportable for machines, which can not be removed, and again vacuum annealing is not a large-scale technical process.

In connection with the idea of the invention on the other hand it was ascertained on hot-rolled, normally annealed dynamo sheets with 2.4% silicon content after five hours re-annealing at 1200* C. in argon of 1 atmosphere pressure that the hysteresis loss had sunk from 1.8

Patented June 9, 1959 .wJkg. (liflwattsikilogramme) to 1.2=w./kg. The siiniil= taneous metallographic test of the annealed material showed, besides a slight coarsening of the grain, the precipitation of. large perlite islands, which in the micro- .structure of the original sheet could not be discefnedeven in -fold' magnification.

The connection between the reduction of. hysteresis loss and the appearance of the perliteislands can only be interpreted as being the result ofthe originally highly dispersed cementite having collected during high-temperature annealing and during the cooling stage precipita-ting itself as coarse perlite. This. provides a new standpoint insofar as according to the foreign body theory agglomerations are less harmful magnetically than highly dispersed distributions and so reduce coercive force and with it hysteresis loss. If following the above mentioned theory the coercive force H is represented in the form in which C is a constant and is the relation of the Blochwall""6 in the Weiss magnetic elementary area to the. size of, particle d, it will be. seen that with increasing coagulation of the foreign bodies the coercive force rapidly sinks.

In contrast to the usual view held by metallurgists, who regard elimination of carbon and other foreign. elements by chemical reactions from the sheetas the sole prerequisite for reducing hysteresis loss, there is shown here the effect of a collective precipitation at the grain bouna r g about. y dif us on du i g a eng hy an: neal, at 1200 C.

A h ame time it as Pos i le t pr v that, the silicon content within narrow limits plays a considerable pa t in prec p i n o ar as an nneal at he m temperature on low-silicon dynamo sheets produces a much smaller result with regard to lowering hysteresis loss than .with high-silicon sheets. As it is knownthat l c n p m t s cry ta za on and. so ac elerates the d splac men oi. th o eig e e ent .t th raln o m arics, theenhanced. nfl c of. h v s l con. on. collecti P c p i n n be. expla n d A fu n proc sse r a c w th. ing, emp r ur ann a ing. empera ure f. 2,5 C. are rec mmen e for. lowilicon he s... for e y h ghli on:- shcets 1100 C. so. on dyna o sheet with. .2%. silicon. c nent it Was. po sible to achieve a hysteresislossofOfZ w.-/ kg. atlZOOlC. after 1. hours... during. whi h t me monoc y tals of 1.0 di met r. ha fonnc lin h she As a sult. of the anneali g. h mago t 'zafion l o increases at all fi'eld strengths by 400 gausses.

The process can also be applied on segments which have been cut out from unannealed panels, so long as the panels have been given an absolutely even surface by finishing.

As high-temperature annealing in a slowly flowing gas has the same effect as in a stagnant protective gas, it can be used as a technical process, for the speed of flow can be reduced at Will so as to substitute a creeping gas flow for the stagnant gas.

In the light of the invention it is suggested that seg ments of hot-rolled dynamo electric sheets should be annealed in slowly flowing, technically pure argon, hydrogen, or nitrogen at temperatures between 1100 C. and 1250 C. for several hours.

The slow flow is a necessary condition, for otherwise undesirable foreign gases such as oxygen and water vapour would be drawn into the container and oxidation ,high speed of flow between ferrosilicon alloys in chemically pure hydrogenywith 1100 C. and 1400" C. achieved high initial permeabilities, which call for absolute purity of the material. It is known that decarburization in electric sheets at high temperatures can only be achieved in quickly flowing hydrogen, for the reaction at a high temperature only becomes effective through constant disturbance of the equilibrium, whereas with slower flow the equilibrium is shifted far to the left and no decarburization takes place.

As a further result of the invention oxidation of the sheets over a short period by oxygenous nitrogen can after cooling to 800 C. be of particular importance, in that by a chemical attack it again neutralizes the brittleness of the sheets.

The practical operation of the process consists appropriately in arranging the sheet segments separated by Al O -powder, in several stacks in an iron container welded or riveted hermetically; an inlet and an outlet pipe are welded to the container, through which the gas flows. The gases developed during heating up pass out through the outlet, while the inlet is closed. From 900 C. onward intermittent cleansing with argon is carried out for a short period, which can cease at 1200 C. when the container has been sealed. To counteract warping of the sheet segments it is advisable to use weight plates suitably arranged in the stacks. A further artifice consists in applying a directed flow, driving the desired gas on to every point of the surface inside the sheet pile. for heating; it can, alternately with cooler hoods, be propped over the various containers on ceramic pillars, whereby a continuous annealing program can be maintained.

It is a matter of course that the annealing container can be used repeatedly when surrounded by a cheap protective gas and kept free from scale.

There is today in electro-mechanical engineering a demand for dynamo sheets with extremely low core loss. The direct way to achieve this is by re-annealing the stamped machine segments at 1100" to 1250 C.,' which not only eliminates cut-hardening but at the same time has an improving elfect on the crystal structure of the iron. In contrast to transformer cores, stamping reintroduces into the deeply incised machine segment with long teeth and wide guide holes, greater tensions, which as cut-hardening again enhances hysteresis loss. At the edges of the cut the coercive force is greatly increased, on the other hand the mean tooth induction is very high, so

Finally the hood furnace has proved useful that in segments with narrow teeth a rise in total core loss of the segment of 15%, in those with 'widetceth of 10% is calculated. For this reason it has already been suggested subjecting the stamped segments to a stress-relieving anneal at 700 C.; this procedure is, however, not economical enough and for that reason alone has never been introduced.

According to the invention, the total reduction in core loss is 45%. I i v This way of re-annealing stamped segments at very high temperature could only be proposed once it had been proved that the expansion due to heat is reversible and also that no appreciablechanges in length remain as a result of the precipitations caused and of the equalization of tension brought about. In practice accuracy of size of '-0.01 mm. in tooth breadth and '-0.3 mm. in segment length is called for, for after assembly a smooth slot surface must essentially be present to take the conductor.

With the aid of microscope micrometers with an accuracy to 1 micron, it was established both on 200 mm. check gauges and on segments, that annealing at 1200" C. causes a shrinkage of only 1.5-10- as long as the panel has been normalized in the rolling mill at 800 C., and a shrinkage of 3 -10- without normalizing having been done. This proves that high-temperature annealing of segments does not influence the accuracy in size-of machine construction and the original dimensions are retained in the slot within limits of '-0.01 mm.

I claim:

A heat treatment for stacks of hot-rolled silicon steel electric sheets for generators to produce segments with very low hysteresis losses and good mechanical properties, which treatment comprises annealing ready stamped segments with a silicon content of 1.5 %3.5% in an in dustrial protective gas between 1100 and 1250" C. for several hours, and then cooling the stack to about 800 C., and then oxidizing the segments for a short time in a stream of strongly oxidizing gas.

References Cited in the file of this patent UNITED STATES PATENTS 7 1,569,355 Cole Jan. 12, 1926 2,158,065 Cole et al May 16, 1939 2,209,684 Crafts July 30, .1954 2,599,340

Littman et a1. -Q June 3, 1952 OTHER REFERENCES The Electric Journal, vol. 21, pages 55 to 61. Published in 1924 at Pittsburgh, Pa.

Stahl and Eisen: Vol. 48, pp. 484 and 485. Publ. in 1928. i y

Alloys of Iron and Silicon, 1st edition, pp. 355 .and 356. Edited by Greiner, Marsh and Stoughton. Published in 1933 by the McGraw-Hill Book Co., New'York,

Ferromagnetism, pp. 92 and 93. Edited by Bozorth. Published in 1951 by D. Van Nostrand Co.