US 2977263 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
March 28, 1961 A. HARENDz'A-HARINXMA 2,977,263
MAGNETIC coREs AND METHODS 0E MAKING TEE sAME Filed Dee. s, 1959 MAGNETIC coREs AND METHODS 0F MAKING-f THE SAME Alfred Harendza-Hanxrna, Chicago, Ill., assig'nor to Western Electric Company, Incorporated, New York, l N.Y., a corporation of New York Filed nec. s, 1959, ser. No. 857,104
4 claims. (ci. 14s-104) The present invention relates generally to magnetic cores and methods of making the same, and more particularly to methods of treating magnetic cores manufactured in accordance with the general principles of A. F. Bandur Patent 2,105,070, dated January 11, 1938, so as to increase the permeability thereof and the resulting magnetic cores. The cores may be used for many purposes, but are designed primarily for use in loading coils employed in voice frequency circuits.
The general objectV of the invention is to providenew and improved magnetic cores, and methods of making the same.
According to the above-noted Bandur patentgmagnetic cores are produced by coating finely divided magnetic particles with an insulating composition consisting of a refractory metal silicate, an alkali metal silicate and magnesium hydroxide. The insulated particles are then compressed into a core and the core is tired at a temperature of 1000 F. to 1300 F.
For many reasons, it is desirable to increase the permeability rating of the cores as much as possible to the extent that this can be done without any substantial ad-y verse effect on the core loss. For example, by increasing the permeability of the cores, it is possible to utilize cores which are smaller in size to accomplish a given result, thereby resulting in a saving both in space and in the materials used in the cores. Of particular importance from an economic standpoint is the saving of nickel used as the predominant constituent of the magnetic particles. In
the alternative, the same size core having an increased permeability may be wound with fewer turns of wire to obtain the same transformer action.
Accordingly, a more speciiic object of the invention is to provide improvements in the cores and methods of making them as generally disclosed in the Bandur patent, which improvements are directed primarily to producing cores having increased permeability without any substantial increase in the core loss or change in any other important property.
The foregoing and other objects are accomplished, according to certain features of the invention, by wetting a tired core produced generally in accordance with the process of the Bandur patent with water, and then retiring the core at a temperature of 1000 to l300 F. With these additional steps, the permeability of the core is increased without any substantial adverse effect on the core loss.
The invention is likewise directed to the novel magnetic core produced in accordance with the above-described method, as a new article of manufacture.` Preferably, the red core is soaked in boiling water for a period of about 30 to 60 minutes and is then retired in a hydroger1- containing atmosphere at a temperature of about l200 F. If it is desired to increase the permeability still further, the soaking and retiring steps may be lrepeated a second time.
Other objects, advantages and features of the invention will appear from the following detailed description of ICC specific embodiments and examples thereof, when taken in conjunction with the accompanying drawing. In the drawing, the single ligure illustratesgraphically the eects of treatment according to the invention on the permeability and core loss at 1800 cycles per second.
THE PRIOR PROCESS While various finely divided magnetic particles may be utilized in the practice of the invention, it is preferred to use lan embrittled metal alloy selected from the group of nickel and iron alloys known as Permalloys Of particular interest is a molybdenum-containing Permalloy consisting essentially of 4about 82% nickel, 16% iron and 2% molybdenum.
Such an yalloy may be treated and comminuted according to the general principles enunciated in C. P. Beath et al. Patent 1,669,649, dated May 15, 1928. According to that patent, the metallic constituents of the alloy are melted together and are oxidized in the molten state to embrittle the alloy. This treatment produces `a line crystalline structure in the solidified alloy that facilitates reduction to a ne powder by conventional grinding and pulverizing techniques.
The resultant magnetic powder is sieved through a 120 mesh screen, and any oversize particles are recycled. The majority of the particles are in the range of about 200 to 300 mesh. The sieved dust is next subjected to an annealing heat treatment at about 1500 to 1600" F. to remove strains introduced into the magnetic material by the grinding operation.
The particles are then given an insulating coating comprising a refractory metal silicate, magnesium hydroxide and an alkali metal silicate. Preferably, the coating is made up of talc (a hydrated magnesium silicate), 1 part by Weight; sodium silicate,` 0.3 to 0.4 part by weight; and magnesium hydroxide, 0.06 to 0.08 part by weight. The
optimum proportions embodied in the commercial process are talc, 1 part by weight; sodium silicate, 0.35 part by weight; and magnesium hydroxide, 0.07 part by weight. The sodium silicate should have a high silicate to soda ratio, preferably about 1.6 to 3.0 parts silicate to one part soda.y Other refractory metal silicates, such as aluminum silicate, may be used as well as other alkali metal silicates, such as potassium silicate.
The coating composition is preferably applied in at least three stages from an aqueous suspension of the constituents, followed by heating to dryness at a temperature of about 270 to 300 F. after each stage as is described more fully in the Bandur patent.
After the particles have been insulated, they are compressed into a core of a suitable shape, suchV as a ring, by molding at a pressure in the neighborhood of 160,000 pounds per square inch. During the application of this pressure, the magnetic particles are again subjected to stresses which impair the magnetic properties thereof. Therefore, the cores are again subjected to an annealing heat treatment by ring at a temperature between about 1000 and 1300 F. The cores are preferably'tired in a hydrogen-containing atmosphere at about 1200" F. During this heat treatment, the insulating material is fully cured.
According to theicommercial embodiment of the abovedescribed process, core rings are produced having an insulating coating las described above containing about 1.25 parts by weight of the insulating material to parts by weight of the magnetic powder. Such core rings have been found to exhibit an average magnetic permeability of 125 and an average core loss at 1800 cycles per second of 0.190 unit. The present manufacturing limits on such core rings have been a permeability of at least and a core lo'ss no higher than 0.240 unit,
3 THE IMPROVED METHOD According to the present invention, the cores formed generally in accordance with the prior process just described are treated by wetting them with water after the iiringstep and are then retired at a temperature of 1000 to 1300 F. With these additional steps, the permeability of the cor is increased without any substantial adverse effect on the core loss. Preferably, the cores are soaked in boiling water fora period of between 30 and 60 minutes and iare retired in a hydrogen-containing atmosphere at a temperature of about 1200 F.
While the initial permeability (prior to the water dip) may vary considerably depending on the process conditions, particularly the properties of the magnetic metal and amount of insulation used, the permeability is increased by the water dip and retiring steps in all cases. The amount of insulation is preferably between about 0.5 and 1.5 parts `by weight based on 100 parts by weight of the magnetic powder, and the process comprehends mixtures of lightly and heavily insulated powder. Over a wide range of process conditions, the improved method has been found effective to raise the permeability between about 18 Iand 24 units, with the average increase being aboutl units. By the same treatment, the core loss decreases slightly in most cases or remains about the same. The average treated core shows a decrease in core' loss of about 0.01 unit.
The soaking and reiiring steps may be repeated a second time to further increase the permeability. In this event, the rise in permeability varies between about 9 and 12 units, with the average increase being about 10 units. However, the core loss rises somewhat, from a value equal to the original to a value of about 0.02 unit higher. Subsequent dipping and retiring is effective to raise the permeability even further, by about 6 to 10 units, ybut the core loss also goes up so that further treatment is not practical in many cases for this reason.
Soaking in water without the rering step or retiring without the soaking step are both ineffective to produce any significant change in the permeability. Likewise, soaking in water after the pressing step but before the initial firing produces no substantial increase in the permeability.
Under the improved method, the wet core may be placed directly in the tiring oven at 1200 F. or, if desired, the core may iirst be dried or treated at a lower temperature. In each case, the permeability increase is substantially the same. While it is preferred to use boiling water in the process, soaking in cooler Vwater is effective to a lesser degree so that the time of soaking must be increased. As a general rule, the hotter the water and the longer the soaking time, the greater will be the increase in permeability.
The drawing illustrates graphically the average effect of the soaking time in boiling water on the magnetic properties, permeability and core loss, in a typical example. Curves A Iand B show respectively the change in permeability and core loss as a result of a first dip of the indicated time. Curve A illustrates that during the first minutes, approximately 90% of the maximum increase of 20 units in permeability has been achieved and that, by one hour, nearly all of the potential increase has ibeen achieved. The preferred time range is 30 to 60 minutes, with 45 minutes representing a good practical average for most applications. Curve B shows a decrease in core loss of 0.01 unit as a result of a dip of one hour followed by retiring. Curves C and D illustrate, respectively, a permeability increase of 11 additional units and an increase in core loss back to about the original value resultant from a second dip of one hour followed by rering. Similarly, curves E and F show that the permeability is increased Vby 7 additional units as a result of a third dip of one hour, while the core loss increases by 0.017 unit.
The invention is also directed to the magnetic cores produced in accordance with the improved process; that is, to the magnetic cores having enhanced magnetic properties as new articles of manufacture. Since the mechanism by which the water treatment and reanneal affect the properties of the cores is not fully understood, it is not possible to state precisely how the improved core rings differ from the prior core rings as to their physical or chemical structures. However, it is apparent that a new article of manufacture is produced by the method since at least one important physical property of the cores, the magnetic permeability, has been materially altered.
Example I According to one specific example of the invention, several core rings produced according to the commercial process described hereinabove were taken asa sample. These core rings were insulated with 1.25 parts by weight of the insulation as previously described, were compressed into core rings at a pressure of 160,000 pounds per square inch, and were tired in a hydrogen-containing atmosphere at a temperature of 1200 F. The average permeability of these core rings was 125 and the average core loss was 0.190 unit.
These tired cores were soaked in boiling water for a period of 45 minutes and the soaked cores were retired, without intermediate treatment, in a hydrogen-containing atmosphere at a temperature of 1200 F. The average permeability of the treated cores was 145, an increase of 20 units or 16%, while the average core loss was 0.176 unit, a decrease of 0.014 unit.
The rered cores were then soaked in boiling water for Van additional 45 minutes and reiired in a hydrogencontaining atmosphere at 1200 F. The average permeability of these cores was 155, a further increase ofI l0 units and Ia total increase of 24% over the original value. The average core loss increased back to the orig-V As a second example, a batch of 2000 off-specification cores producedy commercially were treated according to the improved process. These cores had permeabilities ranging from 103 to 110, and thus were defective as not meeting the minimum standard of 115 units. The cores were otherwise perfectly formed and had acceptable core loss ratings. The entire lot of defective cores was soaked in boiling water for one hour and retired in a hydrogencontaining atmosphere at 1200 F. The iinal permeabilities of the treated cores increased to 123 to 134 units, while the core loss values were still acceptable. Thus, the permeability of this entire batch of defective cores was raised suiiciently in one dipping and retiring operation to meet commercial standards.
Example III In another example, a mixture of insulated powders was used to provide a high permeability, and the core rings were treated with sodium aluminate after pressing and before firing to decrease the core loss. These methods are described more fully in my copending application, Serial No.V 857,087, tiled contemporaneously herewith.
According to the specific example, the insulated particles .were made up in two distinct batches: batch A containing 0.8 part by weight of the insulation per parts of the metal, and batch B containing the standard 1.25 parts by weight ofV insulation. These batches were otherwise produced according to the process described hereinbefore, A series of cores were pressed from a uniform mixture of 74% of the lightly insulated particles from batch A and 26% of the standard particles of batch B. Without further treatment, some of these cores were fired. Cores produced from the mixed powders according to conventional pressing and firing techniques exhibited an average permeability of about 165 and a core loss of 0.28 unit, which is not within the present manufacturing limit of 0.24 unit.
Several cores were similarly pressed from the mixed powders, but were soaked for one hour in a solution of sodium aluminate after pressing and before tiring. These cores were then fired, and the resultant permeability was 165 (no change) while the core loss was reduced to 0.176 unit, which is well within the tolerance limits.
A number of the core rings thus formed were soaked in boiling water for 45 minutes and then retired in a hydrogen-containing atmosphere at 1200 F. The permeability increased 22 units to 187, while the core loss was unchanged. A second water dip o-f 30 minutes followed by reliring increased the permeability to 198, while the core loss was elevated slightly to 0.205 unit.
The subject process may also be used in conjunction with the oil-'absorption process disclosed in my copending application Serial No. 857,087 or the oil-absorption process combined with the aluminate process described hereinbefore. In the oil-absorption process, the insulated particles are treated with a small amount of oil dissolved in a volatile solvent before the pressing operation. The oil-containing particles are then heated to vaporize at least a portion of the oil. The oil vapor is absorbed by and reacts with the sodium silicate to form a composite insulation.
The vforegoing examples illustrate the application of the invention 'to produce improved cores having increased permeability without any substantial change in the core loss. While various specific examples of the invention have been described in detail hereinabove, it will be obvi ous that various modifications may be made from the the core is fired at a temperature of 1000 to 1300 F.;
the improved method which comprises the additional steps of wetting the fired core with water, and then rering the core at a temperature of 1000 to 1300 F.
2. As a new article of manufacture, the magnetic core produced in accordance with the method recited in claim 1, said core having a substantially higher permeability than that resultant without the wetting and retiring steps. 3. The method in accordance with claim l, wherein the wetting step is accomplished by soaking the core in boiling water for a period of at least 30 minutes.
4. In a process of making a magnetic core wherein inely divided magnetic metal particles are given an insulating coating of between 0.5 and 1.5 parts by weight, based on parts by weight of the metal, of a composition consisting essentially of talc, 1 part by weight, sodium silicate, 0.3 to 0.4 part by weight, and magnesium hydroxide, 0.06 to 0.08 part by weight, wherein the insulated particles are then compressed into a core, and wherein the core is fired in a hydrogen-containing atmosphere at a temperature of about l200 F.; the improved method which comprises the additional steps of soaking the fired core in boiling water for a period of between 30 and 60 minutes, and then retiring the core in a hydrogen-containing atmosphere at a temperature of about 1200 F.
References Cited in the tile of this patent UNITED STATES PATENTS 2,105,070 Bandur Ian. 11, 1938