Iron core for transformer and transformer

Abstract

The present invention suppresses an increase in the building factor due to magnetic flux leakage in the stacking direction of an iron core by controlling variation of the insulation coating thickness of a grain-oriented electrical steel sheet having an insulation coating that is used for the iron core. This iron core for a transformer is formed by stacking a plurality of single layers composed of one or a plurality of grain-oriented electrical steel sheets having an insulation coating. Each of the single layers has at least one steel sheet discontinuity part extending in a direction crossing a direction in which a magnetic flux flows in the single layer. A standard deviation σ of the thickness of the insulation coating in a region sandwiched between the steel sheet discontinuity part of one single layer and the steel sheet discontinuity part of another single layer adjacent thereto in the stacking direction is 2 μm to 5 μm inclusive.

Description

Transformer cores and transformers The present invention relates to a transformer core and a transformer. Grain-oriented electrical steel sheets are used as the material for the iron cores of transformers. In these transformers, the heat loss (iron loss) that occurs when grain-oriented electrical steel sheets are magnetized with AC current affects the efficiency of the transformer, so development of grain-oriented electrical steel sheets with low iron loss is underway. Here, the iron loss of grain-oriented electrical steel sheets is mainly composed of hysteresis loss and eddy current loss. One method to improve hysteresis loss is to use the GOSS orientation (110).

[0001] Methods have been developed to highly orient the orientation in the rolling direction of the steel sheet, and to reduce impurities in the steel sheet. Methods for improving eddy current loss have also been developed, such as a method of increasing the electrical resistance of the steel sheet by adding Si, and a method of applying coating tension in the rolling direction of the steel sheet. However, when pursuing further reduction in iron loss of grain-oriented electrical steel sheets, these methods have limitations in terms of production. Therefore, magnetic domain refinement technology has been developed as a method to further reduce the iron loss of grain-oriented electrical steel sheets. Magnetic domain refinement technology is a method to refine the width of the 180° magnetic domain (main magnetic domain) formed along the rolling direction by introducing non-uniformity of magnetic flux into the steel sheet after finish annealing or baking of the tension coating. By refining the magnetic domain, it is possible to reduce the iron loss of grain-oriented electrical steel sheets, especially the eddy current loss. Various methods have been proposed to introduce non-uniformity of magnetic flux, such as forming grooves and introducing local distortion. For example, Patent Document 1 proposes a technique for introducing linear grooves having a width of 300 μm or less and a depth of 100 μm or less into the surface of a steel sheet. This technique improves the iron loss, which was 0.80 W / kg or more, to 0.70 W / kg or less. Patent Document 2 proposes a technique for irradiating a plasma flame in the sheet width direction of the steel sheet surface after secondary recrystallization to locally introduce thermal strain. The steel sheet obtained by this technique has a magnetic flux density (B ８ ) is 1.935T, and the iron loss (W １７／５０ ) was 0.680 W / kg. The method of introducing linear grooves as disclosed in Patent Document 1 is called heat-resistant magnetic domain refinement because the magnetic domain refinement effect does not disappear even when stress relief annealing is performed after core forming. On the other hand, the method of introducing thermal strain as disclosed in Patent Document 2 is called non-heat-resistant magnetic domain refinement because the effect of introducing thermal strain is lost due to stress relief annealing. The iron loss of grain-oriented electrical steel sheets is measured by sinusoidal excitation in the rolling direction, but because the magnetization behavior inside an actual transformer is more complex, it is known that the loss of the transformer increases more than the iron loss of the grain-oriented electrical steel sheets used. This rate of increase is called the building factor (= total transformer iron loss / iron loss of the grain-oriented electrical steel sheets used). Reducing the building factor is essential for manufacturing more efficient transformers. Here, the cores are roughly divided into two types, wound cores and stacked cores, depending on their structure. A typical example of a wound core is the wrap core type. A wrap core type wound core is manufactured by shearing grain-oriented electromagnetic steel sheet for each turn to obtain a single layer, or by stacking multiple sheets of such sheets to form a single layer, and then stacking (wrapping) multiple single layers and inserting them into a coil. In this case, with the wrap core type, the shear points of the steel sheet are periodically changed. When viewed from the side, it is called the wrap core type because the shear points appear to be wrapped for each layer. On the other hand, a stacked core is an iron core constructed by stacking steel sheets cut to a specified shape. Specifically, a plurality of pieces (diagonal bars) obtained by shearing grain-oriented electromagnetic steel sheets are combined in the same plane in a frame shape, or a plurality of such combinations are stacked to form a single layer, and the core is manufactured by stacking a plurality of such single layers. Therefore, in both wound and stacked cores, the shear ends of the grain-oriented electrical steel sheets in each of the above single layers are joined via an air gap. When the core is excited, magnetic flux transfer occurs in the lamination direction of the above single layers near such joints, which increases iron loss due to in-plane eddy current loss. This is known to be one of the causes of the increase in the building factor. In recent years, the demand for large-scale renewable energy farms such as wind and solar power has been increasing. The renewable energy generating areas and the energy demand areas are often far apart, and the demand for DC transmission is also increasing to reduce transmission losses. In DC transmission, transformers that operate at a higher frequency range (100 to 10,000 Hz) than the conventional commercial frequency (50 or 60 Hz) are used. In such transformers, the time change of the magnetic flux inside the iron core is large due to the high frequency, and the increase in the building factor due to the magnetic flux transfer is significant. Various methods for improving the building factor have been studied, and for example, Patent Document 3 proposes subjecting grain-oriented electrical steel sheets near the joints to a magnetic domain refinement treatment using strain. Patent Document 4 proposes inclining the joints in the winding direction to improve iron loss due to magnetic flux transfer. Japanese Patent Publication No. 6-022179 Japanese Patent Publication No. 7-192891 Japanese Patent Publication No. 2020-096100 Japanese Patent Publication No. 2005-150507 Although all of the above proposals are effective, they are proposals related to core design, and therefore have the problem of increasing manufacturing costs. For example, adding a new process for introducing strain or changing the shear direction increases manufacturing costs. The present invention has been made in consideration of the above circumstances, and aims to suppress an increase in the building factor due to magnetic flux transfer in the lamination direction of the iron core by controlling the thickness of the insulating coating of the insulating-coated grain-oriented electrical steel sheet used in the iron core. The inventors have conducted extensive research to solve the above problems. First, in order to investigate the degree of magnetic flux transfer at the joint, a search coil was used to investigate the amount of magnetic flux transfer at the joint. As a result, it was found that the amount of magnetic flux transfer differs depending on the measurement location even near the joint. The eddy current loss generated in the steel sheet is proportional to the square of the magnetic flux density. Therefore, if the amount of magnetic flux transfer varies within the sheet surface, the eddy current loss increases locally, and the iron loss at the joint increases significantly. The researchers conducted a more detailed investigation into this variation in magnetic flux transfer amount. As a result, they found that the variation in magnetic flux transfer amount was caused by the distance between the magnetic bodies in the lamination direction, i.e., the distance between the base steel sheets of the grain-oriented electrical steel sheet, not being constant. The surface of the grain-oriented electrical steel sheet contains a metal oxide layer formed on the surface of the base steel and a tensile coating applied on top of that. They newly discovered that the variation in the distance between the base steel sheets occurs because the shapes of these layers are not flat. The present invention has been made based on the above findings. That is, the gist and configuration of the present invention are as follows: (1) A transformer core made by stacking a plurality of single layers each made of one or more grain-oriented electrical steel sheets with an insulation coating, Each of the monolayers has at least one steel sheet discontinuity extending in a direction transverse to a direction of magnetic flux flow in the monolayer; A transformer core, wherein the standard deviation σ of the thickness of the insulating coating in a region sandwiched between a discontinuous portion of one single layer of steel plate and a discontinuous portion of the other single layer of steel plate adjacent in the stacking direction is 2 μm or more and 5 μm or less. (2) A transformer core according to (1), wherein the distance Δd between the steel plate discontinuity of one adjacent single layer and the steel plate discontinuity of the other adjacent single layer in the direction of magnetic flux flow is 1 mm or more and 20 mm or less. (3) A transformer core according to (1) or (2), which is a wound core obtained by winding the grain-oriented electromagnetic steel sheet in the rolling direction of the steel sheet. (4) The transformer core according to (3), wherein the wound core has a bent portion that is bent in the winding direction. (5) A transformer core according to (1) or (2) above, which is a stacked core formed by stacking a plurality of single layers each formed by combining a plurality of pieces of grain-oriented electromagnetic steel sheet in the same plane. (6) A transformer comprising the transformer core according to any one of (1) to (5). According to the present invention, in a transformer core using grain-oriented electrical steel sheets, it is possible to suppress magnetic flux transfer in the lamination direction and reduce the building factor, and therefore it is possible to provide a transformer core with low core loss. FIG. 1 is a schematic diagram showing an example of a three-phase three-legged wound core; FIG. 2 is a schematic diagram showing an example of a three-phase three-legged stacked core; FIG. 3 is a schematic diagram showing an example of a laminated structure of a core; FIG. 4 is a graph showing the effect of the standard deviation σ of the insulation coating thickness on the building factor (BF); FIG. 5 is a graph showing the effect of the spacing Δd of the joints between the end faces of grain-oriented electrical steel sheets on the building factor (BF); and FIG. 6 is a schematic diagram showing a C-section observation image of the insulation coating obtained by SEM. The experimental results that led to the completion of the present invention will now be described. (Experiment 1) A steel slab having the composition shown in Table 1 was used to manufacture a 0.23 mm thick grain-oriented electrical steel strip as a test material by a general manufacturing process. Single-sheet test pieces were cut out from the test material, and the core loss (W １７／５０ ) was measured. １７／５０ means the core loss when a single sheet test piece is subjected to AC magnetization at 1.7 T and 50 Hz in the longitudinal direction of the sample. Next, the above-mentioned test material was used to prepare a three-phase three-legged wound core as shown in the schematic diagram in FIG. 1. That is, the test material (steel strip) was slit to a width of 100 mm and wound in the rolling direction to form a single layer. A plurality of the single layers were stacked to a stacking thickness of 30 mm to prepare a three-phase three-legged core having a height of 300 mm and a width of 210 mm. Here, two types of wound cores were prepared: a type in which the core corners were wound in an arc shape (Trancocore), and a type in which the corners had two 45-degree bent parts (Unicore). The Trancocore and some Unicores were subjected to stress relief annealing (SRA) at 800° C. for 3 hours in an Ar atmosphere. Furthermore, using the above-mentioned test material, a three-phase three-legged core was fabricated as shown in the schematic diagram of Figure 2. That is, a diagonal bar with a width of 100 mm was cut from the test material, and a plurality of these diagonal bars were combined in the same plane in a frame shape to form a single layer. A plurality of these single layers were stacked to a stacking thickness of 30 mm to fabricate a three-phase three-legged core with a height of 420 mm and an overall width of 420 mm. In each core, at least one discontinuity in the steel sheet exists in each single layer, which extends in a direction intersecting with the flow of magnetic flux. In each single layer, the shear ends of the grain-oriented electrical steel sheets are joined via an air gap, i.e., a discontinuity in the steel sheet. Hereinafter, such a discontinuity in the steel sheet will also be referred to as a joint. Incidentally, in the wound core of FIG. 1, there are one or two joints in a single layer. In addition, in the stacked core of FIG. 2, there are six joints in a single layer. Furthermore, the junctions of the laminated single layers are shifted in a step-like manner in the direction of magnetic flux flow of the single layers so that they do not coincide with each other between adjacent single layers. This structure is called a step-lap structure. For example, FIG. 3 is a diagram showing a schematic structure of a six-layer portion of an iron core. The six layers of single layers 1a to 1f each have joints 2a to 2f. In the five stages of single layers 1a to 1e, the positions of the joints 2a to 2e are shifted by an interval Δd in the direction 3 of magnetic flux flow. In the structure shown in FIG. 3, the positions of the joints are shifted in the five stages of single layers 1a to 1e, while the positions of the joints 2f of single layer 1f and the joints 2a of single layer 1a in the direction 3 are the same. Therefore, this structure is defined as a step-lap structure in which the five stages of single layers 1a to 1e form one period. Note that FIG. 3 shows only six layers out of the many single layers stacked. In the portions not shown, the step-lap structure in which five stages form one period is repeated, similar to the above-mentioned single layers 1a to 1e. In the above-mentioned wound core and stacked core, a step-lap structure with a 7-step period was adopted. Also, the above-mentioned mutual offset (spacing) Δd of the joints between adjacent layers was set to 8 mm in both the wound core and stacked core. Next, using these iron cores, the transformer iron loss was measured when the magnetic flux density in the iron core leg portion was 1.7 T at a frequency of 50 Hz. In the above measurement, the no-load loss was measured using a wattmeter. The building factor (BF) was calculated from the obtained transformer iron loss and the material iron loss measured previously. After the above iron loss measurements, 10 C-section samples were cut out from the region of each iron core corresponding to the above Δd, and the base steel-insulating coating layer was observed with a SEM (scanning electron microscope). From the observation results, the insulating coating layer thickness was calculated using the method described below, and its standard deviation σ was calculated. Figure 4 is a graph plotting the BF values ​​obtained by the above measurements against the standard deviation σ of the insulation coating thickness. For all core configurations, a significant improvement in the building factor was confirmed when the standard deviation σ was in the range of 2 μm to 5 μm. This is thought to be because when the standard deviation σ exceeds 5 μm, the magnetic flux is concentrated in the thin parts of the insulation coating, increasing the in-plane eddy current loss and increasing the building factor. On the other hand, the building factor also increases in areas where the standard deviation σ is less than 2 μm. Although the detailed mechanism behind this phenomenon is not clear, the inventors speculate as follows: In other words, it is believed that the insulating coating layer has a similar thickness on almost all surfaces of the joint, which causes magnetic flux transfer to occur over the entire area of ​​the plate surface, increasing the area where eddy current loss occurs and increasing the building factor. Also, the improvement effect of the building factor was the largest in the Unicore without stress relief annealing (SRA), followed by the Unicore with stress relief annealing. On the other hand, the improvement effect was about the same in the Trancocore and the stacked core. That is, in the Trancocore and the stacked core, the permeability is equal in the core, and the magnetic flux passes through the shortest path and tends to concentrate on the inside of the core. Therefore, it is considered that the magnetic flux transfer concentrates on the inside, and the area where the magnetic flux transfer occurs becomes excessive on the inside, and the eddy current loss increases. On the other hand, in the Unicore without stress relief annealing, the magnetic permeability on the inside of the core decreases due to the distortion of the bent part, so the magnetic flux concentration on the inside is difficult, and the amount of magnetic flux transfer is homogenized in the lamination direction. As a result, it is considered that the influence of the variation in the thickness of the insulation coating is significantly manifested. Also, in the Unicore with stress relief annealing, the distortion of the bent part was not completely removed, so it is considered that the improvement amount was intermediate. From the above results, it was found that in the configuration of the present invention, if the standard deviation σ is 2 μm or more and 5 μm or less, the building factor can be improved more efficiently. More preferably, the standard deviation σ is 2.5 μm or more and 4.5 μm or less. In this specification, the "standard deviation σ" refers to the standard deviation of the thickness of the insulating coating in the region between the steel sheet discontinuity of one single layer and the steel sheet discontinuity of the other single layer adjacent in the stacking direction. (Experiment 2) Further, the above-mentioned interval Δd was examined in detail. That is, a steel slab having the composition shown in Table 1 was used to manufacture a cold-rolled steel strip of a grain-oriented electrical steel sheet as a test material by a general manufacturing process. The surface roughness in the width direction and the length direction of the obtained cold-rolled steel strip was measured using a laser meter. Next, the cold-rolled steel strip was subjected to decarburization annealing, application of an annealing separator, final annealing, and application and formation of a tension coating. Thereafter, the thickness of the forsterite coating and the thickness of the tension coating in each of the width direction and the length direction of the steel strip were measured by fluorescent X-ray analysis. In the measurement, the thickness (weight) of each coating was calculated using a calibration curve from the fluorescent X-ray intensity. The sum of the obtained thickness of the forsterite coating and the thickness of the tension coating was defined as the thickness of the insulating coating. The standard deviation σ of the insulating coating layer thickness was calculated from the distribution of the thickness of the insulating coating. The locations where σ was about 3 μm (3 μm ± 0.2 μm) were selected to be the steel plate discontinuous parts, and a three-phase three-leg unicore wound core with a stacking thickness of 30 mm, height of 300 mm, and width of 210 mm, and a three-phase three-leg stacked core with a stacking thickness of 30 mm, height of 420 mm, and overall width of 420 mm were fabricated. Note that when fabricating the cores, Δd was changed from 0.5 to 25 mm. The core structure was a step-wrap structure with a five-step period. In addition, samples were cut out from the above cold-rolled steel strip, and the core loss (W １７／５０ ) was measured. Next, using these iron cores, the transformer iron loss was measured when the magnetic flux density in the iron core legs was 1.7 T at a frequency of 50 Hz. In the measurement, the no-load loss was measured using a wattmeter. The building factor was calculated from the obtained transformer iron loss and the previously measured material iron loss. FIG. 5 is a graph plotting the measured building factor against Δd. As shown in FIG. 5, it was confirmed that the building factor was further improved when Δd was in the range of 1 mm or more and 20 mm or less. This is thought to be because when Δd was more than 20 mm, the area where eddy current loss occurred due to magnetic flux crossing increased, increasing the building factor. On the other hand, in the area where Δd was less than 1 mm, the area of ​​the magnetic flux crossing portion became too small, and the amount of magnetic flux crossing per unit area increased, increasing the building factor. From the above results, it was found that if Δd was 1 mm or more and 20 mm or less in the configuration of the present invention, the building factor could be improved more efficiently. It is more preferably 3 mm or more. It is also more preferably 15 mm or less. In the above experiments, one grain-oriented electrical steel sheet with an insulating coating was used as a single layer. However, even when a stack of multiple grain-oriented electrical steel sheets with an insulating coating was used as a single layer, similar trends to the results shown in Figures 4 and 5 were obtained. Next, a preferred embodiment of the present invention will be described in detail. However, the present invention is not limited to the configuration disclosed in the present embodiment, and various modifications are possible without departing from the spirit of the present invention. [Grain-oriented electrical steel sheet] In the present invention, the desired effect can be obtained by controlling the standard deviation of the thickness of the insulating coating, and therefore any grain-oriented electrical steel sheet can be used in the present invention. The method for producing the grain-oriented electrical steel sheet is not particularly limited, and the sheet can be produced by a general method. The composition of the steel material (slab) used for producing the grain-oriented electrical steel sheet is not particularly limited, but it is usually sufficient that the composition causes secondary recrystallization. In addition, when an inhibitor is used, for example, when an AlN-based inhibitor is used, a steel material containing Al and N may be used. Similarly, when an MnS / MnSe-based inhibitor is used, a steel material containing at least one of Se and S and Mn may be used. Of course, both inhibitors may be used in combination. When at least one of Al, N, S, and Se is added, the preferred contents of each element are as follows. Al: 0.010 to 0.065 mass%, N: 0.0050-0.0120mass%, S: 0.005 to 0.030 mass%, and Se:0.005~0.030mass% Furthermore, the present invention can also be applied to grain-oriented electrical steel sheets in which the contents of Al, N, S, and Se are limited and no inhibitor is used. In this case, the contents of Al, N, S, and Se are, respectively, Al: less than 0.010 mass%, N: less than 0.0050 mass%, S: less than 0.0050 mass%, and Se: less than 0.0050 mass% It is preferable to suppress the Suitable basic components and optional additional components of the steel material (steel slab) that can be used to manufacture the grain-oriented electrical steel sheet will now be described. C: 0.08mass% or less C can be added to improve the hot-rolled sheet structure. However, if the C content exceeds 0.08 mass%, it becomes difficult to decarburize the steel during the manufacturing process to 50 mass ppm or less, at which magnetic aging does not occur. Therefore, the C content is preferably 0.08 mass% or less. On the other hand, since secondary recrystallization occurs even in steel materials that do not contain C, the lower limit of the C content is not particularly limited and may be 0 mass%. Si:2.0~8.0mass% Silicon is an element that is effective in increasing the electrical resistance of steel and improving iron loss. However, if the content is less than 2.0 mass%, the improvement effect is not fully exerted, while if the content exceeds 8.0 mass%, the workability and threadability are significantly deteriorated and the magnetic flux density is also reduced. Therefore, the Si content is preferably in the range of 2.0 to 8.0 mass%. Mn: 0.005 to 1.0 mass% Mn is an element necessary for improving hot workability, but if the content is less than 0.005 mass%, it is difficult to obtain this effect sufficiently. On the other hand, if the content exceeds 1.0 mass%, the magnetic flux density deteriorates. Therefore, the Mn content is preferably in the range of 0.005 to 1.0 mass%. The composition of the steel material in one embodiment may include the above components, with the balance being Fe and unavoidable impurities. The above component composition may further contain at least one of the following optional additive components which are known to be effective in improving magnetic properties. Ni: 0.03 to 1.50 mass%, Sn: 0.01 to 1.50 mass%, Sb: 0.005 to 1.50 mass%, Cu: 0.03 to 3.0 mass%, P: 0.03 to 0.50 mass%, Mo: 0.005 to 0.10 mass% and Cr:0.03~1.50mass% Ni is an effective element for improving the hot-rolled sheet structure and enhancing the magnetic properties. However, if the content is less than 0.03 mass%, the contribution to the magnetic properties is small, while if the content exceeds 1.50 mass%, the secondary recrystallization becomes unstable and the magnetic properties deteriorate. Therefore, when Ni is added, the Ni content is preferably in the range of 0.03 to 1.50 mass%. In addition, Sn, Sb, Cu, P, Mo, and Cr are also elements that improve magnetic properties, but if the content is less than the lower limit, the effect is insufficient, and if the content exceeds the upper limit, the growth of secondary recrystallized grains is suppressed, resulting in deterioration of magnetic properties. Therefore, it is preferable that the content of these components is within the above range. A steel material (slab) having the above-mentioned composition is hot-rolled and then hot-rolled annealed. Then, one or more cold rollings are performed to finish the steel strip to the final thickness. The steel strip is then subjected to primary recrystallization annealing (decarburization annealing). Next, an annealing separator is applied to the surface of the steel strip, which is then wound into a coil and subjected to final annealing for secondary recrystallization. After the final annealing, the steel strip is subjected to flattening annealing to form a tension coating, resulting in a grain-oriented electrical steel strip. During the first recrystallization annealing, the steel strip is decarburized, and the C content is reduced. Furthermore, during the final annealing, the steel strip is purified, and as a result, Al, N, S, Se, etc. are reduced. As a result, in the finally obtained grain-oriented electrical steel sheet, the contents of C, Al, N, S, and Se are reduced to the impurity level. Therefore, in the composition of the grain-oriented electrical steel strip, C, Al, N, S, and Se can be considered as unavoidable impurities. Typically, the Al content is 0.01 mass% or less, and the contents of C, N, S, and Se are each 0.005 mass% or less. The contents of other components can be considered to be the same as those at the time of the steel slab. After the flattening annealing, a magnetic domain refining step may be further carried out in which thermal distortion is formed on the surface of the grain-oriented electrical steel sheet (steel strip) by irradiating the surface with an energy beam. - Insulating coating layer An insulating coating is formed on the surface of the grain-oriented electrical steel sheet. The insulating coating refers to a layer including either or both of a metal oxide layer (e.g., a forsterite coating) formed on a base steel and a tensile coating applied thereon and then baked with a tensile coating agent. The insulating coating is preferably formed on both sides of the grain-oriented electrical steel sheet. In the present invention, a transformer core is formed by stacking a plurality of single layers each made of one or more grain-oriented electrical steel sheets with an insulating coating. - Standard deviation of insulation coating thickness (variation in insulation coating thickness) As shown in the above experiment 1, if the standard deviation σ of the thickness of the insulating coating in the region between the discontinuous steel sheet of one single layer adjacent to the discontinuous steel sheet of the other single layer in the lamination direction is less than 2 μm, the building factor increases. Therefore, in the present invention, the standard deviation σ is set to 2 μm or more, preferably 2.5 μm or more. On the other hand, if the standard deviation σ is greater than 5 μm, the building factor increases. Therefore, in the present invention, the standard deviation σ is set to 5 μm or less, preferably 4.5 μm or less. In this way, by setting the standard deviation σ to 2 μm or more and 5 μm or less, it is possible to suppress an increase in the building factor due to magnetic flux transfer in the lamination direction of the iron core. The above-mentioned effects of the present invention can be sufficiently obtained even in commercial frequency bands such as 50 Hz and 60 Hz, but as will be shown in Example 2 described later, the higher the frequency, the greater the effect. The reason for this has not yet been clarified, but the inventors believe it to be as follows. That is, under high frequency conditions, the number of times that magnetic flux crossing occurs between layers increases, and the resulting change in magnetic flux density in the out-of-plane direction also becomes larger, increasing the contribution of out-of-plane eddy current loss. In the present invention, the distribution of this magnetic flux crossing is optimized, so that the effect on improving out-of-plane eddy current loss is large. For this reason, it is believed that the higher the frequency, the greater the effect. Therefore, the transformer core of the present invention can be used in any frequency band, but is particularly suitable for use in a high frequency band of 100 Hz or more. However, if the frequency is too high, the magnetic permeability of the grain-oriented electrical steel sheet is significantly reduced, and sufficient effects may not be obtained. Therefore, the transformer core of the present invention is preferably used in a frequency band of 10,000 Hz or less, and more preferably in a frequency band of 5,000 Hz or less. In other words, the transformer core in one embodiment of the present invention may be a high-frequency transformer core of 100 to 10,000 Hz. ・How to control standard deviation σ The value of the standard deviation σ is an index representing the degree of variation in the thickness of the insulating coating. The variation in the thickness of the insulating coating varies depending on the unevenness of the surface of the base steel and the sum of the unevenness of the surfaces of the layers that make up the insulating coating. Therefore, the standard deviation σ can be adjusted by controlling the roughness of the steel sheet surface after rolling, the variation in the amount of annealing separator applied during secondary recrystallization annealing, the variation in the amount of tension coating agent applied during planarization annealing, and the like. The roughness of the steel sheet surface after rolling can be measured, for example, using a laser meter. Specifically, the surface roughness distribution over the entire length and width of the grain-oriented electrical steel strip cold-rolled to the final thickness can be measured using a laser meter. In addition, the basis weight distribution of the insulating coating layer in the finally obtained grain-oriented electrical steel strip can be measured, for example, by fluorescent X-ray analysis. Specifically, the basis weight distribution over the entire length and width of the steel strip after the decarburization annealing, application of the annealing separator, final finish annealing, and formation of the tension coating can be calculated from the fluorescent X-ray intensity of the main component of the insulating coating layer. When manufacturing grain-oriented electrical steel sheets, the application conditions of the annealing separator and tension coating agent may be controlled by performing feedback control based on the results of these measurements. Furthermore, by adjusting the roll roughness during operation and the viscosity of the agent to be applied, it is possible to control the variation at any desired location. Furthermore, as a method for controlling the variation with high accuracy, it is also effective to finely vibrate the steel sheet by irradiating ultrasonic waves when applying the annealing separator or tension coating agent, thereby promoting the spreading of the applied agent. By appropriately combining the above-mentioned various methods from the viewpoints of man-hours, manufacturability, etc., the standard deviation σ can be controlled. - Method for evaluating the standard deviation of insulation coating thickness (variation in insulation coating thickness) The standard deviation σ can be evaluated by image analysis of the cross-sectional (C-section) image of the grain-oriented electrical steel sheet with the insulating coating taken by SEM. At least 10 samples of C-section observation samples are cut from different locations in the direction perpendicular to the rolling direction of the steel strip from which they were taken, from the above-mentioned joint of the iron core and its vicinity, and cross-sectional observation is performed. Here, a schematic diagram of a C-section observation image obtained by SEM is shown in FIG. 6(a). From the C-section observation image, a line profile of the insulating coating (FIG. 6(b)) can be obtained by extracting the contour using the openCV library of Python. From the line profile thus obtained, the thickness t of the insulating coating at each position is obtained (FIG. 6(c)). The thickness t is the shortest distance between the outermost surface of the insulating coating of the grain-oriented electrical steel sheet and the surface of the base steel (the boundary between the metal oxide layer and the base steel). The thickness can be calculated for each position, for example, with one pixel of the SEM photograph as the unit length. This calculation is performed for 10 or more samples. The variation in t in the C-section observation images of all samples obtained in this manner is defined as the variation (standard deviation σ) in the insulation coating thickness in the present invention. The number of samples N is preferably 10 or more, and more preferably 15 or more. By deriving a correlation equation from the insulation coating thickness thus determined, the surface roughness of the steel strip after production, and the coating weight distributions of the annealing separator and tension coating agent, it becomes possible to predict the insulation coating thickness from the above-mentioned surface roughness and coating weight distribution, and the standard deviation σ of the insulation coating thickness at a joint can be determined. ・Spacing Δd is 1 mm or more and 20 mm or less As described above, if the distance Δd between the steel plate discontinuity of one adjacent single layer and the steel plate discontinuity of the other adjacent single layer in the direction of magnetic flux flow is 1 mm or more, the building factor is further improved. Therefore, the distance Δd is preferably 1 mm or more, and more preferably 3 mm or more. On the other hand, if the distance Δd is 20 mm or less, the building factor is further improved. Therefore, the distance Δd is preferably 20 mm or less, and more preferably 15 mm or less. With respect to the steps and manufacturing conditions other than those described above, a known manufacturing method for grain-oriented electrical steel sheet can be appropriately used. The thickness of the grain-oriented electrical steel sheet is not particularly limited, but is preferably 0.01 mm or more, and more preferably 0.05 mm or more. On the other hand, the thickness is preferably 0.35 mm or less, and more preferably 0.30 mm or less. In one embodiment of the present invention, the thickness is preferably 0.25 mm or less, and more preferably 0.23 mm or less. When the grain-oriented electrical steel sheet has a relatively thin thickness, it can be particularly preferably used in the high frequency band (100 to 10,000 Hz). In another embodiment of the present invention, the thickness is preferably 0.10 mm or more, and more preferably 0.15 mm or more. When the grain-oriented electrical steel sheet has a relatively thick thickness, it can be particularly preferably used in the commercial frequency band. The thickness of the insulating coating is not particularly limited, but is preferably 0.5 μm or more, and more preferably 0.8 μm or more, while the thickness is preferably 4 μm or less, and more preferably 3 μm or less. The number of steel sheets constituting one single layer is not particularly limited, and may be any number equal to or greater than one. Typically, the number is preferably four or less, and more preferably three or less. In other words, the single layer, which is the lamination unit constituting the iron core of the present invention, may be a laminate of at least one, and preferably one to four, grain-oriented electrical steel sheets with an insulating coating. On the other hand, the number of single layers to be laminated, in other words, the total number of laminated grain-oriented electrical steel sheets, varies greatly depending on the capacity of the transformer to be manufactured, the sheet thickness of the grain-oriented electrical steel sheets to be used, etc. Therefore, it is not particularly limited and may be appropriately determined depending on the capacity of the transformer, etc. [Iron core] The form of the transformer core in the present invention is not particularly limited, and may be either a wound core or a stacked core. The core in one embodiment of the present invention may be a wound core in which a grain-oriented electromagnetic steel sheet is wound in the rolling direction of the steel sheet. The wound core may have a bent portion that is bent in the winding direction. In another embodiment of the present invention, the core may be a stacked core in which a plurality of single layers are stacked, each of which is made by combining a plurality of pieces of grain-oriented electromagnetic steel sheet in the same plane. The present invention will be specifically described below based on examples. Note that the following examples are merely preferred examples of the present invention, and the present invention is not limited to these examples. The present invention can be modified within the scope of the present invention, and such modifications are also included in the technical scope of the present invention. Example 1 A grain-oriented electrical steel sheet with an insulating coating was produced using a steel slab having the chemical composition shown in Table 2. Specifically, a cold-rolled steel strip was first obtained from the steel slab by a general manufacturing process. Next, the cold-rolled steel strip was subjected to a shot blasting treatment with varying shot particle sizes to roughen the surface of the steel strip. After that, the cold-rolled steel strip was subjected to decarburization annealing, application of an annealing separator, secondary recrystallization annealing, flattening annealing, and insulating coating formation in this order to produce a grain-oriented electrical steel sheet with an insulating coating. A sample with a width of 100 mm and a length of 280 mm was cut out from the obtained steel strip. The sample was used to measure the core loss (W １７／５０ ) was measured. Furthermore, the standard deviation σ was calculated from the surface roughness distribution of the steel sheet and the distribution of the coating weight of the insulating coating layer measured during the sheet passing. Transformer cores were produced using single sheets cut from the steel strip coil so that the standard deviation σ was the value shown in Tables 3 to 5. The following three types of transformer cores were produced, differing in core shape and whether or not they had been subjected to strain relief annealing (SRA). In this case, the wrap length Δd was changed between 0.5 and 22 mm. - Three-phase stacked core model transformer: No SRA (Table 3) ・Wound core model transformer (Unicore): No SRA (Table 4) Wound core model transformer (Unicore): With SRA (Table 5) For the iron core thus obtained, the transformer iron loss was measured when the magnetic flux density in the iron core legs was 1.7 T at a frequency of 50 Hz. In the measurement, the no-load loss was measured using a wattmeter. The building factor (BF) was calculated from the obtained transformer iron loss and the material iron loss measured previously. The results are shown in Tables 3 to 5. As can be seen from the results, the transformer cores satisfying the conditions of the present invention had good building factors (BF). The standard deviation σ of the legs of the iron core is also shown in Tables 3 to 5. As can be seen from the results in Tables 3 to 5, if the standard deviation of the joint is 2 to 5 μm, desired characteristics are obtained regardless of the standard deviation of the legs. Example 2 A grain-oriented electrical steel sheet with an insulating coating was produced using a steel slab having the component composition shown in Table 2. Specifically, a cold-rolled steel strip having a thickness shown in Table 6 was first obtained from the steel slab by a general manufacturing process. Next, the cold-rolled steel strip was subjected to a shot blasting treatment with varying shot particle size to control the surface condition of the steel strip. After that, the cold-rolled steel strip was subjected to decarburization annealing, application of an annealing separator, secondary recrystallization annealing, flattening annealing, and insulating coating formation in this order to produce a grain-oriented electrical steel sheet with an insulating coating. A sample having a width of 100 mm and a length of 280 mm was cut out from the obtained steel strip. The material iron loss was measured using the sample by the single sheet magnetic measurement method described in JIS C2556. The material iron loss was measured at the magnetic flux density, frequency, and excitation waveform shown in Table 6. Furthermore, the standard deviation σ was calculated from the surface roughness distribution of the steel sheet and the distribution of the coating weight of the insulating coating layer measured during the sheet passing. A three-phase stacked core model transformer A was produced using single sheets cut from a steel strip coil so that the standard deviation σ was 3 μm. Similarly, a three-phase stacked core model transformer B was produced using single sheets cut from a steel strip coil so that the standard deviation σ was 6 μm. In both model transformers, the overlap length Δd was 4 mm. The transformer iron loss of the thus obtained core was measured. The measurement of the transformer iron loss was carried out at the magnetic flux density, frequency, and excitation waveform shown in Table 6, similar to the material iron loss. Here, the magnetic flux density is the magnetic flux density at the leg of the core. At this time, the excitation waveform was a normal sine wave, a superimposed waveform with up to the tenth harmonics superimposed, or a square wave. In the measurement, the no-load loss was measured using a wattmeter. From the obtained transformer iron loss and the material iron loss measured earlier, the building factor BF of the model transformer A is calculated. Ａ and the building factor BF of model transformer B Ｂ The obtained BF Ａ and B.F. Ｂ From the values ​​of , the BF improvement rate defined by the following formula was calculated. The obtained values ​​are shown in Table 6. This BF improvement rate can be regarded as an index of the BF improvement effect by changing the standard deviation σ from outside the range of the present invention (6 μm) to within the range (3 μm). BF improvement rate (%) = (BF Ｂ -BF Ａ ) / B.F. Ｂ ×100 As can be seen from the results in Table 6, the higher the frequency, the greater the improvement in BF. Also, a greater improvement in BF was obtained with the superimposed waveform than with the sine wave, and even greater with the square wave. This is thought to be because the higher the frequency and the more complex the waveform, the greater the change in local magnetic flux density dB / dt due to the crossover magnetic flux, and the greater the contribution of eddy current loss caused by the crossover magnetic flux to the iron loss of the transformer. 1a-1f Single layer 2a~2f joint part 3. Direction of magnetic flux flow

Claims

1. A transformer core formed by stacking multiple single layers each made of one or more grain-oriented electrical steel sheets with an insulating coating, wherein each of the single layers has at least one steel sheet discontinuity extending in a direction intersecting the direction of magnetic flux flow in the single layer, and the standard deviation σ of the thickness of the insulating coating in a region sandwiched between the steel sheet discontinuity of one single layer and the steel sheet discontinuity of the other single layer adjacent in the stacking direction is 2 μm or more and 5 μm or less.

2. A transformer core as described in claim 1, wherein the distance Δd in the direction of magnetic flux flow between the steel plate discontinuity of one adjacent single layer and the steel plate discontinuity of the other adjacent single layer is 1 mm or more and 20 mm or less.

3. A transformer core according to claim 1 or 2, which is a wound core in which the grain-oriented electromagnetic steel sheet is wound in the rolling direction of the steel sheet.

4. A transformer core according to claim 3, wherein the wound core has a bent portion that is bent in the winding direction.

5. A transformer core according to claim 1 or 2, which is a laminated core formed by stacking a plurality of single layers each formed by combining a plurality of pieces of said grain-oriented electromagnetic steel sheet in the same plane.

6. A transformer comprising a transformer core according to any one of claims 1 to 5.

Images