Three phase five leg wound iron core

Abstract

To improve the iron loss characteristics of a wound core.SOLUTION: A three phase five leg wound core comprising four cores arranged in a row, each of the four cores comprising a planar portion and a corner portion adjacent to the planar portion, two side cores arranged at both ends of the four cores being subjected to stress relief annealing, Two center cores disposed on an inner side among the four cores each include a bent portion in the corner portion and a joint portion in at least one of the planar portions, and are not subjected to stress relief annealing.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a three-phase five-leg wound core. [Background technology]

[0002] Iron cores are broadly divided into two types based on their structure: stacked cores and wound cores. Stacked cores are made by laminating steel sheets cut to a specific shape. Wound cores, on the other hand, are made by winding steel sheets one on top of the other, and are widely used in reactors, filters, transformers, etc.

[0003] Wound cores can be manufactured in a variety of ways, but the two most common manufacturing methods are to press magnetic steel sheets after winding them together, or to stack magnetic steel sheets that have been bent in advance.

[0004] In the former method, for example, a wound core is manufactured by the following procedure: First, a coil of grain-oriented electromagnetic steel sheet is wound around a circular winding die while the steel sheet is cut into individual turns. Next, a forming die is placed on the inside and outside of the circularly wound steel sheet and pressed to form a roughly rectangular wound core.

[0005] In this method, strain is introduced into the entire wound core during the pressing. To release this strain, stress relief annealing must be performed after pressing. This method also requires the use of a dedicated mold that is tailored to the wound core being manufactured. If the number of wound cores manufactured is small, the proportion of mold manufacturing costs in the total cost increases, making this method unsuitable for high-mix, low-volume production.

[0006] On the other hand, in the latter method, the corners of cut grain-oriented electrical steel sheets are bent in advance, and the bent steel sheets are then stacked to produce a wound core. This method does not require a press using a forming die, making design changes easy. Furthermore, strain is only introduced in the corners where bending is performed, and the area where strain is introduced accounts for a small proportion of the entire core. Therefore, there is almost no iron loss degradation due to strain, and stress relief annealing can be omitted.

[0007] With regard to such wound cores, various techniques have been proposed to improve the iron loss characteristics.

[0008] For example, Patent Document 1 proposes a technique for manufacturing a single core using multiple electromagnetic steel sheets with different magnetic properties. This technique involves using a material with poor magnetic properties on the inner side where magnetic flux tends to be biased, in order to suppress bias (concentration) of magnetic flux that occurs within a single core due to differences in magnetic path length.

[0009] Patent Document 2 also proposes arranging cuts formed at the joints at both ends of the iron core material so that they are sequentially shifted in a stepped manner in the circumferential direction of the iron core material, which allows for smoother flow of magnetic flux, thereby reducing the magnetic resistance of the magnetic path and suppressing increases in iron loss.

[0010] Patent Document 3 proposes distributing cuts formed at the joints at both ends of the core material around the wound core window, thereby suppressing an increase in magnetic flux density near the joints of the core material. [Prior art documents] [Patent documents]

[0011] [Patent Document 1] Japanese Patent Application Laid-Open No. 2007-043040 [Patent Document 2] Japanese Patent Application Laid-Open No. 2001-284136 [Patent Document 3] Japanese Patent Application Publication No. 2016-028406 Summary of the Invention [Problem to be solved by the invention]

[0012] Although the iron loss characteristics of wound cores can be improved by applying the techniques described above, further improvements in iron loss characteristics are desired.

[0013] The present invention has been made in view of the above circumstances, and has an object to improve the iron loss characteristics of a wound core. [Means for solving the problem]

[0014] In order to solve the above problems, the inventors have investigated factors that affect the iron loss characteristics of wound cores. Representative results of experiments conducted for the investigation will be described below.

[0015] <Experiment 1> First, to evaluate the effect of the wound core structure on the magnetic properties, a three-phase three-limbed wound core and a three-phase five-limbed wound core were fabricated and the iron loss and magnetic flux density were evaluated.

[0016] Figure 1 is a schematic diagram showing the structure and dimensions of the fabricated three-phase three-legged core 10. Three-phase three-legged core 10 is composed of two inner cores 11 and one outer core 12, and inner core 11 and outer core 12 each have a joint 13. When three-phase three-legged core 10 is used in a transformer, a coil is wound around each of main legs 14.

[0017] 2 is a schematic diagram showing the structure and dimensions of the fabricated three-phase five-legged core 20. The three-phase five-legged core 20 has four cores arranged in a row, of which two at both ends are side cores 21 and two at the inside are center cores 22. The side cores 21 and center core 22 each have a joint 23. When the three-phase five-legged core 20 is used in a transformer, a coil is wound around each of the three main legs 24, but no coil is wound around the return legs 25 located at both ends.

[0018] The structures shown in Figures 1 and 2 are for wrapped cores with joints 13 and 23, but in order to investigate the influence of the steel plate joints, non-cut cores without joints 13 and 23 were also produced for each of the three-phase three-limbed wound core and the three-phase five-limbed wound core.

[0019] Each wound core was fabricated by press-forming grain-oriented electrical steel sheet, followed by stress relief annealing. The grain-oriented electrical steel sheet used for all wound cores was a 0.23 mm thick grain-oriented electrical steel sheet that had been subjected to heat-resistant magnetic domain refinement treatment. The grain-oriented electrical steel sheet used had a magnetic flux density B8 of 1.93 T at a magnetizing force of 800 A / m, a magnetic flux density amplitude of 1.7 T, and an iron loss W at 50 Hz. 17 / 50 was 0.75W / kg.

[0020] Next, the overall iron loss and building factor BF of each wound core obtained by the above procedure were determined using the following procedure.

[0021] First, the loss P, expressed by the following formula, was calculated for each of the U, V, and W phases, and then the total loss Ptotal was calculated by adding them together. The total loss Ptotal was then divided by the weight of the iron core to obtain the total iron loss (W / kg). The excitation conditions were a frequency of 50 Hz and a magnetic flux density of 1.7 T.

number

[0022] The building factor BF was calculated by dividing the total iron loss by the material iron loss (0.75 W / kg in this case). The total iron loss and building factor BF are shown in Table 1.

[0023] Furthermore, the local iron loss distribution and local magnetic flux density distribution of each wound core were evaluated. In the evaluation, the local iron loss distribution was measured using the thermography method, and the local magnetic flux density distribution was measured using the search coil method. The specific measurement method was as described in IEEJ Transactions on Magnetic Materials, Vol. 141, No. 4, pp. 226-232, 2021. The excitation conditions were a frequency of 50 Hz and a magnetic flux density of 1.7 T.

[0024] The following four values ​​were calculated from the measured local iron loss distribution and are shown in Table 1. Iron loss at the yoke core interface Average iron loss difference in legs ΔW ·Leg average iron loss Iron loss in the joint installation area

[0025] [Table 1]

[0026] The positions of yoke portions 16, 26 and legs 17, 27 are shown in Figure 3(a). Furthermore, yoke portion-core interface 18 refers to a 64 mm x 10 mm area where the yoke portions of inner core 11 and outer core 12 constituting three-phase three-legged core 10 meet, as shown in Figure 3(b). Therefore, iron loss at the yoke portion-core interface was measured only for the three-phase three-legged core.

[0027] The leg average iron loss difference ΔW is a value defined by the following formula. ΔW = Average iron loss in the inner core legs - Average iron loss in the outer core legs (for a three-phase, three-leg core) ΔW = average iron loss in center core legs - average iron loss in side core legs (for a 3-phase 5-leg core) Here, "average iron loss in the legs of the inner core" refers to the average value of iron loss in the legs of the two inner cores (four locations in total), "average iron loss in the legs of the outer core" refers to the average value of iron loss in the legs of the outer core (two locations). Similarly, "average iron loss in the legs of the center core" refers to the average value of iron loss in the legs of the two center cores (four locations in total), and "average iron loss in the legs of the side core" refers to the average value of iron loss in the legs of the two side cores (four locations in total).

[0028] On the other hand, the average leg iron loss is the average value of the leg iron loss in all cores.

[0029] Since conventional wound cores have joints, we first compared the results for wound cores with joints (No. 1 and 2). As a result, we found that the three-phase five-limbed wound core (No. 2) had better overall iron loss and building factor (BF) than the three-phase three-limbed wound core (No. 1).

[0030] Furthermore, in terms of local iron loss distribution, the "iron loss at the yoke core interface" was larger than the "average iron loss in the legs" in the three-phase, three-limbed wound core (No. 1). This is thought to be because, due to the structure of the three-phase, three-limbed wound core, magnetic flux must pass from the inner core to the outer core and back again to the inner core, and large out-of-plane eddy current loss occurs when the magnetic flux passes. Also, the average iron loss difference in the legs, ΔW, was 0.7 in the three-phase, three-limbed wound core (No. 1). A positive ΔW value indicates that the average iron loss in the inner core legs is greater than the average iron loss in the outer core legs. This is thought to be the result of magnetic flux concentrating on the inside, where magnetic resistance is low, due to the difference in magnetic path length.

[0031] The relationship between iron loss and magnetic flux density is not proportional, and the increase in iron loss accelerates as the magnetic flux density increases. For this reason, it is desirable to minimize areas of high magnetic flux density (areas of magnetic flux concentration). In fact, measurements using a search coil confirmed the transfer of magnetic flux between cores and non-uniformity in magnetic flux density (magnetic flux concentration on the inside).

[0032] On the other hand, in the three-phase five-limbed winding core (No. 2), no increase in iron loss due to crossover magnetic flux was confirmed. Also, like No. 1, the average leg iron loss difference ΔW was 0.7. A positive ΔW value indicates that the average leg iron loss in the center core is greater than the average leg iron loss in the side cores. The reason for this is not yet clear, but it is thought that more magnetic flux flows through the center core, and the non-uniformity in magnetic flux density between the cores is the cause of the increased loss in the three-phase five-limbed winding core.

[0033] Reducing the magnetic resistance within the core and making it easier for the magnetic flux to flow is thought to be an effective method for suppressing the unevenness of the magnetic flux between cores. Therefore, we investigated eliminating the joints as a means of reducing the magnetic resistance.

[0034] Comparing the measurement results of a three-phase, three-legged core with a joint (No. 1) and a three-phase, three-legged wound core without a joint (No. 3), the BF was significantly lower with the core without a joint. Looking at the measurement results of iron loss distribution and magnetic flux distribution, it was clear that the increase in iron loss due to the joint had been suppressed. In addition, as expected, the transfer of magnetic flux between cores and the unevenness of magnetic flux density between the outer and inner cores had been suppressed, resulting in a decrease in iron loss at the core interface of the yoke and the average iron loss difference ΔW between the legs.

[0035] Even in the three-phase five-leg wound core, the average leg iron loss difference ΔW decreased when there was no joint (No. 4).

[0036] The iron loss improvement effect of eliminating this joint was greater in the three-phase, three-limbed wound core. However, as mentioned earlier, in actual transformers, the core is opened at the joint and the excitation coil is inserted, so it is difficult to eliminate the joint. Therefore, in order to improve iron loss without placing a large load on the production process, it is necessary to reduce the magnetic resistance inside the core.

[0037] In experiments with joints, the iron loss improvement effect due to reduced magnetic reluctance was greater in the three-phase, three-limbed wound core. However, with a three-phase, three-limbed wound core, magnetic flux transfer is unavoidable due to the core structure and magnetic circuit, and it is difficult to reduce it to zero. On the other hand, the magnetic flux non-uniformity between cores observed in the three-phase, five-limbed wound core is not unavoidable due to the structure or magnetic circuit, and can be made uniform with appropriate measures. Therefore, it was thought that a three-phase, five-limbed wound core would ultimately be able to reduce iron loss to a lower level.

[0038] <Experiment 2> Based on the results of Experiment 1 above, we next investigated a method for homogenizing the magnetic flux in a three-phase, five-limbed wound core. In other words, in order to homogenize the magnetic flux, it is important to control the ease with which the magnetic flux flows in the core. Therefore, we investigated changing the ease with which the magnetic flux flows by introducing distortion into the core.

[0039] As mentioned above, if the core is manufactured by laminating pre-bent magnetic steel sheets, strain is introduced only at the bent corners. Therefore, there is almost no iron loss deterioration, and it is possible to freely select whether or not to perform stress relief annealing. Furthermore, the degree of strain introduced into the core can be controlled by whether or not to perform stress relief annealing.

[0040] Therefore, a three-phase, five-limbed wound core was fabricated using four cores manufactured using the method described above. Specifically, a Unicore manufacturing machine manufactured by AEM was used to fabricate a core (Unicore) with two 45-degree bends (strain introduction sections) at one corner. The dimensions and materials of the core were the same as in Experiment 1. Some of the obtained cores were subjected to stress relief annealing, while the remaining cores were not.

[0041] Four of the obtained cores were used to fabricate a three-phase five-leg wound core. For the center core and side core, some cores had undergone stress relief annealing and some had not, as shown in Table 2.

[0042] [Table 2]

[0043] Next, the iron loss characteristics of the obtained three-phase five-leg wound core were evaluated in the same manner as in Experiment 1. The evaluation results are shown in Table 2.

[0044] Comparing No. 1, in which all cores were stress relief annealed, with No. 4, in which none of the cores were stress relief annealed, the overall iron loss was roughly the same, but the iron loss was slightly higher in the case without stress relief annealing due to the strain introduced. Also, no difference was observed in the iron loss difference ΔW between the center core and side core in No. 1 and No. 4. This is thought to be because the ease with which magnetic flux passes through all four cores was the same. Magnetic flux measurements using the search coil method also confirmed that the magnetic flux was non-uniform.

[0045] On the other hand, No. 3, in which the side cores were stress-relief annealed but the center core was not, had better iron loss characteristics than No. 1, in which all the cores were stress-relief annealed. The core that had been stress-relief annealed had slightly better iron loss characteristics, but the difference in iron loss between the side cores and the center core was very small. This is thought to be due to the uniform magnetic flux density.

[0046] Magnetic flux density measurements using the search coil method also confirmed that No. 3 had a uniform magnetic flux density distribution. In other words, magnetic flux originally tended to concentrate in the center core. This means that magnetic flux passes more easily through the center core. Therefore, by leaving distortion in the center core, which reduces the ease with which magnetic flux passes, and by not leaving distortion in the side cores, which does not reduce the ease with which magnetic flux passes, it is thought that a balance was achieved and the magnetic flux density distribution became uniform.

[0047] On the other hand, looking at No. 2, in which the side cores were not subjected to stress relief annealing but the center core was subjected to stress relief annealing, the total iron loss was the largest.

[0048] The present invention has been completed based on the above findings, and the gist and configuration of the present invention are as follows.

[0049] 1. A three-phase five-leg wound core with four cores arranged in a row, Each of the four cores is a wound core having a flat portion and a corner portion adjacent to the flat portion, Of the four cores, the two side cores located at both ends have been subjected to stress relief annealing. Among the four cores, the two center cores arranged on the inside each have a bent portion at the corner portion and a joint portion at at least one of the flat portions, and have not been subjected to stress relief annealing. 3-phase 5-leg wound core. [Effects of the Invention]

[0050] According to the present invention, a three-phase five-leg wound core with extremely excellent core loss characteristics can be obtained. [Brief explanation of the drawings]

[0051] [Figure 1] FIG. 1 is a schematic diagram showing the structure and dimensions of the three-phase, three-legged wound core used in Experiments 1 and 2. [Figure 2] FIG. 1 is a schematic diagram showing the structure and dimensions of the three-phase five-leg wound core used in Experiments 1 and 2. [Figure 3] FIG. 2 is a schematic diagram showing the positions of a yoke, legs, and a yoke-core interface. DETAILED DESCRIPTION OF THE INVENTION

[0052] The present invention will be described in detail below. Note that the following description is of preferred embodiments of the present invention, and the present invention is not limited to the embodiments described below.

[0053] The wound core of the present invention is a three-phase, five-limbed wound core including four cores 21, 22 arranged in a row, as shown in Fig. 2. Each of the four cores 21, 22 is a wound core including a flat portion F and a corner portion C adjacent to the flat portion.

[0054] Side core Of the four cores, the side cores 21 have been subjected to stress relief annealing. Since the side cores 21 are subjected to stress relief annealing, there are no particular limitations on the core type, and any type can be used. For example, the side cores 21 may be any of a tranco-core, a uni-core, and a duo-core. As shown in FIG. 2 , the side cores 21 are the two cores arranged at both ends of the four cores arranged in a row, and constitute the main leg 24 and the return leg 25.

[0055] Center Core On the other hand, of the four cores, the center core 22 has a bent portion at the corner portion C and a joint portion 23 at at least one of the flat portions F, and is a core that has not been subjected to strain relief annealing. This is because, in the case of a tranco type that generally requires strain relief annealing, if strain relief annealing is omitted to leave strain remaining, the amount of strain introduced would be too great, resulting in a significant deterioration in iron loss.

[0056] For example, a unicore or duocore can be used as the center core 22. As shown in Fig. 2, the center core 22 refers to the two cores arranged on the inside of the four cores arranged in a row, and constitutes the main leg 24 but does not constitute the return leg 25.

[0057] The joining method of the joint portion is not particularly limited and may be any method. As the joining method, an overlap type and a step lap type are common, and either type can be used.

[0058] In a three-phase five-leg wound core, magnetic flux tends to concentrate more easily in the center core than in the side cores. Therefore, by using a core that has not been stress-relief annealed as the center core and a core that has been stress-relief annealed as the side core, this concentration of magnetic flux can be counteracted and the magnetic flux in the legs can be made uniform.

[0059] In the present invention, the discrimination between cores that have been subjected to stress relief annealing and cores that have not been subjected to stress relief annealing is defined as being performed by the following method: the cores are subjected to additional annealing at 800°C for 3 hours in an argon atmosphere, and the iron loss W 17 / 50 If the difference is 0.1 W / kg or less, the core is judged to have been subjected to stress relief annealing, and if it is more than 0.1 W / kg, the core is judged to have not been subjected to stress relief annealing.

[0060] In the above judgment, even though the core was actually subjected to stress relief annealing, the iron loss W 17 / 50 In some cases, the difference between the values ​​of the core loss and the stress relief annealing exceeds 0.1 W / kg. Such a significant decrease in core loss due to additional annealing means that the stress relief annealing was insufficient, and the desired magnetic flux density distribution cannot be achieved. Therefore, such cores are considered to be cores that have not been subjected to stress relief annealing.

[0061] Known methods can be used to manufacture wound cores. For example, a Unicore manufacturing machine manufactured by AEM can be used to manufacture Unicores. When using the Unicore manufacturing machine, the design size is input into the machine, and steel plates are sheared and bent to the size specified in the design, one by one, and the wound core (Unicore) can be manufactured by stacking these processed steel plates.

[0062] The side core may be a tranco core. A tranco core can typically be manufactured by winding a steel plate around a mandrel and then pressing it. Stress relief annealing can then be performed.

[0063] As described above, the present invention can achieve the desired effect by optimizing the strain distribution within the three-phase five-limbed wound core. Therefore, conditions other than those described above, such as the core size, angle of bends, number of bends, material, etc., are not particularly limited. [Example]

[0064] Next, the present invention will be described in more detail based on examples. The following examples are examples of preferred embodiments of the present invention, and the present invention is not limited to these examples. The embodiments of the present invention can be appropriately modified within the scope of the invention, and all such modifications are included in the technical scope of the present invention.

[0065] Magnetic flux density B8 is 1.92T, iron loss W 17 / 50 A three-phase, three-limbed wound core and a three-phase, five-limbed wound core were fabricated using 0.30 mm thick grain-oriented electrical steel sheets with a thermal conductivity of 1.02 W / kg. The capacity was 30 kVA. Unicores and Trancocores were used as the cores constituting the wound cores, in the combinations shown in Tables 3 and 4. The Unicores have bent portions at the corners and one lap portion (joint) at the flat portions. The Trancocores have one joint and are fabricated by wrapping grain-oriented electrical steel sheets around a mandrel and then pressing. Some of the cores were subjected to stress relief annealing at 800°C for 5 hours in a N2 atmosphere (Tables 3 and 4).

[0066] The iron loss and building factor (BF) of the fabricated three-phase three-legged winding core and three-phase five-legged winding core were evaluated, and the results are shown in Tables 3 and 4, respectively.

[0067] When the same core combination was used, the three-phase five-leg wound core showed better iron loss characteristics. This is thought to be due to the suppression of in-plane eddy current loss caused by the crossover magnetic flux.

[0068] Furthermore, in Comparative Examples Nos. 2, 3, 4, 6, 8, 10, 12, 18, 19, 20, 22, 24, 26, and 28, in which Trancocore was used in part of the wound core without stress relief annealing, iron loss was significantly degraded and the building factor was large. This is thought to be because the amount of strain introduced by pressure during pressing was large, and the influence of this strain caused a significant deterioration in iron loss.

[0069] In Comparative Examples 11, 15, 27, and 31, in which the inner or center core was annealed and the outer or side core was a unicore that had not been annealed, the iron loss and building factor were larger than those of the normal combination. This is thought to be the result of magnetic flux being more concentrated in the inner or center core.

[0070] All cores were subjected to stress relief annealing, and Comparative Examples Nos. 5, 9, 21, and 25, in which the types of outer core and inner core, and center core and side core were changed, showed almost the same properties as the normal combinations.

[0071] On the other hand, invention examples Nos. 23 and 30, which satisfied the conditions of the present invention, showed better characteristics than comparison examples Nos. 1, 13, 16, 17, 29, and 32, which were ordinary combinations. From the above results, it can be seen that the three-phase five-limbed wound core of the present invention has excellent iron loss characteristics.

[0072] [Table 3]

[0073] [Table 4] [Explanation of symbols]

[0074] 10 3-phase 3-legged wound core 11 Inner Core 12 outer core 13 Joint 14 Main landing gear 16 York 17 Legs 18 Yoke core interface 20 3-phase 5-leg wound core 21 Side Core 22 Center Core 23 Joint 24 Main landing gear 25 Return Leg 26 York 27 Legs F flat part C corner part

Claims

[Claim 1] A three-phase five-leg wound core having four cores arranged in a row, Each of the four cores is a wound core having a flat portion and a corner portion adjacent to the flat portion, Of the four cores, two side cores located at both ends are subjected to stress relief annealing, Among the four cores, two center cores arranged on the inside each have a bent portion at the corner portion and a joint portion at at least one of the flat portions, and have not been subjected to stress relief annealing. 3-phase 5-leg wound core.

Images