Insulation testing method and insulation testing apparatus

Abstract

To provide an insulation inspection method which can detect surface discharge.SOLUTION: An insulation inspection method for inspecting surface discharge generated through a space in a resin part 5 with metal parts 21 of wire rods of each phase constituting a coil 2 possessed by an armature, as an object where at least a part of the coil 2 is molded with a resin as an inspection object includes the following steps: applying a predetermined impulse voltage to the coil 2, and detecting a current value flowing in a current detector 12 connected to the coil 2; and calculating the total sum of current value spectra of a high frequency band different from a predetermined low frequency band capable of detecting partial discharge generated in a thin film 26 site of the wire materials constituting the coil 2 as a high frequency spectrum area, in a relation between a frequency analyzing the detected current value and the current value spectrum, and inspecting surface discharge generated in the coil 2 according to the size of the high frequency spectrum area.SELECTED DRAWING: Figure 23

Description

Technical Field

[0001] The present invention relates to an insulation inspection method and an insulation inspection device for inspecting the insulation state of a coil included in an armature (i.e., a stator or a rotor) constituting a rotating electric machine.

Background Art

[0002] Conventionally, an insulation inspection method and an insulation inspection device for inspecting the insulation state of a coil included in an armature constituting a rotating electric machine are known. The insulation inspection device described in Patent Document 1 applies an impulse voltage to a motor and inspects partial discharges occurring at locations where the coating of the wire material constituting the three-phase coil is thin. Specifically, in the insulation inspection, the following three inspections are performed. First, it is an insulation inspection between the wire material of a predetermined phase constituting the three-phase coil and the wire material of another phase. Second, it is an insulation inspection between the wire materials of a predetermined phase constituting the three-phase coil. Third, it is an insulation inspection between the wire material of a predetermined phase constituting the three-phase coil and the core.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, among the coils of an armature constituting a rotating electric machine, there is a case where a portion protruding from a slot (i.e., a coil end) is resin-molded. In that case, if voids occur in the resin portion that molds the coil, and further pinholes occur in the coating of the wire material constituting the coil and the metal parts of the wire materials of each phase constituting the coil are connected by a space (i.e., voids and pinholes), surface discharges may occur along the inner wall surface of that space. In the insulation inspection of an armature, it is required to separately detect (i.e., distinguish) surface discharges and partial discharges.

[0005] However, the insulation testing method described in Patent Document 1 above detects both partial discharge and surface discharge together, making it impossible to detect only surface discharge. Thus, there are no conventional insulation testing methods or apparatus that can separate (i.e., distinguish) surface discharge from partial discharge when the object of inspection is an armature with coils molded in resin.

[0006] In view of the above points, the present invention aims to provide an insulation inspection method and an insulation inspection apparatus capable of detecting surface discharge. [Means for solving the problem]

[0007] To achieve the above objective, the invention according to claim 1 relates to the coil (2) of the armature The coil ends protruding from the slots of the stator core The inspection targets items that are at least partially molded with resin, and the metal parts (21) of the wires of each phase constituting the coil are compared. between in Molded coil ends In an insulation testing method for inspecting surface discharge occurring through the space within a resin part, For the coil Of the 1kV to 4.5kV range Applying a predetermined impulse voltage and detecting the current value flowing through the current detector (12) connected to the coil (S10), In the relationship between the frequency and current spectrum analyzed from the detected current value, the partial discharge occurring in the thin coating (26) portion of the wire constituting the coil is different from a predetermined low-frequency band that can be detected. A specified frequency range between 300MHz and 500MHz The sum of the current values ​​in the high-frequency band is calculated as the high-frequency spectrum area, and the creepage discharge generated in the coil is determined by the size of this high-frequency spectrum area. Distinguish from partial discharge This includes conducting inspections (S20-S60).

[0008] According to this method, in the insulation inspection of the coils of an armature, by using a high-frequency band different from the predetermined low-frequency band used in the partial discharge inspection, surface discharge and partial discharge can be inspected separately (i.e., distinguished). Therefore, this insulation inspection method makes it possible to accurately distinguish between armatures with insulation defects that produce surface discharge and armatures that do not produce surface discharge.

[0009] The invention according to claim 3 relates to the coil (2) of the armature The coil ends protruding from the slots of the stator core The inspection targets items that are at least partially molded with resin, and the metal parts (21) of the wires of each phase constituting the coil are compared. between in Molded coil ends In an insulation testing device for inspecting surface discharge occurring through the space within the resin part (5), For the coil Of the 1kV to 4.5kV range An impulse power supply (10) that applies a predetermined impulse voltage, A current detector (12) detects the value of the current that flows when a predetermined impulse voltage is applied to the coil, In the relationship between the frequency and current spectrum obtained by analyzing the current value detected by the current detector, the frequency band differs from a predetermined low-frequency band in which partial discharge occurring in the thin coating (26) portion of the wire constituting the coil can be detected. A specified frequency range between 300MHz and 500MHz. The sum of the current values ​​in the high-frequency band is calculated as the high-frequency spectrum area, and the creepage discharge generated in the coil is determined by the size of this high-frequency spectrum area. Distinguish from partial discharge An insulation testing apparatus comprising a determination device (13) configured to perform an inspection.

[0010] According to this, in the insulation inspection of the coils of an armature, by using a high-frequency band different from the predetermined low-frequency band used for partial discharge inspection, surface discharge and partial discharge can be inspected separately (i.e., distinguished). Therefore, this insulation inspection device can accurately distinguish between armatures with insulation defects that produce surface discharge and armatures that do not produce surface discharge.

[0011] Note that the reference numerals in parentheses attached to each component etc. indicate an example of the correspondence relationship between the component etc. and the specific components etc. described in the embodiments described later.

Brief Description of Drawings

[0012] [Figure 1] It is a perspective view showing a stator as an example of an armature to be inspected by an insulation inspection device according to an embodiment. [Figure 2] It is a schematic configuration diagram showing an example of the connection state between an insulation inspection device according to an embodiment and the coils of a stator. [Figure 3] It is a schematic configuration diagram showing another example of the connection state between an insulation inspection device according to an embodiment and the coils of a stator. [Figure 4] It is a cross-sectional view showing two wire rods having a normal film in the coil. [Figure 5] It is a cross-sectional view showing two wire rods having a thin film in the coil. [Figure 6] It is a cross-sectional view showing two wire rods in a state where pinholes in the film of the wire rod and voids in the resin part are connected in the coil. [Figure 7] It is an enlarged view of part VII of FIG. 6. [Figure 8] It is a flowchart for explaining the determination process executed by the determination device. [Figure 9] It is a flowchart for explaining another determination process executed by the determination device. [Figure 10] It is a diagram showing a state where a test piece is connected to a position far from the neutral point among the wire rods of each phase constituting the coil in a verification test of the effectiveness of the insulation inspection by the insulation inspection device. [Figure 11] In the verification test shown in FIG. 10, the current value detected when a predetermined instruction voltage is applied to the coil is subjected to fast Fourier transform, and it is a graph showing the relationship between the frequency and the current value spectrum. [Figure 12]It is a graph showing the relationship between the indicated voltage applied to each of the coil connected with the test piece and the good coil, and the low-frequency spectrum area in the demonstration test shown in FIG. 10. [Figure 13] It is a graph showing the relationship between the indicated voltage applied to each of the coil connected with the test piece having voids or the like and the good coil, and the high-frequency spectrum area in the demonstration test shown in FIG. 10. [Figure 14] It is a figure showing a state where a test piece is connected to a position close to the neutral point among the wire materials of each phase constituting the coil in the demonstration test of the effectiveness of the insulation inspection by the insulation inspection device. [Figure 15] It is a graph showing the relationship between the frequency and the current value spectrum obtained by performing fast Fourier transform on the current value detected when a predetermined indicated voltage is applied to the coil in the demonstration test shown in FIG. 14. [Figure 16] It is a graph showing the relationship between the indicated voltage applied to each of the coil connected with the test piece and the good coil, and the low-frequency spectrum area in the demonstration test shown in FIG. 14. [Figure 17] It is a graph showing the relationship between the indicated voltage applied to each of the coil connected with the test piece and the good coil, and the high-frequency spectrum area in the demonstration test shown in FIG. 14. [Figure 18] It is a figure showing a state where a good coil is arranged as an inspection target in the demonstration test of the effectiveness of the insulation inspection by the insulation inspection device. [Figure 19] It is a figure showing a state where a test piece is connected to a position far from the neutral point of the coil in the demonstration test of the effectiveness of the insulation inspection by the insulation inspection device. [Figure 20] It is a figure showing a state where a test piece is connected to a position closer to the neutral point of the coil compared to FIG. 19 in the demonstration test of the effectiveness of the insulation inspection by the insulation inspection device. [Figure 21] It is a figure showing a state where a test piece is connected to a position closer to the neutral point of the coil compared to FIG. 20 in the demonstration test of the effectiveness of the insulation inspection by the insulation inspection device. [Figure 22] Figures 18 to 21 show graphs illustrating the relationship between the indicated voltage applied to a good coil and a coil connected to a test piece, and the low-frequency spectrum area, in the demonstration tests. [Figure 23] Figures 18 to 21 show graphs illustrating the relationship between the indicated voltage applied to a good coil and a coil connected to a test piece, and the high-frequency spectrum area, in the demonstration tests. [Modes for carrying out the invention]

[0013] (One embodiment) The following describes one embodiment of the present invention with reference to the drawings. As shown in Figures 1 to 3, the insulation inspection device of this embodiment is a device that inspects the insulation state of a coil 2 provided on a stator 1, which is an example of an armature constituting a rotating electric machine (not shown). The rotating electric machine is used, for example, as a motor generator mounted on an electric vehicle.

[0014] The stator 1 comprises a ring-shaped core 3 (i.e., the stator core) and coils 2 inserted into slots (not shown) provided in the core 3. The coils 2 are three-phase coils that constitute part of a three-phase AC circuit. As shown in Figure 1, the stator 1 also includes a resin part 5 that molds the portion of the coil 2 that protrudes from the slots in the core 3 (i.e., the coil end). The resin part 5 is formed during the manufacturing process by immersing the coil end in molten resin. Power lines 6a, 7a, and 8a, which are electrically connected to the wires of the U-phase coil 6, V-phase coil 7, and W-phase coil 8 that constitute the three-phase coil 2, protrude from a part of the resin part 5 in which the coil end is molded.

[0015] As shown in Figure 2, the insulation testing device includes an impulse power supply 10, a current detector 12, a determination device 13, and the like. The coil 2 of the stator 1 to be tested will be described using a Y-connected coil as an example.

[0016] The impulse power supply 10 applies a predetermined impulse voltage to the coil 2 according to the indicative voltage. Figure 2 shows an example of how to connect the wiring 14 and 15 extending from the impulse power supply 10 to the coil 2. In the connection method shown in Figure 2, one wiring 14 extending from the impulse power supply 10 is connected to the end of a power line 6a extending from a predetermined phase of the coil 2 (for example, the U-phase coil 6). The other wiring 15 extending from the impulse power supply 10 branches off midway and is connected to the ends of two power lines 7a and 8a extending from two other phases of the coil 2 (for example, the V-phase coil 7 and the W-phase coil 8), respectively.

[0017] Note that the method of connecting the wires 14 and 15 extending from the impulse power supply 10 to the coil 2 is not limited to that shown in Figure 2. For example, as shown in Figure 3, the other wire 15 extending from the impulse power supply 10 may be connected to the end of a power line 8a extending from another phase of the coil 2 (for example, the W-phase coil 8).

[0018] When a predetermined impulse voltage is applied to coil 2 from impulse power supply 10, partial discharge and surface discharge may occur between a wire of a predetermined phase connected to a power line connected to one wiring 14 extending from impulse power supply 10 and a wire of another phase connected to a power line connected to the other wiring 15 extending from impulse power supply 10. The insulation testing device detects these partial discharges and surface discharges separately (i.e., distinguishes between them).

[0019] Figures 4 to 7 show cross-sections of a predetermined phase wire (hereinafter referred to as the "+ side wire" for convenience) connected to a power line connected to one of the wirings 14 extending from the impulse power supply 10, and a different phase wire (hereinafter referred to as the "- side wire" for convenience) connected to a power line connected to the other wiring 15 extending from the impulse power supply 10.

[0020] Specifically, Figure 4 shows a wire having a coating 20 (i.e., a normal coating) formed with a thickness within the design range. In this case, even if a predetermined impulse voltage is applied to the coil 2 from the impulse power supply 10, no discharge will occur.

[0021] Figure 5 shows a wire having a coating (i.e., a thin coating 26) that is thinner than the design thickness. In this case, when a predetermined impulse voltage is applied to the coil 2 from the impulse power supply 10, a partial discharge occurs between the metal part 21 constituting the positive wire and the metal part 21 constituting the negative wire, via the thin coating 26.

[0022] Figures 6 and 7 show how the metal part 21 constituting the positive wire and the metal part 21 constituting the negative wire are connected in space (i.e., by the void 23 and pinhole 22) due to a pinhole 22 formed in the wire coating 20 and a void 23 formed in the resin part 5 that molds the wire. In this case, when a predetermined impulse voltage is applied to the coil 2 from the impulse power supply 10, creepage discharge occurs along the inner wall surface of the space (i.e., the void 23 and pinhole 22) between the metal part 21 constituting the positive wire and the metal part 21 constituting the negative wire, as shown by the dashed arrow.

[0023] As shown in Figures 2 and 3, the current detector 12 detects the current value that flows when a predetermined impulse voltage is applied to the coil 2 from the impulse power supply 10. The current value detected by the current detector 12 is transmitted to the determination device 13.

[0024] The determination device 13 consists of a computer having a processor and memory, and an oscilloscope. The determination device 13 is configured so that the processor executes a program stored in memory to determine whether or not surface discharge has occurred in the coil 2.

[0025] The determination process performed by the determination device 13 will be explained with reference to the flowchart in Figure 8.

[0026] In step S10, the determination device 13 acquires the current value detected by the current detector 12 when a predetermined impulse voltage is applied to the coil 2 from the impulse power supply 10.

[0027] Next, in step S20, the determination device 13 analyzes the current value detected by the current detector 12 using the speed of light Fourier transform (FFT) or the like, to obtain a relationship between frequency and current value spectrum, for example, as shown in the graph in Figure 11. The current value spectrum is also called current amplitude or current intensity.

[0028] Next, in step S30 of Figure 8, the determination device 13 calculates the sum (i.e., integral) of the current value spectrum in the high-frequency band in which surface discharge can be detected. In this embodiment, the frequency band in which surface discharge can be detected is set to a higher frequency band than the frequency band in which partial discharge can be detected. The determination device 13 calculates the sum of the current value spectrum in the high-frequency band in which surface discharge can be detected.

[0029] In this specification, the sum of the current value spectra in the high-frequency band set for surface discharge testing is referred to as the "high-frequency spectrum area." Similarly, the sum of the current value spectra in the low-frequency band set for partial discharge testing is referred to as the "low-frequency spectrum area."

[0030] Next, in step S40, the determination device 13 determines whether the high-frequency spectrum area is greater than a predetermined first threshold. The first threshold is set in advance through experiments or other means and stored in the memory of the determination device 13.

[0031] If the high-frequency spectrum area is greater than a predetermined first threshold in step S40, the process proceeds to step S50. In step S50, the determination device 13 determines that surface discharge is occurring in the coil 2.

[0032] On the other hand, if the high-frequency spectrum area is smaller than a predetermined first threshold in step S40, the process proceeds to step S60. In step S60, the determination device 13 determines that no surface discharge has occurred in the coil 2.

[0033] In this way, the determination device 13 can determine whether or not surface discharge has occurred in the coil 2.

[0034] Furthermore, the determination device 13 of this embodiment can also perform a determination process other than the determination process described above.

[0035] Another determination process performed by the determination device 13 will be explained with reference to the flowchart in Figure 9. In this determination process, the determination device 13 can inspect not only the surface discharge occurring in the coil 2 but also the partial discharge occurring in the coil 2 in a single inspection.

[0036] In the flowchart of Figure 9, steps S10 to S60 are the same as those explained with reference to the flowchart of Figure 8.

[0037] In step S70, following step S20, the determination device 13 calculates the sum (i.e., integral) of the current value spectrum in a predetermined low-frequency band in which partial discharge can be detected. In this embodiment, the frequency band in which partial discharge can be detected is a lower frequency band than the frequency band in which creepage discharge can be detected. Therefore, the determination device 13 calculates the sum (i.e., low-frequency spectrum area) of the current value spectrum in the low-frequency band in which partial discharge can be detected.

[0038] Next, in step S80, the determination device 13 determines whether the low-frequency spectrum area is greater than a predetermined second threshold. The second threshold is set in advance through experiments or other means and stored in the memory of the determination device 13.

[0039] If the low-frequency spectrum area is greater than the second threshold in step S80, the process proceeds to step S90. In step S90, the determination device 13 determines that a partial discharge is occurring in coil 2.

[0040] On the other hand, if the low-frequency spectrum area is smaller than the second threshold in step S80, the process proceeds to step S100. In step S100, the determination device 13 determines that no partial discharge has occurred in the coil 2.

[0041] Thus, the determination device 13 can inspect not only the surface discharge generated in the coil 2 but also the partial discharge generated in the coil 2 in a single inspection.

[0042] (Demonstration test) Next, we will describe the demonstration test conducted to demonstrate the effectiveness of the insulation test using the insulation testing device described above. In this demonstration test, a test piece 25 was connected to the coil 2 of the stator 1. As shown in Figures 6 and 7, the metal parts 21 of the two wires of the test piece 25 were connected in space (i.e., through the void 23 and pinhole 22) due to a pinhole 22 in the insulation 20 of the wire and a void 23 in the resin part 5 that molds the wire.

[0043] (Demonstration Test 1) As shown in Figure 10, in demonstration test 1, a power line 6a extending from a predetermined phase (e.g., U-phase coil 6) of the coil 2 of the stator 1 was connected to one wire of the test piece 25, and a power line 7a extending from another phase (e.g., V-phase coil 7) of the coil 2 was connected to the other wire of the test piece 25. In this state, a predetermined impulse voltage was applied to the coil 2 from the impulse power supply 10, and the current value detected by the current detector 12 was subjected to a speed-of-light Fourier transform (FFT) to analyze the relationship between frequency and current value spectrum. The results are shown by the dashed line A in the graph of Figure 11.

[0044] Furthermore, in this demonstration test 1, even for stator 1 coils 2 that did not have test pieces 25 connected to them (i.e., good products), a predetermined impulse voltage was applied to coil 2 from the impulse power supply 10, and the relationship between frequency and current spectrum was analyzed by performing a Fourier transform (FFT) on the current value detected by the current detector 12. The results are shown by the solid line B in the graph of Figure 11.

[0045] Then, for the stator 1 connected to the test piece 25 shown by dashed line A in Figure 11, and for the good stator 1 shown by solid line B, the sum of the current value spectrum in a predetermined low-frequency band (e.g., 20 to 200 MHz) in which partial discharge can be detected (i.e., the low-frequency spectrum area) was calculated. In addition, for the stator 1 connected to the test piece 25 shown by dashed line A in Figure 11, and for the good stator 1 shown by solid line B, the sum of the current value spectrum in a predetermined high-frequency band (e.g., 300 to 500 MHz) in which creepage discharge can be detected (i.e., the high-frequency spectrum area) was calculated.

[0046] In Demonstration Test 1, the impulse voltage applied from the impulse power supply 10 to the coil 2 was increased from a predetermined indicator voltage (e.g., 1kV) to a higher indicator voltage (e.g., 4.5kV), in increments of 100V, and each increase was performed 10 times. The analysis results shown in the graph of Figure 11 are for when the indicator voltage is V_1 in the graphs of Figures 12 and 13, which will be explained next.

[0047] The graph in Figure 12 shows the indicator voltage (hereinafter simply referred to as "indicator voltage") of the impulse voltage applied from the impulse power supply 10 to the coil 2 on the horizontal axis, and the low-frequency spectrum area on the vertical axis.

[0048] The solid line C in Figure 12 shows the relationship between the indicated voltage and the low-frequency spectrum area in a test conducted using a test piece 25 connected to the coil 2 of the stator 1.

[0049] The shaded area between solid lines D and E in Figure 12 represents the average ±4 × variation range (hereinafter referred to as the "4σ range") of 10 tests conducted using a stator 1 with no test piece 25 connected to its coil 2 (i.e., a good product). Specifically, solid line D is +4σ and solid line E is -4σ.

[0050] As shown in Figure 12, when examining the low-frequency spectrum area, the solid line C, which represents the test results when the test piece 25 is connected to coil 2, is close to the ±4σ range of the test results when using a good product between the indicated voltages V_2 to V_4. At indicated voltage V_3, solid lines C and D are in contact. Therefore, even when calculating the low-frequency spectrum area, it is difficult to distinguish between a stator 1 with coil 2 that produces surface discharge and a good stator 1. Consequently, in inspection methods that calculate the low-frequency spectrum area (i.e., conventional partial discharge inspection methods), there is a risk that a stator 1 with coil 2 that produces surface discharge may be mixed in with a good stator 1.

[0051] In contrast, the graph in Figure 13 shows the indicated voltage on the horizontal axis and the high-frequency spectrum area on the vertical axis.

[0052] The solid line F in Figure 13 shows the relationship between the indicated voltage and the high-frequency spectrum area in a test conducted using a test piece 25 connected to the coil 2 of the stator 1.

[0053] The shaded area between solid lines G and H in Figure 13 represents the average ±56 × variation range (hereinafter referred to as the "56σ range") of 10 tests conducted using a stator 1 with no test piece 25 connected to its coil 2 (i.e., a good product). Specifically, solid line G is +56σ and solid line H is -56σ.

[0054] As shown in Figure 13, when examining the high-frequency spectrum area, the solid line F, which represents the test results using a coil 2 connected to a test piece 25, deviates significantly from the ±56σ range of the test results using a good product between the indicated voltages V_1 to V_4. This demonstrates that by applying an impulse voltage to the coil 2 between the indicated voltages V_1 to V_4 and calculating the high-frequency spectrum area, it is possible to reliably distinguish between a stator 1 having a coil 2 that produces surface discharge and a good stator 1. Therefore, the inspection method that calculates the high-frequency spectrum area can prevent a stator 1 having a coil 2 that produces surface discharge from being mixed in with a good stator 1.

[0055] (Demonstration Test 2) Next, as shown in Figure 14, in demonstration test 2, one wire of the test piece 25 was connected to a point near the neutral point 9 in a predetermined phase (e.g., U-phase coil 6) of the stator 1, and the other wire of the test piece 25 was connected to a point near the neutral point 9 in another phase (e.g., V-phase coil 7) of the coil 2. In this state, a predetermined impulse voltage was applied to the coil 2 from the impulse power supply 10, and the relationship between frequency and current value spectrum was analyzed by performing a Fourier transform (FFT) on the current value detected by the current detector 12. The results are shown by the dashed line I in the graph of Figure 15.

[0056] In this demonstration test 2, a predetermined impulse voltage was applied to the coil 2 of the stator 1 from the impulse power supply 10 (i.e., a good product) without the test piece 25 connected to it. The current value detected by the current detector 12 was then subjected to a speed-of-light Fourier transform (FFT) to analyze the relationship between frequency and current value spectrum. The results are shown by the solid line J in the graph of Figure 15.

[0057] Then, for the stator 1 with the test piece 25 connected as shown by the dashed line I in Figure 11, and for the good stator 1 shown by the solid line J, the sum of the current value spectrum in a predetermined low-frequency band (e.g., 20-200 MHz) in which partial discharge can be detected (i.e., the low-frequency spectrum area) was calculated. In addition, for the stator 1 with the test piece 25 connected as shown by the dashed line I in Figure 11, and for the good stator 1 shown by the solid line J, the sum of the current value spectrum in a predetermined high-frequency band (e.g., 300-500 MHz) in which creepage discharge can be detected (i.e., the high-frequency spectrum area) was calculated. Note that the closer the test piece 25 is connected to the neutral point 9 in each phase coil 2, the more the impulse voltage drops due to the resistance of the wire, making detection more difficult.

[0058] In demonstration test 2, the impulse voltage applied from the impulse power supply 10 to the coil 2 was changed from a predetermined indicator voltage (e.g., 1kV) to a larger indicator voltage (e.g., 4.5kV) in increments of 100V, for example, 10 times for each increment. The graph in Figure 15 above is for when the indicator voltage is V_21, as shown in the graphs in Figures 16 and 17 which will be explained next.

[0059] The graph in Figure 16 shows the indicated voltage on the horizontal axis and the low-frequency spectrum area on the vertical axis.

[0060] The solid line K in Figure 16 shows the relationship between the indicated voltage and the low-frequency spectrum area in a test conducted using a test piece 25 connected to the coil 2 of the stator 1.

[0061] The shaded area between solid lines L and M in Figure 16 represents the average ±4 × variation range (hereinafter referred to as the "4σ range") of 10 tests conducted using a stator 1 with no test piece 25 connected to its coil 2 (i.e., a good product). Specifically, solid line L is +4σ and solid line M is -4σ.

[0062] As shown in Figure 16, when examining the low-frequency spectrum area, the solid line K, which represents the test results obtained by connecting the test piece 25 to the coil 2, overlaps with or is close to the ±4σ range of the test results obtained using a good product between the indicated voltages V_20 and V_25. This indicates that if a creepage discharge occurs near the neutral point 9 of the coil 2, it is difficult to distinguish between a stator 1 with a coil 2 exhibiting creepage discharge and a good stator 1, even when calculating the low-frequency spectrum area. Therefore, in inspection methods that calculate the low-frequency spectrum area (i.e., conventional partial discharge inspection methods), there is a risk that a stator 1 with a coil 2 exhibiting creepage discharge may be mixed in with a good stator 1.

[0063] In contrast, the graph in Figure 17 shows the indicated voltage on the horizontal axis and the high-frequency spectrum area on the vertical axis.

[0064] The solid line N in Figure 17 shows the relationship between the indicated voltage and the high-frequency spectrum area in a test conducted using a test piece 25 connected to the coil 2 of the stator 1.

[0065] The shaded area between solid lines O and P in Figure 17 represents the average ±15 × variation range (hereinafter referred to as the "15σ range") of 10 tests conducted using a stator 1 with no test piece 25 connected to its coil 2 (i.e., a good product). Specifically, solid line O is +15σ and solid line P is -15σ.

[0066] As shown in Figure 17, when examining the high-frequency spectrum area, the solid line N, which represents the test results using a coil 2 connected to a test piece 25, deviates significantly from the ±15σ range of the test results using a good product between the indicated voltages V_21 and V_24. This demonstrates that even when there is a location where surface discharge occurs near the neutral point 9 of coil 2, applying an impulse voltage to coil 2 between the indicated voltages V_21 and V_24 and calculating the high-frequency spectrum area can reliably distinguish between a stator 1 with a coil 2 exhibiting surface discharge and a good stator 1. Therefore, the inspection method that calculates the high-frequency spectrum area can prevent stators 1 with a coil 2 exhibiting surface discharge from being mixed in with good stators 1.

[0067] (Summary of the demonstration test) As shown in Figures 18 to 21, the same tests as those described in demonstration tests 1 and 2 above were performed by connecting test pieces 25 to various positions in each phase of the coil 2 of the stator 1. As mentioned above, the closer the test piece 25 is to the neutral point 9 in each phase of the coil 2, the greater the voltage drop in the impulse voltage due to the resistance of the wire, making detection more difficult. In the following explanation, the coil shown in Figure 18 will be called a "good product," the one shown in Figure 19 will be called "easy to detect," the one shown in Figure 20 will be called "medium to detect," and the one shown in Figure 21 will be called "difficult to detect."

[0068] Figure 22 shows the indicated voltage on the horizontal axis and the low-frequency spectrum area on the vertical axis. The solid line Q indicates a "good product," the dashed line R indicates an "easy detection difficulty," the dashed line S indicates a "medium detection difficulty," and the dashed line T indicates a "difficult detection difficulty."

[0069] As shown in Figure 22, even when calculating the low-frequency spectrum area, it is difficult to distinguish between a stator 1 having a coil 2 that produces surface discharge and a good stator 1. Therefore, in inspection methods that calculate the low-frequency spectrum area (i.e., conventional partial discharge inspection methods), there is a risk that a stator 1 having a coil 2 that produces surface discharge may be mixed in with a good stator 1.

[0070] In contrast, Figure 23 shows the indicated voltage on the horizontal axis and the high-frequency spectrum area on the vertical axis. The solid line U indicates a "good product," the dashed line V indicates an "easy detection difficulty," the dashed line W indicates a "medium detection difficulty," and the dashed line X indicates a "difficult detection difficulty."

[0071] As shown in Figure 23, by applying an impulse voltage to coil 2 at an indicator voltage V_31 or higher and calculating the high-frequency spectrum area, it is possible to reliably distinguish between a stator 1 having a coil 2 that produces surface discharge and a good stator 1. Therefore, the inspection method that calculates the high-frequency spectrum area can prevent a stator 1 having a coil 2 that produces surface discharge from being mixed in with a good stator 1.

[0072] The insulation testing method and insulation testing apparatus of this embodiment described above have the following configuration and provide the following effects.

[0073] (1) The insulation inspection method of this embodiment is a method for inspecting surface discharge occurring in a coil 2, with at least a portion of the coil 2 of a stator 1 being molded with a resin part 5 as the inspection target, and includes the following: a predetermined impulse voltage is applied to the coil 2 and the current value flowing through a current detector 12 connected to the coil 2 is detected; the detected current value is analyzed and the relationship between frequency and current value spectrum is analyzed, and the sum of the current value spectra in a high-frequency band different from a predetermined low-frequency band in which partial discharge occurring in the thin film 26 portion of the wire material constituting the coil 2 can be detected is calculated as the high-frequency spectrum area, and the surface discharge occurring in the coil 2 is inspected based on the size of the high-frequency spectrum area.

[0074] According to this, in the insulation inspection of the coil 2 of the stator 1, by using a high-frequency band different from the predetermined low-frequency band used for partial discharge inspection, it is possible to inspect surface discharge and partial discharge separately (i.e., distinguish between them). Therefore, this insulation inspection method makes it possible to accurately distinguish between a stator 1 with insulation defects that produce surface discharge and a good stator 1 that does not produce surface discharge.

[0075] (2) The insulation inspection method of this embodiment further includes, in addition to inspecting surface discharge occurring in the coil 2 based on the size of the high-frequency spectrum area, calculating the sum of the current value spectra in a predetermined low-frequency band as the low-frequency spectrum area, and inspecting partial discharge occurring in the coil 2 based on the size of the low-frequency spectrum area.

[0076] According to this, in the insulation test of the coil 2 of the stator 1, partial discharge and surface discharge can be detected separately (i.e., distinguished) in a single test.

[0077] (3) The insulation inspection device of this embodiment comprises an impulse power supply 10, a current detector 12, and a determination device 13. The impulse power supply 10 applies a predetermined impulse voltage to the coil 2. The current detector 12 detects the current value that flows through the current detector 12 when a predetermined impulse voltage is applied to the coil 2. The determination device 13 is configured to calculate the sum of the current value spectra in a high-frequency band different from a predetermined low-frequency band in which partial discharge occurring at the thin coating 26 portion of the wire constituting the coil 2 can be detected, based on the relationship between the frequency and current value spectrum obtained by analyzing the current value detected by the current detector 12, as the high-frequency spectrum area, and to inspect the surface discharge occurring in the coil 2 based on the size of the high-frequency spectrum area.

[0078] According to this, the insulation testing device can distinguish between surface discharge and partial discharge by using a high-frequency band different from the predetermined low-frequency band used for partial discharge testing. Therefore, this insulation testing device can accurately distinguish between a stator 1 with insulation defects that produce surface discharge and a good stator 1 that does not produce surface discharge.

[0079] (4) In this embodiment, the determination device 13 is configured to inspect surface discharge occurring in the coil 2 based on the size of the high-frequency spectrum area, and to calculate the sum of the current value spectra in a predetermined low-frequency band as the low-frequency spectrum area, and to inspect partial discharge occurring in the coil 2 based on the size of the low-frequency spectrum area.

[0080] According to this, the insulation testing device can detect partial discharge and surface discharge separately (i.e., distinguish between them) in a single test.

[0081] (Other embodiments) (1) In each of the above embodiments, the armature to be inspected was described as a stator 1, but it is not limited to this, and the armature to be inspected may be a rotor having coils 2.

[0082] (2) In the above embodiment, the coil 2 of the stator 1 to be inspected was described as being Y-connected, but it is not limited to this, and may be Δ-connected, ΔY-connected, YΔ-connected, ΔΔ-connected, YY-connected, etc.

[0083] The present invention is not limited to the embodiments described above, and can be modified as appropriate within the scope of the claims. Furthermore, the embodiments and parts thereof are not unrelated to each other, and can be combined as appropriate, except in cases where the combination is clearly impossible. In addition, it goes without saying that the elements constituting the embodiments are not necessarily essential, except in cases where they are explicitly stated to be particularly essential or where they are clearly considered essential in principle. Furthermore, in the embodiments, when numerical values ​​such as the number, numerical values, quantities, or ranges of the components of the embodiments are mentioned, the invention is not limited to those specific numbers, except in cases where they are explicitly stated to be particularly essential or where they are clearly limited to a specific number in principle. Furthermore, when the shapes, positional relationships, etc., of the components, etc., are mentioned in the embodiments, the invention is not limited to those shapes, positional relationships, etc., except in cases where they are explicitly stated to be particularly essential or where they are clearly limited to a specific shape, positional relationship, etc., in principle. [Explanation of symbols]

[0084] 1. Stator (armature) 2 coils 5. Resin part 10 Impulse Power Supply 12 Current detector 13 Judgment device 21 Metal parts 26 Thin film

Claims

1. In an insulation testing method that inspects surface discharge occurring between the metal parts (21) of the wires of each phase constituting the coil via the space within the resin part (5) in which the coil end is molded, with the coil (2) of the armature having a coil (2) in which at least a portion of the coil end protruding from the slot of the stator core is molded with resin, A predetermined impulse voltage between 1 kV and 4.5 kV is applied to the coil, and the current value flowing through the current detector (12) connected to the coil is detected (S10). An insulation inspection method comprising: analyzing the relationship between the frequency and current spectrum obtained from the detected current value; calculating the sum of the current spectrums in a predetermined high-frequency band between 300 MHz and 500 MHz, which is different from a predetermined low-frequency band in which partial discharge occurring in the thin coating (26) portion of the wire constituting the coil can be detected, as the high-frequency spectrum area; and inspecting the creepage discharge occurring in the coil by distinguishing it from partial discharge based on the size of the high-frequency spectrum area (S20 to S60).

2. In addition to inspecting the surface discharge generated in the coil based on the size of the high-frequency spectrum area, The insulation inspection method according to claim 1, further comprising calculating the sum of the current value spectrums in a predetermined low-frequency band between 20 MHz and 200 MHz as the low-frequency spectrum area, and inspecting the partial discharge occurring in the coil in distinction from surface discharge based on the size of the low-frequency spectrum area (S70 to S100).

3. In an insulation testing device that inspects surface discharge occurring between the metal parts (21) of the wires of each phase constituting the coil via the space within the resin part (5) in which the coil end is molded, with the coil (2) of the armature having a coil (2) in which at least a portion of the coil end protruding from the slot of the stator core is molded with resin, An impulse power supply (10) applies a predetermined impulse voltage between 1 kV and 4.5 kV to the coil, A current detector (12) detects the current value that flows when a predetermined impulse voltage is applied to the coil, An insulation inspection apparatus comprising: a determination device (13) configured to analyze the relationship between frequency and current value spectrum obtained by analyzing the current value detected by the current detector, calculate the sum of current value spectra in a predetermined high-frequency band between 300 MHz and 500 MHz, which is different from a predetermined low-frequency band in which partial discharge occurring in the thin coating (26) portion of the wire constituting the coil can be detected, as the high-frequency spectrum area, and to distinguish creepage discharge occurring in the coil from partial discharge and inspect it based on the size of the high-frequency spectrum area.

4. In addition to inspecting the surface discharge generated in the coil based on the size of the high-frequency spectrum area, the determination device also: The insulation inspection device according to claim 3, configured to calculate the sum of the current value spectrums in a predetermined low-frequency band between 20 MHz and 200 MHz as the low-frequency spectrum area, and to distinguish and inspect partial discharges occurring in the coil from creepage discharges based on the size of the low-frequency spectrum area.

Images