Method for testing electronic component and electronic component having testing device
Through ultrasonic incentives combined with the meter signs, the manufacturing errors and aging problems in electronic components are identified, and the reliable quality monitoring and aging prediction of electronic components has been realized, and the aging of electronic components that are difficult to detect in existing technologies.
Patent Information
- Application Number
- CN202380045477.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-06-08
- Filing Date
- 2023-06-02
- Publication Date
- 2025-01-21
AI Technical Summary
The prior art is difficult to reliably identify and detect manufacturing errors and aging problems in electronic components, especially circuit board bending, large-area cavity, layering of conductor material and poor laser welding connections of metal load joints.
Ultrasonic incentives are used to combine the meter signs to measure and analyze the impedance spectrum of electronic components, and identify the material changes and aging signs of materials on the component.
This enables simple, low-cost and reliable quality monitoring of electronic components, enabling continuous aging measurements during operation to derive necessary maintenance measures and avoid unexpected failures.
Smart Images
Figure CN119343606A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for testing an electronic component and to an electronic component having a device for carrying out such a method. Background Art
[0002] When manufacturing electronic components, in particular after the assembly of the circuit carrier for the electronic component, the electronic components still need to be quality-monitored for manufacturing errors. The manufacturing errors that usually occur are defects. These defects are, in particular, large cavities in the soldered connections, circuit board bending, insufficient solder joint contact, delamination of conductor materials in the circuit board, flux residues, chip distortion and poor laser soldering of metal load terminals.
[0003] It is known to check electronic components, such as, in particular, assembled circuit carriers, for possible defects by means of optical imaging and / or electrical testing. However, for some defects, optical quality monitoring by means of automatic imaging is not reliable enough. In particular, circuit board bends are difficult to detect by means of optical imaging.
[0004] Large-area cavities, delaminations of conductor material in printed circuit boards and bad laser-welded connections of metallic load terminals cannot yet be detected optically and electrically with sufficient reliability.
[0005] However, quality monitoring is still required during the operation of electronic components, especially power electronic components, because faults may occur due to operation or aging. In particular, power electronic systems, such as power modules, age due to temperature cycles caused by operation. Thermally heavily loaded power electronic systems and their power switch modules, such as IGBT modules, will age significantly due to normal operation within their specified operating limits. For example, IGBT modules are used in power electronic converters.
[0006] This aging is mainly caused by a large number of load cycles of operating heating and cooling, which leads to the gradual destruction of the chip solder layer under the semiconductor chip and fatigue of the bonding wires and chip metallization that conduct the current into the chip for switching and rectification. Common failure mechanisms are bond detachment (i.e. lifting until the bonding wire breaks), heel cracks at the bottom of the solder joint, and solder fatigue of the chip and system solder (including the bottom solder of the ceramic carrier). In particular, the problem is that the point in time when the failure occurs is strongly and in a complex way related to the past loads, so that even advanced prediction models are difficult to predict accurately.
[0007] Furthermore, aging can also occur in the chip periphery due to temperature fluctuations. Thus, on the one hand, the circuit carrier (DCB, AMB) can be destroyed due to different expansion coefficients (ceramic / copper) above the conchoidal crack surface, and on the other hand, the drive control located thereon can also be destroyed, especially in future integrated power modules.
[0008] Currently, for example, model-based estimation methods are used, but these are very inaccurate. Another possibility is impedance spectroscopy, which determines the thermal impedance of the module structure. Here, electronic switches are used as heating elements and sensors. Such measurements can only be made when the system is not in operation. Another option is the distortion measurement of electrical excitations caused by interfaces and transitions. These measurements are also carried out during operation using high-frequency signals. Since these signals are very weak, the results need to be interpreted.
[0009] In areas of use that are difficult to access or that are in permanent operation, where converter failures must be avoided absolutely, such as offshore wind turbines or in traction applications, there are currently different approaches to avoid unexpected operating stops as systematically as possible. Spontaneous stops can therefore be counteracted by redundancy of power electronics or by measures at the maintenance level, such as preventive replacement of power modules.
[0010] In this case, the expected and already used service life can be inferred with the aid of advanced thermal models based on previous accelerated service life tests. However, due to the high complexity of the dependencies, these calculations always have very large errors.
[0011] Some power module manufacturers offer IGBT power switch modules in which a small measuring diode is integrated as a temperature sensor in the IGBT. At a given operating current, the (solder) aging can thus be indirectly inferred through the increased temperature rise of the IGBT. However, this is accompanied by a loss of chip area and higher chip costs. Summary of the invention
[0012] Therefore, against the background of the prior art, the technical problem to be solved by the present invention is to provide an improved method for testing electronic components and an electronic component having a device for implementing the improved method. In particular, the method for testing electronic components according to the present invention can be used to perform simple, low-cost and reliable testing of electronic components.
[0013] The above object of the invention is achieved by a method for testing an electronic component having the features specified in claim 1 and by an electronic component having a device for carrying out such a method for testing having the features specified in claim 11 .
[0014] Preferred embodiments of the invention are specified in the associated dependent claims, the following description and the drawings.
[0015] The method according to the invention for quality testing of electronic components, in particular power electronic components, comprises checking at least one component of the electronic component by means of an electrical characterization method, wherein the component being checked is excited by means of ultrasound, so that a change in a signal of the electrical characterization method is caused thereby. The change in the electrical signal caused by the ultrasound is evaluated by means of a signal analysis. This indicates a material change on the component.
[0016] The method according to the present invention thus realizes condition monitoring, on which the remaining service life determination can be based. Accidental failures are avoided, and predictive maintenance is enabled. In addition, the following advantage is obtained, namely, a commercial solution can be provided for the condition monitoring of the entire power module in the converter, which can be used for service life prediction. The errors to be advantageously detected by the given quality monitoring are particularly defects, such as material defects, but these defects may occur after manufacturing, especially due to aging. In particular, damage caused by excessive component temperature, especially temperature changes, can be detected very reliably using the described method. Therefore, although the method according to the present invention is also very suitable for component or assembly inspection after manufacturing, it is also advantageously used for aging prediction and failure risk prediction after manufacturing based on the results of the inspection. The method is particularly advantageous in that it can be used when electronic devices or electronic components are running. The inspection that can be carried out during operation and the inspection that can be repeated bring the special advantage that aging prediction or failure risk prediction can be derived therefrom.
[0017] An electronic component is to be understood in particular as a power electronic component, such as a power module. An electronic component comprises at least one component, in particular a semiconductor device, such as a power switch. The structure of the electronic component therefore comprises different layers and generally different materials, which are also produced and combined using different manufacturing processes. Material changes may be understood as damage, delamination, cracks and general material aging.
[0018] Typical components of power electronic devices are, for example: bipolar power transistors, such as switching power supplies and DC / DC converters, power MOSFETs, IGBTs, in particular switching power supplies, motor controllers and converters, as well as thyristors, in particular converters, semiconductor relays, pulse current sources, thyristor controllers, GTO thyristors (these are all high-power converters), as well as triacs, in particular dimmers or semiconductor relays, as well as diodes for rectification and as freewheeling diodes, such as Schottky diodes or silicon diodes, and finally power capacitors.
[0019] Fields of use of power electronics devices may be, for example, switched mode power supplies, frequency converters, high-frequency generators, DC regulators, phase interface controls, high-voltage DC transmission, inverters or semiconductor relays.
[0020] Another advantage of the method according to the invention is that, for example, measurements can be made not only during operation and repeatedly during operation, but even continuous aging measurements are possible. Thus, the state of the entire module structure can be measured continuously during operation and necessary measures can be derived therefrom. Such a solution can be used commercially for the state monitoring of the entire power module in a converter. The quality inspection method presented is reliable with regard to the prediction of the service life.
[0021] The method enables an improved, non-destructive, more reliable detection of aging-related damage or delamination on a power electronics assembly and on individual components of the assembly, in particular power modules of a power electronics assembly, which can be performed during operation of the component.
[0022] The method according to the invention for quality testing of power electronics components has the advantage, among other things, that optically inaccessible regions of the components and components of the components can also be checked for production errors after assembly.
[0023] Impedance spectroscopy is preferably used as the electrical characterization method. In an advantageous embodiment of the method, at least one component of the power electronics assembly is therefore checked by measuring at least one impedance spectrum of the component and a change in the impedance spectrum due to ultrasonic excitation. The impedance spectra are compared by means of a signal analysis, wherein the determined changes indicate damage or delamination on the component. In particular, the signal analysis is functionally coordinated with the measurement: the frequency-dependent impedance is measured in a comparison, once with additional ultrasonic (US) excitation and once without additional ultrasonic excitation. The difference signal is evaluated.
[0024] In electrical distortion measurements, due to the nonlinear electrical, thermal and mechanical properties of high-frequency transmission in power electronics systems, high-frequency signals are generated that deviate significantly from their original frequency. In this case, frequency signals are generated that correspond to multiples of the original frequency or, when multiple frequencies are fed in, to the difference frequency or the sum frequency.
[0025] In an advantageous embodiment of the method, the ultrasound waves are generated by means of electrical excitation. For example, the ultrasound signal is coupled into the component or assembly to be characterized by means of electrical excitation. Electrical distortions are then generated at the interface of the connecting layers (welding, bonding, etc.) due to nonlinear effects caused by local temperature, mechanical delamination and transition resistance.
[0026] In another advantageous embodiment of the method according to the invention, the ultrasonic signal component is modulated into an intermodulation electrical signal. The electrical signal is the output signal of the electrical characterization method. This results in the advantage that errors in the tested structure are first amplified, thereby making it possible to better identify these errors.
[0027] In the method according to the invention, in one of its embodiments, a signal analysis is advantageously performed, by means of which the thermo-mechanical and mechanical components of the wear are separated. This is possible because the mechanical signal, which indicates the presence of a defect, is amplified in particular by the ultrasonic signal.
[0028] In another advantageous embodiment of the method according to the invention, an electronic component is checked which has a planar structure and generates ultrasound waves by means of an ultrasound generator placed directly on the plane AVT. AVT is to be understood as a planar structure in mounting and connection technology. In this case, it is particularly preferred to use an ultrasound generator which is directly integrated in the planar structure, in particular as a piezoelectric layer, in the method. In one of the described embodiments, ultrasound waves are coupled into the power electronic component as near-surface waves, for example.
[0029] Alternatively or additionally, in another advantageous embodiment of the method according to the invention, ultrasound waves can be generated as volume waves in the electronic component. In particular, ultrasound waves are generated at multiple locations of the power electronic component for this purpose. For this purpose, multiple ultrasound generators are present, in particular in a planar structure.
[0030] In a further advantageous embodiment of the method, the electronic component is exposed to ultrasonic vibrations continuously or once or at discrete points in time.
[0031] The electronic component according to the invention, in particular a power electronic component, in particular a circuit carrier, is designed to implement the method according to the invention described above. For this purpose, the electronic component comprises: at least one ultrasonic source, which is configured and arranged to subject the electronic component to ultrasonic vibrations; a measuring device, which is designed to perform electrical characterization, in particular frequency-dependent impedance measurement; and an evaluation device, which is configured to infer defects of the electronic component by means of the detected changes in the electrical signal.
[0032] The component is in particular a power electronics component, in particular a circuit carrier, ie at least one component of the component is a power switch.
[0033] For example, a spectrum analyzer or a lock-in amplifier can be used as a measuring device designed for frequency-dependent impedance measurement.
[0034] In an advantageous embodiment of the invention, the electronic assembly according to the invention has a planar structure and comprises an ultrasonic generator placed directly on the plane AVT. For example, a piezoelectric actuator or an ultrasonic probe can be used for radiating ultrasonic signals close to the surface.
[0035] In an alternative advantageous design of the present invention, the electronic component according to the present invention has a planar structure and comprises an ultrasonic generator integrated in the planar structure, in particular as a piezoelectric layer. Preferably, a silicon carbide semiconductor or a silicon carbide MOS (Metal oxide semiconductor) or gallium nitride is used as the ultrasonic generator.
[0036] In another advantageous embodiment of the invention, the electronic component according to the invention alternatively or additionally has an ultrasonic generator, in particular integrated as a piezoelectric layer in a component of the power electronic component, which has piezoelectric properties. In other words, it comprises at least one ultrasonic generator, which in another functional aspect is a component of the power electronic component, which comprises a piezoelectric layer. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Refer to the attached figure Figure 1 and Figure 2 Examples and embodiments of the present invention are further described in an exemplary manner:
[0038] Figure 1 Schematically shows a flow chart of the measurement signal evaluation of the method according to the invention,
[0039] Figure 2 An assembly with a corresponding device for carrying out the method according to the invention is schematically shown. DETAILED DESCRIPTION
[0040] In the method according to the invention for testing an electronic component 10, the electronic component 10 is exposed to ultrasonic vibrations 110. The ultrasonic vibrations 110 extend regularly in the frequency range of 10 to 300 kHz. At the same time, the electronic component 10 is checked by means of an electrical characterization method, in particular an impedance spectroscopy method. Preferably, in the method according to the invention, an impedance spectrum Z of the electronic component 10 or only one electronic connection 80, 90 or one component 50, 60 that is changed by the ultrasonic vibrations 110 is collected and used. us [Ω], to infer defects. In order to evaluate 210, as in Figure 1 The flow chart schematically shows that the impedance spectrum Z us [Ω] Impedance spectrum Z with and without ultrasonic excitation o[Ω] is compared. The measurement signal evaluation is preferably performed in an evaluation unit 210, which is designed to receive and evaluate the electrical measurement signal. With the method provided, defects of the electronic component 10, such as, in particular, local defect locations 95, which cannot be detected using optical or electronic methods known from the prior art, can advantageously be checked.
[0041] Preferably, in the method according to the invention, the electronic component 10 is exposed to ultrasonic vibrations 110 once. For this purpose, the electronic component 10 is excitably excited by means of a temporally limited ultrasonic train and thus vibrates the electronic component 10. Alternatively and also preferably, the electronic component 10 can be repeatedly exposed to ultrasonic vibrations at discrete time points. For example, a plurality of temporally localized ultrasonic trains are used for this purpose. In this way, the gradually weakening ultrasonic vibrations 110 of the electronic component 10 can also be evaluated. In particular, information about local attenuation and self-resonance can be obtained from the gradually weakening ultrasonic vibrations 110. Alternatively or additionally, the electronic component 10 can also be continuously exposed to ultrasonic vibrations 110. Therefore, in particular, the change in the resonant frequency of the self-vibration of the electronic component 10 or the components 50, 60, 80, 90 of the electronic component can be taken into account to infer the presence of a defect and / or the location 95 of such a defect. Preferably, in the method according to the invention, the electronic component 10 is disposed in ultrasonic vibrations 110 in a plurality of regions 120 of the electronic component 10. Thus, exposing the electronic component 10 to the ultrasonic vibrations 110 in a plurality of regions of the electronic component 10 can provide information about the spatial localization of the defect 95. Furthermore, it can advantageously be ensured that all relevant self-vibrations of the electronic component 10 are excited when the electronic component 10 is not only exposed to the ultrasonic vibrations 110 in a single region. Thus, in this embodiment, the reliability of the method according to the invention is further increased.
[0042] In particular, defective connections 80, 90 of different materials lead to a change in the vibration behavior of the electronic component 10 during ultrasonic vibrations 110. In particular, large-area soldered connections 80, 90 of components 50, 60 or laser welding of load connections have a changed vibration behavior in the event of a defect. In addition, with the method according to the invention, in particular, defective metallizations 40 of the circuit carrier 20 or bending of the circuit carrier 20 can also be easily detected. Therefore, with the method according to the invention, an especially reliable inspection of the electronic component 10 is possible.
[0043] For example, the electronic component 10 is a circuit carrier, or the electronic component has a circuit carrier 20. The electronic component 10 often consists of different components 50, 60 or parts, and correspondingly consists of different materials, in particular structural and connection technology materials, in which defects in the form of cavities, cracks or disconnected connections often occur. In this way, especially in the important case where the electronic component 10 includes a circuit carrier 20, it is possible to reliably and effectively test for defects.
[0044] Figure 2 The schematic diagram of an electronic component 10 is shown schematically, which is arranged and configured for carrying out the method according to the invention for testing an electronic component 10. The electronic component 10 comprises, for example, an ultrasonic source 120, which is configured and arranged for subjecting the electronic component 10 to ultrasonic vibrations 110. Furthermore, the electronic component 10 comprises a device 200 for performing an impedance measurement or electrical contacts for connecting the device 200 for performing an impedance measurement. The evaluation device 210 (which is arranged in Figure 2 10) by means of the impedance spectrum Z influenced by ultrasound us [Ω] A defect 95 of the electronic component 10 is inferred.
[0045] Preferably, the method is performed to test an electronic component in the form of a power module 10. The power module 10 has a circuit carrier in the form of a circuit board 20, which is formed as a flat part with a flat side 30. A copper metallization in the form of a copper layer 40 connected in a planar manner is located on the flat side 30. Semiconductor components 50, 60 and load connections 70 are connected to the copper layer 40. The semiconductor components 50, 60 are connected to the copper metallization 40 by means of solder layers 80, 90. The load connection 70 is soldered to the copper metallization 40 by means of laser welding 100. In the solder layers 80, 90 for connecting the semiconductor components 50, 60 to the copper metallization 40 and in the laser welding 100 for connecting the load connection 70 to the copper metallization 40, defects such as cavities or delaminations occur. In order to test the power module 10 for these defects, the power module 10 is subjected to ultrasonic vibrations 110. For this purpose, an ultrasonic source 120 is provided, which radiates ultrasonic waves 130 on the circuit board 20 in a direction 140 perpendicular to the flat side 30 to the side of the circuit board 20 facing away from the flat side 30. Here, the ultrasonic source 120 is arranged at the ends of the flat side 30 of the circuit board 20 that are away from each other, as seen in the planar extension direction of the flat side 30. That is, the power module 10 is exposed to ultrasonic vibrations 110 at multiple regions.
[0046] Aging in the power module can be seen by changes in the vibration behavior of the bonding wires 70 and the soldered chips 50, 60, because these connections 80, 90 are loosened due to delamination. When delamination occurs and increases, the vibration load caused by the ultrasound 110 acts on this delamination, amplifies the signal, and in particular generates additional signals. Therefore, for example, in the event of a crack, the US component can be modulated into an electrical signal. The thermal component and the mechanical component can be separated by signal analysis 210. This can be used particularly advantageously in planar mounting and connection technology (AVT) because the US transmitter 1250 can be easily coupled into a planar structure. The US transmitter 120 is then located on the planar AVT and can couple the ultrasound 110 into the planar structure. The US transmitter 120 includes piezoelectric layers, which can also be integrated into the planar mounting and connection technology. Here, it should be noted that the ultrasound 110 can be coupled not only as a near-surface wave, but also as a volume wave in the component to be inspected.
[0047] The power components 50 , 60 themselves can be used particularly advantageously as US transmitters 120 and receivers, since SiC and GaN semiconductors in particular have piezoelectric properties.
[0048] The ultrasonic transmitter 120 is then also integrated in the power module 10. By correspondingly conducting electrical excitation, the components 50, 60 can emit ultrasonic signals 110 and detect feedback signals. In the case of planar mounting, additional mechanical stimulation by means of ultrasound 110 can be particularly easily achieved. Delamination and cracks can thus be better identified. Compared with previous inspection methods (such as RF reflectometers), the proposed method is advantageously wired at significantly lower frequencies. The piezoelectric actuator 120 for US stimulation can be directly integrated into the planar mounting technology, or the function can be performed by the power components 50, 60 themselves. Another advantage is high measurement accuracy, possibly with spatial resolution, depending on how many ultrasonic generators 120 and sensors are used and how the ultrasonic generators 120 and sensors are distributed. In addition, the proposed measurement technology is relatively inexpensive, and the proposed method enables early detection of the scope and type of thermomechanical aging.
[0049] Reference numerals list
[0050] 110 Ultrasonic vibration
[0051] 120 Ultrasonic wave source, for example, a piezoelectric layer capable of being electrically excited
[0052] 130 Ultrasound
[0053] 140 Direction of Ultrasonic Waves
[0054] 10 Electronic components, power modules
[0055] 20 Circuit Board
[0056] 30 Flat side
[0057] 40 Copper Layer
[0058] 50 Semiconductor devices
[0059] 60 Semiconductor devices
[0060] 70 Bonding wire, solder connection
[0061] 80 solder layer
[0062] 90 Solder layer
[0063] 95 Local defect location
[0064] 100 Laser welding
[0065] Z us [Ω] Impedance during ultrasonic excitation
[0066] Z o [Ω] Impedance without ultrasonic excitation
[0067] 200 receiving unit, impedance measurement unit
[0068] 210 evaluation units
Claims
1. A method for quality testing of an electronic component (10), in which at least one component of the electronic component (10) is checked by means of an electrical characterization method, wherein: The component being examined is excited by means of ultrasound (130), so that a change in a signal of an electrical characterization method is caused thereby, the change in the signal is evaluated by means of a signal analysis (210), and the change in the signal indicates a material change on the component.
2. The method according to claim 1, wherein: The electrical characterization method is impedance spectroscopy, wherein impedance spectra are compared by means of the signal analysis (210), wherein changes in the determined impedance spectra indicate damage or delamination on the component.
3. The method according to claim 2, wherein: A frequency-dependent impedance measurement is performed, and wherein the signal analysis functionally cooperates with the measurement such that the frequency-dependent impedance measurement is performed in a comparison, once with and once without additional ultrasonic excitation, and the difference signal is evaluated.
4. The method according to claim 1, 2 or 3, wherein: The ultrasonic waves (130) are generated by means of electrical excitation.
5. A method according to any one of the preceding claims, wherein: The ultrasonic signal component is modulated into an intermodulation electrical signal. 6 . The method according to claim 1 , wherein a signal analysis is performed, by means of which the thermomechanical and mechanical components of the wear are separated.
7. A method according to any one of the preceding claims, wherein: The electronic component (10) has a planar structure, and the ultrasonic wave (130) is generated by means of an ultrasonic generator (120) placed directly on the plane AVT.
8. The method according to claim 7, wherein: The ultrasonic generator (120) is integrated into the planar structure, in particular as a piezoelectric layer.
9. The method according to claim 7 or 8, wherein: Ultrasonic waves (130) are coupled into the electronic component (10) as near-surface waves.
10. A method according to any one of the preceding claims, wherein: The ultrasonic wave (130) is generated as a volume wave within the electronic component (10).
11. A method according to any one of the preceding claims, wherein: The electronic component (10) is exposed to ultrasonic vibrations (110) either continuously, once or at discrete points in time.
12. An electronic component (10), in particular a circuit carrier, which is designed to carry out the method according to any one of the preceding claims, The electronic component comprises at least one ultrasonic source (120) which is configured and arranged to expose a power electronic component (10) to ultrasonic vibrations (110), a measuring device (200) which is designed to carry out electrical characterization, in particular frequency-dependent impedance measurement, and an evaluation device (210) which is designed to infer defects in the electronic component (10) by means of changes in detected electrical signals.
13. The electronic component (10) according to claim 12, which has a planar structure and has an ultrasonic generator placed directly on the plane AVT. 14 . The electronic component ( 10 ) as claimed in claim 12 , having a planar structure and having an ultrasonic generator integrated into the planar structure, in particular as a piezoelectric layer.
15. The electronic component (10) according to any one of the preceding claims 12 to 14, comprising an ultrasonic generator which is integrated in a component of the electronic component (10), in particular as a piezoelectric layer, the component having piezoelectric properties.