Life estimation system
The lifespan estimation system addresses inaccuracies in conventional methods by maintaining a constant temperature of the test object, thereby achieving precise lifespan predictions despite variations in self-heating from non-thermal loads.
Patent Information
- Application Number
- JP2024098176
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-06-18
- Publication Date
- 2026-01-06
AI Technical Summary
Conventional lifespan estimation methods for electronic components are inaccurate due to variations in actual temperature loads caused by self-heating from non-thermal loads, leading to inconsistent and unreliable lifespan predictions.
A lifespan estimation system that includes a temperature measurement unit, a temperature control unit, a heating and cooling unit, an application unit for non-thermal loads, and a detection unit, which maintains a constant temperature of the test object by controlling the heating and cooling units based on measured temperature data to accurately estimate lifespan.
The system provides accurate lifespan estimation by accounting for individual self-heating variations, ensuring consistent and reliable predictions across test objects.
Smart Images

Figure 2026000697000001_ABST
Abstract
Description
[Technical Field]
[0001] The present invention relates to a lifespan estimation system. [Background technology]
[0002] The lifespan of electronic elements such as semiconductor circuits and capacitors is generally very long, and it is not realistic to evaluate the lifespan of a newly developed product under normal use conditions. Therefore, when estimating the product's lifespan, it is common to conduct so-called accelerated tests in which loads other than temperature, such as environmental temperature, voltage, and current, are set higher than those during normal use.
[0003] Then, tests are conducted by varying the intensity of temperature load due to the environmental temperature and loads other than temperature loads, the lifespan is calculated from a Weibull plot under each test condition, an acceleration factor is calculated using the Eyring model or the Arrhenius model, and the lifespan under normal use environments is generally estimated from the acceleration factor (see Patent Document 1). [Prior art documents] [Patent documents]
[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2003-302436 Summary of the Invention [Problem to be solved by the invention]
[0005] However, because the application of voltage and current, which are load factors other than the ambient temperature, causes heat to be generated in the test object, the actual temperature load on the test object is higher than the ambient temperature. Therefore, if the lifespan is estimated based on the ambient temperature, the estimated lifespan will be at a temperature that differs from the actual temperature load. Furthermore, the amount of heat generated varies depending on the characteristics of each test object, and the amount of heat generated changes as the deterioration progresses. Therefore, the actual temperature load on each product is not necessarily constant, even for lots set under the same test conditions. This results in insufficient accuracy for accelerated evaluation, resulting in problems with the accuracy of lifespan estimation.
[0006] The present invention has been made in view of the above circumstances, and has an object to provide a life estimation system that can accurately estimate the life of a test object. [Means for solving the problem]
[0007] A life estimation system according to one aspect of the present invention comprises a temperature measurement unit that measures the temperature of a test object, a temperature control unit, a heating and cooling unit that heats and cools the test object, an application unit that applies a load other than temperature to the test object, and a detection unit that detects failures in the test object, wherein the load other than temperature is an electrical load, and the temperature control unit controls the temperature obtained from the temperature measurement unit to a temperature to be used for life estimation. [Effects of the Invention]
[0008] A lifespan estimation system according to one aspect of the present invention can accurately estimate the lifespan of a test object.
[0009] A life estimation method according to another aspect of the present invention includes measuring the temperature of a test object, applying a load other than temperature to the test body, and controlling the temperature of the test object, and applying the temperature of the test object to a life estimation formula.
[0010] A lifespan estimation method according to another aspect of the present invention can accurately estimate the lifespan of a test device. [Brief explanation of the drawings]
[0011] [Figure 1] FIG. 1 is a schematic diagram of a lifespan estimation system according to one aspect of the present invention. [Figure 2] FIG. 2 is a schematic diagram of a conventional lifespan estimation system. [Figure 3] FIG. 3 is a schematic diagram of a Weibull plot obtained using a lifespan estimation system according to one aspect of the present invention. [Figure 4] FIG. 4 is a schematic diagram of a Weibull plot obtained by a conventional lifespan estimation method. DETAILED DESCRIPTION OF THE INVENTION
[0012] First, an overview of the lifespan estimation system disclosed in this specification will be described.
[0013] (1) In one aspect of the present invention, a life estimation system includes a temperature measurement unit that measures the temperature of a test object, a temperature control unit, a heating and cooling unit that heats and cools the test object, an application unit that applies a load other than temperature to the test object, and a detection unit that detects failures in the test object, and the temperature control unit controls the temperature obtained from the temperature measurement unit to a temperature used for life estimation.
[0014] The life estimation system described in (1) above has high accuracy in estimating the life of the test object. The reasons for this are considered to be as follows. Generally, life estimation is performed as follows: The test device is placed in a test chamber such as an oven, and a temperature load is applied by setting the ambient temperature higher than that under normal use. A so-called accelerated test is then performed, in which non-thermal loads, such as electrical loads (e.g., voltage and current) stronger than those under normal use, are applied to the test device. The time from the start of the test to failure of the test device is then measured. The obtained results are plotted as a Weibull plot, and the life under those test conditions is calculated from the Weibull mean. Next, the load intensity of temperature and non-thermal loads, such as electrical loads (e.g., voltage and current), is varied, and accelerated tests are performed under at least two load conditions. The Eyring model and Arrhenius model are used to calculate the respective acceleration coefficients from the test results, and the life under normal use conditions is estimated. When a load other than temperature, such as an electrical load, is applied to the DUT, the DUT's resistance component generates heat, and the temperature load applied to the DUT becomes a temperature that is the sum of the ambient temperature and the self-heating, which is higher than the load temperature set by the tester. The amount of heat generated varies depending on the characteristics of each DUT and also changes depending on the degree of deterioration of the DUT. Therefore, with conventional life estimation methods, even for lots set under the same test conditions, the temperature load actually applied to the test object is different for each test object because the temperature generated by self-heating is superimposed on the environmental temperature. In other words, even though the test conditions are not the same and the temperature load actually applied to the test object is higher than the accelerated temperature used for life estimation, these factors are not taken into account at all when calculating life estimation. As a result, the accuracy of the calculated estimated life is not high. In contrast to this, in the life estimation system described in (1) above, the temperature of the test object itself is measured and controlled to be constant, so the accuracy of the calculated estimated life is high.
[0015] (2) In the lifespan estimation system described above in (1), the load other than the temperature load may be an electrical load.
[0016] (3) A life estimation method according to another aspect of the present invention includes measuring the temperature of a test object, applying a load other than temperature to the test object, and controlling the temperature of the test object.
[0017] (4) In the lifespan estimation method described in (3) above, the load other than the temperature load may be an electrical load.
[0018] The configuration of a lifespan estimation system according to one embodiment of the present invention and other embodiments will be described in detail. Note that the names of the components (elements) used in each embodiment may differ from the names of the components (elements) used in the background art.
[0019] <Test Subject> The test object may be any object that can be subjected to the above-mentioned accelerated test, such as a passive component, an active component, a battery, etc.
[0020] The passive component is not particularly limited, and may be, for example, a resistor, a capacitor, or an inductor.
[0021] The resistor is not particularly limited and may be, for example, a fixed resistor, a semi-fixed resistor, or a variable resistor. The material of the resistor is also not particularly limited and may be, for example, a carbon coating, a metal coating, a metal oxide coating, a metal grace coating, a metal wire (winding), a metal thin film, or a metal plate.
[0022] The capacitor is not particularly limited, and may be, for example, an electrolytic capacitor, a ceramic capacitor, a film capacitor, a mica capacitor, an air capacitor, a glass capacitor, an oil capacitor, a supercapacitor, or the like.
[0023] The inductor is not particularly limited and may be, for example, a leaded inductor or a surface-mounted inductor. The leaded inductor is not particularly limited and may be, for example, an axial leaded inductor or a radial leaded inductor. The surface-mounted inductor is not particularly limited and may be, for example, a wire-wound inductor, a multilayer inductor, or a thin-layer inductor.
[0024] Examples of active components include vacuum tubes and semiconductor components. The semiconductor components are not particularly limited and may be, for example, diodes or transistors, or may be so-called integrated circuits (ICs) that are formed by integrating these components.
[0025] Examples of diodes include general rectifier diodes, switching diodes, fast recovery diodes, Schottky barrier diodes, step recovery diodes, Zener diodes, avalanche diodes, tunnel diodes, light-emitting diodes, photodiodes, and laser diodes.
[0026] Examples of the transistor include a bipolar transistor, a field effect transistor (FET), and an insulated gate transistor.
[0027] The bipolar transistor may be either an NPN type or a PNP type.
[0028] The field effect transistor (FET) may be a junction transistor (JFET) or a metal oxide semiconductor FET (MOSFET). The metal oxide semiconductor FET (MOSFET) may be an enhancement type or a depletion type. Either field effect transistor (FET) may be an N-channel type or a P-channel type.
[0029] The integrated circuit (IC) is not particularly limited, and may be an analog IC, a digital IC, or a mixed signal IC.
[0030] Examples of analog ICs include operational amplifiers, comparators, voltage regulators, interface ICs, data conversion circuits, etc. Examples of digital ICs include microcomputers, memories, standard logic ICs, dedicated logic ICs, programmable logic devices (PLDs), etc.
[0031] The battery is not particularly limited and may be a primary battery or a secondary battery. Examples of primary batteries include manganese dry batteries and alkaline dry batteries. Examples of secondary batteries include lead-acid batteries, nickel-cadmium batteries, nickel-metal hydride batteries, and lithium-ion batteries.
[0032] <Temperature measurement part> The temperature measurement unit measures the temperature of the test object. A known measurement method can be selected depending on the temperature range of the test object to be measured. For example, a thermometer may be attached to the test object. The thermometer may be, for example, a thermocouple or a resistance temperature detector. The thermocouple may be appropriately selected from types conforming to JIS-C-1602 (2015). The resistance temperature detector may be a metal type or a thermistor. The thermistor may be a PTC (Positive Temperature Coefficient) or an NTC (Negative Temperature Coefficient).
[0033] When the test object includes a container, a sealing resin, or the like, the thermometer may be provided inside them.
[0034] If the device under test is a semiconductor device, the temperature of the device under test may be measured from the temperature characteristics of the forward voltage of a PN junction included in the semiconductor device under test.
[0035] If the test object has a temperature gradient due to self-heating or the like, the temperature at the point where the temperature is highest may be measured.
[0036] The temperature data of the test object measured in this way is provided to a temperature control unit (described later). The temperature data may be captured in a data logger, for example, and provided to the temperature control unit (described later). As the data logger, for example, a Keysight 34970A or the like may be used.
[0037] The interface between the temperature measurement unit and the temperature control unit (described later) is not particularly limited, and for example, GPIB, RS232C, RS485, USB, etc. can be used, and these may also be converted for connection.
[0038] <Heating and cooling section> The heating and cooling unit heats and cools the test specimen to a temperature set for use in lifetime estimation under the control of a temperature control unit (described later). For heating and cooling, a known heating and cooling method may be appropriately selected, and for example, a commercially available thermostatic bath, hot plate, or the like may be used.
[0039] The interface between the heating / cooling unit and the temperature control unit described below is not particularly limited, and for example, GPIB, RS232C, RS485, USB, etc. can be used, and these may also be converted for connection.
[0040] The heating / cooling unit may be set to a temperature that is higher than the temperature at which the test object is normally used. The temperature to be set is not particularly limited and may be any temperature between 30°C and 1000°C. More specifically, the temperature may be, for example, 55°C, 60°C, 65°C, 70°C, 85°C, 100°C, 110°C, 125°C, 150°C, 155°C, 175°C, 200°C, or 250°C. If the accelerated test to be performed complies with a predetermined test standard, the temperature may be set to a temperature specified by the test standard.
[0041] <Temperature control unit> The temperature control unit controls the heating and cooling unit to adjust the temperature of the test object obtained from the temperature measurement unit to a set temperature used for life estimation. The control method may be appropriately selected from known control methods. For example, a personal computer equipped with control software W4000D VEE Pro 9.33 manufactured by Keysight may be used.
[0042] <Applying unit> The application unit applies a load other than a temperature load to the device under test. The application of the load other than a temperature load causes the device under test to heat up. The load other than a temperature load is not particularly limited as long as it causes the device under test to heat up when applied, and an example of such a load is an electric load. The electric load may be a voltage load or a current load. In either case, it may be a direct current (DC) load, an alternating current (AC) load, or an AC load superimposed on a DC load. The waveform of the AC load is not particularly limited, and may be, for example, a sine wave, a triangular wave, or a square wave. The frequency of the AC load can be selected arbitrarily.
[0043] <Detection section> The detection unit detects a failure in the test object and records the time from the start of the accelerated test to the occurrence of the failure. The definition of a failure in the test object may be determined according to the test object. For example, if the test object is a ceramic capacitor, a failure may be defined as a short circuit in the ceramic capacitor under test, and a short circuit may be defined as the insulation resistance IR of the ceramic capacitor under test falling below a certain level. If the test object is a film capacitor, a failure may be defined as the capacitance of the film capacitor under test falling below a predetermined initial value. Furthermore, if the failure mode of the test object can be any of a short circuit, an open circuit, or a threshold voltage fluctuation, the occurrence of any of these may be defined as a failure.
[0044] <Lifespan Estimation System> The life estimation system performs an accelerated test to estimate the life of the test object. Figure 1 is a schematic diagram of a life estimation system according to one aspect of the present invention. A feedback loop is configured to measure the temperature of the test object and control it to a constant temperature used for life estimation. If the test object generates heat due to a load other than temperature, such as an electrical load, the temperature of the test object, rather than the environmental temperature, is controlled to a constant temperature by, for example, suppressing the heating output of the heating / cooling unit. For example, when a direct current (DC) voltage load V0 is applied to a capacitor under test, the capacitor generates a power loss P0 expressed by the following equation 1 according to its insulation resistance IR. P0=V0 2 / IR 1 The capacitor under test generates self-heating ΔT0 in response to the power loss P0. The temperature measurement unit included in the life estimation system acquires temperature data including the heat generation ΔT0, and the temperature data is provided to the temperature control unit. The temperature control unit controls the heating / cooling unit to reduce the heating output or increase the cooling output in response to the heat generation ΔT0, thereby maintaining a constant temperature of the capacitor under test. Furthermore, when an alternating current (AC) load I1 is applied to a capacitor that is the device under test, the capacitor generates a power loss P1 expressed by the following equation 2 according to its equivalent series resistance ESR. P1=I1 2 ×ESR 2 The capacitor under test generates self-heating ΔT1 in response to the power loss P1. The temperature measurement unit included in the life estimation system acquires temperature data including the heat generation ΔT1, and the temperature data is provided to the temperature control unit. The temperature control unit controls the heating / cooling unit to reduce the heating output or increase the cooling output in response to the heat generation ΔT1, thereby maintaining a constant temperature of the capacitor under test.
[0045] The life estimation system may be configured by combining the above-mentioned commercially available products for each component (each element), or may be configured using a dedicated device.
[0046] <Lifespan estimation method> Using a life estimation system according to one aspect of the present invention, an accelerated test is performed by applying a temperature load and a load other than a temperature load to a test object to determine the time until the test object fails. For each load, the load is applied under two or more conditions, and the acceleration coefficient for each load is determined by performing the accelerated test.
[0047] The number of test specimens used in the accelerated test may be set appropriately depending on the reliability level and lot acceptability defective (LTPD) required for the accelerated test, and may be, for example, 2, 5, 9, 11, 13, 18, 22, 31, 32, 45, 76, etc.
[0048] 3 is a schematic diagram of a Weibull plot obtained by using a life estimation system according to one aspect of the present invention to determine a temperature acceleration coefficient by varying the temperature load at three points: T0-T1, T0, and T0+T1, while keeping the electrical load (i.e., a load other than the temperature load) constant. From the Weibull plot obtained in this manner, the temperature acceleration coefficient is calculated using a standard method, and the life under desired temperature conditions is estimated.
[0049] Figure 4 is a schematic diagram of a Weibull plot obtained by a conventional method, that is, a method in which the environmental temperature is controlled to T0-T1, T0, and T0+T1 rather than controlling the temperature of the test object.
[0050] A comparison of Figures 3 and 4 reveals that when accelerated testing is performed using a life estimation system that includes a temperature measurement unit that measures the temperature of the test object, a temperature control unit, a heating and cooling unit that heats and cools the test object, an application unit that applies loads other than temperature to the test object, and a detection unit that detects failures in the test object, and the temperature control unit controls the temperature obtained from the temperature measurement unit to a temperature to be used for life estimation, temperature variations due to heat generation caused by individual differences in the test object are suppressed, the slope is linear, and the shape parameters under each test condition are consistent, making it an appropriate accelerated test. On the other hand, when accelerated testing is performed using the conventional method, that is, by controlling the environmental temperature rather than the temperature of the test object, the self-heating of the test object is superimposed on the environmental temperature, and the actual temperature load on the test object is higher than the set environmental temperature, resulting in a shorter Weibull mean life η than that shown in Figure 3. In addition, because the self-heating temperature varies depending on the individual characteristics of the test object, the slope is not linear for each test object and there is variation in the shape parameters of each test condition, which shows that this is not an appropriate accelerated test.
[0051] In addition, when the temperature load is kept constant at T0 and an electrical load other than the temperature load is applied to the above-mentioned test object, the same results are obtained as when the electrical load other than the temperature load is kept constant and the temperature load conditions are varied. [Industrial Applicability]
[0052] As described above, the life estimation system of the present invention is extremely useful as it can accurately estimate the life of an object under test such as an electronic component. [Explanation of symbols]
[0053] 1, 11 Heating and cooling section 2, 12 Temperature measurement part
Claims
[Claim 1] a temperature measurement unit for measuring the temperature of the test object; A temperature control unit; a heating and cooling unit for heating and cooling the test object; an application unit that applies a load other than temperature to the test object; a detection unit for detecting a fault in the device under test; Equipped with The loads other than the above temperature loads are electrical loads, The temperature control unit controls the temperature obtained from the temperature measurement unit to a temperature used for life estimation. Lifespan estimation system.
Citation Information
Patent Citations
Quality control method of electronic parts
JP2003302436A