MTBF and reliability of DC/DC converter

Modular power supply (BMP) is widely used in industrial automation, relay protection, distribution network automation, rail transit, automotive electronics, aerospace and other high-reliability and high-performance fields because of its high efficiency, safety, and reliability. Especially with the rapid development of the rail transit industry, higher requirements and challenges have been placed on the service life and reliability of the module power supply.

Usually, the power supply manufacturer will provide the MTBF value of the product in its product manual, such as 1 million hours, 1.5 million hours. MTBF, the mean time between failures, the full English name is "Mean Time Between Failure", is the average value of the time from the new product starting to work under the specified working environment conditions to the first failure. The longer the MTBF, the higher the reliability and the stronger the working ability, and the unit is "hour". It reflects the time quality of the product and reflects the ability of the product to maintain its function within the specified time.

So, how is the value of MTBF obtained? Assuming that the MTBF of the DC/DC module power supply is 1 million hours, is it detected that the DC/DC module power supply runs continuously for 1 million hours?

The answer is no, if it did, it would take more than 100 years of testing to complete the product. Power supply manufacturers cannot wait for more than 100 years to measure the actual failure rate and the rate of change of the failure rate of the product to obtain enough valid data to improve the reliability of the product. In fact, the common method in the industry is to calculate MTBF and reliability analysis prediction according to relevant standards. At present, the most common authoritative standards are MIL-HDBK-217, and Bellcore, which are used for military products and civilian products respectively. Among them, MIL-HDBK-217 was proposed by the U.S. Department of Defense Reliability Center and Rome Laboratory and became an industry standard, which is specially used for the calculation of MTBF value of military products; Bellcore was proposed by AT&T Bell Laboratory and became the MTBF value of commercial electronic products. The industry standard for computing. However, there is still a considerable gap between the values estimated by these methods and the actual mean time between failures.

M

The main consideration in the TBF calculation is the failure rate of each device in the product. However, due to the fact that the failure rate of the device will be very different under different environments and different conditions of use, for example, the reliability value of the same product in different environments is different; for example, a MOSFET with a rated voltage of 80V is The failure rate under the actual voltage of 40V and 60V is definitely different. The same DC/DC module power supply works at an ambient temperature of 25°C and works at an ambient temperature of 70°C, and its failure is definitely different. Therefore, when calculating the reliability index, the above factors must be considered.

According to the MIL-HDBK-217 standard, there are generally two methods for predicting reliability, Part Stress Analysis/PSA and Parts Count Analysis/PCA. The PSA method requires a lot of specific information and is mostly used in the later stages of product design, as measured data and preliminary results can be used in reliability models. The PCA method requires very little information such as part quantity, quality grade and application environment. The biggest advantage of MIL-HDBK 217 technology is that the PCA method can predict reliability based only on the bill of materials (Bill of Materials/BOM) and expected usage quantities, so the reliability of a product that has not been produced can be predicted as follows:

λP = (Σ NC λC) (1 + 0.2 πE) πF πQ πL

Formula 1 Failure rate calculation formula

Here: NC element count (each element)

λC Reliability of each component (basic value taken from database)

πE Environmental stress factor (application specific)

πF mixed functional stress (additional stress due to interactions between elements)

πQ screening level factor (standard or pre-screened based on component tolerances)

πL maturity factor (known and tested designs or new attempts)

The characteristics of each element can be obtained by calculation. The total reliability can be obtained by adding up all the individual results.

failure rate

Failure rate is either defined by the time between two failures (unit: hours), the mean time between failures (Mean Time Between Failures/MTBF), or by the time it takes for the first failure to occur (Mean Time To Failure/MTTF) ) to define.

The curve of the standard failure rate can be described by the well-known "bathtub curve". The curves are shown in Figure 1, and the shape of the curves for all components and systems is approximately the same - only the elongation in the direction of the time axis is different. It can be divided into three areas: early failure period (I), occasional failure period (II), wear failure period (III). MTTF includes regions I and II, while MTBF only includes region (II).

Zone I is the early failure period of a component, where product failure is usually caused by a potential material failure or a manufacturing defect that was not detected during final product inspection prior to shipment. The early failure usually lasts for a short time. According to empirical statistics, most of the early failures of the DC/DC module power supply will occur within 24 hours of use. 24 hours may be very short for a DC/DC module power supply with a shelf life of three years, but from another perspective, taking Density Power's typical DC/DC module power supply operating frequency of 350KHz as an example, MOSFETs and transformers are working for 24 hours. It will be operated more than 30.24 billion times during the hour. According to actual statistical data and experience, most of the component defects will fail during this period. Therefore, in order to detect and eliminate early defects and failures as soon as possible, high reliability The DC/DC module power supply that meets the requirements will undergo a 24-hour high temperature aging test before leaving the factory.

The failure rate continues to stabilize at a lower level in the occasional failure period region II of the useful operating life stage. The second transition time (T2), from the effective operating life stage to the end of the product's life, is affected by many factors, such as the design and the quality of the components used, the quality of the assembly at the time of production, and the environmental stress of the application. Region III indicates the end of the product life cycle, during which product performance declines due to wear, chemical degradation of materials, and sudden failures.

It is well known that operating temperature has a great impact on reliability, and reliability will decrease as the temperature rises. As early as 1898, the Swedish chemist Arrhenius proved that the rate of chemical reaction is affected by temperature. For every 10 degrees increase in temperature, the rate of chemical reaction doubles, and the acceleration factor of component aging will also increase exponentially. According to Table 2, we can see that the aging effect accelerates when the ambient temperature rises from 25°C to 50°C, and the acceleration factor is 6 times. If the temperature continues to rise from 25°C to 70°C, the aging effect will be accelerated by more than 20 times. Therefore, when choosing a high-reliability DC/DC module power supply, not only should pay attention to the MTBF value in the product manual, but also the temperature conditions corresponding to the actual calculation of this MTBF value.

MTBF is only a measure of reliability. These calculations are actually the first step. More importantly, according to the calculation process of the MTBF index, it is necessary to discover key components and links that affect product reliability, and further improve product design and production. Ultimately achieve the purpose of improving the reliability of the product.

From the calculation theory and method of MTBF, it can be known that the failure-free time of the product is evaluated by the number of components and the failure rate of the parts; this method assumes that the components of the product are all working under the expected working stress. In fact, due to unpredictable factors , the product may have momentary overstress. There is also a situation where some of the stresses that affect product life are difficult to fully evaluate. In addition, the influence of the quality and reliability of different suppliers of similar devices of the product, the production process of the product, and human factors on the reliability of the product is not fully considered.

So, is there any scientific and effective test method to verify the MTBF of the product?

The answer is yes. For example, the MTBF of a DC/DC module power supply calculated according to MIL-STD-217 is 1 million hours. In fact, all the test samples fail during the test, and the required test time may be as long as several decades, or even a century. This method is impractical for products with high reliability and long service life. In fact, according to reliability principles and related standards, DMTBF test (also called Accelerated Aging & Life Test) is an effective and feasible test method. On the premise of not affecting the reliability of the reliability test, the purpose of shortening the reliability verification time is achieved by increasing the AF (acceleration factor) and the number of times of switching on and off during the actual test. Corresponding to the MTBF of 1 million hours of DC/DC module power supply, working for a long time at an ambient temperature of 70 °C, 60 samples, working continuously for 38 days without any failure, it proves that the product has a confidence level of 0.9 and can reach 100 10,000 hours of MTBF.

To sum up, for the DC/DC module power supply used in high reliability fields, especially in the application fields of rail transit, automotive electronics and power operating systems, in addition to MTBF calculation, reliability analysis and prediction, the DMTBF in the product verification test stage is 100% aging test in the test and production process is very important and necessary to ensure the long-term reliability of the product and eliminate the early defects and failures of the product.