Don't mistake power-supply MTBF for life expectancy

For machine system designers needing to evaluate which power supplies will best fit a specific machine design or application, it's standard practice to compare each unit's mean time between failures (MTBF) rating. Most people believe that MTBF is the number of operating hours that will elapse before a unit fails.

12/01/2005


For machine system designers needing to evaluate which power supplies will best fit a specific machine design or application, it's standard practice to compare each unit's mean time between failures (MTBF) rating. Most people believe that MTBF is the number of operating hours that will elapse before a unit fails. In reality, MTBF is the total functional life of a system component divided by the number of failures - a measurement of reliability. Examining reliability of a power supply is important because the MTBF can have a significant impact on overall system performance and efficiency. Yet, engineers and designers often make the mistake of assuming that MTBF can automatically be equated with a unit's life expectancy.

In reality, there is no direct correlation between MTBF and the actual operating life of a product. In fact, it's possible to find a power supply with extremely high MTBF but very low operating life span, depending on the types of components used and actual operating conditions.

The expected life of a power supply is often misunderstood, because operating life can be affected by many factors such as the average load rate, vibration and ambient temperature. A power supply's MTBF also is closely linked to the quality and life of a unit's internal electrolytic capacitors - the devices that store energy and filter out electrical variations.

Capacitors can be used to guard against sudden losses of voltage in circuits. So, for example, while a certain power supply may have a calculated MTBF of more than 600,000 hours, its electrolytic capacitors may have a MTBF rating of as little as 50,000 hours. Yet, because manufacturers cannot predict how their products will be used in the end user's operating environment, the industry still relies heavily on MTBF ratings.

Realistic picture of life expectancy

To get a better idea of the actual life span of a power supply, engineers and system designers evaluating different options examine which method the manufacturer uses to create its MTBF ratings. It's important to find out, for example, whether or not the power supply was rated at its full load.

If a power supply is tested at less than full load under continuous operation, it will likely operate cooler and have no thermal cycling, yielding a much longer life than would likely occur in normal operating conditions. Similarly, MTBF ratings should be conducted at a high ambient temperature, such as 104 F (40 C), to duplicate actual operating conditions.

Most manufacturers today rely on one of two standards to determine MTBF ratings: the Siemens-norm SN 29500 and the MIL-HDBK-217F prediction method. Because factors such as temperature and power load can have such a dramatic impact on capacitor life span, the SN 29500 can be particularly helpful for evaluating MTBF rating. The SN 29500 method takes into account the following key factors: