A fundamental law of physics is that for every 10°C that you are able to keep the power supply’s environment lower than 40°C, you double the mean time between failures (MTBF). Conversely, for every 10°C your power supply’s ambient temperature increases, your MTBF cuts in half (that is, your power supply is half as reliable). Many, but not all, of the failure mechanisms on this list are related to temperature.
More and more, we’re seeing the use of end-equipment plastic chassis compared to the metal chassis that have been used since time began, which impacts thermals as well as EMC. Anything you can do to enhance thermal management around your power supply in the system is of critical importance.
1. Fans
Fans are the number one failure mechanism of power supplies, as found by both military MTBF simulations as well as Belcore standards and as
A fanless system can be sealed, which also eliminates other issues, including ingress of moisture. In the case of outdoor applications, such as digital signage, a sealed system can keep out leaves, bugs, twigs, and bird nests, as well as rain and moisture and, in the case of maritime applications, salt and fog.
Removing the fan increases reliability by 25% and is the best solution for avoiding failure. A good design that keeps the efficiency of the power supply high enough makes fans unnecessary.
2. Capacitors
Despite popular thought, a lot of progress is being made in capacitor technologies every year; however, they are prone to failure if overstressed or if substitutes are made in production or by counterfeiting.
Capacitors, especially