Power system applications, such as street lighting, typically have a 10 year warranty. When these systems include electrolytic capacitors, it is important to choose a supplier that meets these requirements. Traditional lifetime testing of electrolytic capacitors to ascertain their life expectancy requires specialized equipment, is time consuming, labor intensive, and for most OEMs, is ultimately cost prohibitive. Electrolytic capacitors with the same capacitance and voltage ratings from different suppliers may be rated to the same lifetime, but historical data confirms that they can have significantly different operational expected lives. An accelerated testing methodology is needed to compare the reliability of electrolytic capacitors from different suppliers. DfR has developed an approach that reduces test times from thousands of hours to several weeks by taking advantage of two key behaviors of electrolytics. The first involves the rate at which capacitors lose electrolyte, which is fairly predictable at a given temperature and electrical stress. The second key behavior is the dependence of the equivalent series resistance (ESR) of electrolytic capacitors on the volume of liquid electrolyte. The approach that will be described in this paper will demonstrate a means of comparing the time to failure for comparable capacitors from different suppliers under the same conditions. Case studies will demonstrate how this method avoids the extended testing that is typically required.

This research compared the lifetime of similar aluminum electrolytic capacitors from different manufacturers using an accelerated life test, which consisted of critical weight loss testing and rate of weight loss testing. In critical weight loss testing, capacitors are perforated to speed up electrolyte evaporation and the equivalent series resistance (ESR) and weight are measured periodically to determine their relationship. In rate of weight loss testing, capacitors are subjected to final operating conditions (i.e. voltage and ripple current are applied) and the weight is periodically measured over the course of 500 hours. After test completion the relationship between ESR and weight loss is used to calculate the critical weight loss that occurs at datasheet-defined failure, which is typically a 200% increase in ESR. The rate of weight loss is extrapolated to the critical weight to estimate a time to failure that can be compared to other capacitors tested using the same accelerated approach. In this research, testing compared 450 V, 68 μF capacitors from Manufacturer A and Manufacturer B, and results indicated Manufacturer A had a significantly longer lifetime. Therefore, capacitors from Manufacturer A are more reliable than capacitors from Manufacturer B.