Restoring power to data centers can result in costly maintence and downtime for service providers. In 2016, the national average cost of downtime was almost $9000 per minute. Often, this power failure can be attributed to preventable hardware failures. In 2016, the Ponemon Institute conducted a study to quantify the cost of downtime and identify the most common causes of data center outages. According to the study, the most common reasons for unplanned outages are UPS system failure, cybercrime, accidental human error, water or heat failure, weather-related disasters, generator failure, and IT equipment failure. The least common reason for unplanned outages was IT equipment failure, coming at 4%. However, when the study examined the total cost of each cause of failure, IT equipment failure ranked the most expensive. Hardware failures can stem from many places, including cooling fans, hard disk drives, and busbars.
An integrated cloud technology company noticed that the backup servers in its data centers were being used more than normal. After a little digging, the team realized that their primary servers were experiencing power failures, which caused the system to rely on the backup servers. Upon inspection, they discovered that the busbar and crown clip connection for the primary server had corrosion buildup. This buildup was attributed to fretting corrosion, or micro-motions that wear contacts and expose fresh layers of metal to oxidation. This eventually created an open connection and, ultimately, power failure. The provider determined that the micromotion occurred during both shipping and regular operation. The data center provider needed a solution that would protect future manufactured connectors from fretting corrosion and restore reliable connectivity to damaged connectors in the field.