Transferring to bypass in an uninterruptible power supply occurs when there’s a fault such as an overload or a component failure.
Unless it’s for routine maintenance purposes, transferring to bypass should be seen as a last resort as it switches the critical load from the UPS and batteries to the raw mains electricity supply.
When designing a power protection scheme for a data centre or similar facility, operators need to consider their availability needs, management capabilities, and financial constraints.
These factors will influence their choice as to whether they use a centralised or decentralised bypass architecture:
- Centralised UPS Bypass: rated for the required capacity of the entire UPS system. Has a single common static bypass switch and components in a separate bypass cabinet.
- Decentralised (Distributed) UPS Bypass: each UPS in the power protection system has its own internal static bypass, rated to the maximum capacity of the individual unit. When required to transfer to bypass, the static switch in each UPS turns on simultaneously to share the load.
In the following whitepapers, you’ll learn about the factors that can impact the choice of UPS bypass, such as overall system reliability, power balancing, cost, and required footprint.
Due to the varied nature of UPS system installations (i.e. capacity or redundancy), one whitepaper focuses solely on parallel capacity (N) configurations, where the total load is met without any provision for redundancy.
The other document explores similar themes for parallel redundant systems where if a UPS fails, the remaining units have the ability to continue supporting the load.