One of the key strategies employed in critical infrastructure design is redundancy, which involves the implementation of multiple uninterruptible power supplies (UPS) to mitigate the risk of downtime. There are varying levels of redundancy from the basic ‘N’ UPS configuration to the more reliable N+1, N+X and 2N configurations.
'N' UPS Configuration
Level of UPS redundancy:
Low
Risk to the load:
High
Also known as a ‘capacity’ in an N configuration a single UPS system, or multiple UPS systems are set up in parallel to match the required capacity for supporting the electrical load, not for adding redundancy. One small UPS in an office is an N configuration, likewise, three 100kW UPS units in parallel supporting a 300kW load would also be ‘N’. If one UPS fails, the remaining units continue to provide power, although the overall capacity may decrease below the 300kW requirement.
When maintenance is carried out on an N configuration, the UPS must be bypassed to allow for a safe shutdown. This configuration is often used for situations where the load is not critical to business operations.
'N+1' UPS Redundancy
Level of redundancy:
medium
Risk to the load:
medium
N+1 redundancy typically refers to a system configuration where there is one additional backup unit beyond what is required to support the load. It is sometimes referred to as a ‘Parallel Redundant’ configuration and should not be confused with an ‘Isolated Redundant’ configuration*
As it is not advised to consistently operate a UPS system above 50% of its capacity, an N+1 set up would see each UPS in the configuration sharing the critical load evenly. For example, if a data centre requires three UPS units to handle its power needs (N), then an N+1 configuration would involve installing a fourth UPS unit (+1). This extra unit provides redundancy in case one of the primary units fails, ensuring uninterrupted power supply to critical equipment. Healthcare is another industry where N+1 redundancy is required to meet industry guidelines because the UPS are supporting critical equipment which requires continuous power.
All UPS systems in an N+1 are either part of a common output bus meaning they are synchronised with one another, or they have a function embedded within the module itself. The number of UPS systems that can be paralleled is dictated by the manufacturer.
During a system maintenance, one UPS can be taken offline with the remaining system/s being required to immediately support the extra load, temporarily giving that UPS system/s 100% of the load.
'N+X' UPS Redundancy
Level of redundancy:
high
Risk to the load:
low
N+X Redundancy: Similar to N+1, but with more redundancy built-in. Instead of just one extra unit (N+1), there are multiple redundant units (N+X). This configuration offers higher fault tolerance and scalability by adding more redundancy beyond the minimum required.
'2N' UPS Redundancy
Level of redundancy:
Very high
Risk to the load:
Very low
Also known as ‘Distributed Redundant’, a 2N configuration involves duplicating the entire UPS system to create two fully independent systems with separate input and output feeders and two supplies out. Each UPS system is capable of supporting the entire load independently with no power connections between each. If one system fails, the other can seamlessly take over without any interruption in power supply. PDUs (Power Distribution Units) are used for distributing the electrical power between the critical load and independent UPS units.
With two independent distribution systems the critical load equipment is always available so there is less risk for the electrical independence of the distributed redundant system, ensuring the load fault on one UPS will not propagate to the other. This means that the critical load will still be supported by the unaffected UPS.
All downstream loads can be switched to one of the UPS units allowing the others to be taken offline for maintenance.
*Isolated redundancy:
This is also sometimes called an N+1 configuration but is significantly different to a parallel redundancy. There’s a main UPS system that feeds the load and a secondary UPS that feeds its static bypass. If the primary UPS fails, the second UPS consumes 100% of the load.