UPS Sizing, Battery Runtime & Critical Power Design Lab

MEPVirtual Engineering Labs - an interactive training and reference tool (module E-113) for electrical, HVAC and ELV engineers covering UPS topologies, sizing and battery runtime calculators, battery technologies, static transfer switches and bypass schemes, data center tier concepts, parallel/redundant UPS architectures, harmonics, grounding and bonding, commissioning and a scored self-assessment.

Critical power / building services scope - verify against project spec, adopted code edition and manufacturer data before design or sign-off

What "critical power" means

Critical power is the part of a facility's electrical supply that a UPS, its batteries and the surrounding distribution architecture exist to protect from any interruption the connected load cannot tolerate, not just from a total blackout.

Load classification

Not every load behind a switchboard needs the same protection. A useful, widely used way to group loads by how badly an interruption hurts is:

Critical

Zero tolerance for any break, sag or waveform disturbance. Loss causes immediate data loss, process failure or safety risk. Always fed from an online (double-conversion) UPS with battery backup. Example: core compute/storage, process control systems.

Essential

Must keep running to support the critical load or the space around it, but can usually tolerate a very short, sub-second break. Example: UPS room cooling, some control and monitoring systems.

Non-essential

Desirable to keep running but can be shed for a period during a contingency without an immediate operational failure. Example: general office IT, non-critical displays.

Non-critical

Ordinary building load with no UPS backing at all; it may still sit on the generator, but never on the UPS. Example: general lighting, general small power.

Getting this classification right up front drives almost every later decision in this tool: which UPS topology to specify, how much redundancy to design in, and where the boundary between "battery-backed" and "generator-backed only" actually sits on the single-line diagram.

Redundancy concepts: N, N+1, 2N, 2N+1

Redundancy describes how much spare capacity or how many spare paths exist beyond what is strictly needed to carry the load, so that a single failure or a piece of planned maintenance does not interrupt the critical load.

ConfigurationWhat it meansSurvives one module failure?Survives planned maintenance without a break?Relative cost
NExactly the capacity needed to carry the load, no more. All installed capacity is required simultaneously.NoNoLowest
N+1One extra module of the same size beyond what N requires, sharing a common bus.Yes, one moduleYes, one module can be pulled for serviceModerate
2NTwo entirely independent, fully rated systems (or buses), each able to carry 100% of the load alone.Yes, an entire systemYes, one whole system can be shut downHigh (double capacity)
2N+1Two independent systems as in 2N, but each of those systems is itself built with N+1 redundancy internally.Yes, an entire system plus a module in the surviving systemYes, with margin to also service a module in the remaining systemHighest

Simple way to picture it: N is "just enough", N+1 adds one spare part to a shared pool, 2N duplicates the whole pool, and 2N+1 duplicates the whole pool and still adds a spare part to each copy. Tab 08 walks through N+1 and 2N side by side with an interactive failure simulation.

Single points of failure (SPOF)

A single point of failure is any one component whose failure alone takes the critical load down, regardless of how much redundancy exists elsewhere in the chain. Designing out SPOFs means tracing the full path from utility to the load's power cord and checking every link:

  • Single utility feed - one substation or one incoming feeder with no second independent source or generator backup.
  • Single UPS module with no redundancy - an N-only UPS system where any internal fault drops the load straight to bypass or to nothing.
  • Single battery string - one string feeding the whole DC bus, so a single failed string (or a maintenance disconnect) removes all battery backup.
  • Single static transfer switch or bypass path - if every module ultimately depends on one shared switch or one shared distribution board, that shared component is the new SPOF even after the UPS modules themselves are made redundant.
  • Single-corded load - even a perfect 2N power system fails the load if the equipment itself only has one power supply cord plugged into only one of the two independent paths.

Verify before design: the required redundancy level, the acceptable load classification for a given space, and where the critical/non-critical boundary sits are project-specific decisions driven by the client's risk tolerance, applicable codes and (where targeted) data center tier or similar certification requirements. Confirm against the current project specification, not against this training material.

About this lab: UPS Sizing, Battery Runtime & Critical Power Design Lab

UPS Sizing, Battery Runtime & Critical Power Design Lab

Explore UPS sizing, battery runtime, operating topologies and redundancy arrangements using interactive diagrams and critical power calculations.

Topics you can explore

  • UPS battery runtime
  • UPS in MEP
  • Critical power redundancy

Practical engineering example

An illustrative 5 kW load for 30 minutes needs 2.5 kWh at the load before losses and battery allowances. Use the lab to explore inputs, then verify runtime with manufacturer discharge data.

Developed by Eng. Mannan Maqsood. Verify design outputs against current project requirements and the applicable authority and manufacturer data.

© 2026 Eng. Mannan Maqsood | Engineering Excellence Through Interactive Labs

VERIFY ALL FIGURES AGAINST CURRENT MANUFACTURER DATASHEETS AND LOCAL AUTHORITY REQUIREMENTS BEFORE DESIGN SIGN-OFF.