Not every facility needs the same level of power protection. A retail shop that closes at 9pm has very different requirements from a hospital, a data centre, or a cold storage warehouse that runs continuously. For genuinely 24/7 operations, an uninterruptible power supply is not a convenience feature. It is core infrastructure, and planning it correctly from the start determines whether it does its job when you need it most.
This guide walks through the key decisions: load sizing, runtime targets, UPS topology, redundancy, battery selection, and the maintenance practices that most facilities get wrong.
Start With Load Assessment
The VA or watt rating is the most important consideration when selecting a UPS. Get this wrong, and nothing else matters.
Load assessment means calculating the total power draw of everything connected to the UPS: servers, networking equipment, medical devices, industrial control systems, SCADA, refrigeration monitoring, or whatever your critical load actually is. From that figure, you select a UPS rated to handle it.
Power factor matters here. UPS systems are rated in VA (volt-amperes), but your equipment draws watts. These are not the same number. A server drawing 900W with a 0.9 power factor has an apparent power of 1000VA. Size based on watts alone and you will undersize the system.
Before selecting any equipment, establish:
- Current critical load in VA and watts
- Expected load growth over the next 3 to 5 years
- Which loads are genuinely critical versus which can tolerate interruption
- Whether any loads have high inrush current at startup
That last point matters for facilities with large motors, compressors, or industrial equipment where startup current can be several times the running current.
Understand What Runtime Actually Means
There is a common misconception about how long a UPS is supposed to run. Most UPS systems designed for critical facilities are sized for 5 to 15 minutes of runtime. That is not a limitation. That is the design intent.
In facilities with standby diesel generators, the UPS has one job: bridge the gap between a mains failure and the moment the generator accepts load. That transition takes minutes. The UPS does not need to run the facility independently for hours. It needs to hold the load long enough for the generator to take over.
If your facility has no generator, the calculation changes completely. You are sizing for standalone runtime, and battery capacity requirements increase significantly.
Decide early whether your UPS is part of a generator-backed power chain or operating as a standalone backup. That single decision drives everything downstream.
Choose the Right UPS Topology
Not all UPS systems work the same way. For any genuine 24/7 operation, double-conversion online UPS is the standard.
In a double-conversion system, the load runs continuously on the inverter output. Mains power feeds the rectifier, which charges the battery and powers the inverter. If mains power fails, the load does not notice. The DC bus draws from the battery and the inverter keeps running. Transfer time is zero.
This matters for sensitive equipment. Servers, medical devices, and industrial control systems can be sensitive to even brief power interruptions. Any transfer time introduces a window of risk, even if that window is measured in milliseconds.
For lower-criticality environments, line-interactive UPS systems offer a cost-effective alternative. They regulate voltage without constantly running the inverter, which improves efficiency. But they do have a brief transfer time when switching to battery. Whether that is acceptable depends on what is connected.
Plan Your Redundancy Level
Once you have the right topology, the next question is how much redundancy the facility needs. Two configurations cover most 24/7 operations:
N+1 redundancy adds one additional UPS module beyond the minimum required to carry the full load. If any single module fails, the remaining modules absorb the load and the facility keeps running. This is the standard for most critical facilities: data centres, manufacturing operations, larger commercial buildings.
2N redundancy takes it further. Two completely independent UPS systems, each rated to carry 100% of the load, operate in parallel. If one entire system fails, the other takes over. This level is used where any single-system failure must not reach the load: hospitals, tier IV data centres, financial infrastructure.
The right choice depends on your risk tolerance and budget. N+1 is appropriate for most operations. 2N is for environments where the consequences of any interruption are severe enough to justify the cost of a fully independent backup system.
VRLA or Lithium-Ion Batteries?
Battery selection has a significant impact on footprint, total cost of ownership, and maintenance requirements. Most UPS installations use one of two technologies:
| VRLA (Lead-Acid) | Lithium-Ion | |
| Design life | 3 to 5 years | 10 to 15 years |
|
Energy density |
Baseline | 3 to 5x higher |
| Physical footprint | Larger | 40 to 60% smaller |
| Upfront cost | Lower | Higher |
| Mid-life replacement | Required | Not typically needed |
VRLA (valve-regulated lead-acid) is the established technology. Lower upfront cost, widely available, and well-understood. The tradeoff is a shorter lifespan that typically requires a full battery replacement partway through the UPS system's working life.
Lithium-ion costs more to install but delivers 3 to 5 times the energy density of VRLA, resulting in substantially smaller battery cabinets. A design life of 10 to 15 years means most lithium-ion installations will not need a mid-life replacement cycle, which changes the total cost picture significantly over a 10-year horizon.
For space-constrained facilities or operations planning to hold equipment for a decade or more, lithium-ion is worth serious consideration. For facilities with tighter upfront budgets, VRLA remains a capable option provided the replacement cycle is planned and funded in advance.
The Maintenance Traps That Catch Facilities Out
Two things consistently cause UPS failures that should not happen.
Battery degradation goes unnoticed until it is too late. Lead-acid batteries degrade silently. A battery bank rated for 8 minutes of runtime when new may have dropped to 5 or 6 minutes after a few years of service. The next step is not a gradual slide to fewer minutes. It can go from reduced capacity to zero with almost no additional warning. Annual battery testing is the minimum required to catch this before a real outage exposes it.
Maintenance bypass removes all protection. When a UPS is placed in maintenance bypass for servicing, the battery is taken out of the circuit entirely. The load runs directly from raw utility power with no protection against outages, surges, or sags. This is necessary for certain maintenance tasks, but it is a window of complete exposure. Operations teams need to know when their systems are in bypass, plan those windows carefully, and ensure it is always a deliberate decision.
Both of these are planning failures, not equipment failures. The equipment does what it is designed to do. The question is whether the people responsible for it have visibility into its actual state.
Getting the Design Right from the Start
A UPS that is correctly sized, appropriately redundant, and properly maintained will do its job for years without incident. One that was sized on incomplete load data, missed a battery replacement cycle, or was never properly tested will fail at the worst possible moment.
If you are planning power backup for a new facility or reviewing an existing installation, speak to the team at UPS Solutions about sizing, topology, and the right UPS system for your operational requirements.