Nigeria's grid power environment creates a set of risks for IT equipment that are largely absent in markets with stable, clean electricity supply. Voltage sags, spikes, complete outages, frequency variation, and harmonic distortion are routine, not exceptional. IT equipment, servers, network switches, workstations, storage systems, is sensitive to all of these. Unprotected equipment fails earlier, performs inconsistently, and creates data integrity risks during uncontrolled shutdowns.
The purpose of a UPS (Uninterruptible Power Supply) is not simply to bridge outages. That is one function. The broader purpose is to place a layer of clean, stable power between Nigerian grid realities and the equipment that depends on it. This article covers how to think about that protection requirement and how to plan for it across an office environment.
What Nigerian power actually does to IT equipment
Before sizing or selecting any UPS, it helps to understand the specific power problems that equipment faces:
Voltage sags and surges: Momentary drops below nominal voltage (sags) and rises above it (surges) are frequent. Sags are particularly common during generator start-up and load switching. Sustained under-voltage, sometimes called brownouts, occurs when demand exceeds supply on the local distribution network.
Complete outages: The most visible power event. Unprotected equipment shuts down without warning. Servers that shut down without a clean shutdown sequence risk database corruption, file system damage, and component stress from thermal cycling.
Frequency variation: Nigerian grid frequency nominally runs at 50Hz, but variation outside acceptable limits is common, particularly on generator supply during the run-up phase and under varying loads.
Harmonic distortion: Non-linear loads, variable speed drives, switch-mode power supplies, electronic ballasts, introduce harmonic distortion into the power supply that affects other equipment on the same circuit. In a building with multiple generators and large electrical loads, harmonics can be significant.
Transients: Very short, very high-voltage spikes caused by switching events, lightning strikes on the distribution network, and generator switching. Transients are the category of power event most likely to cause immediate, catastrophic component damage.
Each of these can be mitigated by an appropriately specified UPS. The key phrase is "appropriately specified", the wrong type of UPS for the environment provides less protection than it appears to.
UPS types and what they protect against
Offline (standby) UPS
The most basic type. Under normal conditions, utility power passes directly through to the equipment. When utility power fails completely, the UPS switches to battery. Switchover time is typically 5–25ms.
What it protects against: complete outages only. What it does not protect against: voltage sags and surges, harmonic distortion, frequency variation, transients that do not cause complete failure.
Offline UPS is appropriate for very low-stakes applications, a single workstation, a printer, where the primary concern is graceful shutdown rather than clean power. It is not appropriate for servers, networking equipment, or any device for which power quality matters.
Line-interactive UPS
More capable than offline. Under normal conditions, it uses an automatic voltage regulator (AVR) to correct voltage sags and surges without switching to battery. On complete failure, it switches to battery like an offline unit.
What it protects against: complete outages, moderate voltage sags and surges. What it does not protect against: harmonic distortion, frequency variation, severe transients, power problems that don't cause complete failure but degrade quality continuously.
Line-interactive UPS is appropriate for office workstations, network switches, and mid-tier equipment in environments where grid supply is inconsistent but not severely so. In Nigerian conditions with generators, it provides meaningful protection for workstations.
Online double-conversion UPS
The most complete protection. Utility power is continuously converted to DC (charging the battery bus) and then reconverted to clean AC output. The equipment connected to it never sees utility power directly. When utility fails, there is zero switchover time, the battery bus is always active.
What it protects against: all categories of power disturbance: outages, sags, surges, harmonic distortion, frequency variation, and transients.
Online double-conversion UPS is the correct specification for servers, storage systems, core network equipment, and any device where continuous clean power matters. In a Nigerian server room or data room, this is the only appropriate type.
The naming of UPS products is not standardised: some products marketed as "online UPS" are actually line-interactive. Before purchasing, ask specifically: is this true online double-conversion? Ask for the technical datasheet.
Sizing UPS capacity correctly
UPS capacity is measured in VA (volt-amperes) and watts. The watt rating is the real power capacity, this is what you use for sizing.
Step 1, Calculate the connected load
List every device that will connect to the UPS and find its power consumption in watts. This information is on the device specifications or its nameplate. For devices that list only amps, multiply by the voltage (220V in Nigeria) to get VA, then multiply by the power factor (typically 0.8 for most IT equipment) to get watts.
Sum the total wattage of all connected devices.
Step 2, Apply a loading factor
UPS efficiency decreases and battery life shortens when a UPS is operated near its rated capacity. A UPS should be sized so that the connected load is 60–80% of its rated capacity. Divide your total connected load by 0.75 to get the minimum UPS watt rating.
Step 3, Factor in runtime
Runtime, how long the UPS runs on battery during an outage, depends on battery capacity and connected load. Manufacturers publish runtime tables: at 50% load, the UPS runs for X minutes; at 100% load, for Y minutes.
For server rooms, target 15–20 minutes at full load, enough to bridge a generator start with margin. For office workstations, 5–10 minutes is typically enough for a controlled shutdown if the generator does not start. If the organisation genuinely needs to operate through extended outages on battery alone, external battery modules can extend runtime significantly.
Planning UPS across an office environment
A single large UPS at the building level may seem simpler than multiple smaller units, but distributed UPS planning is typically more practical and more resilient.
Server room / network room
One or more online double-conversion UPS units covering all servers and core network equipment. Sized for the full IT load with 30–50% headroom. Battery runtime of 15–20 minutes minimum at full load. Consider dual-feed configuration for critical servers, two UPS units, each on a separate circuit, with the server drawing from both through a dual-power-supply or automatic transfer switch.
Workstation areas
Line-interactive UPS units at the workstation level, sized to cover the workstation and monitor (typically 300–600W per station). Office workstation UPS serves two purposes: providing runtime for a controlled shutdown during extended outages, and providing voltage stabilisation that protects equipment from sags and surges during the generator switching period.
Active network equipment outside the server room
Corridor switches, wireless access points, IP phone systems, and similar equipment distributed through the building should be on UPS protection sized for the relevant load. These devices are often overlooked in UPS planning, creating a situation where the servers stay up during an outage but the network they connect to goes down.
Battery management and lifecycle
Batteries are the most maintenance-intensive and reliability-critical component of any UPS. A UPS with degraded batteries is a UPS that will fail at the worst moment, precisely when power goes out.
Key battery management practices:
- Regular testing: Most enterprise UPS units include a self-test function that discharges batteries partially to verify health. Run this on a schedule (monthly or quarterly) rather than discovering battery failure during an actual outage.
- Temperature: Battery life is strongly temperature-dependent. Batteries operating at 25°C last roughly twice as long as batteries at 35°C. Server rooms and network rooms should be cooled to protect both equipment and batteries.
- Replacement: Lead-acid batteries in UPS units typically have a useful life of three to five years in a properly maintained environment. Plan and budget for replacement before batteries reach end of life, not after they fail.
- Manufacturer batteries: Replacing UPS batteries with generic or off-brand alternatives to save cost is a false economy. The UPS is monitoring and charging based on the rated characteristics of its original battery chemistry. Mismatched batteries charge incorrectly and fail unpredictably.
For infrastructure solutions support including UPS sizing, installation, and battery maintenance programmes, a qualified infrastructure partner provides both the initial design expertise and the ongoing support that keeps the system working.
Frequently asked questions
How do we know if our current UPS is providing adequate protection?
Measure the load on the UPS (most units display this) and compare it to the rated capacity. If the load is above 80% of the UPS capacity, it is undersized. Also review the battery age and last test results, these are available on the UPS display or management interface on most enterprise units. If neither is available, that itself is a sign the UPS is not being actively managed.
Is it better to use one large central UPS or multiple smaller ones?
Both approaches are used. A central UPS simplifies monitoring and battery management but is a single point of failure. Multiple smaller units distributed by area or by equipment type are more resilient, a failed UPS affects a subset of equipment rather than everything, but require more management. For most Nigerian offices, a hybrid approach works well: one online double-conversion UPS for the server room and separate line-interactive units for workstation areas.
What should we do when the power comes back after an extended outage?
After a long outage, resist the instinct to immediately restore everything. NEPA/PHCN power returning after an extended outage sometimes arrives with elevated voltage or instability in the first minutes. A UPS in bypass or low battery may not provide protection in this window. If possible, allow the generator to continue powering the building for a period after grid returns, 10–15 minutes, to confirm stability before switching over.
How do we handle UPS capacity when we add new equipment?
Before adding significant new equipment, a new server, additional workstations, an expanded network room, recalculate the UPS load to confirm headroom remains adequate. Many organisations find their UPS headroom erodes over time as equipment is added without any load review. A UPS that was appropriately sized five years ago may now be overloaded. This review should be a standard part of any IT procurement process.
