Electrical Systems for Data Centres & Server Rooms: A Planning Guide

Whether you operate a dedicated data centre in Greater Manchester or maintain a modest server room in your Cheshire office building, the electrical infrastructure supporting your IT equipment is arguably the most critical system in your organisation. A single power interruption, voltage fluctuation, or cooling failure can result in catastrophic data loss, extended downtime, and costs that run into tens of thousands of pounds. After more than 25 years designing and installing electrical systems for commercial and industrial clients across North West England, we have seen first-hand how proper planning separates resilient operations from vulnerable ones.

Understanding Power Demand and Load Calculations

Server rooms and data centres present unique challenges when it comes to electrical load calculations. Unlike a standard commercial space where loads are relatively predictable, IT environments must account for current equipment draw, planned expansion, and the substantial additional load created by cooling systems.

A common mistake we encounter is sizing the electrical supply based solely on existing server hardware. Best practice demands that you plan for at least 30% additional capacity above your current peak demand. This headroom accommodates future growth without requiring a costly infrastructure overhaul every time you add a new rack.

• Measure actual consumption: Use power monitoring at the rack level to establish genuine baselines rather than relying on manufacturer nameplate ratings, which often overstate real-world draw.
• Account for cooling loads: Cooling systems typically consume 30 to 50% of total data centre power. These must be factored into your overall electrical capacity from the outset.
• Plan your power density: Modern high-density racks can draw 10kW or more per unit. Ensure your distribution boards and cabling can handle concentrated loads without overheating.

Power Redundancy and UPS Systems

Redundancy is the cornerstone of data centre electrical design. The industry-standard tier classifications provide a useful framework. Most commercial server rooms in the Merseyside and Greater Manchester area should aim for at least N+1 redundancy, meaning one additional power path beyond the minimum required to run the facility.

Uninterruptible Power Supply (UPS) systems are non-negotiable for any serious IT environment. There are three main types to consider:

• Online double-conversion UPS: The gold standard for data centres. These continuously convert incoming AC power to DC and back again, providing complete isolation from mains supply irregularities. We recommend these for mission-critical environments.
• Line-interactive UPS: A cost-effective option for smaller server rooms where brief switchover times (typically 2 to 4 milliseconds) are acceptable.
• Static transfer switches: For larger installations, these enable instantaneous switching between dual power feeds, ensuring continuity if one supply fails.

Battery runtime should be calculated to bridge the gap until standby generators reach full load. We typically recommend a minimum of 10 to 15 minutes of battery autonomy, though this varies depending on generator start-up times and the criticality of your operations.

Standby Generation and Dual Supply Feeds

For facilities where downtime is simply not an option, standby diesel generators provide the extended backup that UPS batteries alone cannot deliver. The electrical integration of generators requires careful attention to automatic transfer switch (ATS) design, ensuring seamless changeover without transient voltage spikes that could damage sensitive IT equipment.

Many of our clients across Cheshire and Greater Manchester have benefited from arranging dual incoming mains supplies from the local Distribution Network Operator. This provides an additional layer of resilience before your on-site backup systems even come into play. While this involves coordination with the DNO and additional infrastructure cost, it is increasingly common for businesses whose operations depend entirely on continuous IT availability.

Electrical Distribution and Cable Management

Inside the data centre or server room, the quality of your electrical distribution directly affects both reliability and maintainability. We recommend dedicated distribution boards serving only IT loads, completely separated from general building circuits. This isolation prevents electrical noise from other building systems affecting sensitive equipment and simplifies maintenance shutdowns.

Cable management deserves particular attention:

• Segregate power and data cabling: BS 7671 (the IET Wiring Regulations) requires adequate separation between power and data cables to prevent electromagnetic interference. In practice, we use separate containment routes or shielded trunking.
• Use appropriately rated cables: Low Smoke Zero Halogen (LSZH) cables are essential in enclosed server environments. In the event of a fire, standard PVC cables release toxic gases that damage equipment and endanger personnel.
• Label everything meticulously: Every circuit, breaker, and connection should be clearly labelled. During an emergency, your facilities team needs to identify and isolate circuits within seconds, not minutes.

Cooling System Electrical Requirements

Cooling is inseparable from electrical planning in data centre environments. Precision air conditioning units, in-row coolers, and chilled water systems all place significant electrical demands on your infrastructure. The Power Usage Effectiveness (PUE) metric, which measures total facility power divided by IT equipment power, provides a useful benchmark. A well-designed facility should target a PUE of 1.4 or lower.

We work closely with HVAC specialists to ensure that cooling system electrical supplies are properly sized, independently circuited, and backed up by UPS or generator systems. Losing cooling during a power event can be just as damaging as losing power to the servers themselves. Modern hot aisle/cold aisle containment strategies can significantly reduce cooling energy consumption, and the electrical infrastructure should be designed to support these configurations from the start.

Compliance, Testing, and Ongoing Maintenance

Data centre electrical installations must comply with BS 7671 and should be carried out by NICEIC-approved contractors who can certify the work to the highest standards. Beyond the initial installation, regular inspection and testing is essential. We recommend Electrical Installation Condition Reports (EICRs) at intervals no greater than three years for data centre environments, with more frequent thermal imaging surveys to identify potential connection failures before they cause outages.

UPS batteries require regular testing under load to verify their actual runtime matches the design specification. Battery degradation is gradual and often undetected until a real power event exposes the shortfall. A planned maintenance programme should include quarterly UPS health checks and annual battery discharge tests.

Regular thermal imaging of distribution boards and UPS connections can identify hotspots caused by loose terminations or overloaded circuits, preventing failures before they happen.

At DRM Electrical, we have designed and maintained electrical systems for server rooms and data environments across Cheshire, Greater Manchester, and Merseyside for businesses ranging from financial services firms to logistics companies. Every installation is NICEIC-certified and built around the specific resilience requirements of your operation. If you are planning a new server room, expanding your existing facility, or simply want a professional assessment of your current electrical resilience, get in touch with our team to arrange a consultation.