Data centre electrical design is the engineering of a power system whose single overriding requirement is that the IT load never loses power — not during a fault, not during maintenance, and not during a utility outage. In Australia it combines high-voltage distribution, standby generation, UPS systems and redundant architectures, all designed to a defined reliability tier and proven through power system modelling. It is unlike standard industrial electrical design because redundancy, not efficiency, is the governing constraint.
Here’s what data centre electrical design involves, the standards that apply in Australia, and how the design gets proven rather than just drawn.
What reliability tiers apply to data centres?
Data centres are designed to a reliability tier that defines how much redundancy the electrical system carries:
- N — exactly enough capacity to run the load, with no redundancy.
- N+1 — one spare unit beyond what’s needed, so any single component can fail or be serviced with no loss of supply.
- 2N — fully duplicated, independent systems (A and B), so an entire path can be lost with no impact on the load.
These map to the Uptime Institute Tier classifications (Tier I to Tier IV), where Tier III adds concurrent maintainability (any component serviceable without downtime) and Tier IV adds fault tolerance (the facility survives a single unplanned failure anywhere). The tier is a commercial commitment as much as a technical one — it’s what the facility is sold on — so the electrical design has to demonstrably deliver it.
How is power distributed in a data centre?
Most Australian data centres take a high-voltage or medium-voltage (HV/MV) utility supply and step it down through on-site transformers to feed the switchrooms. The distinguishing features of the distribution design are:
- Segregated A/B distribution — two physically and electrically independent paths carried all the way to the dual-corded IT load, so power arrives from two directions at once.
- Multiple independent supply paths feeding each critical board, sized so any one can carry the full load alone.
- Modular, repeatable blocks — a validated distribution unit replicated across data halls, which suits the scale and delivery speed of hyperscale builds.
How do UPS and standby generation fit in?
Two systems bridge the gap between a utility outage and continuous supply:
UPS (uninterruptible power supply) carries the critical load instantaneously through any supply disturbance, holding it up on stored energy for the seconds it takes standby generation to pick up. UPS topology and its fault contribution have to be built into the model, because it behaves very differently from a utility source.
Standby generation — typically diesel generator sets — provides sustained backup during a prolonged utility outage. The design has to cover generator sizing, paralleling, transfer schemes and, critically, the protection coordination that holds up while the source of fault current changes from utility to generator.
Why do data centres need power system studies?
Because the redundancy claims are the product, they have to be proven, not assumed. Data centre electrical design leans heavily on power system studies:
- Load flow across normal and failure-mode configurations, confirming that when a path is lost the remaining path isn’t overloaded.
- Fault level studies to rate equipment and set protection for supply from utility, generator or UPS.
- Protection coordination that maintains discrimination across multiple parallel supply paths — a genuinely harder problem than a single radial feed, because fault current can arrive from several directions.
- Arc flash assessment for personnel safety across every operating and failure state.
What standards apply to data centre electrical design in Australia?
The core Australian standards are:
- AS/NZS 3000 (the Wiring Rules) — the baseline requirements for electrical installations.
- AS 3009 — electrical installations for standby power / emergency supply, directly relevant to generator and transfer design.
- Relevant AS/NZS 3008 cable sizing, network operator connection requirements, and the Uptime Institute Tier standard where the client is certifying to it.
Designing to the Wiring Rules is the floor; delivering a Tier III or Tier IV facility requires layering redundancy and concurrent maintainability on top, and proving it through modelling.
Proof: the DigiCo SYD1 data centre
JMB Engineering applied exactly this approach on the DigiCo SYD1 data centre, directing the same power system modelling and protection rigour we use on complex mine reticulation at the uptime and redundancy demands of a critical data centre. The protection coordination across redundant supply paths — keeping discrimination in every operating and failure state — is the kind of problem where modelling isn’t optional.
The bottom line
Good data centre electrical design combines two things that don’t always sit together: the discipline of critical-infrastructure redundancy and the modelling rigour to prove it. The single-line diagram has to be correct, but more importantly the availability it promises has to be demonstrable across every failure mode. That’s the difference between a facility that’s drawn to a tier and one that actually delivers it.
Planning a data centre build or expansion in Australia? Explore our electrical design and drafting service or the underlying power system studies capability.
FAQ
What is the difference between N+1 and 2N redundancy in a data centre?
N+1 provides one spare component beyond the minimum needed to run the load, so any single unit can fail or be serviced without loss of supply. 2N fully duplicates the entire system into two independent A and B paths, so an entire path can be lost with no impact. 2N is more resilient and more expensive, and typically maps to higher Uptime tiers.
What Australian standards apply to data centre electrical design?
The primary standards are AS/NZS 3000 (Wiring Rules) for the installation baseline and AS 3009 for standby and emergency power. Cable sizing follows AS/NZS 3008, and facilities certifying to a reliability tier also design to the Uptime Institute Tier standard.
Why do data centres need power system studies?
Because their reliability tier is a commercial promise that has to be proven. Load flow, fault level, protection coordination and arc flash studies demonstrate that the design keeps the load supplied across normal operation and every credible failure mode — not just under ideal conditions.
Does data centre protection coordination differ from standard industrial work?
Yes. With multiple parallel supply paths and sources (utility, generator, UPS), fault current can flow from several directions, and the protection scheme must stay coordinated in every operating and failure state. That is significantly harder than grading a single radial feed and relies heavily on an accurate power system model.
—
JMB Engineering Pty Ltd — electrical engineering consultancy specialising in power system studies, protection, earthing and electrical design for data centres, mining, renewables and heavy industry. Thornton, NSW.
