The difference between an automatic transfer switch (ATS) and a static transfer switch (STS) in data center power design — how fast each one switches, what kind of failure each protects against, and why a facility typically needs both, not one or the other.
Two Different Switches, Two Different Jobs
An automatic transfer switch (ATS) and a static transfer switch (STS) both move electrical load from one power source to another, but they solve different problems, operate on very different timescales, and are typically both present in the same data center rather than being alternatives to each other. Confusing the two — or assuming one can do the other's job — is a common source of misunderstanding about how data center power switching actually works.
The ATS: Mechanical Switching Between Utility and Generator
An automatic transfer switch is an electromechanical device that switches a facility's power source between the utility connection and onsite backup generation, and it's the switch responsible for transferring load onto generators after a utility outage (see Data Center Generator Systems Explained). Because it involves physically moving mechanical contacts, an ATS switching operation takes on the order of tens of milliseconds to a few cycles — fast by everyday standards, but far too slow to protect sensitive electronic loads from even a brief interruption. This is fine for the utility-to-generator transfer specifically, because the UPS is already carrying the load without interruption throughout that entire sequence; the ATS's mechanical switching speed simply doesn't matter for IT continuity in that scenario.
The STS: Electronic Switching Between Redundant UPS Outputs
A static transfer switch uses solid-state electronic switching — no moving mechanical contacts — to transfer a load between two live, redundant power sources, typically the outputs of two independent UPS systems feeding the same downstream equipment in a 2N architecture (see AI Data Center Power Distribution Explained). Because it's purely electronic, an STS can switch in a fraction of a single AC cycle — fast enough that sensitive IT equipment sees no interruption at all, even mid-cycle. This speed is what makes an STS suitable for protecting single-corded equipment (servers with only one power input) that needs to draw from a redundant A/B power architecture without itself having dual power supplies to do the job.
Why a Facility Typically Needs Both
The ATS handles the relatively slow, infrequent transition between utility and generator power — a transition the UPS's own battery runtime already covers for continuity, so ATS switching speed isn't the limiting factor there. The STS handles a different problem entirely: giving single-corded equipment the same protection dual-corded equipment gets automatically from drawing on two power paths simultaneously, by switching fast enough between two live UPS outputs that the equipment never notices a gap. A well-designed data center power architecture uses both devices at the stages where each one's specific switching characteristics actually matter, rather than treating them as interchangeable.
What to Look For on a Single-Line Diagram
On a facility's single-line diagram (see AI Data Center Single-Line Diagram Explained), an ATS typically appears near the generator connection, positioned to switch the facility's main power source between utility and generation. An STS typically appears further downstream, closer to the IT load, positioned between the outputs of redundant UPS systems and the equipment it protects. Spotting both — and understanding which specific failure each one is protecting against — is part of correctly reading a data center's full redundancy architecture rather than just its high-level N+1/2N label.