Eaton has announced a new edge-based solution that enables its Power Xpert Quality (PXQ) event analysis system to detect large energy fluctuations—called “AI power bursts”—associated with artificial intelligence computing infrastructure. Available through a remote firmware update, this solution helps data center operators identify and address potential subsynchronous oscillations (SSO) before these fluctuations can damage infrastructure or disrupt operations.
According to Eaton, SSO events can result from the sudden, significant load spikes generated by GPU servers in AI data center environments. These power bursts pose substantial risks to both data center hardware and the surrounding grid, potentially leading to transformer overheating, ferroresonant damage, or equipment failure if left unchecked.
The PXQ power quality meter, typically deployed within switchgear, switchboards, and power distribution units, now features advanced edge analytics designed to streamline detection and analysis of SSO. Eaton says the firmware upgrade empowers existing PXQ users to proactively monitor, detect, and take preventive actions on emerging power quality issues—including sags, swells, transients, and harmonics—without deploying additional hardware.
“The energy demands of AI workloads surpass anything data centers and the grid have encountered before, with load fluctuations that can exceed the limits of existing infrastructure,” said JP Buzzell, vice president and chief data center architect at Eaton. “By enabling customers to harness their existing PXQ technology in new ways, we’re delivering a market-first capability to effectively respond to AI power bursts. Our SSO detection solution marks a major milestone in our grid-to-chip strategy, which empowers data center operators with innovative tools to address and manage the energy challenges posed by AI.”
Eaton reports that this update aligns with its broader grid-to-chip strategy, which includes intelligent power distribution, backup systems, and digital tools aimed at optimizing white and gray space in AI-centric data centers. The company also references recent collaborations with NVIDIA focused on 800 volt direct current (VDC) power infrastructure for 1 megawatt racks and with Siemens Energy for fast-tracked onsite power integration.
Eaton will demonstrate its grid-to-chip portfolio at Yotta 2025 from September 8–10 in Las Vegas, booth 319, and will feature expert sessions on meeting AI data center energy challenges during the event.
Source: Eaton
