1 MW Solid State Transformer: 98% Live Feeder Test Validated by NC State, NYPA and EPRI
Project Background: From Lab Prototype to Live Distribution Feeder
In August 2026, North Carolina State University (NC State), the New York Power Authority (NYPA) and the Electric Power Research Institute (EPRI) jointly announced the completion of a milestone field test at EPRI’s power delivery laboratory in Lenox, Massachusetts. A 1 MW solid state transformer (SST) operated on a live 13.8 kV distribution feeder with approximately 98% end-to-end conversion efficiency. The demonstration is the first independently verified, megawatt-class SST validation on a live utility distribution network anywhere in the world.
1.1 The Real Positioning of the Three-Party Collaboration
The NC State team, led by Srdjan Lukic (Lampe Distinguished Professor of Electrical and Computer Engineering), was responsible for the design, fabrication and deployment of the prototype. EPRI led the test plan and data acquisition. NYPA contributed engineering support, application scenario definition, and downstream grid interconnection evaluation. Initial funding for the prototype came from the U.S. Department of Energy’s Transportation Technologies Office, while NYPA underwrote assembly and field-testing labor.
1.2 Engineering Significance of the Test Site
The EPRI Lenox laboratory operates a complete 13.8 kV distribution feeder interface that can replicate North American utility grid voltage flicker, harmonics and short-circuit conditions. Compared with previous vendor-led megawatt-class SST validations on purpose-built closed-loop platforms, this test is the first to connect a prototype to a real utility network that researchers cannot fully control.
For related coverage of the broader DOE-funded SST program that complements this milestone, see our profile of the Hybrid Solid-State Transformer Concept Wins DOE Feeder Prize.
Beyond the hardware itself, the test produced operating data that closed a long-standing credibility gap for SST technology. Industry analysts have repeatedly pointed out that megawatt-class SST efficiency numbers published before 2026 were generated on controlled test benches and did not necessarily reflect operation on utility feeders with their wider voltage variation, harmonic background, and unbalanced loading. The Lenox demonstration addressed that critique directly: data were collected across multiple load points and under real feeder events, not synthesized from ideal waveforms.
Core Technical Architecture of the 1 MW SST
2.1 Functional Integration: Transformer Plus Rectifier
Conventional medium-voltage AC to low-voltage DC conversion requires two pieces of equipment: a line-frequency transformer for voltage step-down, plus an active or passive rectifier to produce DC. The SST compresses both functions into a single device using high-frequency switching (typically 10 to 100 kHz). A high-frequency isolation transformer replaces the line-frequency transformer, and a high-frequency rectifier module replaces the discrete diode or thyristor rectifier.
The resulting volume reduction is order-of-magnitude. According to the dimensions disclosed by the NC State team, the 1 MW SST occupies roughly 30% to 40% of the footprint of a comparable power rating conventional transformer-plus-rectifier solution, with corresponding weight reduction.
2.2 Semiconductor Power Device Selection
For broader context on power-electronic and solid-state device developments across the grid, see the transformer industry developments tag archive.
Megawatt-class SST designs typically use 3.3 kV, 6.5 kV, or 10 kV SiC MOSFET modules, with some prototypes employing 15 kV SiC IGBTs or GaN HEMTs. The NC State team deployed a 3.3 kV SiC MOSFET series-parallel topology at approximately 20 kHz switching frequency, balancing switching losses against magnetics volume.
2.3 Control and Protection Architecture
The SST control core uses a layered structure. The bottom layer implements nanosecond-scale gate drive and protection (short-circuit, overcurrent, overtemperature) on FPGA and MCU hardware. The middle layer runs voltage and power regulation loops on DSP. The top layer provides IEC 61850 and DNP3 communication interfaces to SCADA and DERMS platforms. NYPA emphasizes that SSTs natively support bidirectional power flow, making them a unified gateway for distributed energy resource interconnection, energy storage interfaces, and V2G charging stations.
A key design choice for the Lenox prototype is the use of a modular building block. Each power stage is built from identical 100 kW sub-modules that can be paralleled to reach 1 MW. This modularity has two practical consequences. First, a faulted sub-module can be bypassed and replaced without taking the entire SST offline, which is the kind of graceful degradation utilities require for reconfiguration. Second, the same sub-module can be applied across a wide product portfolio, from 250 kW commercial-building units to multi-megawatt utility installations, which compresses non-recurring engineering cost across the roadmap.
Field Test Data and Efficiency Performance
For an industry perspective on conductor and winding-wire selection criteria that complement the SST control architecture, see the related coverage on conductor selection criteria for transformer winding design.
3.1 Efficiency Verification Results
The overall end-to-end conversion efficiency disclosed by the test team is approximately 98% (AC-DC-AC or AC-DC full link). Conventional transformer-plus-six-pulse rectifier systems at the same power rating typically achieve 92% to 94%, so the SST gains 4 to 6 percentage points. Considering the internal losses for auxiliary supplies, control boards, and forced-air or liquid cooling, an end-to-end 98% figure is engineering-meaningful.
3.2 Harmonics and Power Factor
The EPRI test report shows that across the 50% to 100% load range, output-side current total harmonic distortion (THD) remains below 3%, meeting IEEE 519 medium-voltage interconnection limits. Input-side power factor is dynamically adjustable between 0.95 and 1.0, eliminating the need for additional reactive compensation equipment.
3.3 Key Operating Parameters Comparison
| Parameter | Conventional Transformer + Rectifier | 1 MW SST (This Test) |
|---|---|---|
| End-to-end efficiency | 92% to 94% | ~98% |
| Volume (relative) | 100% (baseline) | 30% to 40% |
| Bidirectional power flow | Not supported | Natively supported |
| Voltage regulation | On-load tap changer (mechanical) | Millisecond power-electronic regulation |
| Fault ride-through | Limited | Programmable LVRT/HVRT |
Application Scenarios: Why Utilities Care About SST
4.1 Data Center Power Distribution
Both NYPA and EPRI highlight data center applications in the test report. The 800 VDC distribution architecture requires efficient, low-footprint conversion from front-end medium-voltage AC to cabinet-level DC. The SST can output 800 VDC directly, eliminating the line-frequency transformer, the standalone rectifier, and the DC-DC boost stage present in conventional architectures, thereby simplifying the chain and improving efficiency.
4.2 Urban Distribution and Load-Dense Areas
Load-dense areas such as New York City and Boston face substation siting difficulties. Conventional 500 kVA to 2.5 MVA pad-mount substations are large and noisy. The compact nature of the SST enables deployment in underground substations, building distribution rooms, and even parking-lot footprints. The same logic applies to electric vehicle ultra-fast charging stations: a single site with over 1 MW of demand can be powered by a prototype-class SST on its own.
4.3 Distribution-Level Regulation Capability
NYPA’s briefing identifies four distribution-grid values for SSTs: bidirectional power flow (supporting DER back-feed), voltage regulation (replacing or supplementing OLTC), renewable energy integration (smoothing wind and solar fluctuations), and high-load growth backup power. The millisecond-level controllability of SSTs is the fundamental differentiator from line-frequency transformers.
Funding Structure and Industrial Promotion Path
5.1 DOE Funding Logic
The DOE Transportation Technologies Office’s involvement reveals another promotion path for SSTs: transportation electrification. EV ultra-fast charging, V2G charging stations, and port shore power systems all require efficient medium-voltage AC to low-voltage DC conversion, which is exactly the gap SSTs fill. DOE funding signals that SSTs are no longer purely academic; they are part of the national energy-transition toolkit.
5.2 NYPA’s Promotion Path
As the largest state-level public power organization in the United States, NYPA’s participation signals utility engineering acceptance of SSTs. NYPA plans to conduct long-term grid-connected trials at its own substations and EV fast-charging stations and incorporate the results into equipment procurement technical specifications.
5.3 EPRI’s Follow-Up Plan
EPRI will publish a field-test methodology file based on this demonstration, enabling other utilities to replicate it. EPRI will also drive the relevant IEEE working group to incorporate SST test procedures into the standards.
The methodology document is expected to cover feeder selection criteria, instrumentation for high-bandwidth voltage and current capture, baseline load profile characterization, fault injection procedures, and acceptance criteria for utility-side verification. Standardizing these steps is a prerequisite for cross-vendor comparability of SST efficiency and reliability numbers, which in turn is needed for utilities to write procurement specifications that do not have to be rewritten per vendor.
Industry Significance and Outlook
6.1 A Key Milestone in SST Industrialization
This test closes the final gap in SST industrialization: independent third-party verification data on a live utility network. ABB, Siemens and GE have previously announced traction-class SST products, but those were predominantly closed- or semi-closed-loop tests. This demonstration provides the foundation for subsequent commercial tenders, procurement specifications, and reliability statistics.
6.2 Alignment with the SiC/GaN Roadmap
Volume supply of 3.3 kV SiC MOSFETs and the production ramp of 6.5 kV, 10 kV and 15 kV devices are the key drivers of SST cost reduction. Wolfspeed, II-VI, Onsemi and Rohm are expanding 8-inch SiC wafer capacity through 2026. Unit cost per kW for SSTs is expected to drop 30% to 40% by 2027 to 2028.
6.3 Coupling with New Distribution Architectures
800 VDC distribution, SST medium-voltage interfaces, and solid-state circuit breakers (SSCB) together form the hardware foundation of next-generation distribution grids. The 1 MW SST tested here can serve as the front-end supply for 800 VDC data centers and AI compute facilities, with the potential to scale in data center power distribution over the next 5 to 10 years.
Conclusion
The 1 MW solid state transformer field test at the EPRI Lenox laboratory marks the inflection point for SST technology moving from research prototype to utility-grid product. The collaboration model among NC State, NYPA and EPRI (academic R&D plus utility application plus independent testing) provides a replicable engineering template for subsequent 1 MW solid state transformer industrialization. Across the four main lines of data center distribution, urban substations, EV ultra-fast charging, and renewable integration, the 1 MW solid state transformer class now has the engineering foundation for real-world deployment.