SiC modules are most valuable in high-power online double-conversion UPS systems, where the rectifier, inverter, and battery converter process electrical power continuously. Their lower switching and conduction losses improve efficiency, reduce cooling effort and increase power density. The commercial business case should be validated against electricity cost and lifetime expectation. The system configuration depends on topology, load profile, and operating conditions.

This article answers some of the frequently asked questions around SiC modules for UPS. 

Frequently asked questions about SiC modules for UPS systems

1. How much can SiC modules improve UPS efficiency?

SiC modules improve UPS efficiency by reducing switching and conduction losses in both the rectifier and inverter stage. The total efficiency gain depends on factors, such as topology, switching frequency, cooling design, and load profile.

Under the evaluated conditions, the CoolSiC™ inverter demonstrated a 0.69 percentage-point higher calculated stage efficiency than the silicon-based NPC2 topology.

These figures represent inverter-stage calculations and not certified end-to-end UPS efficiency values.

Comparing SiC-based UPS solutions at the system level is the right approach, rather than considering the module price alone. The business case should include semiconductor cost, energy loss, cooling energy, heatsinks, magnetic components, filter components, cabinet space, converter module count, maintenance cost, expected operating lifetime, and overall system uptime.

Example: A 1 MW UPS operating at 50% load over five years is estimated to incur loss-related energy costs of €84,000 for a two-level silicon design, €76,000 for a three-level silicon design, and €45,000 for two-level SiC design.

The calculation assumes €0.10/kWh and a 1.2 cooling-energy factor; it is an illustrative scenario, not a universal payback claim.

The higher cost of SiC is often justified in medium- and high-power UPS systems, where efficiency, energy consumption, and space utilization directly affect operating and installation costs.

The strongest return is typically achieved in data center and critical infrastructure applications that run continuously and benefit from reduced cooling demand and higher system efficiency.

SiC modules dissipate less heat than conventional silicon solutions. As a result, system designers can reduce cooling requirements or use the available thermal margin within the existing cooling architecture to increase power density.

Higher switching frequencies also enable the use of smaller inductors, transformers, and filter elements, helping reduce the overall UPS footprint without compromising performance.

SiC modules can improve both two-level and three-level UPS architectures, but they are particularly effective in two-level topologies, where their low switching losses enable higher efficiency without the added complexity of additional switching devices.

The main decision criteria include DC-link voltage, efficiency target, switching frequency, and overall system architecture.

Key specifications include:

  • Voltage class and DC-link transient margin
  • Continuous, peak and reverse current
  • RDS(on) across junction temperature
  • Switching energy at the intended current, voltage and gate resistance
  • Package topology: half-bridge, 4-pack, 6-pack, booster, or 3-level
  • Thermal resistance and transient thermal impedance
  • Gate-charge, gate-voltage, and short-circuit limits
  • Isolation, creepage, temperature sensing, and terminal technology
  • Safe operating area, qualification, and product lifecycle

Engineers should evaluate these parameters against the UPS mission profile rather than selecting devices solely based on power rating. Datasheet values provide an initial basis for selection and should be validated through simulation and laboratory testing under the intended operating conditions.

For most high-power online UPS systems, 1200 V SiC modules are a common choice as they provide adequate voltage margin for typical DC-link voltages while enabling efficient two-level converter designs.

Multilevel architectures may use a combination of different chip voltage classes within one module package depending on system requirements.

SiC devices switch much faster than IGBTs, making gate-driver performance more essential. Designers must carefully manage gate resistance, propagation delay, common-mode transient immunity, and protection functions such as DESAT detection and active Miller clamp to achieve reliable operation and prevent unwanted switching events.

The fast voltage and current transitions of SiC devices can increase electromagnetic interference if the power stage is not designed appropriately. EMI control begins with proper power-stage layout. Designers can control these emissions through compact current paths, low-inductance interconnections, optimized gate-drive settings, appropriate shielding, and suitable input and output filters.

Yes. Modern SiC modules undergo extensive reliability testing, including high-temperature operation, humidity exposure, power cycling, and high-voltage stress conditions. When used within their specified operating limits and integrated into a properly designed UPS system, SiC modules can support the long service life and continuous operation required in AI data center and other critical power applications.

A full SiC upgrade is not a simple drop-in replacement. The higher switching speeds of SiC devices can affect gate-drive requirements, protection settings, EMI behavior, and thermal performance. To fully realize the benefits of SiC technology, existing UPS platforms often require some degree of power-stage redesign.

There is no universal answer. SiC power modules provide a higher level of integration than discrete SiC MOSFETs, simplifying assembly, reducing parasitic inductance, and easing thermal management. These advantages often make modules the preferred choice for UPS and other high-power systems.

Discrete SiC MOSFETs can be advantageous in highly modular designs. However, assembling many individual devices can increase design and manufacturing complexity. Factors such as production capabilities, assembly processes, and workforce expertise can also influence the choice between modules and discrete devices.

Silicon IGBT modules remain a strong choice when minimizing upfront cost is the primary objective and constraints related to space, cooling and efficiency are less critical. In cost-sensitive or lower-switching-frequency UPS designs, the incremental benefits of SiC may not justify the additional semiconductor investment.

SiC modules require careful optimization of gate-drive design, PCB and busbar layout, thermal management, and EMI control to fully benefit from their fast-switching capabilities. Designers should evaluate switching frequency, protection strategy, and cooling architecture early in the design process to maximize efficiency, power density, and long-term reliability.

Takeaway

In UPS systems, SiC power modules are more than a device-level upgrade; they are a system-level enabler for higher efficiency, lower cooling effort, and increased power density. By reducing switching and conduction losses, SiC helps online double-conversion UPS designs improve energy performance and enables compact power architecture for AI data centers, critical infrastructure, and industrial power systems.

Want to improve UPS efficiency, power density and reliability?

Explore UPS solutions for data centers

Need the right power modules for your UPS design?

Optimize UPS power stages with CoolSiC™ modules

Additional resources