SPICE but Better: Qorvo’s Free Circuit Simulation Software for Power Engineers

As power and analog engineers increasingly face more complex hardware and software systems that impact their efficiency and productivity, Qorvo brings a unique combination of modern schematic capture and fast mixed-mode simulation to the table to help system designers face these complicated challenges.

 

QSPICE circuit simulation software. Image used courtesy of Qorvo

 

Historically, power engineers worked with analog control loops and silicon MOSFETs. Today, digital control algorithms and silicon carbide are areas of intensive research and development. 

“Analog SPICE engines can’t simulate complex digital circuits. In comparison, QSPICE integrates fast digital simulation with a benchmark-setting SPICE engine to accelerate innovation around these new technologies,” Jeff Strang, general manager for Qorvo’s Power Management business, said in an interview with EE Power.

System simulations integrating both dense digital and analog circuits have traditionally been the exclusive domain of expensive proprietary tools. QSPICE, Qorvo’s new generation of free circuit simulation software, allows system designers to simulate complex digital circuits and algorithms, resulting in improved simulation speed, functionality, reliability, and, ultimately, higher productivity.

 

A Step Up From SPICE: Improving Upon an Industry Standard

SPICE – developed in 1973 at the University of California Berkeley and arguably the go-to industry standard in circuit simulation – has been improved over time in terms of speed, functionality, and reliability for designers. 

In its latest iteration, QSPICE benchmarking tests performed by Qorvo produced reduced overall runtimes and a 100% completion rate, compared to a failure rate of up to 15% with the same test circuits using comparative SPICE simulators.

 

A guide to QSPICE. Video used courtesy of Qorvo

 

Advantages of QSPICE

QSPICE brings important improvements to the table regarding energy and power simulation.

Strang told EE Power that Qorvo has developed a native device type for its silicon carbide FETs, improving simulation speed and accuracy for FETs widely used in high-power and energy storage applications.

QSPICE also supports large amounts of digital logic through its built-in C++ and Verilog compilers, allowing power engineers to model complex circuit behavior, such as processor-based battery management algorithms and brushless motor control. 

For higher speed and accuracy, QSPICE provides an upgraded simulation engine, including a GPU-rendered user interface and SSD-aware memory management, that uses advanced numerical methods optimized for today’s computing hardware.

“The optimizations and improvements QSPICE introduces solve the most common problems in simulation,” Strang said. “Because of these improvements, QSPICE successfully completes 100% of a collection of ‘stress test’ circuits that serve as a benchmark for SPICE engine robustness. Of course, we’ll continue to innovate and add new capabilities to QSPICE, and we’ll be talking about those updates in the future.”

QSPICE is available free of charge and provides support for advanced analog and digital system simulations like those used in AI and machine-learning applications. Qorvo QSPICE’s model library is regularly updated with Qorvo's advanced power management and silicon carbide solutions in support of engineers as they evaluate and design. 

 

QSPICE in the Real World

In real-world terms, QSPICE can be used to solve current problems. For example, engineers testing machine learning algorithms for motor control or battery management can use QSPICE to import a RISC-V Verilog model, tie it to an analog model of the battery or motor, and quickly iterate through different concepts. 

Strang notes that analog control for power supply design is here to stay, and Qorvo is still developing technologies for applications such as smartphone power based on analog loops and silicon FETs. QSPICE supports advanced analog and digital system simulations like those used in AI and machine-learning applications.

“Silicon carbide becomes very interesting in higher power applications. When you’re at or above 1 kW, and you’re using 600 V or higher rated power switches, silicon carbide has clear advantages in efficiency, size, and cost compared to silicon-based switches,” Strang told EE Power. “For example, electric vehicle traction inverters are widely adopting SiC because of the longer range they offer due to their higher efficiency, as well as their smaller solution size due to higher switching frequency.”

Digital control loops are also useful in various applications like controlling a brushless, sensorless, three-phase electric motor.

“In order to implement field-oriented control and other advanced algorithms, designers must use a digital engine, such as a microcontroller. You can’t practically implement it using an analog loop,” Strang said. “Another example is the core power rail for a GPU or AI processor. These chips require hundreds or thousands of amps of current, and managing that power today is largely accomplished through advanced digitally-controlled multiphase regulators.”