Simulation Sparks Electronics Innovations

As electronics and communications are embedded in more and more products, and the density and complexity of high-performance electronic systems increases, design workflows have become more complicated and reliance on physical testing is no longer sustainable on its own. Electrical, thermal, mechanical and other considerations now must be considered from a system-wide perspective, and simulation—including multiphysics simulation approaches—is critical for validating designs.

“No longer can the various parts of an electronic system be designed in silos and cobbled together before prototyping,” says Sherry Hess, senior product management group director for multiphysics system analysis at Cadence Design Systems. “Designers now must consider issues of electromagnetic interference (EMI) and electromagnetic compatibility (EMC), as well as power integrity (PI), signal integrity (SI) and thermal integrity. To accelerate the design process and meet aggressive delivery schedules, engineers need to be able to perform cross-fabric and multiphysics analysis to model, simulate and analyze these effects on system-level designs.”

In addition, Hess says, many companies have adopted a “shift left” practice, where simulation and analysis are now an integral part of the design cycle from the beginning, rather than for verification at the end of the process. Multiphysics analysis technologies are becoming an integral part of each phase of the design process at the chip, package, board and complete system level, Hess says.

Electronics density is also putting more pressure to analyze designs for thermal and mechanical impacts.

“There is this constant drive to pack more into less space, and as you put things closer together and generate heat, that heat starts to build up,” says Matt Commens, director of product management at Ansys. “You can have mechanical failures, solder balls can fail, things can delaminate, and that can be a really challenging problem to test for and design against.”

This convergence and complexity has also made systems-level simulation more important, particularly in applications like automotive and aerospace. This can improve design quality, while also driving down costs and saving time.

“On the electronics side, simulation has always been important, but it was isolated to a specific SOC or a specific application,” says Tom De Schutter, senior vice president of the product management and markets group at Synopsys. “What’s changing now is a lot of larger systems have a lot of electronics inside of them. Instead of simulating one SOC, you need to look at different SOCs and chips that need to communicate with each other. The complexity of electronics within larger products has increased the need to expand from, say, digital twins focused on mechanical properties to those that take into account electronics and communications, and the interaction between the two of them.”

“With increasing demand for miniaturization and higher operating frequencies, effects such as electromagnetic interference, parasitic capacitance and thermal hotspots—often involving two or more simultaneous physics phenomena—must be accurately modeled,” says Bjorn Sjodin, senior vice president of product management at multiphysics specialist COMSOL. “The expectation for longer-lasting electronics components is supported by detailed physics models, including those accounting for complex and nonlinear multiphysics effects like electromigration and electric discharges. Generally speaking, recent advancements in algorithms and more powerful computers have enabled detailed simulation models that can be used to identify issues with a design before building physical prototypes, thereby avoiding unnecessary costs.”

Challenging Designs

According to Sjodin at COMSOL, electronics design is unique in that it requires simultaneous consideration of electrical, thermal and mechanical behaviors, making multiphysics simulation invaluable. “For instance, thermal runaway in power electronics can be detected and mitigated early through simulation,” he says. “The net results are improved reliability, reduced failure rates and the ability to design for extreme operating conditions without extensive physical testing.”

For example, to simulate an advanced driver assistance system in a car, or to simulate a fully self-driving vehicle or a robotics system, engineers need an accurate representation of not just the multiphysics of the vehicle, but also the electronics inside, as well as the effect on power requirements that result from these interactions.

“In so many cases, battery power becomes increasingly important to simulate the functional behavior of the devices,” De Schutter says. “Some of the new capabilities we need to simulate in electronics include the interaction between different components, the ability to react quickly enough to events that may be affecting the system, but we also need to simulate power consumption and how that affects function based on the power available.”

Advancements in computing and processing hardware are helping organizations tackle these expanding physics challenges.

“There has been an exponential growth in the size of the problems customers are trying to solve, and we are engaging more often with customers who want to solve a bigger chunk of the entire system,” says Commens at Ansys. High-performance computing hardware, graphics processing unit acceleration, and the ability to leverage multiple processors has made this easier. Ansys has also focused on helping customers bring designs from different software tools together for system-level simulations.

Those types of efforts are important, as electronics design has typically been done with various individual point tools that require manual import and export manipulation, which adds to design cycle time and introduces the possibility of errors.

“Today, complete and comprehensive systems design and analysis offerings encompass not only the IC [integrated circuit], PCB [printed circuit board], and IC package design platforms but also their inherent multiphysics phenomena as well,” says Hess. “This includes SI/PI [signal integrity/power integrity], EM [electromagnetic], thermal and computational fluid dynamics (CFD), as well as RF [radiofrequency]/microwave effects. Together, these integrated and interoperable workflows for complex, electrically mechanical, and/or heterogeneous systems deliver unconstrained capacity for analysis and optimization of the complete system.”

Industry Convergence

As a result of the increasing presence of electronics across different systems and products, mechanical and electronic simulation vendors are beginning to converge as well. Synopsys and Ansys, for example, are in the midst of having their proposed merger approved, and Cadence has acquired simulation specialist BETA CAE Systems, OpenEye Scientific (which focuses on molecular modeling) and Invecas, which provides embedded software solutions.

The shift toward stacking chiplets on top of each other to save space in these systems has also put more of an emphasis on multiphysics and mechanical simulation, as these now 3D-focused electronics designs can affect mechanical properties and vice versa.

According to Hess, heterogeneous integration and cross-domain team collaborations are shaping the future of semiconductors via systems innovation and enabling new advancements.

“The mechanical aspects of multiphysics in system simulations are critical for 3D-IC, multi-die and chiplet design,” Hess says. “Today’s products are not only feats of electronic engineering but of mechanical as well, as the electronics find themselves in new and novel forms such as foldable phones where drop testing, a mechanical simulation, becomes essential. Here, engineering domains must co-exist and collaborate to bring about the best end products possible. The world of electronics and its vast array of end products is pushing us beyond electronic design to be more broadly minded to whole product/systems engineering in order to develop not only heterogeneous products but heterogeneous engineering teams as well.”

Ansys, for example, has made it easier for electronics engineers to access Ansys CFD and FEA tools within their existing workflows. “In our HFSS platform, we have a design instance that uses Ansys Mechanical solver technology,” Commens says. “That’s tailored toward the electrical engineer, but once the early analysis is done, you can easily send it to a colleague who can use it in the Ansys Mechanical workflow.”

Cadence offers easy collaboration capabilities for mechanical and electrical engineers in its own products, and last year announced an integration with Dassault Systèmes’ 3DEXPERIENCE Works portfolio design and simulation tools. The Synopsys Saber suite for virtual prototyping of power electronics also encompasses cross-domain capabilities.

AI Powers Electronics Design

Simulation software providers industrywide are working to incorporate artificial intelligence (AI) capabilities in their products in various ways, and all companies interviewed for this story have made an investment in AI and/or machine learning.

“Recent advances in AI and related technologies are already benefiting the electronics space in several ways,” COMSOL’s Sjodin says. “Programmatically driven simulations now automate tedious manual modeling tasks, while chatbot technology significantly enhances the productivity of engineers working with such simulation code; chatbots are good at programming and debugging. Additionally, surrogate models—trained on outputs from traditional simulation models—serve as fast-evaluating components in system simulations or as the engine behind ready-made user-friendly simulation apps.”

In 2023, Synopsys announced its Synopsys.ai Copilot for semiconductor design. De Schutter says the company is leveraging AI to help guide which simulations are required for specific projects. “AI will continue to play a role in helping to understand what to simulate to get higher verification coverage, and to get there more quickly,” he says.

He adds that AI can play a role in helping write SystemC modeling to reduce barriers to generating models quickly, and automate other time-consuming and error-prone programming and modeling tasks.

“In chip design, AI can help explore architectural possibilities, including fine-tuning P&R (place and route) settings to achieve better power, performance and area (PPA),” says Cadence’s Hess. Cadence has released a number of generative AI tools for design. “AI technology has the potential to transform many industries by optimizing results that couldn’t have been achieved previously, as well as automating, thus increasing user productivity substantially.”

Although simulation is growing in its capabilities and importance, it still has not been embraced completely across the electronics design industry. Designers and engineers can benefit by incorporating simulation across the design cycle. “A lot of users aren’t aware of what they can do with simulation now,” Commens says. “Because of the complexity of designs, you do have to bring in the full system to make design decisions earlier. Otherwise, you will compromise performance. If you use simulation, you can break some traditional design rules in ways that help improve these complex designs.

More Ansys Coverage

More Cadence Coverage

More COMSOL Coverage

More Synopsys Coverage