Key Drivers for Power Module Density

Scale with low and high density cubes
Evaluating key drivers for power module density goes far beyond simple rules of thumb, such as the inverse proportionality of power solution switching frequencies to overall size and density. Power densities often change at different rates than the loads driving system density; therefore, it makes sense to analyze the subsystems and associated components separately. Advanced packaging and three-dimensional power packaging (3DPP) technology help maintain power module density in line with the systems, applications, and loads they serve.

Power vs. Volumetric Density

Power density measures how much power a solution delivers relative to its own volume, while volumetric density describes how efficiently the entire system uses space. In other words, power density focuses on the power module itself, whereas volumetric density reflects the overall system architecture and all supporting components. Power solutions are notorious for being the long pole in the tent—for influencing overall system size, volumetric efficiency, system bill-of-materials costs, and power density. Typically, these are broken down into common figures of merit (FOM) for a system, such as its size, weight, and power (SWaP) characteristics. When combined with a cost metric, this is referred to as SWaP-C factors [1].

Power density is generally a function of total available power versus overall solution volume, which is why component size tends to inversely relate to power density. The power density metric becomes more critical when combined with overall solution mass (typically translated to weight on Earth), which can be a key FOM in nontethered applications, as discussed from multiple perspectives in the content that follows.

It is also important to differentiate power density from volumetric density. Power density can be specifically characterized in the context of the power solution, which is a subset of the overall system volume. In general, power density always increases, whereas volumetric density can decrease as major system loads shrink in size (and possibly in power requirements) and/or increase functionality to perform more work in the same volume from generation to generation. This can create a trend different from that observed directly for power solutions. The industry has attempted to normalize these discrepancies with oversimplified metrics, such as dollar per watt ($/W), which make little sense unless comparing highly similar power supplies.

As with any aspect of evaluating power solutions and their technical and financial impact, it is crucial to look beyond first-order analysis. Power consumption and energy efficiency often resemble a “whack-a-mole” scenario, where optimizing one subsystem may reduce performance in another area, leaving the effective system-level impact the same or worse. Classic examples include when the enhanced power density of a wide-bandgap power switch, such as gallium nitride or silicon carbide, allows for a physically smaller power train (even with increased power-handling capability) by leveraging higher switching frequencies that reduce some power components.

However, this may also necessitate a larger (and potentially costlier) thermal mitigation solution to manage denser power dissipation in smaller geometries or even require liquid cooling. Often, seemingly minor features can disproportionately impact solution size and cost. For example, connectors (especially blind-mate types) and fans can significantly affect all FOMs in a SWaP-C analysis since they can be large, and electromechanical components can be bottlenecks for maximizing system quality and reliability.

Power solutions do not scale at the same rate as items on the load side, such as those influenced by Moore’s Law and microelectromechanical system devices. Consequently, system roadmaps cannot expect an exponential reduction in power solution size (or exponential increase in power density) solely from year-over-year process node improvements. That said, a power solution can help keep pace with enhanced load size and performance by meeting increasing load demands in its own way [2].

How the Feature Set Influences Power Density

Open computer power supply unit with cables
Looking at this power supply (above) with its chassis cover removed, does most of the volume appear to be occupied by actual power supply components, or do connectors, wiring, fans, heat sinks, and enclosures take up the majority of the volume (with empty space)? Often, it is surprising how little the power train affects the overall solution volume and, therefore, the maximum achievable power density. This illustrates why a metric like $/W is inappropriate for evaluating power supply designs unless all features are nearly identical, with power rating being the main difference.

Requirements driven by safety certification and support for high-voltage inputs (e.g., increased 2D and 3D spacing requirements) and harsh operating environments can significantly impact solution density. If stringent electromagnetic compatibility (EMC) or shock-and-vibe levels must be met, such as for network equipment certification, bulk and volume are consumed by larger filter components and enhanced mechanical support for securing heavy components. This may require adhesives or sealants (like room-temperature vulcanizing (RTV) silicone), strapping, and even full potting (immersing the solution in epoxy/polymer) to facilitate thermal transfer and electrical/environmental insulation. These larger components and retrofitted materials for safety, certification, and thermal/environmental support contribute to the solution’s overall weight and density metrics.

Considering the comprehensive quality and accelerated life testing power solutions undergo, along with functional electrical bench qualification tests, these setups and pass requirements should be accounted for in the design phase and test planning. Long-term, expensive qualification tests aim to pass on the first attempt, but frequent failures of larger or complex designs, and accessibility to assemblies/components for failure analysis, can affect corrective actions.

SWaP Improvement Opportunities in Density-Driven Power Design

The largest contributors to SWaP metrics are also the primary opportunities to improve associated FOMs. These include filter components, electromechanical components, and support for their mass. Identifying these factors and isolating component contributions enables designers to focus optimization on subtasks with dedicated validation testing.

Filter components selected to meet EMC requirements are often top candidates for optimization. Large capacitors and denser magnetics are frequent offenders but may receive less attention since many designers lack comfort with filter design. While filter design can be subjective and complex, it is recommended that designers involved in power design/qualification gain basic training on filter design and optimization [3] [4] [5].

Key trade-offs exist between filter component FOMs (better performance usually with larger/heavier components) and acceptable compliance levels (typically emissions). The best strategy for undesired energy at a given frequency is mitigation: optimize design to reduce noise sources before relying on filtering. For instance, a power driver/controller using spread-spectrum clocking distributes energy across a broader frequency spectrum, reducing the need for heavy-duty filtering [6].

Disaggregating power subsystems can also improve density. While counterintuitive in integration-driven density discussions, overloading a single solution can reach diminishing returns. For example, a system power rail requiring wide input range support and isolated/tightly regulated output may be better served by separate solutions optimized individually. Similarly, splitting a large single-phase converter into smaller multi-phase converters reduces component size, electrical/thermal stress, and allows higher switching frequencies to enhance FOMs.

Optimizing filter values, individual components, or disaggregated solutions enables designers to meet targets while leveraging state-of-the-art (SOTA) and commercial off-the-shelf (COTS) solutions. Major advancements in the area of three-dimensional power packaging (3DPP®), particularly for lower-voltage DC/DC converters, are highly effective. Advanced packaging integrates SOTA technologies into dense components. Filters, in particular, are heterogeneously integrated into power modules using planar magnetics, overmold packaging, and multichip modules. 3DPP® maximizes SWaP benefits while maintaining access to COTS solutions.

Conclusion

There is no Moore’s Law for power solutions, especially considering energy storage devices that dominate SWaP-C metrics and influence power and system density. Packaging is a key driver that enables modules and ready-made products to keep up with load-side density advancements. Chasing power density solely for improved metrics (e.g., W/m3) can be costly, with tradeoffs in cost, development time, efficiency, and reliability. It is crucial to evaluate the real impact of desired features on cost, space, and efficiency relative to the application.

That said, leveraging advanced packaging and 3DPP® techniques and automation can improve SWaP-C metrics. Increased design complexity can raise risks, such as decreased manufacturing yield or extra rework, but automating assembly may enable highly integrated solutions with controlled processes, enhancing reliability and power module density. The growing use of planar magnetics exemplifies this.

Higher power density also challenges thermal mitigation strategies. More heat in smaller spaces complicates effective heat transfer, raising component temperatures and reducing reliability. It is essential to consider holistic power design impacts to ensure power module density does not compromise other SWaP-C targets. Density can influence warranty analyses and support costs, particularly for thermals and product lifetime predictions.

References

[1] “Power Supply Design for maximum Performance,” RECOM Blog, Oct 21, 2022, https://recom-power.com/en/rec-n-power-supply-design-for-maximum-performance-229.html (accessed February 15, 2023).
[2] “Power Modules are Catching up with Moore’s Law,” RECOM Blog, Nov 11, 2022, https://recom-power.com/en/company/newsroom/blog/rec-n-power-modules-are-catching-up-with-moores-law-235.html (accessed March 13, 2023).
[3] “Specifying line inductors for power converter noise filters,” RECOM White Paper, https://recom-power.com/en/support/technical-resources/whitepaper/whitepaper-specifying-line-inductors/whitepaper-specifying-line-inductors.html.
[4] “Very low noise filter for isolated DC/DC converters,” RECOM Blog, Mar 4, 2019, https://recom-power.com/en/company/newsroom/blog/rec-n-very-low-noise-filter-for-isolated-dc!sdc-converters-46.html (accessed March 13, 2023).
[5] S. Roberts, “DC/DC BOOK OF KNOWLEDGE – Practical tips for the User,” Fifth Edition, RECOM Engineering, 2021.
[6] Wikipedia contributors, “Spread spectrum,” Wikipedia, The Free Encyclopedia, https://en.wikipedia.org/w/index.php?title=Spread_spectrum&oldid=1138317993 (accessed March 13, 2023).