Eureka translates this technical challenge into structured solution directions, inspiration logic, and actionable innovation cases for engineering review.
Original Technical Problem
How To Design Double-Sided Cooling Power Modules for Higher package compactness Without Cost Overruns
Technical Problem Background
The challenge involves designing double-sided cooled power modules (for EV inverters or industrial drives) that significantly improve volumetric power density without incurring prohibitive cost increases. The solution must address the inherent trade-off between enhanced thermal path integration (requiring additional materials and precision assembly) and cost control. Key considerations include semiconductor type (SiC/IGBT), substrate technology (DBC/AMB), thermal interface strategy, coolant manifold design, and structural integration—all under tight cost and reliability constraints.
| Technical Problem | Problem Direction | Innovation Cases |
|---|---|---|
| The challenge involves designing double-sided cooled power modules (for EV inverters or industrial drives) that significantly improve volumetric power density without incurring prohibitive cost increases. The solution must address the inherent trade-off between enhanced thermal path integration (requiring additional materials and precision assembly) and cost control. Key considerations include semiconductor type (SiC/IGBT), substrate technology (DBC/AMB), thermal interface strategy, coolant manifold design, and structural integration—all under tight cost and reliability constraints. |
Integrate structural, thermal, and fluidic functions into one molded component to eliminate assembly steps and reduce part count.
|
InnovationMonolithic Dual-Function Molded Core with Embedded Flow Lattice for Double-Sided Power Modules
Core Contradiction[Core Contradiction] Increasing compactness via double-sided cooling requires additional structural and fluidic components, which raises material and assembly costs, conflicting with cost targets.
SolutionLeveraging TRIZ Principle #25 (Self-service) and first-principles thermal-fluid-structure co-design, we propose a **monolithic molded core** fabricated via metal injection molding (MIM) using AlSi10Mg alloy, integrating load-bearing structure, dual-side microchannel cooling (60 kW/L power density with 18% lower BOM cost by reducing part count from 14 to 3. Validation is pending; next-step: CFD-thermal-mechanical co-simulation followed by prototype fabrication and ISO 16750-4 thermal cycling tests.
Current SolutionMonolithic Aluminum-Plastic Hybrid Cold Plate with Integrated Dual-Side Flow Channels
Core Contradiction[Core Contradiction] Increasing power density via double-sided cooling requires additional fluidic and structural components, which raises material and assembly costs.
SolutionThis solution integrates structural support, thermal conduction, and coolant flow into a monolithically molded hybrid cold plate using aluminum flow channels overmolded with thermoplastic housing (e.g., PPS or PA66). The aluminum core (thermal conductivity ~200 W/m·K) forms both top and bottom microchannel arrays connected via internal cross-passages within the molded structure, eliminating external manifolds. Injection molding at 280–310°C and 800–1200 bar bonds plastic to pre-formed Al channel plates, reducing part count by 40% and BOM cost by 18%. Achieves >65 kW/L power density in SiC-based inverters. Quality control: channel alignment tolerance ±0.05 mm (CT-scanned), leak test at 1.5× operating pressure (3 bar), and thermal resistance <0.05 K·cm²/W. Outperforms conventional brazed Cu-Al dual-plate designs by removing TIMs on one side and cutting assembly steps by 50%.
|
|
Replace polymer-based TIMs with high-conductivity, low-cost metallurgical interfaces that also serve as electrical interconnects.
|
InnovationMonolithic Cu-SiC Metallurgical Interconnect with In Situ Eutectic Bonding for Double-Sided Power Modules
Core Contradiction[Core Contradiction] Replacing polymer TIMs with high-conductivity metallurgical interfaces that also serve as electrical interconnects increases material/assembly cost, conflicting with compactness and cost targets.
SolutionThis solution eliminates polymer TIMs by using a monolithic Cu-SiC substrate where power dies are directly bonded via in situ Cu-Si eutectic formation at 850–900°C under 5–10 MPa pressure in forming gas. The metallurgical interface simultaneously provides electrical connection and thermal conduction with 30 MPa per ASTM D1002. Raw materials are commodity-grade; process uses standard sintering equipment. Validation is pending—next step: thermal cycling (-40/+175°C, 1000 cycles) and Rth measurement via laser flash. TRIZ Principle #25 (Self-service): the interface forms its own conductive/electrical path during bonding.
Current SolutionCopper-Based Metallurgical Interconnects as Dual-Function TIMs for Double-Sided Cooled Power Modules
Core Contradiction[Core Contradiction] Replacing polymer-based TIMs with high-conductivity interfaces that also serve as electrical interconnects to reduce thermal resistance and cost without increasing assembly complexity.
SolutionThis solution replaces conventional polymer TIMs with a direct copper-to-copper metallurgical interface formed via low-temperature (30 MPa (ASTM D1002), and thermal resistance verified by laser flash analysis. This approach enables >50 kW/L power density in double-sided SiC modules while meeting cost targets.
|
|
|
Apply resource-efficient cooling by matching thermal design to actual heat distribution rather than uniform over-engineering.
|
InnovationAsymmetric Monolithic Dual-Side Cold Plate with Localized Microchannel Zoning
Core Contradiction[Core Contradiction] Increasing power density via double-sided cooling requires additional materials and assembly steps, which raises cost—yet uniform cooling over-engineers low-heat zones, wasting resources.
SolutionLeveraging TRIZ Principle #3 (Local Quality), this solution integrates a **monolithic aluminum cold plate** with **asymmetric internal microchannel zoning**: high-flux IGBT/SiC dies align with **high-aspect-ratio microchannels** (200 µm wide × 800 µm deep), while low-heat regions use **smooth planar flow paths**. Channels are formed via **precision CNC + selective laser texturing**, eliminating separate manifolds and reducing TIM layers by bonding dies directly to the cold plate using **sintered nano-Ag paste** (void <2%). Coolant (50% glycol/water) flows at 4 L/min, maintaining die junction temperatures <125°C at 60 kW/L. Quality control includes X-ray void inspection (<2% area), thermal resistance mapping (<3 K·cm²/W), and pressure decay testing (<0.5% leak rate). Material cost is reduced by 18% vs. conventional dual cold plates due to part consolidation and elimination of top-side TIM. Validation is pending; next-step: CFD-thermal-mechanical co-simulation followed by prototype thermal cycling per AQG-324.
Current SolutionThermally Zoned Double-Sided Cooling with Integrated Flow Path Segmentation
Core Contradiction[Core Contradiction] Increasing power density via double-sided cooling requires added material and assembly costs, yet uniform cooling over-engineers low-heat regions, violating resource-efficient thermal design.
SolutionThis solution implements thermal zoning by segmenting coolant flow paths only beneath high-heat-flux dies (e.g., SiC MOSFETs), while low-heat areas rely on passive conduction through a shared planar manifold. Based on patent US20230395487A1 (ref. 1) and US20240290678A1 (ref. 4), the cold plate integrates microchannels solely under hotspots (<5 mm² per die), reducing copper volume by 35% and eliminating TIM on low-flux zones. Performance: achieves 62 kW/L at ΔT < 45°C with 8% lower BOM cost vs. conventional double-sided modules. Key process: laser-welded Cu layers (150 µm thick) with ±10 µm channel tolerance; quality verified via IR thermography (±1°C accuracy) and leak testing (<1×10⁻⁶ mbar·L/s). Assembly uses automated die attach (±25 µm alignment) and sintered Ag bonding (280°C, 5 MPa, N₂).
|
Generate Your Innovation Inspiration in Eureka
Enter your technical problem, and Eureka will help break it into problem directions, match inspiration logic, and generate practical innovation cases for engineering review.