news

Thermal Grease vs Thermal Pad: Engineering Selection Guide for CNC Machined Heat Sinks

Thermal grease and thermal pads serve the same fundamental purpose—filling microscopic air gaps between a heat source and a heat sink—but they achieve this with dramatically different mechanical and thermal properties. For most high-performance applications, thermal grease (paste) is the superior choice due to its lower thermal resistance (0.01 to 0.05 °C·cm²/W) and thinner bond line, while thermal pads are preferred for vibration-prone, high-volume, or field-serviceable assemblies where reworkability outweighs peak thermal performance. This article provides a data-driven comparison to help you select the correct interface material for your CNC machined, stamped, or spring-loaded thermal assemblies.

Thermal Interface Material Fundamentals: Why Gap Filling Matters

The microscopic roughness of a CNC machined aluminum heat sink base is typically Ra 0.8 to 1.6 µm, and a bare CPU or IGBT die surface is Ra 0.2 to 0.5 µm. When two such surfaces are clamped together, the actual contact area is only 5% to 10% of the apparent area. The remaining 90% to 95% is filled with air, which has a thermal conductivity of just 0.026 W/m·K. A thermal interface material (TIM) displaces this air with a medium that conducts heat at least 10 to 100 times better. The effectiveness of any TIM is quantified by its thermal impedance, expressed in °C·cm²/W, which accounts for both bulk conductivity and contact resistance at the two interfaces. Lower thermal impedance means lower junction temperature for the same heat flux, directly impacting component lifespan—a 10 °C reduction in junction temperature typically doubles the expected life of silicon-based semiconductors.

Thermal Grease: Composition, Performance Limits, and Application Method

Thermal grease is a viscous paste composed of a silicone or hydrocarbon carrier fluid loaded with thermally conductive fillers such as aluminum oxide (Al₂O₃), zinc oxide (ZnO), boron nitride (BN), or silver particles. The filler loading ranges from 70% to 90% by weight, producing a thermal conductivity of 1.0 to 8.5 W/m·K for commercial grades, with premium liquid metal pastes reaching 40 to 80 W/m·K. The key performance advantage is the achievable bond line thickness (BLT). With proper application using a stencil, screen printer, or automated dispenser, the BLT can be held to 25 to 50 µm. This thin layer minimizes the thermal path, yielding thermal impedance values of 0.01 to 0.05 °C·cm²/W for standard pastes and as low as 0.005 °C·cm²/W for liquid metal alloys. For a 100 W heat source on a 25 mm x 25 mm die (heat flux 16 W/cm²), a typical grease with 0.03 °C·cm²/W impedance produces a temperature rise of only 0.48 °C across the TIM layer. Application methods include manual spreading with a metal spatula, screen printing for high-volume production, and automated needle dispensing for precise dot or line patterns on CNC machined parts. The primary disadvantages are pump-out under thermal cycling (the grease migrates away from the contact zone), dry-out over time, and messiness during rework.

Thermal Pad: Construction, Compressibility, and Reliability Trade-offs

Thermal pads are pre-formed solid sheets made from silicone, polyurethane, or acrylic matrices filled with ceramic, boron nitride, or graphite fillers. They are supplied in thicknesses from 0.5 mm to 10 mm, with thermal conductivities ranging from 1.0 W/m·K for standard silicone pads to 15 W/m·K for graphite-based or phase-change enhanced pads. The critical parameter for pads is compressibility, expressed as a percentage of deflection at a given pressure. Most pads require 10% to 30% compression to achieve optimal contact, meaning the pad thickness must be 0.1 to 0.3 mm greater than the gap it fills. For a typical application with a 0.5 mm gap, a 0.8 mm thick pad compressed to 0.5 mm (37.5% deflection) will exhibit a thermal impedance of 0.15 to 0.40 °C·cm²/W—roughly 5 to 10 times higher than grease. However, pads offer distinct advantages: they are cut to shape using rotary dies or water jets, they do not flow or pump out, they withstand vibration and thermal cycling without degradation, and they allow easy disassembly without cleaning residue. Pads also provide electrical insulation in many formulations (dielectric strength of 5 to 15 kV/mm), eliminating the need for a separate mica or ceramic insulator. The main drawbacks are higher thermal resistance, a minimum practical thickness of 0.5 mm, and the requirement for pressure management—too little pressure leaves air gaps, too much pressure can crack brittle heat sink bases or deform thin stampings.

Comparative Analysis: Thermal Impedance, Cost, and Long-Term Stability

The decision between grease and pad hinges on quantifying the trade-offs. Below is a direct comparison based on typical production-grade materials available from BQUQ’s supply chain partners, tested on our own CNC machined aluminum heat sinks (6061-T6, Ra 1.0 µm finish) with a 25 mm x 25 mm test die.

For a 100 W power module dissipating 16 W/cm², the temperature drop across the TIM layer is: grease (0.03 °C·cm²/W) = 0.48 °C, liquid metal (0.008) = 0.13 °C, standard pad (0.25) = 4.0 °C, graphite pad (0.15) = 2.4 °C. In a system with a 50 °C ambient and 0.5 °C/W heat sink-to-ambient resistance, the junction temperature would be: grease = 100.5 °C, liquid metal = 100.1 °C, standard pad = 104.0 °C, graphite pad = 102.4 °C. The 3.5 °C difference between grease and standard pad may not be critical for consumer electronics but is significant for IGBT modules in automotive inverters where the junction limit is often 125 °C and headroom is minimal.

Application-Specific Recommendations for CNC Machining and Stamping Environments

For BQUQ’s CNC machined heat sinks used in LED lighting (typical 20 to 50 W), server CPUs (65 to 280 W), and power electronics (IGBT modules up to 600 W), the selection matrix is as follows. Use thermal grease when the design requires maximum thermal performance, the heat sink is mounted with screws or spring clips providing consistent 10 to 30 psi clamping force, and the product is not subject to frequent rework. This applies to high-end server CPUs, GPU cooling, and laser diode modules. Use thermal pads when the heat sink is part of a vibration-prone assembly (automotive under-hood, industrial motors), when the gap between the heat source and heat sink varies by more than 0.2 mm due to manufacturing tolerances (common with stamped aluminum parts), or when the assembly must be disassembled in the field for maintenance. Pads also excel in applications requiring electrical isolation between the heat source and the heat sink, such as in power supplies where a bare TO-247 package touches a grounded heat sink. For spring-loaded heat sinks (BQUQ manufactures custom springs with load ranges of 5 to 50 N), a pad is often mandatory because the spring’s deflection tolerance of ±10% would crush a grease layer or squeeze it out entirely.

Cost and Manufacturing Efficiency: Total Applied Cost Analysis

The raw material cost per cm² favors grease by a factor of 2 to 3, but the total applied cost includes labor, dispensing equipment, and rework. Automated grease dispensing adds US$15,000 to $40,000 in capital equipment (positive displacement pumps with 0.01 ml accuracy), while pad application requires only a manual or semi-automatic placement fixture costing $500 to $2,000. For a production run of 10,000 units per month, the per-unit labor cost for grease is $0.03 to $0.06 (including stencil printing and inspection), versus $0.01 to $0.02 for pad placement. However, grease-based processes yield a higher scrap rate (2% to 5%) due to voids or contamination, while pad placement has a scrap rate below 0.5%. For low-volume prototyping (under 500 pieces), pads are almost always the more economical choice because they require no dispenser calibration and no curing time. For high-volume production (over 50,000 pieces per month) with a stable design, grease provides a lower total cost per unit (savings of $0.02 to $0.04 per unit) but requires tighter process control and more rigorous incoming inspection of the heat sink surface flatness, which should be under 0.05 mm per 25 mm for grease to achieve its rated performance.

Frequently Asked Questions: Practical Tips for Design Engineers

How do I choose the correct pad thickness? Measure the maximum gap between the heat source and heat sink after mounting, then add 0.2 to 0.3 mm for 20% to 30% compression. Never specify a pad thinner than 0.5 mm because it cannot absorb surface roughness variations. For grease, the ideal BLT is 25 to 50 µm; use a stencil with a 0.1 to 0.2 mm aperture to control dispensed volume, which should be 0.02 to 0.05 ml per cm² of die area. What is the maximum operating temperature for each? Standard silicone grease is rated to 200 °C continuous, but above 150 °C the carrier oil evaporates, increasing thermal resistance by 20% per year. Pads with ceramic fillers are stable to 200 °C, while graphite pads handle 250 °C but must be coated to prevent electrical conductivity. Can I combine a pad and grease? It is not recommended because the two materials have different compressibilities, and the grease will migrate into the pad’s pores, reducing its compressive strength and creating a non-uniform thermal path. How do I test the quality of the TIM joint? Use a thermocouple embedded in the heat sink base (drilled 1 mm from the surface) and compare the measured temperature rise to calculated values. A deviation greater than 15% indicates poor wetting, excessive pressure, or a void. For production, use transient thermal testing (T3Ster or equivalent) to measure thermal impedance in under 30 seconds per part.

Conclusion: Final Selection Criteria and BQUQ Engineering Support

Select thermal grease when your design demands the lowest thermal resistance, you can control clamping force to 10 to 30 psi, and the assembly is permanent. Select thermal pads when you need electrical insulation, vibration resistance, or easy rework, and when a 2 to 4 °C higher junction temperature is acceptable. For most CNC machined heat sinks in power electronics, the decision is driven by the operating temperature headroom: if the calculated junction temperature with a pad exceeds the component’s maximum rating by more than 5 °C, you must switch to grease or increase the heat sink surface area by 15% to 20%. BQUQ’s engineering team can simulate your thermal interface using finite element analysis (FEA) and provide sample kits with both grease and pads for your specific heat sink geometry. We machine heat sinks to Ra 0.4 µm on the contact surface, which reduces the required BLT and improves TIM performance by 10% to 15%. For a free thermal interface recommendation and a quote within 12 hours, email your CAD drawing to sc@bquq.com, contact us on WhatsApp at +86 13713157787, or visit www.bquq.com.

Related Articles

• How to Attach a Heat Sink Without Screws: Thermal Adhesive Guide
• Thermal Management for High Brightness LEDs with Heat Sinks
• New energy vehicle IGBT heat dissipation, heat sink reliability engineering, heat pipe and temperature plate integrated heat sink, surface treatment and heat radiation enhancement: anodic oxidation, black coating and micro-nano structure, graphene/3D printing/intelligent thermal management: 2030 heat sink technology roadmap