Member Center
    Exit
    qr Code Url

    Scan qrcode to view mobile website

    image English
    • image English
    • image 日本語
    • image Français
    Home /News /THERMAL PASTE /Router Thermal Paste Selection for an Eight Year Life Without a Fan /

    Router Thermal Paste Selection for an Eight Year Life Without a Fan

    author: CHACE / Tousen Thermal Management Engineering Team
    2026-11-19
    {当前产品的产品关键词轮巡使用}

    Thermal Paste in a Router: A Thin Joint That Must Last Eight Years Without a Fan

    An access point drawing 18 W in a sealed plastic enclosure, standing in a 45 °C cabinet for eight years, is a different problem from a laptop that bursts for thirty seconds. There is no fan to raise, no service visit, no heatsink added later. Heat leaves through the same plastic wall for roughly 70,000 hours, and the radio front end sits centimetres from the processor producing most of it. That is why router thermal management is a materials question before it is an airflow question.

    Where the watts actually go

    A mainstream Wi-Fi 6 or 7 router spreads its budget across four to six parts, each with a different flux. Treating the box as one 18 W source is the first simplification that produces a wrong joint specification in router thermal management: the smallest parts carry the highest density, not the processor.

    Path segment Typical flux What sits there Why it matters
    SoC die to shield can 3–6 W/cm2 10 x 10 mm die at 5 W Thin joint possible, low clamp
    Front-end die to can 6–9 W/cm2 1.5–2 W on a 5 x 5 mm die Worst flux in the box
    Can to aluminium plate 1–2 W/cm2 Stamped can, 0.2 mm Can warps, gap varies
    Plate to plastic housing 0.3–0.6 W/cm2 Air gap or thin pad Spreading, not conduction
    Housing to room air Natural convection Vertical wall, no fan The real limit

    Two conclusions follow. The joint that matters most is not under the processor but under the front-end modules, where flux is highest and clamp load is lowest. And the last two segments dominate the total rise, so a thermal paste that saves 0.4 K at the die must not cost anything on the housing side.

    Why a 0.5 mm pad is the default, and where it stops being right

    Most routers ship with a gap pad because the stack is loose: a stamped can, a plastic bracket, a screw every 20 mm, tolerances adding to half a millimetre. A pad absorbs that, and at 2 W/cm2 a 0.5 mm pad costs about 2 K, which the design can afford. The problem appears when the same pad is carried over to a joint whose real gap is 0.1 mm and whose flux is 6 W/cm2, where no thermal paste is needed to beat it by a factor of ten.

    Option Bond line Rise at 5 W/cm2 Router verdict
    Thermal paste, 0.03 °C·cm2/W 10–20 µm 0.15 K Best where a screw clamps the can
    Thermal paste, 0.080 °C·cm2/W 20–40 µm 0.4 K Forgiving of slight can warp
    Thermal paste, 0.201 °C·cm2/W 30–60 µm 1.0 K Fine on the SoC, not on the FEM
    Gap pad, 0.5 mm at 5 W/m·K 500 µm 5 K Correct only for real gaps
    Gap pad, 1.0 mm at 3 W/m·K 1,000 µm 16.7 K Never under a 6 W/cm2 part
    Note: the numbers scale with flux. A joint costing 0.15 K at 5 W/cm2 costs 0.3 K at 10 W/cm2, and a 1 mm pad that looks acceptable at 2 W/cm2 becomes a 30 K penalty at 6 W/cm2.

    Five things a datasheet does not describe

    Eight years is a creep test, not a viscosity test. A router sits at 60–75 °C at the joint and cycles gently. The failure mode is slow loss of contact area at the edge, not fast pump-out. A thermal paste picked for the best day-one impedance on a soft base can be dry in year three; a slightly stiffer grade at 0.080 °C·cm2/W is still there in year eight.

    Volatiles reach the antenna. Silicone-based pastes release low-molecular siloxane, and in a sealed plastic box that vapour has nowhere to go but onto the antenna elements and the matching network. The result is not a thermal failure but an RF one: a fraction of a dB of insertion loss and a slow VSWR drift nobody attributes to the thermal interface material, which makes it the most under-tested interaction in router thermal management.

    Wall mounting changes the joint. Many access points hang vertically on a wall or pole, and a vertical joint at 70 °C drains a low-viscosity paste over years, leaving a dry region at the top of the die. Either keep the joint horizontal in the layout, or accept a stiffer grade and a thicker bond line.

    Stamped cans are not flat. A 0.2 mm shield can warp by 0.1–0.2 mm across a 40 mm face, and no thermal paste bridges that. The answer is a screw pattern that flattens the can locally, or a pad where the warp is largest — not a thicker paste layer everywhere.

    Service access is real. Routers return for RF tuning, firmware recovery and refurbishment, and a non-curing thermal paste lets a can be lifted and re-applied. A cured or phase-changed film does not, and a pad compressed twice reads worse than a fresh print.

    Selecting the grade for a sealed box

    The TSAD Series from TOUSEN is a solvent-free, non-curing thixotropic gray paste covering 1.2–6.5 W/m·K, with measured impedance of 0.03 to 0.201 °C·cm2/W at 30 psi, specific gravity 2.5–2.9 g/cm3 and a 12-month shelf life. In a router the choice of a thermal interface material is narrower than in a phone or a server: pick the grade whose bond line survives eight years at 70 °C, then confirm the volatile fraction is low enough for the radio.

    Grade W/m·K Impedance @ 30 psi (°C·cm2/W) Density (g/cm3) Typical router position
    TSAD60 1.2 0.201 2.5 Low-flux PHY and DDR under a can
    TSAD66 3.5 0.080 2.7 Can to plate, long-life joints
    TSAD66-P 4.0 0.070 2.6 Cans with slight warp, thicker print
    TSAD83 PLUS 5.2 0.050 2.6 Mainstream Wi-Fi 6/7 processor
    TSAD88 PLUS 5.2 0.030 2.9 Front-end modules at 6–9 W/cm2
    TSAD90 6.5 0.030 2.7 High-power FEM and 10G PHY
    TSAD100 6.5 0.030 2.8 Fanless enterprise AP, large plate

    Dose by mass, because the joint is 10 to 20 µm

    At 2.7 g/cm3 a 15 µm joint needs 4 mg per cm2, so a 40 x 40 mm can takes about 65 mg. A 30 percent over-dose is then 0.6 K over the whole part, and at 8 W/cm2 that separates a compliant front-end module from a marginal one. A syringe with a defined shot size beats an operator judging a cross pattern by eye.

    At that scale the thermal interface material is being dosed in fractions of a milligram per part: for a 5 x 5 mm front-end die the shot is under 1 mg, which no manual dispenser holds repeatably; stencil printing at 0.1 mm with 80 percent coverage is the practical route, and it keeps the paste off the matching components beside the die.

    Clamp load decides whether paste is allowed at all

    A thermal paste joint needs pressure, and pressure is what most router brackets lack. With M2.5 screws at 0.4–0.6 N·m on a 40 mm can the average is 3 to 8 psi, enough for a 20 µm joint but not for 10 µm. Where a can is held by clips only, centre pressure may fall below 1 psi and a 0.5 mm low-modulus pad outperforms any paste. That is not a contradiction; it is the design boundary.

    Where paste is the wrong part

    Between the aluminium plate and the plastic housing, around the DDR and switch fabric under a wide can, and at the housing interface of a wall-mounted unit, no thermal interface material in paste form closes the gap, and a pad is correct: those gaps are 0.3 to 1.5 mm and exist to absorb tolerance. Thermal paste there is squeezed into the void without closing the gap, and adds volatile mass inside a sealed box for no benefit.

    Qualification that matches an eight-year life

    Router thermal management is judged on units nobody will open, so the plan must compress time rather than measure a first article. Run the board at rated load in a 45 °C chamber for 1,000 hours, logging temperature at the processor, the hottest FEM and the plate edge, then add 200 cycles from -10 °C to 60 °C, 500 hours at 70 °C and 85 % RH, an oil separation test at 150 °C and an outgassing check by weight loss. Then measure RF insertion loss and VSWR on the same sample before and after: if the antenna match has moved, the paste is part of the cause. Finish with drop and vibration on the packed unit.

    Assembly rules that survive a factory audit

    • Clean the can and plate with isopropyl alcohol and verify visually: residue of thermal paste under 15 µm is a hot spot no spreader can average out.
    • Torque the can screws diagonally in two passes and never re-torque after the joint settles.
    • Use one dated syringe per model or a printed deposit; topping up a used syringe changes solids content and volatiles.
    • Never reuse paste lifted from a can, and never mix two grades to chase a number.
    • Store cartridges closed below 30 °C away from flame; the product is non-hazardous, but an open cartridge loses solvent balance within months, and that loss ends up on the antenna.

    Conclusion

    In router thermal management the thermal paste joint is thin in dimension and large in consequence: 15 µm of thermal paste instead of a 1 mm pad is worth 16 K on a front-end module, which is the difference between a warm case and a radio that throttles. The discipline is to treat the box as five segments, to put paste only where a screw supplies pressure, to choose the grade by eight-year stability rather than first-day impedance, and to treat siloxane volatility as an RF parameter. A sealed fanless enclosure then holds its temperatures for a decade without a service visit.

    Impedance data, grade table and packaging options for the thermal paste grades described here are published on the TSAD Series thermal paste product page, and samples for stencil trials on a shield can can be requested there.

    Share:

    Thermal Paste Selection for Server Thermal Management

    Related Article

    image
    Server thermal management is a fleet problem: at 17 W/cm2 a thermal paste joint adds 0.5 K, a 0.5 mm pad adds 17 K, and only the lid to heatsink bond line is set by the assembly line.
    Thermal Paste Selection for Server Thermal Management
    2026-11-18
    image
    Smartphone thermal management is a burst problem: at 25 W/cm2 a thermal paste joint adds 0.75 K where a 0.5 mm pad adds 12.5 K, and the joint is the one layer assembly still controls.
    Thermal Paste Selection for Smartphone Thermal Management
    2026-11-17
    image
    Tablet thermal management is a passive problem: thermal paste at 8.5 W/cm2 adds 0.3 K where a 0.5 mm gap filler adds 8.5 K, and the bond line remains the one layer the assembly process still controls.
    Thermal Paste Selection for Tablet Thermal Management
    2026-11-13
    image
    Laptop thermal management lives in a 20 to 50 micron interface: thermal paste at 35 W/cm2 adds under 2 K, while a 0.3 mm gap filler adds 17 K to the same joint between die and cold plate.
    Thermal Paste Selection for Laptop Thermal Management
    2026-11-12
    SiteMap

    Home

    Products

    Application

    News

    Download

    HOW CAN I HELP YOU

    TEL YANAGI: +86 18566122282 WhatsApp: +81 80 7029 9037

    Email: ZDLIU@ITOUSEN.COM SALES@ITOUSEN.COM  Europe@itousen.com

    If you have any questions related to thermal management, including material selection and the design of thermal management systems and solutions, please feel free to contact us. Together, we can ensure the stable and reliable operation of our equipment and machines!

    Copyright © 2021 King Theme v2. Powered by web Jinggong Network Security No. 32058302002032

    Terms Of Use Privacy Cookies Terms & Conditions

    CONTACT US

    • You can only upload 1 attachments at most
    • Supported formats: PDF,JPG,PNG,EXCEL,WORD,STP,IGS,DWG,DXF,PDF,STEP
    • The size of the uploaded file should not exceed 10MB
    (384096)
    0