Custom Electronics Thermal Interface Pads, Gap Fillers & Graphite Heat Spreaders

Because the joint sees thermal resistance, not bulk conductivity. The W/m·K on the front of the datasheet is a material property; the assembly adds contact resistance at both faces, and the total is thermal impedance measured per ASTM D5470 at a stated pressure. Contact resistance often dominates, so a pad that wets the surface well at your pressure can outperform a nominally higher-conductivity pad that does not. Compare candidates on their D5470 curve at your real clamping force, and state the target thermal resistance on the drawing. [1]

A lot. Thermal impedance falls steeply with pressure: Sil-Pad® TSP 900 improves from 1.82 to 1.13 to 0.54 °C-in²/W as pressure rises from 10 to 50 to 200 psi on its TDS, and a soft gap filler's impedance changes with deflection percentage. A snap-fit shield can at a few psi and a screwed heat sink at 50 psi are different thermal joints with the same part. State the real clamping pressure (or the deflection a soft filler reaches) so the impedance is read where the joint lives. [9]

Two dimensions, two jobs. Thickness is set by the nominal gap and its tolerance: the filler must stay in contact at the largest gap and not over-compress at the smallest. Hardness (Shore 00) is set by the allowable closure force: a softer grade wets and conforms at low force but adds bulk resistance, a firmer grade needs more force. Gap Pad® grades span roughly 1.0–7.0 W/m·K, and the impedance you get is read at the deflection your design achieves (TGP 6000ULM: 0.34/0.29/0.26 °C-in²/W at 10/20/30%).

Send the gap nominal and tolerance, not a single thickness. [10]

Whenever siloxane outgassing would matter. Silicone TIMs can release low-molecular-weight siloxane under heat and time; it migrates and condenses on nearby surfaces, fogging a camera lens or optical sensor as an oily film, or fouling an electrical contact. Cameras, optical modules, MEMS devices, and low-outgassing sealed enclosures are the classic silicone-free cases.

Default to filled silicone for its thermal and mechanical performance, and switch to a silicone-free gap filler where the design is contamination-sensitive. Flag any optics or sensors on the drawing so the material track is chosen up front. [11]

It depends on the problem. A through-plane pad crosses a stacked joint (die to heat sink) and is specified by D5470 impedance. A graphite spreader moves a concentrated hot spot in-plane to a larger area and is specified by in-plane conductivity. They are not interchangeable, because graphite is anisotropic: SpreaderShield™ SS-1500 conducts ~1500 W/m·K in-plane but only ~3.4 through-plane.

A spreader laid across a clamped joint is a poor TIM, and a soft pad laid over a hot die spreads almost nothing. Where a design needs both, draw them as separate layers. [7]

Because the problem in a thin, fanless product is a concentrated hot spot, not a stacked joint. Pyrolytic graphite sheet (PGS-class, ~700–1000 W/m·K in-plane, very thin and pliable) and natural-graphite spreaders take the heat from a small SoC or camera-compute die and spread it laterally across the largest available area, so the case never develops a burning-hot spot. The sheet is extremely thin, so it fits high-density mounting, and it is to a footprint that reaches from the die to the coolest part of the structure. [8]

Thermal vias carry heat through the board itself, and they change what "board conductivity" means. Copper conducts around 385 W/m·K while FR4 is around 0.3, so a via array under a hot pad raises the board's apparent (effective) through-thickness conductivity far above bulk FR4, moving heat to a plane or the far side where a TIM and heat sink take over.

The board's effective conductivity is a composite, not a bulk material number, which is the same lesson as reading D5470 impedance rather than bulk W/m·K: the system value is what matters. The vias are a board-design item (IPC-2221 context); the surface TIM is what H-O converts. [6]

Usually a thin insulating pad. An LED array drives heat into its metal-core board (MCPCB), which must then reach the heat sink or luminaire chassis across an interface that often has to isolate electrically. A Sil-Pad®-class insulating pad gives low thermal resistance and a built-in dielectric barrier on one TDS; where the luminaire interface is larger or less flat, a soft gap filler bridges the tolerance.

Because LED lumen maintenance and color are temperature-driven, read the D5470 curve at the real mounting pressure to keep the junction where the LED maker specified it. [9]

By conduction to the case plus passive convection and radiation off it. With no vents and no fan, the TIM and spreader are the cooling: a graphite spreader moves the SoC hot spot in-plane, a soft gap filler bridges the board to the case wall at low force (so the sealed housing is not distorted), and a compliant layer both seals and carries some heat. Where a camera or sensor is present, a silicone-free filler protects the optics. H-O the whole path, one kit per unit, on liner in assembly order. [11]

Many do, by grade. UL 94 is a material-level flammability class: Sil-Pad® TSP 900 and eGRAF® HITHERM™ grades list a V-0 class on their TDSs, and BISCO® silicone sponges list their class per grade. The class belongs to the specific grade and thickness, so cite it by designation from the individual TDS, not as a blanket claim across a family. H-O converts materials available with UL 94 classes per the vendor TDS; H-O does not itself certify or list materials. [5]

Yes. H-O die-cuts, kiss-cuts, slits, laminates, and kits thermal interface pads, gap fillers, graphite spreaders, and dielectric films to the customer drawing, with liner and PSA options and pull tabs for line assembly. The whole thermal path for a module or a sealed enclosure can ship as a kitted set, one kit per unit, parts on liner in assembly order, with lot-code TDS records per material.

H-O converts made-to-order as an ISO 9001:2015 certified organization; it does not mold or extrude these materials in its own plant, and it is not a stocking distributor of finished parts.

The joint, not just the part: the gap nominal and tolerance, the clamping pressure or closure force, whether the interface must electrically isolate (and to what voltage), whether optics or sensors demand silicone-free, and whether the problem is a hot spot to spread or a joint to cross. Plus the footprint and keep-outs, any adhesive or liner needs, the target thermal resistance if known, and quantities for prototype and production. "Recommend the material" is a valid callout: that is what the thermal review is for.