In a 1,200 V SiC half-bridge module assembled on an AlN direct-bonded copper substrate, the encapsulant is dispensed at a cavity temperature of 70 °C under a chamber pressure of 5 mbar to 10 mbar after the resin component has been degassed at 25 °C for 15 min. The mixed system typically exhibits a viscosity of 2,500 mPa·s to 5,000 mPa·s at 25 °C when measured on a Brookfield RV spindle #6 at 20 rpm, and a pot life of 3 h to 6 h at 25 °C. Filler settling velocity follows Stokesian behaviour and becomes a process conflict when fused alumina with a median particle size of 5 µm to 10 µm is loaded above 75 wt%; holding the mixed material under vacuum longer than 10 min can produce a sediment layer at the feed line inlet, altering thermal conductivity from 1.2 W/m·K to 0.6 W/m·K in the dispensed shot. The cured compound is evaluated under UL 746B for relative thermal index, with high-temperature electronic/EL grades typically requiring an RTI of 180 °C to 200 °C, and flame retardancy is verified to UL 94 V-0 at 3.0 mm thickness. Stage curing is set at 90 °C for 1 h, then 150 °C for 2 h, then 175 °C for 1 h, because a single-step cure above 150 °C can produce an internal exotherm greater than 220 °C in a shot mass above 50 g, leading to cracked corners and wire-bond lift on 250 µm aluminium wires. Thermal shock per IEC 60068-2-14 Test Na from -55 °C to 175 °C for 500 cycles is commonly used as an acceptance test, with delamination measured by scanning acoustic microscope and allowed only below 1 mm from the DBC edge. The CTE of the cured compound below its glass transition temperature is targeted at 25 ppm/°C to 30 ppm/°C by thermomechanical analysis per ISO 11359-2, which is sufficiently close to the 17 ppm/°C expansion of copper and 4.5 ppm/°C expansion of AlN to reduce tensile stress at wire-bond interfaces but still requires alumina levels near 70 wt%, at which point vacuum infiltration under large-area chips becomes sensitive to die attach voiding.

Can On-Board Charger Potting Compounds Survive 85°C/85% RH Biased Humidity Without Delamination?

An 11 kW on-board charger housed in a die-cast aluminium enclosure is potted after the power board is mounted at a 6 mm standoff from the base plate, leaving a 12 mm maximum fill depth over TO-247 switching devices. The two-part epoxy is metered through a 2K meter-mix dispenser with a 24-element static mixer at 40 °C to reduce mixed viscosity to below 1,200 mPa·s, then the filled housing is transferred to a vacuum chamber at 20 mbar for 3 min to extract air from beneath surface-mount inductors and gate-drive transformers. The main process conflict is outgassing of the anodised aluminium base plate: residual sealing solution in the anodised pores can expand under vacuum and produce 50 µm to 150 µm bubbles at the metal interface, so production lines commonly pre-bake the enclosure at 120 °C for 2 h when relative humidity exceeds 60%. Curing is performed at 85 °C for 1 h followed by 150 °C for 2 h. The cured material must pass biased humidity ageing at 85 °C/85% RH for 1,000 h with a 100 V DC bias applied across a 0.5 mm gap, and insulation resistance must remain above 1 GΩ when measured at 500 V DC per IPC-TM-650 method 3.5.1. Thermal cycling per ISO 16750-4:2010 from -40 °C to 150 °C for 1,000 cycles is the reference acceptance sequence for under-hood electronics; the failure criterion is a shift in capacitance of Y-capacitors of more than 5% or visible delamination at the FR-4/aluminium interface. The glass transition temperature after post-cure is typically 165 °C to 185 °C by differential scanning calorimetry per ISO 11357-2, and the coefficient of thermal expansion below Tg is held near 35 ppm/°C to avoid cracking of ceramic capacitors. Lap shear adhesion to anodised aluminium is measured per ASTM D1002 and should exceed 2.5 N/mm² after ageing; values below 1.5 N/mm² correlate with field failures at the seal groove. For high-voltage potting in OBC circuits, an IEC 60664-1 creepage distance of 6.3 mm is maintained by the encapsulant thickness over the switching node, and the material must show a comparative tracking index above 600 V per IEC 60112 to avoid carbon track formation under polluted conditions.

Because electroluminescent lamp drivers produce sinusoidal outputs from 110 V to 220 V at frequencies between 400 Hz and 4,000 Hz, the potting compound around the step-up transformer and MOSFET drain nodes is subjected to partial discharge stress rather than simple dielectric breakdown. The EL-grade encapsulant is formulated with extractable chloride content below 5 ppm, sodium below 3 ppm, and ammonium below 5 ppm by ion chromatography after aqueous extraction per IPC-TM-650 method 2.3.28; early field failures in EL power supplies were traced to electrochemical migration across 0.2 mm conductor gaps when chloride concentrations exceeded 15 ppm under 85 °C/85% RH bias. The mixed system is supplied at 800 mPa·s to 1,500 mPa·s to permit flow into 0.4 mm gaps around toroidal ferrite cores without breaking 0.1 mm polyester-insulated wire. Vacuum degassing is performed at 1 mbar until foam collapses, normally within 2 min; holding under vacuum longer than 5 min can strip reactive diluent and shift the gel time from 45 min to 60 min. The cure schedule of 85 °C for 30 min followed by 125 °C for 1 h limits peak exotherm to 145 °C, which is below the thermal shrinkage threshold of polyester film dielectric in EL lamps. Partial discharge inception voltage measured per IEC 60270 on a 1 mm gap is required above 1.8 kV at 60 Hz, and the extinction voltage is required above 1.5 kV to prevent sustained discharges after a transient overvoltage. Volume resistivity after 7 days at 85 °C/85% RH is measured at 500 V DC per ASTM D257 and must remain above 1×10^14 Ω·cm. Bare copper traces require a benzotriazole passivation step or the use of non-amine cure systems, because free amine catalysts dissolve copper and create mobile copper-amine complexes that reduce insulation resistance under bias. The potted EL driver is subjected to a 2-hour burn-in at 85 °C under 120 V AC input to identify latent voids; a 1% loss in minimum luminance is the typical factory acceptance criterion.

High-Temperature Potting Failure Modes in Measurement-While-Drilling Electronics Above 200°C

Measurement-while-drilling modules are potted inside stainless steel barrels and are exposed to 175 °C to 220 °C at hydrostatic pressures up to 140 MPa for 500 h to 1,000 h. The dominant failure mode in epoxy systems is not dielectric breakdown but adhesion loss at the polyimide/stainless steel interface, which allows drilling vibration to abrade 0.05 mm magnet wire inside the sensor basket. For high-temperature vacuum potting, the compound is degassed at 5 mbar and 70 °C, then flooded into the sensor annulus; viscosity must remain below 800 mPa·s during the first 10 min to penetrate 0.2 mm varnish gaps between windings. Fumed silica at 2 wt% to 4 wt% is used to produce a yield stress of 15 Pa to 40 Pa on a parallel-plate rheometer, which restricts filler settling during the long flow path while still allowing vacuum impregnation. After cure, the potting compound is aged at 220 °C for 1,000 h in a forced-air oven; hardness retention is measured per ASTM D2240 with a rejection limit of more than 15 Shore A points loss for silicone systems or more than 10 Shore D points loss for rigid epoxies. Outgassing is evaluated per ASTM E595-15; a total mass loss below 1.0% and collected volatile condensable material below 0.1% are required to prevent deposition on downhole optical windows and connector contacts. Thermal conductivity is measured at 200 °C by the guarded heat flow method per ASTM D5470, but published data for exact high-temperature conductivity remain limited; most supplier documentation reports values at 25 °C and 100 °C, so a linear interpolation to 200 °C is not valid without hot-disk or laser-flash verification. The process conflict in silicone potting is platinum catalyst poisoning by sulfur-cured elastomers in cable jackets and O-rings; tooling and fixtures must be verified to be free of sulfur, tin, and amine residues. This application requires a continuous operating temperature rating above 200 °C, which is confirmed by thermogravimetric analysis per ASTM E1131 showing less than 5% weight loss after 100 h at 250 °C in nitrogen.

Across a transmission speed sensor line, the ferrite core and coil assembly is inserted into a PBT 30% glass-fiber connector body, and the potting compound is dispensed at a shot volume of 0.8 mL to 1.2 mL around the coil to a depth of 4 mm. The material must resist automatic transmission fluid at 150 °C for 3,000 h with hardness change less than 5 Shore D points and volume swell below 3% per ASTM D471. Adhesion to PBT 30% GF is verified by a destructive torque test on the terminal pin; torque retention above 0.3 N·m after oil ageing is required to prevent terminal loosening under connector insertion. Vibration qualification follows ISO 16750-3 with a random vibration profile at 27.8 m/s² RMS for 8 h per axis; the acceptance criterion is no cracking detected by 30× stereomicroscopy and no electrical open longer than 1 µs during the final 30 min of each axis. The high-volume curing requirement creates a process conflict: a snap cure at 150 °C for 3 min provides tack-free surface but leaves residual stress and a reduced degree of conversion; therefore a post-cure at 130 °C for 30 min is used to raise the glass transition temperature beyond 165 °C and to reduce post-mould shrinkage. Needle dispensing equipment with ±2% volumetric accuracy is required because shot volumes below 0.7 mL leave the ferrite end face exposed to oil ingress, while volumes above 1.4 mL overflow into the connector latch recess and create interference during mating. The compound must also withstand hot oil immersion at 150 °C and simultaneous 10 kV electrostatic discharge pulses applied to the signal terminals following ISO 10605; this requires the cured material to maintain a dielectric strength above 15 kV/mm per ASTM D149 after oil exposure. Incompatibility with zinc stearate mould release on PBT connectors is a known adhesion risk; precleaning with a low-pressure plasma source at 50 W for 60 s is used where bond-line delamination occurs at the connector wall.

When Avionics Potting Must Control Partial Discharge at 15 kPa Ambient Pressure

Aerospace flyback transformers operating at 15,000 m equivalent altitude are potted under vacuum because the reduced pressure lowers the Paschen minimum for air in trapped voids; a 0.5 mm void becomes a partial discharge site at approximately 500 V peak, and the discharge can erode the potting compound until the transformer fails. Potting under vacuum at 10 mbar absolute is insufficient if the moulded housing outgasses during resin gelation, so polybutylene terephthalate and glass-filled polyamide housings are pre-dried at 120 °C for 4 h and the resin is degassed at 5 mbar for 10 min before filling. The assembled flyback transformer is heated to 60 °C to reduce viscosity to 900 mPa·s and placed in a pressure/vacuum vessel; after resin introduction, the vessel is raised to 2 bar absolute during gelation to collapse residual bubbles below 50 µm diameter. Partial discharge performance is evaluated per ASTM D1868-20 at 15 kPa and 60 Hz; the acceptance threshold is a partial discharge inception voltage above 1,000 V peak for a 2 mm potting thickness. Thermal cycling from -55 °C to 125 °C is performed per RTCA DO-160G Section 4.5.2, with 3 cycles of 2 h dwell at each extreme; delamination after cycling is inspected by C-mode scanning acoustic microscopy and is not acceptable if the void area exceeds 2% of the winding surface. The material must satisfy flammability requirements per UL 94 V-0 at 1.5 mm and toxic gas emission limits for cabin compartments when applicable. A major process incompatibility is platinum-catalysed silicone in contact with sulfur-cured EPDM vacuum chamber seals; sulfur migration poisons the hydrosilylation reaction and produces a permanently tacky surface. Production records for avionics potting must document resin lot, hardener lot, vacuum level, pressure ramp, and oven load to satisfy FAA and EASA part manufacture traceability requirements. The electrical insulation is measured after ageing per ASTM D257 at 500 V DC, and the material must maintain volume resistivity above 1×10^13 Ω·cm at 125 °C at the end of the cycling sequence.

Catalyst Poisoning and Heat Transfer Limits in Induction Heating Coil Encapsulation

In induction heating work coils operating at surface temperatures from 150 °C to 220 °C, the encapsulant must dissipate heat without developing hot spots that accelerate thermal oxidation and reduce coil lifetime. A two-part addition-cure silicone with thermal conductivity of 0.8 W/m·K to 1.2 W/m·K via ASTM D5470 is dispensed around the copper winding; flame retardancy per UL 94 V-0 at 6 mm thickness is typical for industrial drives and induction heating power supplies. The mixed system has a pot life of 90 min at 25 °C and cures at 125 °C for 30 min; deep sections above 30 mm require a step cure at 80 °C for 1 h to prevent foaming from moisture absorbed in the copper oxide layer. Solder flux residues containing tin or amine compounds inhibit platinum hydrosilylation and must be removed with isopropanol and lint-free wipes before dispensing; a simple verification is cure response testing by differential scanning calorimetry per ISO 11357-5, where a residual exotherm above 5 J/g after the standard cure indicates partial inhibition. Thermal cycling under IEC 60068-2-14 Test Na from -40 °C to 200 °C for 1,000 cycles is used as an acceptance test, and adhesion loss at the copper interface is measured by pull-off per ASTM D4541 with a minimum requirement of 1.0 MPa after cycling. Published data for exact thermal conductivity at 220 °C are limited; most supplier thermal conductivity values are generated at 25 °C and 100 °C, and extrapolation to continuous operating temperature should be verified by laser flash analysis per ASTM E1461 before a design freeze. The material must also withstand 10 kV transient overvoltage pulses on the coil terminals; dielectric strength is verified per ASTM D149 on a 2 mm cured sheet, with acceptance above 18 kV/mm. In high-frequency induction systems operating at 50 kHz to 100 kHz, dielectric loss becomes a secondary heat source; the dissipation factor is measured per ASTM D150 at 1 MHz and is maintained below 0.02 to limit additional temperature rise. A critical processing boundary is the presence of silicone inhibitor residues on copper from previous RTV gasketing: even 10 ppm of tin-containing condensation catalyst on the substrate can stop the addition-cure reaction at the interface, leaving a gel-like layer that causes immediate bubble formation during thermal cycling.