A potted power supply fails in the field. You have two choices: scrap it or recover it. If recovery is possible—and it often is—the first challenge is safely removing the failed potting without destroying the underlying PCB, components, and solder joints. This is where many technicians fail. Aggressive removal methods—power tools, excessive heat, harsh solvents—can damage the board faster than the original potting failure ever did. Yet patient, methodical removal using the right techniques and materials can restore a potted assembly to a state where it can be reworked or salvaged. Understanding Why Potting Fails and How That Affects Removal Before you start removing potting, understand why it failed. This shapes your removal strategy. Thermal cycling cracks: If the potting failed due to thermal stress—cracked and delaminated from repeated heating and cooling—the bond between the potting compound and the PCB is already compromised. This is a good scenario for removal; the potting often comes away cleanly because it's no longer well-adhered. Moisture ingress and corrosion: If moisture penetrated the potting and caused corrosion on component leads or solder joints, the potting is likely still well-bonded to the board, but the underlying electronics are compromised. Removal is necessary for inspection and potential repair, but be prepared for the potting to be firmly attached. Mechanical failure or delamination: Some potting compounds delaminate from the enclosure walls or from large component bodies (heat sinks, connectors) while remaining well-bonded to the PCB. Identify these delaminated areas first; they're starting points for mechanical removal. Mechanical Removal: Patience and Small Tools Mechanical removal—chiseling, scraping, cutting—is slow but often the safest approach for preserving the PCB underneath. Tools: - Small flat-blade chisels or scrapers (1–3 mm width) - Utility knife or hobby knife for scoring and peeling - Small pry bars or fine precision picks for gentle leverage - Rotary tools with small grinding wheels or cutting discs, used with extreme caution and only on non-critical areas - Wire brush for cleaning cured potting residue Technique: Start by identifying the edges and boundaries of the potting mass. Look for areas where the potting is cracked, delaminated, or separated from the enclosure. These are natural starting points. Use a small flat chisel held at a shallow angle (nearly parallel to the PCB) to work under the edge of the potting mass. Tap gently—never strike hard. The goal is to slowly separate the potting from the substrate rather than hack through it with force. Work incrementally around the perimeter. Do not apply force straight down into the potting toward the PCB. Always work at a shallow angle, trying to separate the potting from its attachment surface. For large potted assemblies, this process can take hours. Accept it. Rushing accelerates damage. Thermal Methods: Controlled Heat for Bond Breaking Heat weakens the bond between potting compound and the PCB, making mechanical removal easier. But excessive heat damages components and solder joints. Heat gun method: Heat the external potting surface to 60–80°C (warm to the touch, not hot) using a heat gun…

A 48 VDC power supply running in a manufacturing environment with 30–80°C ambient swings depends on one often-overlooked component: the potting compound protecting its internals. Get that selection wrong, and thermal cycling, vibration, and humidity turn a routine encapsulation into a warranty liability. Why Industrial Electronics Demand Specialized Potting Standard potting compounds suitable for consumer electronics rarely hold up in industrial settings, for a few clear reasons: Heat generation: Power supplies, motor drives, and switching controllers generate continuous heat that the potting material must conduct away without trapping it against sensitive components. Thermal cycling: A power supply in a machine tool shop starts at 20°C in the morning, rises to 70°C during the workday, and cools overnight. Over months and years, that cycling stresses both the potting compound and the components inside it. Vibration: Industrial machinery creates constant vibration, typically 10–50 Hz. Potting that's too brittle cracks under this load; potting that's too flexible fails to support component leads and solder joints, inviting fretting corrosion and wire breakage. Moisture and chemical exposure: Humid, salt-air, or chemically active environments demand potting that resists moisture absorption and chemical attack while protecting the PCB from electrolytic corrosion. No single material excels at all four demands — selection means balancing them against the specific application, a trade-off examined further in this comparison of silicone, epoxy, and polyurethane potting compounds. Getting the balance wrong shows up months later, not on day one: a compound that looks adequate on the bench can still fail in the field once thermal cycling and vibration accumulate over a full duty cycle. Thermally Conductive Epoxy for Power Supplies For most industrial power supplies — AC-DC converters, DC-DC modules, three-phase rectifiers — thermally conductive epoxy is the standard choice. These compounds are filled with ceramic particles (aluminum oxide, boron nitride, or aluminum nitride) that improve heat transfer 5–15 times over unfilled epoxy, reaching 1.0–3.0 W/m·K versus 0.2–0.3 W/m·K unfilled. A power supply potted with thermally conductive epoxy typically keeps internal hot spots only 10–20°C above enclosure surface temperature, versus 40–60°C higher with unfilled epoxy. Since every 10°C rise in operating temperature roughly halves the lifespan of electrolytic capacitors and semiconductors, that margin translates directly into field reliability and fewer warranty claims. The trade-off is cost, higher viscosity, and more exothermic heat during cure — proper technique, including vacuum degassing and staged cure profiles, is essential to avoid voids and localized overheating during the cure itself. Engineers should also weigh installation altitude. Convective cooling drops as air density decreases, so a supply installed at 2,000 meters or higher runs hotter for the same load than one at sea level — an argument for specifying the higher end of the conductivity range rather than the minimum that passes bench testing. Polyimide and Silicone for Extreme Conditions For sustained operation above 150°C, polyimide potting outperforms epoxy in thermal stability, which is why it shows up in aerospace power supplies and industrial furnace controllers. It's difficult to work with, though — elevated-temperature cure (150–200°C) and…

A potting project needs 2 hours of assembly time, but the compound needs 48 hours to fully cure — production is blocked for two days. Accelerated heat cure can cut that to 4 hours, but risks incomplete cure, residual stress, or thermal damage if rushed. Cure time is both a production constraint and a quality factor, and understanding the kinetics prevents both outcomes. Cure Chemistry by Potting Type Two-part epoxy potting, the most common system, mixes resin and hardener on demand: pot life 30 minutes to 3 hours, gel time 2–4 hours, full cure at 77°F in 24–48 hours. Thermally-activated hardener epoxy cures slowly at room temperature but is built for elevated-temperature cure — extended pot life of 4–8 hours before gelling, then 1–4 hours in an 80–120°C oven. Polyurethane cures via moisture-triggered reaction, slower than epoxy at room temperature (1–2 hour pot life, 24–72 hour full cure depending on humidity). Silicone cures by room-temperature condensation: 2–4 hour pot life, 24–48+ hour full cure for thick sections. Formulation also affects bubble entrapment — see why potting compound bubbles during curing if voids show up in finished parts. Room-Temperature Cure: The Standard Path Most epoxy potting is formulated for 70–77°F cure without external heat. The typical timeline: gel in 0–2 hours (pourable liquid to putty), intermediate cure through 2–8 hours (partially cured, still tacky), functional cure by 24 hours (80–85% of final properties), full cure by 48 hours (95–99%), with continued maturation for 1–2 weeks in some formulations. This path needs no special equipment and suits large pours and heat-sensitive components. The tradeoff is a 24–48 hour lead time before testing or deployment, plus residual stress that can relax over weeks and redevelop under cycling. Accelerated Heat Cure: Faster, With Real Risk Heating dramatically speeds the reaction — room temperature takes 24 hours to reach 85% cure, 80°C takes 2–4 hours, 120°C takes just 30 minutes. That cuts lead time for aggressive schedules, and some formulations even develop better properties (higher Tg, improved strength) under thermal cure. But the risks are real. Exotherm risk grows because elevated temperature accelerates the reaction, generating more heat in turn — large pours can spike past 200°C internally, degrading the resin. Incomplete cure risk appears if oven temperature is too low or the ramp too fast, letting outer regions cure before the interior traps unreacted hardener. Rapid heating can also thermal-shock embedded components, covered in potting compound under rapid temperature changes. Thermal cure is also sensitive enough to small variations (±10°C, ±30 minutes) to shift final properties batch to batch. Cure Milestones and Pot Life Gelation is the point where the compound stops flowing and the assembly must be immobilized. Handling time typically arrives 4–8 hours after pour; demolding, if using a mold, at 12–24 hours. Test-ready time for mechanical testing is generally 24 hours, improving further to 48. Deployment-ready time for thermal cycling and field use is conservatively 48 hours for room-temperature cure or 4–6 hours after elevated-temperature cure. Pot life governs the workflow…

A power supply potted with "high-temperature epoxy" fails within a year in a 140°C engine bay. Investigation turns up the reason: that epoxy's glass transition temperature is only 160°C — marginal, not conservative. One specification mistake, read off a datasheet without digging deeper, produced a field failure and a warranty bill. Most high-heat potting failures trace back to a handful of recurring mistakes: misreading material specifications, ignoring how thermal cycling differs from steady-state heat, and skipping real-world validation before committing to production. Misreading the Temperature Rating A datasheet claiming "rated to 250°C" almost always means glass transition temperature, not continuous service temperature — safe continuous operation runs 50–100°C below Tg, not at it. Treating the headline number as the operating limit routinely produces inadequate thermal margin and failures within a year or two. Specify the minimum Tg your application needs, and confirm the datasheet number is genuinely Tg rather than a peak or decomposition temperature. For 150°C continuous operation with cycling, that typically means requiring Tg of 230°C or higher. A related trap is designing to average PCB temperature instead of peak component temperature. A board averaging 130°C might carry a MOSFET self-heating to 160°C — fine on paper, dangerously close to the potting's real limit. Add 20–30°C to any hot-spot measurement for uncertainty and aging, then require Tg at least 70°C above that adjusted peak. Confusing Steady-State Strength With Cycling Endurance Potting that passes a single-temperature strength test — say, 5,000 psi shear at 150°C — can still fail under thermal cycling, because cycling stress behaves differently from sustained load. Material that looks adequate at a fixed temperature can fail after a few hundred cycles once component temperature repeatedly crosses its plastic-deformation threshold. Thermal cycling data per ASTM D4169 or IPC standards, validated through 500+ cycles on an actual prototype, is the only way to catch this before customers do. Aging compounds the problem: potting fresh off the cure oven can lose 10–20°C of Tg after two to three years of continuous 150°C service due to oxidation, quietly turning adequate margin into inadequate margin. Designing with 80–100°C of margin above peak component temperature, rather than the 50°C that looks sufficient on day one, absorbs that degradation before it becomes a field problem. Selecting Material Without Matching It to the Application Specifying unfilled potting under a high-power component is a common shortcut that backfires quickly. A 100W supply potted with unfilled "high-temperature epoxy" (0.3 W/m·K) traps heat instead of dissipating it, pushing peak component temperature from 160°C to 180°C — past the potting's own Tg margin, in a feedback loop where trapped heat accelerates the degradation that reduces thermal performance further. Past roughly 20W of dissipation, thermally conductive potting (2–4 W/m·K) is worth the cost premium; it typically cuts peak temperature 10–20°C and breaks the loop before it starts. CTE mismatch causes a related but distinct failure: standard potting at 60 ppm/°C against a 17 ppm/°C copper PCB creates interfacial shear stress under thermal cycling severe enough to initiate delamination…

After 200 thermal cycles from −30°C to +150°C, potting compound begins separating from the PCB surface — starting at the edges, then working inward. By 500 cycles, whole regions have come loose, leaving the traces underneath exposed to moisture and vibration. Delamination undermines the entire purpose of potting encapsulation, and its root causes are well understood enough to design around. CTE Mismatch Is the Root Cause Delamination happens when the CTE mismatch between potting and PCB generates interfacial shear stress that exceeds the adhesive bond strength. Copper PCB runs 17–18 ppm/°C; standard potting runs 50–70 ppm/°C — a 32–53 ppm/°C gap that, across a 100°C swing, produces 320–530 micrometers of differential strain over a 100mm length. That strain concentrates at the interface as shear stress, and repeated cycling fatigues the adhesive bond until it lets go, typically starting at whatever point already carries the highest stress concentration. Those high-stress points are predictable: PCB edges, where the potting-PCB interface is a free surface unsupported by interior potting mass; component leads, which act as stress risers where CTE mismatch between lead, solder, PCB, and potting creates local stress 2–5x the average; sharp corners, which geometrically multiply interfacial shear by 1.5–3x versus a filleted transition; and rigid embedded components — a low-CTE ceramic capacitor surrounded by high-CTE potting forces strain into the adjacent material rather than absorbing any of it itself. Once It Starts, It Accelerates A tiny separated region admits moisture, which travels along the interface by capillary action and swells the material locally — adding more interfacial stress and widening the delamination further. Ionic contamination already present from flux residue or salt dissolves into that moisture, accelerating electrochemical corrosion at the same interface. And mechanically, the void itself concentrates stress once it exists, so what starts as a micro-delamination visible only under magnification typically becomes visible to the naked eye within another 100–200 cycles. Material and Design Factors Behind Poor Adhesion Some potting simply bonds poorly to PCB materials — silicone's naturally low surface energy gives it inherently poor wetting to epoxy or polyimide substrates unless adhesion promoters are added, and mold-release additives in some formulations actively work against adhesion. Manufacturing residue — oil, flux, dust, moisture — left on the board before potting prevents chemical bonding entirely, leaving only a weak mechanical interface that fails quickly under cycling shear. Certain PCB finishes (immersion gold, electroless nickel) bond less reliably to some potting families than bare copper does. And the CTE mismatch discussed above isn't just a stress problem — it directly degrades adhesion itself, since excessive interfacial stress works against the bond regardless of chemistry. Reducing CTE Mismatch First Low-CTE potting (30–45 ppm/°C) cuts interfacial stress 40–60% against standard formulations, typically pushing delamination onset from 200–400 cycles out to 1,000–2,000, for a $20–30/lb premium — a strong return for most high-temperature applications. Ultra-low-CTE potting (15–25 ppm/°C), matched closely to copper and aluminum, all but eliminates the stress and drops delamination risk under 5% even past 5,000 cycles, though the $50–100/lb…

A datasheet claims: "High-Temperature Potting Compound, rated to 250°C." Your design calls for 180°C continuous operation, comfortably under that number, so you assume the compound is adequate. Six months later, solder joints start failing anyway. The temperature rating on a datasheet is not the same thing as usable operating temperature, and understanding the gap between the two prevents reliability failures that carry real warranty cost. Four Numbers That Get Confused as One Potting datasheets specify maximum temperature in at least four distinct ways, and suppliers don't always make clear which one they mean. Continuous service temperature (Ts) is the actual operating limit — the highest temperature the material tolerates indefinitely with acceptable properties. Glass transition temperature (Tg) is the point where resin shifts from rigid to rubbery; the material doesn't fail at Tg, but it softens and loses strength, and Tg is deliberately higher than Ts because operating at Tg itself is undesirable. Peak temperature (Tpeak) is what the material briefly tolerates — seconds to minutes — before irreversible degradation starts, and isn't a usable operating figure at all. Thermal decomposition temperature (Tdeg) is where the material begins to chemically break down, far above any practical operating point. Most datasheets specify Tg, not Ts, without always saying so explicitly. A compound with 250°C Tg typically has a continuous service temperature of 200–220°C — a meaningful gap from the number on the label. If the same "250°C" instead referred to Tpeak, safe continuous operation could be as low as 150–200°C. The only way to know which is being quoted is to ask directly. Why Failure Accelerates Near Tg Below Tg, potting is rigid and thermal expansion is elastic — the material returns to its original dimensions after each cycle, and solder joints see cyclic stress proportional to CTE mismatch but within elastic limits. At and above Tg, potting becomes viscoelastic: expansion and contraction stop fully recovering, and that non-recovery leaves residual stress that accumulates with every cycle. After 100–300 cycles through Tg, residual stress typically exceeds the solder joint's fatigue limit and cracks begin propagating. The numbers make this concrete. A compound with 180°C Tg, run continuously at 150°C with seasonal −20°C to +150°C cycling, imposes roughly 0.2% elastic strain per cycle below Tg — well tolerated, with 5+ year service life. Push operation up to 180°C, right at Tg, and each cycle adds 0.05–0.1% plastic strain on top of the elastic component; after 300 cycles, cumulative strain exceeds 15–30% and solder joints fail. At 200°C, above Tg, plastic strain dominates outright and service life can drop to 6–12 months. Building In Real Margin Reliable long-term performance (5+ years) generally requires continuous operation 50–80°C below Tg: a 200°C-Tg compound supports roughly 120–150°C continuous, a 220°C-Tg compound 140–170°C, and a 240°C-Tg compound 160–190°C. That margin isn't padding — it accounts for specification tolerance (actual Tg can run 5–10°C below the rated value), local hot spots (peak component temperature commonly runs 20–30°C above average assembly temperature), and the plastic strain accumulation that concentrates…

An automotive power supply potted with standard polyurethane survives 18 months before thermal cycling cracks its solder joints. The identical design, potted with high-temperature epoxy instead, survives five years or more. Material family choice matters as much as formulation quality within that family. Three families compete for high-temperature potting duty — epoxy, silicone, and polyurethane — and each brings a distinct set of trade-offs that no amount of formulation tweaking fully erases. Epoxy: The High-Temperature Default Epoxy's glass transition temperature runs 150–240°C depending on formulation, with thermal conductivity from 0.3 W/m·K unfilled up to 3 W/m·K with conductive filler, and a CTE of 40–60 ppm/°C. At $20–100/lb it's also the most cost-effective family at scale. Its strengths are thermal stability well above 150°C continuous, excellent mechanical strength for constraining components under stress, and strong resistance to oils, fuels, and solvents — which is why it dominates automotive under-hood and industrial high-heat applications. Its weaknesses are equally specific: standard epoxy is brittle and cracks under vibration or mechanical shock without 8–12% elastomer toughening added in, which increases both cost and formulation complexity. Unfilled, low-cost variants absorb 1–3% moisture (high-quality formulations cut that to under 0.5%), and large pours generate enough exotherm to risk cure defects without careful process control. Silicone: Built for Flexibility and Weathering Silicone's Tg range (120–220°C) varies more widely than epoxy's, with thermal conductivity of 0.3–2 W/m·K and CTE from 30–80 ppm/°C depending on filler — at a premium of $40–150/lb. Its defining trait is inherent elasticity, even unfilled, which absorbs vibration and thermal-cycling strain without cracking. Environmental resistance is outstanding: silicone tolerates UV, ozone, and weathering better than either alternative, retaining properties in outdoor deployments for a decade or more, and its shrinkage (under 1%) is the lowest of the three families, meaning less residual cure stress to begin with. The trade-offs run the other direction on temperature: standard silicone's Tg (120–160°C) is marginal for continuous operation above 140°C without a high-temperature-specific formulation, and softness that helps with vibration hurts mechanical support — components move more under thermal cycling than they would in a stiffer matrix. Silicone also absorbs more moisture than its reputation suggests (0.5–2%), and its naturally low surface energy can mean poor adhesion to some substrates unless the formulation includes adhesion promoters. Polyurethane: Lower Cost, Lower Ceiling Polyurethane's Tg (80–160°C) is the lowest of the three families, with thermal conductivity of 0.2–0.8 W/m·K, CTE of 50–100 ppm/°C, and a mid-range cost of $25–60/lb. It shares silicone's flexibility and impact resistance, adds genuinely hydrophobic moisture resistance from its closed-cell structure, and holds dielectric strength consistently across its (narrower) temperature range. But most polyurethane formulations top out below 150°C continuous, disqualifying them from the high-heat applications this series otherwise covers, and their CTE — the highest of the three families — generates the most thermal-cycling stress per degree of temperature swing. UV exposure yellows and degrades unprotected polyurethane within 12–24 months, and its chemical resistance lags epoxy and silicone against industrial solvents. It remains a reasonable…

An unencapsulated PCB sits in an industrial environment: 95°C continuous, 90% relative humidity, machinery vibration running underneath it. Within eight months, the first solder joint cracks. Within eighteen, corrosion has taken out several more. The board design was solid — the encapsulation decision wasn't. High-temperature potting compound has to defend against three threats that arrive together — heat-driven mechanical fatigue, moisture-accelerated corrosion, and vibration-induced stress — and understanding how they interact matters as much as addressing each individually. Heat: Thermal Cycling Fatigues Solder Joints Cycling from 50°C ambient to 95°C operating imposes 0.1–0.5% strain on solder joints each cycle, driven by CTE mismatch between components, board, and solder. After 500–2,000 cycles, accumulated strain initiates a fatigue crack that eventually propagates to failure. Potting fights this on three fronts: mechanical constraint reduces solder strain by 50–80% by immobilizing components and limiting board flex, elastomer-toughened formulations absorb cyclic strain internally instead of passing it to the solder, and thermally conductive potting flattens temperature gradients, cutting peak temperatures 10–20°C and reducing thermal-cycling amplitude proportionally. A properly potted board typically survives 5–10x more thermal cycles than an unencapsulated one. Moisture: A Slow, Invisible Accelerant Moisture doesn't announce itself. FR-4 absorbs 0.5–1.0% water by weight at 85°C/85% RH — the standard durability benchmark — and that absorption accelerates to 2–3% at 95°C/90% RH over 6–12 months. Once absorbed, moisture migrates along copper traces under applied voltage (electro-osmosis), concentrates at interfaces, and combines with ionic contaminants — flux residue, salt — to drive electrochemical corrosion and whisker growth that eventually shorts adjacent traces. The insidious part: once moisture reaches a solder joint, corrosion can continue for years even after ambient humidity drops, as residual moisture trapped in the PCB resin slowly migrates outward. Cured potting is essentially a vapor barrier — moisture absorption through quality potting runs under 0.1% per year, against 0.5–3% for bare boards, per ASTM D570, the standard method for water absorption of plastics. By sealing component leads and trace access points, potting blocks the pathway moisture would otherwise use to reach solder joints, keeping a board dry for 5+ years in an environment that would corrode an unencapsulated one within 6–12 months. Vibration: The Accelerant for Everything Else Machinery hum, mechanical impacts, and random vibration flex unencapsulated boards continuously. Tall components — electrolytic capacitors, connectors — act as cantilevers that oscillate at resonance, generating peak solder-joint stress up to 10–20G. Copper traces crack under the same flex, especially where thermal stress compounds it. Critically, vibration doesn't usually initiate solder cracks on its own — thermal cycling does that — but it dramatically accelerates crack growth once a micro-crack exists, sometimes propagating a defect invisible without magnification to complete failure within 1,000 cycles. Potting rigidly immobilizes components, shifting resonance frequency out of the excitation range entirely or eliminating it, while elastomer toughening damps vibration energy through particle deformation, cutting oscillation amplitude 80–95%. See Incure's dedicated analysis of whether potting compound prevents vibration damage for the mechanics in more depth. Why the Combination Is…

The question seems straightforward: doesn't adding an insulating layer of potting reduce heat dissipation? The intuitive answer — yes, potting must add thermal resistance — is technically correct but practically misleading. Potting's effect on heat dissipation depends entirely on formulation and application. Poor choices trap heat and hurt reliability; optimized potting actually improves overall thermal performance. The Local Thermal Resistance Calculation Heat flows from hot components through the potting matrix to the assembly exterior. The potting layer adds thermal resistance proportional to thickness and inversely proportional to conductivity: R = thickness / (conductivity × area). A 5mm layer of unfilled epoxy (0.5 W/m·K) across a 10cm × 10cm surface adds about 0.025°C/W — for a 50W heat source, roughly a 1.25°C rise. The same 5mm layer at 3 W/m·K (thermally-filled potting) adds only 0.0042°C/W, raising temperature just 0.2°C. At first glance, even unfilled potting looks like a minor penalty. But that analysis ignores how potting changes system-level thermal behavior. The System-Level Picture Without potting, heat dissipation is distributed and uncontrolled: hot spots conduct to nearby air through slow natural convection, cool regions stay cool while hot regions overheat, vibration and flex disrupt contact with mounting surfaces, and moisture-enabled corrosion increases electrical resistance in ways that generate still more heat. With optimized potting, heat is distributed through thermally-conductive material to cooler regions and the exterior, contact with external heat sinks is controlled and consistent rather than dependent on a few high points touching, and the whole system runs more efficiently despite the added material layer. Real-world measurements often show properly-potted assemblies running 10–20°C cooler than unencapsulated designs. Unfilled vs. Thermally-Conductive Potting Unfilled potting (epoxy, polyurethane, generic formulations) runs 0.2–0.5 W/m·K, acts as a thermal insulator, and can leave peak temperature 20–40°C higher than an unencapsulated design — acceptable only below roughly 10W. Aluminum-oxide-filled, thermally-conductive potting runs 1–4 W/m·K, actively conducts heat away from hot spots, flattens gradients, and often runs 10–20°C cooler than the unencapsulated equivalent — required above roughly 20W. The crossover sits around 15–25W: below it, air cooling dominates and potting choice matters less; above it, thermally-conductive potting meaningfully improves thermal performance. For a broader comparison of how conductivity trades off against other properties across compound families, see our comparison of silicone, epoxy, and polyurethane potting. The Gradient Problem in Practice An unencapsulated 50W power supply with a MOSFET generating 10W locally might see that hot spot reach 120°C while the PCB edge sits at 90°C — a 30°C gradient that stresses nearby solder joints and drives uneven convection currents. With thermally-conductive potting, that gradient drops to 8–12°C: peak temperature falls 15–20°C, cycling stress on solder eases, overall convective cooling improves from more uniform surface temperature, and cooler surfaces absorb less moisture. The potted assembly, despite its added insulating layer, ends up running cooler at the point that matters. External Heat Sink Contact If the assembly mounts to a metal chassis or thermal spreader, potting's effect is even more pronounced. An unencapsulated assembly has irregular contact — high points…

A 100W LED driver sits in a 150°C environment. Its MOSFET junction reaches 170°C, and the inductor core runs at 140°C continuous. Without proper potting, solder joints crack within months, and a driver designed for a 10-year service life fails in eighteen. LED driver potting in extreme heat demands thermal management beyond standard high-temperature practice, because power dissipation concentrates in a handful of components separated by only millimeters. Why LED Drivers Run Hotter Than They Look Power MOSFETs, inductors, and electrolytic capacitors all generate heat in close proximity, creating steep gradients across a small board — an unencapsulated driver can show a 30°C swing between the hottest MOSFET and the PCB edge. That gradient alone drives internal stress and accelerates fatigue. Potting with poor thermal conductivity makes this worse, not better: it traps heat and can raise peak component temperature by 10–30°C compared to leaving the board unencapsulated. Thermally insulating potting, in other words, becomes a liability the moment power dissipation crosses roughly 25–30W. Incure's guide on choosing between conductive and insulating potting covers that threshold in more detail. Conductive Fillers Do the Real Work Aluminum oxide at 60–75% loading is the workhorse filler: 2–4 W/m·K, cost-effective, and widely available. Boron nitride costs more but stays electrically insulating at 2–5 W/m·K, useful where aluminum oxide would be electrically or chemically inappropriate. Silicon carbide reaches 5–10 W/m·K but its hardness abrades mixing equipment, so it's reserved for the highest-power designs where the extra conduction is worth the wear. For most LED driver work, 60–70% aluminum oxide loading delivers 3–4 W/m·K without sacrificing too much mechanical toughness — a balance validated against ASTM D5470, the standard method for measuring thermal transmission properties of thermally conductive electrical insulation materials. High filler loading has a cost, though: above roughly 60% by weight, the cured compound starts behaving like a ceramic — excellent conductivity, poor tolerance for the vibration LED drivers experience from load switching. Elastomer toughening (8–12% rubber particles) recovers vibration resistance at a 10–15% conductivity penalty. Optimized filler particle-size distribution can hit the same conductivity at 60–65% loading instead of 70–75%, reducing brittleness without additives. Hybrid filler systems — aluminum oxide for conduction, hollow microspheres for toughening — split the difference. Building a Heat Path, Not Just a Heat Sink Potting technique matters as much as filler chemistry. Identify the MOSFET, inductor, and any other concentrated heat source before pouring, then apply a thin layer of thermally conductive paste directly beneath those components to create a preferential path to the potting surface. If the assembly mounts to an external heat sink or chassis, align potting thickness so heat conducts efficiently to that sink rather than dead-ending in a thick, insulating pour. For very high-power designs, embedded copper or aluminum foils along the heat path outperform relying on the filled matrix alone. Enclosure design compounds these gains: a thin, wide housing radiates and convects better than a compact cube, and any trapped air pocket — air is a poor conductor — undoes careful thermal…