Figure 1. A heat-cure-only epoxy qualification is correct under controlled laboratory conditions. When production oven capacity is shared across multiple SKUs, the actual cure received by each batch may not match the qualification basis.
The qualification document says 100°C × 2 hours. The production floor runs three product lines through the same oven. On the day a transformer batch was loaded, the oven was already mid-cycle on a different SKU. The operators waited until the current cycle finished, loaded the transformers, and started the cure. The actual dwell at temperature: 90 minutes. This happens on Tuesday, Thursday, and most Fridays. The qualification has not changed. The material has not changed. What changed is what was actually produced.
Under-cure caused by scheduling compression is the single most common production non-conformance in epoxy potting that escapes incoming inspection. It does not appear on the surface. It does not fail hi-pot at shipment. It appears as dielectric failure or interface delamination 12 to 36 months into field service - at which point the production batch that caused it has been fully shipped and the cure records, if they exist, do not capture what actually happened to each unit.
What Under-Cure Actually Does to an Epoxy Potting System
Cross-linking in an epoxy potting system is not binary - it does not jump from uncured to fully cured at a defined time. The degree of cross-link conversion is a function of time, temperature, and the specific catalyst or hardener system. A heat-cure-only formulation is designed so that, at its rated cure temperature and duration, the system reaches a target degree of conversion - typically defined by the resulting Shore D, Tg, and dielectric strength values on the TDS.
Figure 2. Under-cure from abbreviated heat cycles is not visible on the surface. The cross-link network remains incomplete below the surface hardness reading, leaving the matrix susceptible to moisture absorption and reduced dielectric strength.
When cure time is shortened or temperature is lower than specification, the degree of conversion falls below target. The result is a polymer network with:
Lower cross-link density - the matrix is softer than the specified Shore D. The surface of the cured part may feel firm, but subsurface cross-link density in thick sections may be substantially lower than the surface reading.
Reduced Tg - the glass transition temperature drops in direct proportion to under-cure. A system rated at Tg 90°C may produce a cured part with effective Tg of 65–75°C under a shortened cycle. Above the actual Tg, thermal softening and accelerated creep begin.
Reduced dielectric strength - incomplete cross-linking leaves polar groups in the matrix that attract moisture. Moisture absorption in the cured epoxy reduces bulk resistivity and creates localized conductive paths.
Weakened adhesion at the substrate interface - the early stages of cross-linking at the substrate-epoxy interface are particularly sensitive to thermal history. Under-cure at the interface reduces adhesion strength, which, combined with thermal cycling stress, initiates delamination.
None of these changes are detectable by visual inspection. A dielectric hi-pot test at initial shipment will usually pass under-cured specimens, because the reduction in dielectric strength from under-cure is gradual and the safety margin built into hi-pot voltage levels typically absorbs it. The failure appears later.
How the Failure Presents in the Field
Under-cure field failures in epoxy-potted assemblies follow a recognizable pattern, though it is rarely identified correctly at first investigation:
Timeline: Assemblies ship with no quality escapes. Early field units function normally. Between 12 and 30 months of service, a cluster of returns begins - not a single failure mode, but a mix of intermittent opens, tracking failures on high-voltage surfaces, and occasional physical cracking at interfaces.
Failure distribution: The failures are not random across the product line. They correlate with production dates - specifically, with batches that were manufactured during periods of high production throughput when oven scheduling was under pressure. This correlation is almost never identified unless someone specifically maps return dates against production batch dates and oven logs. In most factories, oven logs do not capture actual achieved temperature and dwell time for each specific batch - only the oven setpoint and the programmed cycle duration.
Root cause misidentification: The failure is initially attributed to the component - a capacitor lot, a PCB surface finish, a connector. Under-cure of the potting compound is not in the standard failure investigation checklist because the potting material is assumed to have cured correctly. Cross-sectional analysis of the returned units may reveal softer-than-expected Shore D in the potting, but only if someone measures it on the returned unit and compares it to a reference. This rarely happens.
The consequence is that the root cause - production scheduling compression - remains unaddressed, and the next batch produced under similar conditions repeats the failure.
Why Single-Path Heat-Cure Qualification Does Not Cover Production Reality
When an epoxy potting compound is qualified for heat cure only, the qualification covers one specific condition: a defined temperature, a defined dwell time, and an assumed oven load. The UL certification under which the flame classification was granted was obtained on specimens prepared under controlled laboratory conditions - not under the variable thermal history of a shared production oven.
This creates a structural mismatch. The qualification document is a statement about the material under controlled conditions. It says nothing about what the material will do when those conditions are not met - because the qualification process does not model production variability. A single-path heat-cure epoxy system, when used in a production environment where the cure cycle cannot be reliably guaranteed, is being operated outside its qualification basis on a statistically predictable number of production days. The qualification is not wrong. The application of it is.
The correct engineering approach is to select a material whose qualification basis matches the production environment's actual capability - not to assume the production environment will conform to the material's requirements.
How Dual-Path Cure Qualification Resolves the Exposure
A potting system that is formally qualified under both room-temperature and heat-accelerated cure profiles removes the scheduling dependency from the compliance equation. Both cure paths - if specified and validated - produce a part that meets the material's rated properties. The operator can choose the path that is available. If the oven is occupied, the part cures at ambient. If the oven is available, heat cure accelerates throughput. Neither choice produces an out-of-compliance part, provided the selected cure schedule is followed correctly.
This qualification structure matters for several reasons beyond the immediate convenience:
UL compliance is maintained on both paths. The UL 94 V-0 flame classification is not path-dependent - it applies to the cured material regardless of which qualified cure schedule was used. The certification is on the material, and both schedules produce the same cured material.
Production records are simplified. Instead of tracking "did this batch receive the correct oven cycle," the production record only needs to confirm which of two qualified schedules was applied. The compliance gate is the schedule choice, not the oven log.
New operator training is reduced. There is no decision to make about whether a delayed batch needs special handling - the RT cure path is the default for any batch that cannot be heat-cured within the production window.
The limitation is that RT cure requires a controlled ambient temperature over a 7-day window. Floor temperature fluctuations during that window are a process variable that must be controlled - they are not background noise. This is frequently the gap in RT cure implementations: the cure is initiated, the assembly is moved to a shelf, and temperature control at that shelf location is not monitored. Subsurface under-cure from ambient temperature variation during the RT cure window is a real failure mode, separate from the oven scheduling problem it was intended to solve.
Identifying Whether Your Current Process Is Exposed
The following conditions indicate that a production line may be operating a heat-cure-only epoxy outside its qualification basis:
The oven is used for more than one product type and cycles are scheduled in sequence.
Cure dwell time is set by the oven program but actual temperature at the part location has not been validated with a thermocouple in the potted section.
Production records show cure start time but not confirmed part temperature during the cure hold.
Batch sizes loaded into the oven vary - larger thermal mass requires longer ramp time to reach target temperature, which reduces effective dwell time if the timer starts at oven loading rather than at part temperature.
Field returns show a statistical correlation with production throughput periods (high-volume weeks show disproportionate return rates 12–24 months later).
None of these conditions individually confirm under-cure. Together, they indicate process capability that should be formally assessed before assuming the current qualification covers actual production output.
Engineering Limits of Dual-Path Systems
A dual-cure-path epoxy is not interchangeable with a thermally optimized single-cure system in applications where thermal performance is the primary design driver. Flexible-cure formulations are typically not the highest-Tg or highest-RTI options in a product group. The scheduling flexibility involves engineering trade-offs:
RTI rating - a system qualified for RT cure will typically carry a lower RTI than a fully heat-developed high-Tg system. Continuous operating temperature above the RTI is outside the material's insulation life rating. This must match the application's operating temperature requirement.
Minimum UL thickness - the flame classification is thickness-dependent. Verify that the design potting thickness meets or exceeds the certified minimum for the specific colorway specified.
RT cure ambient control - if RT cure is used as the primary production path, ambient temperature must be monitored and documented as a process parameter. A 7-day cure at 18°C produces a different degree of conversion than 7 days at 25°C.
Related Product for Shared-Oven Production Environments
E532/H532 is a two-component, UL 94 V-0 flame-retardant epoxy potting compound evaluated under UL File E120665, qualified for both room-temperature cure (7 days at 25°C) and heat-accelerated cure (50°C × 1 hr + 100°C × 2 hrs). RTI is 90°C for Electrical, Mechanical Impacted, and Mechanical Strength. Minimum certified thickness is 6.0–6.6 mm across all colorways.
It is appropriate for assemblies where the primary production constraint is cure schedule variability, operating temperatures are continuously below 90°C, and the potting section thickness falls within the UL-certified range. It does not address thermal conductivity requirements - designs where the potting layer must conduct heat should evaluate E533/H533 (1.5 W/m·K).
→ 🔗E532/H532 Product Page - Technical Data, UL Certification, Application Notes
Key Engineering Questions
If the batch received 90 minutes instead of 120 minutes at cure temperature, how different is the result?
The answer depends on the specific formulation and the temperature achieved at the core of the potted section - not just the oven surface temperature. A 25% reduction in dwell at 100°C may result in a 10–20°C reduction in effective Tg, depending on the kinetics of the hardener system. This is not visible in visual inspection or initial dielectric testing. The only reliable verification is to produce witness specimens at the shortened cycle and measure Shore D and Tg on those specimens directly.
Can the under-cure be corrected by post-cure after the assembly has been put into service?
No. Once the assembly is in the field, re-curing requires removing it from the application and subjecting it to the corrective thermal cycle - which is impractical in most service environments. Under-cure correction must occur before the assembly is shipped. This reinforces the importance of identifying the exposure during production, not after field returns begin.
Does UL certification require the manufacturer to use a specific cure schedule?
UL component recognition certifies the material properties of the cured compound, tested under controlled laboratory conditions. The certification does not mandate a specific production cure schedule - it certifies what the material is capable of when properly cured. The production process qualification, including cure cycle validation, is the responsibility of the assembly manufacturer. If the production cure schedule does not match the material's validated cure parameters, the resulting cured part may not replicate the certified properties, regardless of whether the material itself holds UL certification.
What is the minimum documentation required to demonstrate cure process compliance?
At a minimum: a documented and validated cure schedule (temperature, ramp rate, dwell time, and confirmation method), a record of actual cure condition for each production batch (not just the programmed setpoint), and periodic verification of cured properties on production witness specimens. In UL-listed end products, the listing body may require specific process controls as a condition of the end-product listing. This should be confirmed with the relevant certification body before finalizing the production process specification.
Next Steps - Contact Fong Yong Chemical
