Estimating Film Capacitor Service Life from Self-Healing Events and dV/dt Stress

Date:2026-9-30 Share to:

A Practical Lifetime Model for Metallized Film Capacitors

Metallized film capacitors fail differently from electrolytic or ceramic types. Instead of a sudden dielectric rupture, they experience self-healing events, also called clearing. During a clearing, the metallization around a defect vaporizes in microseconds, isolating the fault and restoring insulation resistance. The capacitor loses a small amount of electrode area, which translates directly into a capacitance drop. A single clearing event typically removes a few picofarads of capacitance, but repeated events accumulate over time.

The conventional lifetime model treats these capacitors as wear-out devices. The key metric is the number of clearings a capacitor can survive before capacitance loss reaches a defined end-of-life threshold, usually a 2% to 5% drop from initial value. The rate of clearing is governed by the applied voltage stress, the temperature at the core, and the peak current during switching transients. Higher dV/dt events deposit more energy into the defect site, making each clearing larger and more destructive.

For engineering purposes, the lifetime can be expressed as:

L = Lref × 2(Tref − Tcore) / 10 × (Vref / Vapp)n

where Lref is the rated life at rated voltage and rated core temperature, Tcore is the actual hotspot temperature, Vapp is the applied DC voltage, and n is a voltage exponent typically between 7 and 9 for metallized polypropylene film. The Arrhenius term doubles life for every 10 °C reduction in core temperature, which is why thermal management matters more than any other single factor.

Temperature and Ripple Current Derating: Working Numbers

The core temperature of a film capacitor is the sum of ambient temperature and self-heating from ripple current. For a typical metallized polypropylene capacitor rated at 85 °C and 450 V DC, the expected life at full rating is around 100,000 hours. Operating at 75 °C instead of 85 °C doubles that figure to roughly 200,000 hours. Conversely, operating at 95 °C halves it to about 50,000 hours.

Ripple current produces I²R losses in the electrode and contact resistance. The allowable ripple current at a given ambient temperature is specified in the datasheet, but a common derating rule is to reduce the rated ripple current by 1.5% per degree Celsius above 85 °C, and to derate voltage by 2% per degree above 85 °C. At 105 °C, the permissible ripple current falls to about 70% of the 85 °C rating, and the working voltage should be reduced to roughly 80% of the rated value.

Self-heating from ripple is not uniform across the winding. The hotspot sits near the center of the wound element, where thermal resistance to the case is highest. A rule of thumb is that the internal hotspot runs 5 °C to 10 °C above the case temperature for axial-lead parts, and 3 °C to 5 °C above for radial-lead box types with larger surface area. Measuring case temperature with a thermocouple under full load gives a conservative estimate of the core if you add the appropriate thermal gradient.

Operating Conditions vs. Expected Life: A Reference Table

Core Temp (°C) Applied Voltage (% of rated) dV/dt (% of rated limit) Estimated Life (hours) Failure Mode
85 100% 100% 100,000 Gradual capacitance loss
75 100% 100% 200,000 Gradual capacitance loss
85 80% 100% 180,000 Gradual capacitance loss
85 100% 60% 140,000 Gradual capacitance loss
95 100% 100% 50,000 Accelerated clearing
105 80% 60% 45,000 Accelerated clearing
105 100% 100% 25,000 Rapid capacitance loss, short circuit risk

The table above assumes a metallized polypropylene capacitor with a rated life of 100,000 hours at rated conditions. The voltage exponent n is taken as 8, and the dV/dt derating follows a linear reduction in clearing energy. These values are representative; actual figures vary by dielectric type, metallization pattern, and manufacturer. Polyester film capacitors, for example, have a lower voltage exponent and shorter life at high temperature due to higher dielectric losses.

Design Rules for Reliable Operation

When selecting a film capacitor for a switching application, the dV/dt rating is not a secondary concern. It defines the maximum rate of voltage change the capacitor can withstand across its terminals during a switching transient without internal clearing events becoming destructive. Exceeding the dV/dt rating by even 20% can triple the clearing rate, reducing life by half or more. The design rule is simple: select a capacitor with a dV/dt rating that is at least 1.5 times the maximum slew rate expected at the application’s worst-case switching condition.

For procurement engineers, the practical takeaway is to verify the dV/dt rating against the actual switching waveform, not the nominal operating frequency. A capacitor rated for 100 V/µs at 10 kHz may fail prematurely if the circuit produces 150 V/µs spikes during inrush or fault conditions. Similarly, the ripple current specification must account for the harmonic content of the waveform, not just the fundamental frequency.

Cross-border sourcing adds a compliance dimension. Film capacitors destined for EU markets should be RoHS and REACH compliant, with documentation available from the distributor. Lead time is another factor: standard metallized polypropylene capacitors with dV/dt ratings above 500 V/µs are often made to order, with lead times of 8 to 12 weeks. Stock-holding distributors with global shipping capability can reduce that to days for common values and voltage classes.

Finally, do not overlook the end-of-life behavior. A metallized film capacitor that has lost 5% of its capacitance is often still functional but may cause the circuit to fall out of tolerance. In snubber or DC-link applications, monitoring capacitance drift during preventive maintenance provides an early warning. Plan your replacement schedule around the estimated life from the table above, and verify actual core temperatures during commissioning to confirm the assumptions used in the calculation.

Copyright:https://www.shgopi.com Please indicate the source when reprinting