Film Capacitor THB Testing: 6 OEM Checks

A film-capacitor THB qualification must define temperature, relative humidity, electrical bias, duration, acceptance limits, and post-test recovery before a result can be compared. “85/85 tested” is not a complete purchasing specification. An OEM should connect the exact test condition to the proposed series, document the starting measurements, and approve the first article using traceable capacitance, dissipation factor, insulation resistance, and visual records.

Key Takeaways

What procurement and engineering should align before sampling

• Write all six THB inputs; temperature and humidity alone are insufficient.

• Use model-specific evidence because voltage, construction, and protective design change the result.

• Define capacitance, DF, insulation resistance, and visual limits before testing.

• Record recovery time and measurement conditions so before/after values are comparable.

• Keep supplier-series examples attributed; never turn them into universal LORIDA ratings.

Why THB Is More Than an “85/85” Label

Humidity, heat, and electric field interact inside the capacitor

Temperature-humidity-bias testing accelerates moisture-related electrical and metallization risks while the capacitor is energized. TDK's general film-capacitor guidance compares three atmospheres: 40°C/93% RH contains about 47 g/m³ of water vapor and is normalized to severity 1.0; 65°C/95% RH contains about 151 g/m³ and is rated 10.5; 85°C/85% RH contains about 294 g/m³ and is rated 47.3. The figures show why equal-looking RH percentages do not represent equal moisture stress.

TDK also distinguishes reversible humidity effects from long-term irreversible damage. Moisture can temporarily change electrical behavior, while extended exposure may promote corrosion and loss of metallization. Bias matters because it adds an electric field to the thermal and moisture stress. A valid comparison therefore needs the exact applied voltage, polarity or waveform, test duration, sample construction, and measurement sequence—not just the chamber set points.

Six OEM Checks for a THB RFQ

Checks 1–3: temperature, humidity, and bias

1. Temperature: State the chamber temperature, tolerance, ramp or stabilization method, and whether the rating is an operating or test condition.

2. Humidity: State RH, tolerance, condensation controls, and whether specimens are preconditioned.

3. Bias: State AC or DC, magnitude, polarity, frequency when relevant, connection method, and how the applied value relates to the exact part rating.

Checks 4–6: duration, limits, and recovery

4. Duration: State exposure hours and whether interruptions, inspections, or electrical cycling are allowed.

5. Acceptance limits: Set allowable capacitance change, DF change, insulation-resistance floor, dielectric checks, and visual criteria before the test starts.

6. Recovery: Define chamber exit, discharge, cleaning if permitted, recovery temperature and humidity, waiting time, and the instruments used for final measurement.

Published Examples Must Stay Series-Specific

Manufacturer data shows that one THB statement cannot cover every film capacitor

TDK reports that its B3277*H rugged DC-link series was tested for 1,000 hours at 60°C/95% RH with rated voltage applied, while special types were tested for 1,000 hours at 85°C/85% RH with rated voltage. The same announcement lists that family across 450–1,100 V DC and 1.5–120 µF, with a maximum operating temperature of 105°C. These values describe that named family, not all DC-link capacitors.

TDK separately describes a high-humidity AC-filter family tested for 1,000 hours at 85°C/85% RH with rated voltage, including variants rated at 460 V RMS / 650 V peak and 2–50 µF. KEMET's R53 X2 page identifies humidity robustness Class IIIB and a 1,000-hour 85°C/85% RH test at rated 310 V AC and 560 V DC. KEMET's C4AK DC-link announcement likewise states 1,000 hours at 85°C/85% RH under rated voltage for that series. These examples prove the need to name the exact series, bias, and acceptance basis in an RFQ.

When comparing proposals, request the report revision, sample quantity, construction identification, actual voltage at the specimen, chamber tolerances, interruption history, and individual before/after readings. Confirm whether the stated duration is continuous exposure and whether measurements were made immediately or after a controlled recovery. A pass statement without this context cannot show that two test programs are equivalent. The goal is not to demand the harshest number on a marketing page; it is to select a documented test that represents the application and can be repeated on the delivered construction.

RFQ and First-Article Evidence Table

Use the same structure from quotation through approval

First-Article Validation Workflow

Make the comparison traceable and repeatable

Start from the proposed film capacitor part number and drawing. Photograph labels and terminals, identify the lot, stabilize the samples under the agreed starting condition, and record capacitance, DF, and insulation resistance with instrument IDs and measurement settings. Load the chamber only after the electrical connections and bias monitoring have been reviewed. Preserve the chamber profile and applied-voltage record for the full exposure.

After exposure, follow the specified discharge and recovery sequence once. Record final values at the agreed time, compare each sample with its own baseline, and document visual changes such as case cracking, swelling, corrosion, seal changes, or terminal damage. Connect the IR review to the film-capacitor insulation-resistance checklist. If any condition or instrument record is missing, classify the result as incomplete rather than silently treating it as a pass.

Frequently Asked Questions

What does 85/85 THB mean for a film capacitor?

It usually means exposure at 85°C and 85% relative humidity while an electrical bias is applied, but the phrase alone is incomplete. You still need the bias magnitude and waveform, duration, sample count, preconditioning, acceptance limits, recovery period, and measurement method. Two suppliers can both state “85/85” while using different voltage stress, test hours, component construction, and pass/fail criteria. Ask for the exact test report or series specification and ensure it matches the proposed part number rather than a different family.

Is 1,000 hours at 85°C/85% RH a universal requirement?

No, 1,000 hours at 85°C/85% RH is a common published qualification point, not a universal requirement for every film capacitor or application. The correct condition depends on the applicable component standard, end-use environment, voltage category, safety class, customer specification, and exact series design. Some manufacturers publish 60°C/95% RH conditions for one family and 85°C/85% RH for another. Treat each value as part-number-specific evidence, then define the OEM test that represents the actual risk and approval plan. Record the governing document and revision in the purchase specification.

Which measurements should be taken before and after THB?

At minimum, record capacitance, dissipation factor, insulation resistance, and a detailed visual inspection before and after THB. Add dielectric withstand or leakage checks only when the applicable standard and safety process permit them. Use the same instruments, frequency, voltage, temperature, humidity, and recovery timing for comparable readings. Keep raw data by sample, not just averages, because one outlier may reveal a sealing, metallization, or terminal problem. Include lot identity, instrument calibration status, photographs, and the chamber and bias logs in the final report.

Can a supplier's THB result be transferred to another capacitor series?

No, a THB result should not be transferred automatically to another capacitor series. Moisture behavior depends on film and metallization systems, winding and edge design, encapsulation, sealing, terminals, case geometry, voltage stress, and manufacturing controls. Even closely related series may use different constructions or ratings. A supplier may use similarity evidence within a controlled qualification family, but the OEM should request the rationale, construction comparison, worst-case selection, and applicable report. When the relationship is not documented, qualify the exact proposed part or keep the evidence status open.

Sources and Next Step

Verified technical references used here

Exact examples were checked against TDK's general film-capacitor technical information, B3277*H rugged DC-link announcement, high-humidity AC-filter announcement, and KEMET's R53, X2 high-humidity, and C4AK technical pages. All values remain attributed examples and are not LORIDA-wide guarantees.

Last Updated: October 10, 2026