Capacitor Capacitance Loss: Causes and Limits
Capacitor capacitance loss means an independently measured capacitance has declined relative to the correct reference value. In industrial film power capacitors, progressive dielectric or electrode damage can be one cause, but reduced bank kvar does not by itself prove that capacitance was lost. Voltage changes, disabled stages, open fuses, switching faults and measurement differences can produce similar operational symptoms. Establish the component identity, a repeatable offline measurement and the operating context before deciding whether aging or abnormal duty requires replacement. This guide concerns low-voltage compensation capacitors rather than batteries or electrolytic electronics components.
Define the comparison before diagnosing aging
A useful trend compares the same unit, terminal arrangement and approved method with a reliable baseline. A catalogue number, old controller reading and present handheld measurement are three different kinds of evidence. They should not be treated as directly comparable capacitances. Record the original nameplate value and clarify whether it describes an internal element or the equivalent measured at accessible terminals.
If no commissioning measurement exists, use the documented manufacturer expectation with its tolerance and measurement conditions. Identify the uncertainty of the instrument before interpreting a small difference. A result that is numerically lower can still overlap the comparison range. Preserve the actual readings and their conditions rather than reporting an unsupported percentage as a confirmed deterioration rate.
Why a film capacitor can lose useful area
Self-healing film designs can clear small local dielectric defects by isolating a tiny electrode area around a fault. That protective behaviour does not restore the removed area. Repeated or extensive local damage can reduce the effective capacitance. Other construction-specific deterioration mechanisms may also matter, so the supplier’s technical assessment is needed when identifying the actual cause in a removed unit.
Do not assume that every low reading represents benign self-healing. A disconnected element, open internal path or incorrect terminal measurement can have a different explanation. Likewise, a normal-looking case cannot establish internal condition. The CNBYG cylindrical BSMJ capacitor family is an appropriate product reference for an enquiry, but model-specific construction and acceptance data must be confirmed for the installed unit.
Separate capacitance from the bank’s output
For an ideal directly connected capacitor at a fixed frequency, reactive output follows Q = 2πfCV² when C and V describe the same single-phase branch. This means both capacitance and voltage affect output. Use the actual internal connection before extending the relationship to a three-phase assembly. The basic capacitor reactance relationships are explained by OpenStax’s simple AC circuits lesson.
Consider an illustrative uncomplicated branch whose capacitance is unchanged but whose voltage falls to 95% of the original value. Its ideal kvar falls to 0.95², or about 90.3% of the original, without a capacitance failure. This is a calculation example, not a permitted operating-voltage range. A detuned branch introduces reactor effects, so its output requires the coordinated branch model rather than this bare-capacitor shortcut.
Rule out disappearing stages first
Compare the controller’s command with the physical stage response. A commanded stage may have an open fuse, an unclosed contactor or an unavailable switching device. The resulting correction gap can resemble a weak capacitor. Conversely, a welded contactor can leave output connected when the controller expects it to be absent. Retain actual branch currents and switching evidence alongside the displayed stage status.
Check all relevant phases under the approved operational measurement procedure. An asymmetry may direct attention to a fuse, contact, connection or internal element. It does not identify the exact cause without additional evidence. Investigate a sudden step change differently from a gradual trend: abrupt loss often warrants reviewing events and switching paths before assigning a long-term aging explanation.
Use a structured evidence table
The table separates possible causes from measurements that can distinguish them. It is not a universal acceptance standard. IEC 60831-1 identifies the self-healing low-voltage shunt capacitor standards family; the applicable product documentation supplies the relevant test and acceptance requirements.
| Observed change | Alternative explanation | Useful discriminating evidence |
|---|---|---|
| Lower kvar at the bus | Lower voltage or fewer connected stages | Voltage, stage state and fundamental branch current recorded together |
| Low offline terminal-pair reading | Different connection or measurement method | Original diagram, expected terminal equivalent and repeat test |
| Unequal phase response | Fuse, switch, connection or internal imbalance | Phase-by-phase circuit investigation under the approved procedure |
| Slow repeatable capacitance decline | Progressive component deterioration | Comparable history plus thermal and electrical duty |
| Several units decline together | Shared stress or changed test conditions | Common duty, batch identity and instrument-method review |
| Result varies between repeats | Unstable fixture or instrument limitation | Range, leads, external paths and measurement uncertainty |
Make the offline measurement repeatable
The qualified procedure must first establish isolation and control stored energy. Pratiques de travail en électricité de l'OSHA describe hazardous stored-energy release and verification in their jurisdiction. A capacitance function does not verify deenergization. Do not apply it to energized equipment or improvise a discharge method from a tutorial.
After safe access is established, identify external parallel paths and the internal discharge arrangement. Record terminal pairs and the condition of any floating terminal or neutral. Use an instrument with suitable range and documented accuracy. The capacitor-bank safety guide provides the assembly context. Retest anomalies consistently and retain unsuccessful or unstable readings so uncertainty remains visible in the investigation.
Relate deterioration to thermal duty
Compare the capacitance trend with cabinet temperatures, ventilation changes and demanding production periods. Filters, fans, spacing and neighbouring equipment can change the thermal environment. A normal room temperature does not establish the temperature at the capacitor. Look for a repeatable relationship rather than assuming that one warm photograph proves the cause.
Separate local terminal heating from overall case or cabinet heating. Connection resistance may create a concentrated hotspot and require its own repair. A shared cooling problem may affect multiple stages. Document both the identified fault and the verification of the repair. Without representative follow-up operation, the maintenance team cannot tell whether the stress was removed or merely absent during a brief test.
Review harmonic and switching stress
Collect voltage and current spectra at relevant operating states, with the capacitor switching configuration recorded. Total RMS current and fundamental current answer different questions when harmonics exist. Review changes in drives, transformers, generation and compensation equipment. A changed network can alter branch duty even when the capacitor and controller settings remain unchanged.
For reactor-equipped stages, investigate the capacitor and reactor as a matched branch. A decrease in capacitance changes the branch relationship and can affect tuning. The reactor-in-capacitor-bank guide explains the purpose of the coordinated assembly. Switching event history also matters: compare stage cycling, reclosure logic and device condition with the original design rather than assuming all deterioration is calendar aging.
Set replacement criteria from evidence
Do not invent a universal allowable percentage loss. The decision depends on the supplied product requirements, circuit design, measurement uncertainty, imbalance, physical condition and operational needs. Distinguish a confirmed out-of-criteria component from a trend that needs observation. A unit with visible damage or another failed safety-related function may require removal regardless of a seemingly acceptable capacitance result.
When replacement is justified, retain the voltage, frequency, internal connection, stage duty, matching reactor, mounting and discharge requirements. A different capacitance can alter the branch; a higher voltage rating may also change the kvar available at the actual operating voltage. Ask the responsible engineer to approve the coordinated replacement and define its baseline measurements before returning the stage to service.
Preserve a useful trend for the next inspection
Use a stable unit identifier and keep the original records when a component is replaced. Mixing old and new units under one unnamed stage can hide the actual history. Include instrument identity, terminal arrangement, readings, acceptance source, ambient context and operational measurements. Record any method changes so later reviewers do not mistake them for a physical improvement or decline.
Set the next review around the suspected stress mechanism. A cooling issue needs follow-up during demanding thermal conditions; a switching concern needs event evidence; a network change needs representative spectra and stage configurations. Add those checks to the capacitor-bank maintenance records. A useful trend links measurements to decisions rather than collecting numbers without an action threshold.
Educational video: capacitance fundamentals
The Engineering Mindset explains capacitor construction and charge storage. The lesson provides conceptual background; it is not a test or discharge procedure for an industrial compensation bank. The installed unit’s documents and qualified work procedure govern the investigation.
Foire aux questions
Does lower bank kvar prove capacitance loss?
No. Voltage, switching, fuse continuity, phase condition and the branch model can change output. Confirm a repeatable capacitance measurement with the correct comparison.
Can a film capacitor lose capacitance without swelling?
Yes. Internal deterioration can occur without a clear exterior change. A normal-looking case does not replace the appropriate electrical assessment.
Is a fixed percentage loss always the replacement limit?
No. Use product-specific criteria, measurement uncertainty and the coordinated circuit requirements. Do not import a generic threshold from a different capacitor type.
What should a capacitance trend include?
Unit identity, test method, terminal arrangement, instrument, raw results, comparison source and relevant duty. Preserve method changes and replacement history so the trend stays interpretable.