Capacitance Measurement of Power Capacitors: Purpose, Method and Evaluation

What it is

A capacitance measurement determines the real capacitance of a capacitor and compares it with the value marked on its nameplate. The equipment under test is typically a power factor correction capacitor or a bank of such units, a filter or harmonic capacitor, or the capacitor stack of a coupling capacitor voltage transformer, where the stack divides line voltage down to a level the electronics can use. A capacitance meter, an LCR bridge or a power factor test set applies a small alternating signal to the terminals and reports the capacitance. Measurements are taken between each pair of terminals and, where the construction allows, between terminals and the case. Because a capacitor unit is built from many internal elements arranged in series and parallel groups, the capacitance is a direct sign of how many of those elements are still doing their job. The test is non-destructive and quick, and it is paired with insulation tests to give a fuller picture of the unit's health.

Why it is done

Capacitors fail in small steps. An internal element can break down and short out, or an internal fuse can blow and disconnect an element. Each event changes the capacitance of the unit, and in a bank it changes the balance between phases or between series groups, which raises the voltage stress on the remaining elements. The result is often a cascade, in which one failure makes the next more likely, ending in a case rupture, a fire, or a bank that no longer corrects power factor or filters harmonics as intended. On a capacitor voltage transformer, a changed stack capacitance alters the voltage division, so the transformer delivers incorrect readings to meters and protective relays. Measuring capacitance finds these conditions while they are small. At acceptance the test confirms that shipment and installation did no harm and that units match their nameplates. In service it is a periodic check on banks, and it is also used after an alarm or unbalance trip to find the unit responsible.

The principle

Capacitance is set by geometry and by the dielectric between the plates. A capacitor unit is a network of many small capacitors, so the total depends on how many are connected in parallel and how many groups are in series. When an element shorts, a series group loses part of its capacitance and the total of the unit rises, and when an internal fuse removes an element, the total falls. The change per failed element is small but follows a predictable pattern, so a careful measurement can reveal which kind of failure has occurred. The measurement itself is an alternating-current one: the instrument applies a signal of known frequency, measures the resulting current and the phase relationship, and calculates capacitance. Temperature and frequency affect the value somewhat, and stray capacitance in leads and surroundings can distort a measurement of a small unit, which is why guarded connections and consistent setup matter. For capacitor stacks the principle is the same, with each section of the stack compared against the others.

How it is performed

The capacitor is de-energized, isolated from the system and grounded under the owner's procedures, and it is then left alone long enough for its stored charge to bleed off through its internal discharge resistors. Because stored charge can persist, terminals are shorted and grounded again and a test for residual voltage is made before touching them. The nameplate data are noted for later comparison. Bank connections, fuses and bus links that would put other capacitors in parallel with the unit are removed, otherwise the instrument reports the total of everything connected. A calibrated capacitance meter or power factor test set is connected between each pair of accessible terminals, and between terminals and case or ground where the unit design calls for it, using short leads and, where a guard is available, the guard connection. Several readings are taken and compared to confirm they repeat. For a capacitor stack, each section is measured in turn using the manufacturer's procedure. The instrument's calibration, the ambient temperature and the setup are recorded, and the same method is used at later tests so that results are comparable. Afterward the terminals are again shorted and the bank is restored.

What is recorded

The record contains the unit identification and its nameplate data, including the rated capacitance and rating of each unit or section, the measured capacitance for each terminal combination that was tested, the instrument and its calibration, the test frequency, and the temperature of the unit and the surroundings. In a bank, the position of each unit and the phase it belongs to are recorded so that balance can be assessed. The as-found condition is noted, including any bulging of the case, leaking fluid, blown fuses, signs of overheating or evidence of discharge, along with the results of companion tests such as insulation resistance. These details allow a later reviewer to tell a real change in the capacitor from a difference in setup or temperature.

How results are judged

A capacitance reading is judged by comparison, and this page gives no limit. The primary comparison is between the measured capacitance and the nameplate value, with a tolerance defined by the manufacturer and the governing standard; a result beyond that tolerance points to a failed or changed element. The second is comparison between like units, such as the three phases of a bank or the sections of a capacitor stack, where imbalance indicates a developing problem. The third is trending against earlier measurements of the same unit, because a gradual drift matters even when each reading is inside the tolerance. The fourth is correlation with other evidence, such as an unbalance alarm, a bulged case, insulation test results or, for a voltage transformer, ratio and accuracy checks. ANSI/NETA ATS-2025 covers acceptance and ANSI/NETA MTS-2023 covers maintenance testing of capacitors and related equipment, and installation rules such as NFPA 70 govern how capacitors are applied. For the actual criteria, refer to your own licensed copies of the standards and to the manufacturer's data.

Variations

At acceptance the test confirms that each unit matches its nameplate and establishes a baseline. In maintenance it is used to find failed elements and to explain an unbalance alarm or a drop in power factor correction. The equipment class changes the details. For shunt power factor correction banks, the focus is on each unit and on balance between phases. For filter and harmonic capacitors, the capacitance affects the tuned frequency of the filter, so drift can degrade performance. For capacitor voltage transformers, the stack sections are measured individually and the result bears on ratio accuracy, and the work is usually coordinated with ratio and burden checks of the instrument transformer. For internally fused units, a lower reading can mean an element was lost without any outward sign, while for externally fused banks a blown fuse is often visible. The test is combined with insulation resistance or power factor tests to cover both the capacitance and the dielectric condition.