EK-G668A Leakage Current Testing Guide: Baselines, Harmonics & Troubleshooting

Leakage current is one of the earliest measurable signs that an electrical system is losing insulation integrity, accumulating contamination, or developing unintended paths to earth. The challenge is that useful leakage measurements are often far below ordinary load current. A technician may need to distinguish tens of microamps or milliamps while working beside energized conductors, variable-frequency drives, filters, long cables, and electromagnetic interference. A general clamp meter can show that current is flowing, but a high-sensitivity leakage clamp is designed to reveal the small residual current that helps maintenance teams find developing faults before protection trips or equipment fails.

This application guide explains how to plan, perform, and interpret leakage current measurements with the EK-G668A high-accuracy clamp leakage current meter. It covers conductor selection, jaw placement, baseline development, filter and true-RMS use, harmonic interpretation, interference control, trending, and troubleshooting. The method is relevant to industrial plants, commercial buildings, substations, data centers, renewable-energy systems, large motors, cable networks, and maintenance programs that need repeatable non-invasive measurements.

Safety note: Leakage testing is commonly performed on energized systems. Only qualified personnel should work near live conductors. Use appropriate personal protective equipment, approach boundaries, lockout procedures where applicable, and instruments suitable for the installation category and voltage. A leakage clamp does not prove that a conductor is de-energized and must not replace an approved absence-of-voltage procedure.

What Leakage Current Represents

Ideally, all current leaving a source through the intended live conductors returns through the intended circuit. In real equipment, small currents flow through insulation, capacitive coupling, filters, shielding, contamination, protective components, and parasitic paths. When those currents reach protective earth or another unintended return path, they appear as leakage current.

Some leakage is normal. Long cables have capacitance. Motor windings couple electrically to frames. Electromagnetic compatibility filters intentionally connect capacitors to protective earth. Power electronic converters produce common-mode voltage. Surge protective devices and monitoring circuits can also contribute. A single reading therefore cannot be interpreted without understanding the circuit and establishing what is normal for that asset.

Abnormal leakage can result from moisture, dust, conductive deposits, insulation aging, damaged cable jackets, overheated windings, incorrect wiring, deteriorated filters, defective heating elements, or multiple small loads whose combined leakage exceeds a protective-device threshold. The best maintenance decisions come from combining leakage current with insulation resistance, temperature, visual condition, operating state, and historical trend.

Why a High-Accuracy Leakage Clamp Is Different

An ordinary current clamp is optimized for load current. Its lower range, jaw design, magnetic material, noise floor, and shielding may not support stable milliamps in a strong external field. The EK-G668A is designed for high-sensitivity AC leakage measurement from 0.00 mA up to 2000 A with 0.01 mA resolution. Its rectangular 80 mm × 80 mm jaw accommodates grouped conductors and larger installations while permalloy construction and magnetic shielding help reduce the effect of surrounding electromagnetic fields.

The meter uses true-RMS measurement, which is important because leakage waveforms are not always sinusoidal. It also provides a selectable 40–70 Hz filter band, wider 40 Hz–1 kHz measurement when the filter is off, maximum, minimum, and average functions, distortion indication, harmonic analysis through the 21st harmonic above the applicable current threshold, data hold, backlight, spotlight, and Bluetooth storage through the mobile application.

These functions support different questions. True RMS answers “what is the effective total leakage?” The filter helps separate fundamental-frequency leakage from higher-frequency components. Harmonic data helps investigate waveform content. Minimum, maximum, average, and stored records reveal variability that a single snapshot misses.

Residual Measurement Versus Single-Conductor Measurement

The conductor arrangement determines what the clamp measures. If the jaw surrounds one live conductor, the instrument measures the current in that conductor, including load current. This can be useful for direct protective-earth conductor measurements or for verifying a specific leakage path, but it does not automatically isolate residual current.

To measure residual leakage on a single-phase circuit, clamp around both the line and neutral conductors together. Their intended load currents flow in opposite directions and cancel magnetically. Any difference represents current returning by another path. Do not include the protective-earth conductor in that same jaw because doing so can cancel the leakage current that you are trying to detect.

For a three-phase circuit without a neutral, place all three phase conductors inside the jaw. For a three-phase four-wire circuit, include all three phases and the neutral. Again, exclude the protective conductor. The vector sum of the intended currents should approach zero; the remaining signal is residual current.

When the protective-earth conductor is accessible and carries no normal load current, clamping it alone gives a direct view of earth current. This is often easier to interpret, but parallel bonding paths can divide current between conductors. A low reading on one earth conductor does not prove that total leakage is low if current is returning through cable armor, structural steel, piping, communication shields, or multiple bonding conductors.

Pre-Test Planning and Asset Information

Before opening a panel, define the purpose of the test. Are you responding to nuisance RCD trips, comparing similar machines, investigating a suspected insulation fault, establishing a commissioning baseline, or monitoring a critical feeder? The purpose determines where to clamp, how long to record, which operating modes to capture, and what follow-up tests are needed.

Record the asset identifier, circuit, nominal voltage, protective-device type and rating, conductor arrangement, equipment operating state, connected loads, ambient conditions, recent maintenance, and previous leakage results. For variable loads, note production rate, speed, temperature, heater stage, drive frequency, and the number of operating power supplies. Without this context, two technically correct readings may not be comparable.

Inspect the conductor grouping and available jaw clearance. Confirm that every intended current-carrying conductor can be enclosed while the protective conductor remains outside. Avoid forcing the jaw around tightly packed cables or positioning the meter where it can move into hazardous parts. The EK-G668A’s large rectangular jaw is useful for grouped cables, but the measurement is still most repeatable when conductors are centered and the jaw faces close completely.

Step-by-Step Leakage Current Measurement Procedure

1. Inspect and Zero the Test Setup

Check the enclosure, jaw surfaces, hinge, battery, display, and any accessories. Remove magnetic debris and ensure the jaw closes without a gap. Keep the meter away from high-current busbars until the measurement begins. With no conductor enclosed, observe the display and follow the instrument procedure for zeroing or stabilization if required.

2. Select the Correct Conductor Group

For residual current, enclose all intended live conductors of the circuit. For earth-conductor current, enclose only the selected protective conductor. Verify visually that no parallel circuit has been included. If multiple outgoing circuits pass through the jaw, the result represents their combined residual current and cannot identify the faulty branch.

3. Center the Conductors and Close the Jaw

Position the conductors near the center of the magnetic core and confirm the jaw is fully closed. Avoid pressing the jaw against another current-carrying conductor or magnetic surface. Large neighboring currents can influence a low-level measurement, even when the clamp has magnetic shielding.

4. Start With the Fundamental-Frequency Filter

When investigating conventional 50 or 60 Hz insulation leakage, begin with the 40–70 Hz filter enabled. Record the stable true-RMS value. Then disable the filter and compare the wider-band result. If the unfiltered reading is significantly higher, high-frequency common-mode current, switching noise, harmonics, or interference may be contributing.

5. Capture Minimum, Maximum, and Average

Allow the equipment to operate through a representative cycle. Use minimum, maximum, and average functions or Bluetooth logging to capture changes during startup, heating, speed changes, valve operation, or power-supply loading. A short peak may explain intermittent protection trips even when the steady-state value is low.

6. Repeat at Successive Branches

If the feeder result is elevated, move downstream. Measure major branches, then individual loads, while preserving the same conductor-grouping method. This divide-and-locate approach narrows the source without disconnecting every asset. Record each location and avoid comparing residual and direct-earth readings as if they were the same quantity.

Using Filtered and Unfiltered Results

The filter is not simply a way to make an unstable number look smaller. It changes the frequency content included in the result. A filtered reading emphasizes fundamental-frequency components that are commonly associated with conventional insulation leakage. An unfiltered reading includes higher-frequency current up to the specified bandwidth and can reveal the effect of drives, inverters, switching power supplies, and EMC capacitors.

Consider a motor feeder that measures 8.4 mA unfiltered and 1.7 mA with the filter enabled. The difference suggests that most of the measured current is above the fundamental band. That does not prove the motor insulation is healthy, but it changes the investigation. Check drive output cabling, cable length, shield termination, switching frequency, motor common-mode behavior, and filter configuration before concluding that moisture or winding damage is the primary cause.

If filtered and unfiltered values are similar, the leakage is dominated by fundamental-frequency current. Follow-up may focus on insulation condition, connected heating elements, moisture, contamination, or wiring errors. Always compare equivalent operating conditions because switching equipment can change its waveform with load.

Harmonics and Distortion in Leakage Current

Harmonic analysis helps distinguish a nearly sinusoidal leakage waveform from one influenced by nonlinear loads. The EK-G668A can analyze harmonics through the 21st order when the current is above the required threshold. Harmonic percentages should be interpreted together with the total RMS value; a large percentage of a very small current may be less important than a modest percentage of a high residual current.

Third and other triplen harmonics deserve attention in systems with many single-phase electronic loads because they can accumulate in neutral and common-mode paths. Higher-order components may be associated with switching converters, drive carrier effects, or filter networks. Harmonic presence is diagnostic evidence, not an automatic fault classification.

Record the fundamental, dominant harmonic orders, total distortion indication, filtered RMS, and unfiltered RMS at the same location. Compare with similar equipment and with earlier measurements. A change in harmonic pattern can reveal a changed power supply, failed filter component, modified grounding arrangement, or developing insulation path even if total RMS has not yet crossed an alarm threshold.

Building a Useful Leakage Baseline

A universal “safe leakage current” value is not appropriate for every asset. Product standards, protective devices, circuit design, filters, installation rules, and manufacturer limits differ. Build a baseline for each meaningful asset class and measurement location.

Commissioning is the best time to measure. Record feeder residual current, branch residuals, protective-earth current where accessible, filtered and unfiltered values, operating load, and environmental conditions. For repetitive assets, measure several healthy units and calculate the typical range. Preserve the raw readings rather than only assigning a pass/fail label.

Trend absolute value and rate of change. A stable machine that moves from 2 mA to 3 mA over years may be less urgent than one that rises from 2 mA to 8 mA in a week. Establish alert and action levels using equipment requirements, protective-device margin, measurement uncertainty, and maintenance consequences. Review thresholds when equipment or filters are modified.

Troubleshooting Nuisance RCD or Ground-Fault Trips

Begin upstream and measure the residual current of the protected circuit under normal load. Compare it with the device rating and any manufacturer guidance for continuous leakage margin. Do not assume that a breaker marked with a residual-current rating will trip at exactly that value under every waveform and condition.

Switch loads or branches in a controlled manner and observe the change. Multiple electronic devices may each contribute normal leakage that becomes excessive in aggregate. A single defective load may create a step change. Intermittent trips require maximum capture or logging through the operating events that precede the trip.

For circuits with variable-frequency drives, compare filter-on and filter-off results and review drive leakage recommendations. Long shielded cables and high switching frequency can raise common-mode current. Verify shield and protective-earth termination, but do not disconnect protective conductors to make leakage disappear. Any design change must be evaluated by qualified engineering personnel.

Motors, Cables, Heaters, and Electronic Loads

Motor leakage can change with winding temperature, moisture, drive frequency, cable length, and bearing or grounding arrangements. Measure at consistent speed and load. If residual current rises with temperature, combine the finding with offline insulation resistance, polarization behavior where appropriate, and thermal inspection.

Long cables produce capacitive leakage even when insulation is sound. Compare similar cable types and lengths, and separate cable behavior from connected equipment where an approved outage allows. A sudden change after physical work can indicate jacket damage, moisture ingress, or incorrect shield termination.

Heating elements frequently develop moisture-related leakage after storage or shutdown. Trend during warm-up: leakage may initially be high and fall as moisture is driven out, or it may rise as a damaged element expands. Follow equipment procedures and do not repeatedly energize a suspected unsafe heater merely to obtain a trend.

Switch-mode power supplies and EMC filters can create expected high-frequency earth current. Use filtered/unfiltered comparison and branch measurements. When many devices share one protective device, installation design may need review even though no individual device is defective.

Controlling Measurement Error

Low-current clamp measurements are sensitive to setup. A partially open jaw, off-center conductor, external magnetic field, moving cable, depleted battery, or contaminated core can change the result. Close the jaw gently, repeat the measurement, rotate or reposition the clamp where safe, and compare readings for consistency.

To test susceptibility to an adjacent load conductor, take a no-residual configuration around the outgoing and return conductors and observe whether the indication changes as the clamp position changes. A large position-dependent result suggests external field influence or incomplete cancellation. Increase distance from high-current conductors where possible.

Measurement uncertainty should include the meter specification, resolution, repeatability, conductor position, external field sensitivity, frequency content, and calibration status. When the result is near an action threshold, repeat under controlled conditions and confirm with an appropriate alternate method rather than making a high-consequence decision from one display value.

Bluetooth Logging and Trend Analysis

The EK-G668A supports Bluetooth communication and storage of up to 2000 groups through the application. Logging is especially useful for intermittent loads, nuisance trips, heaters, cyclic machinery, and commissioning comparisons. Select an interval that captures the process without producing an unmanageable file.

Synchronize the meter time with maintenance records. Note operational events such as motor starts, batch changes, rain, cleaning, and drive faults. A chart is most useful when technicians can connect current changes to physical events. Preserve the measurement location and conductor arrangement in the file name or record.

Review maximum, average, rate of change, and time above the internal alert level. Do not focus only on the single highest sample; check whether it is repeatable, physically plausible, and associated with a known event. Maintain raw data so future analysts can apply new thresholds without repeating the test.

When to Use Other Instruments

A leakage clamp is ideal for energized, non-invasive current measurement, but it does not directly measure insulation resistance. If leakage is elevated and an outage is available, use an appropriate insulation resistance tester to evaluate isolated equipment according to the manufacturer’s procedure. High-voltage assets may require PI, DAR, step-voltage, ramp, or dielectric-discharge analysis with a tester such as the EK-C352C 15 kV insulation resistance tester.

Use a power quality analyzer when the investigation requires synchronized three-phase voltage, current, power, harmonics, unbalance, flicker, sag, swell, and trend data. The EK-F523A power quality analyzer supports three voltage and four current channels, harmonic analysis through the 51st order, and long-term records.

Use a conventional clamp multimeter when the primary task is load current, voltage, resistance, capacitance, frequency, or temperature rather than low-level residual current. The EK-G645 clamp multimeter covers AC/DC current up to 1200 A together with common multimeter functions.

Frequently Asked Questions

Should I clamp one conductor or all conductors?

Clamp one conductor to measure its current. Clamp all intended live conductors together to measure residual current. Exclude the protective-earth conductor from a residual measurement. Clamp the protective conductor alone to measure the current in that specific earth path.

Why is the unfiltered reading higher?

The wider bandwidth includes higher-frequency current from drives, switching power supplies, filters, and interference. Compare operating states and harmonic content before deciding whether the difference is normal or abnormal.

Can leakage current replace insulation resistance testing?

No. Leakage current describes energized behavior in a particular configuration. Insulation resistance testing applies a controlled DC test to isolated equipment. The methods complement each other but answer different questions.

What reading should trigger maintenance?

Use equipment standards, manufacturer limits, protective-device margin, and the asset’s baseline. Trend and rate of change are often more useful than a universal number. Escalate sudden changes, persistent growth, or readings that reduce protection margin.

How often should leakage be measured?

Base the interval on asset criticality, environment, prior trend, and failure consequences. Critical or moisture-sensitive systems may justify continuous or monthly monitoring; stable low-risk assets may be checked during scheduled inspections.

Final Field Checklist

• Define the diagnostic question and measurement location.
• Confirm the correct live-conductor grouping and exclude protective earth for residual tests.
• Inspect and close the jaw completely with conductors centered.
• Record filtered and unfiltered true-RMS values.
• Capture minimum, maximum, average, and operating state.
• Use harmonic information to investigate nonlinear or switching sources.
• Move downstream systematically to locate the contributing branch.
• Compare with the same asset baseline and similar healthy equipment.
• Control conductor position, external fields, and environmental effects.
• Use insulation resistance or power quality testing when the question extends beyond leakage current.

The strongest leakage-current program combines a capable instrument with a repeatable measurement location, consistent conductor grouping, frequency-aware interpretation, and historical data. Used this way, the EK-G668A becomes more than a sensitive clamp meter: it supports early fault detection, faster branch isolation, protection-margin assessment, and evidence-based maintenance decisions.