An insulation resistance reading of several hundred megohms looks reassuring on a test sheet, and in many cases it deserves that trust. Still, the number describes the insulation at one moment, under one voltage, and at one temperature.
Some forms of deterioration change how insulation behaves under test long before they change the headline value. Insulation resistance testing becomes more informative when attention shifts from one figure to its behavior over time, across voltages, and after the source is disconnected. This article walks through those additional views and the questions each of them helps answer.
Why One Resistance Value Tells Only Part of the Story
A spot insulation resistance test is a common starting point in maintenance work, and its simplicity is its main strength. The three points below show where that simplicity limits what the number can say.
A Spot Reading Reflects One Moment Under One Set of Test Conditions
A spot reading captures insulation resistance at a single point, usually after 60 seconds. The value depends on the circumstances of that moment as much as on the insulation itself.
Several factors shape the result:
- Test voltage — sets the electrical stress the insulation receives.
- Temperature — reflects how warm the winding or cable was during the test.
- Humidity — can open additional leakage paths across surfaces.
- Connected components — cables and accessories left in the circuit add resistance in parallel.
Insulation Behavior Can Change While the Test Voltage Remains Applied
Insulation does not settle the instant voltage is applied, and its resistance reading usually keeps rising for several minutes as the material charges and polarizes. Readings taken at 30 seconds, 1 minute, and 10 minutes on the same winding can therefore produce three different values, each correct for its own moment.
Similar Spot Readings Can Hide Different Resistance-Time Behavior
Two assets can show the same one-minute value while their insulation is in different condition. Consider two motors that each read 800 MΩ at 60 seconds:
- In the first, resistance continues to climb and reaches several gigaohms by the 10-minute mark.
- In the second, the value barely moves from its one-minute level.
The rising pattern is typical of clean, dry insulation. The flat curve suggests that steady leakage dominates, which often points to moisture or contamination.
What the Resistance-Time Curve Can Reveal
Where a spot test gives one point, the resistance-time curve shows the path the insulation takes to reach it. Reading that path starts with the currents that produce it.
Charging, Absorption, and Leakage Components Shape the Early and Later Readings
The current measured during a DC insulation test is the sum of several components, and each fades at its own rate:
- Capacitive charging current — large at first, then close to zero within seconds.
- Absorption current — decays over several minutes as molecules in the dielectric align with the field.
- Conduction current — a steady flow through the bulk of the insulation.
- Surface leakage current — a steady flow across damp or contaminated surfaces.
In the first seconds, charging and absorption dominate, so the reading starts low. As they decline, the steady components take over and the curve levels off.
DAR and PI Compare Resistance at Defined Times During the Test
Both ratios condense the curve into one figure by dividing a later reading by an earlier one:
- The dielectric absorption ratio (DAR) divides the 60-second reading by the 30-second reading.
- The polarization index (PI) divides the 10-minute reading by the 1-minute reading.
Because each ratio compares the insulation with itself, it depends less on the size of the asset than an absolute value does. A frequently cited scale rates a PI below 1 as poor, 1 to 2 as questionable, 2 to 4 as acceptable, and above 4 as good. DAR offers a quicker indication when a 10-minute test does not fit the outage window.
Timed Ratios Still Need Context From the Asset and Insulation System
A ratio can look healthy or weak for reasons unrelated to deterioration:
- High absolute resistance — IEEE Std 43 notes that when the 1-minute value exceeds 5,000 MΩ, the PI may not reliably indicate condition.
- Insulation material — absorption behavior differs between insulation systems, so a ratio typical of one may be unusual for another.
- Asset history — a sudden PI rise of more than 20% without maintenance is a warning sign that earlier records make visible.
Why Insulation May Respond Differently as Test Voltage Increases
Voltage is the second dimension of insulation behavior. Raising the stress in a controlled way shows whether resistance holds as the electric field grows.
Step-Voltage Testing Adds Electrical Stress in Controlled Increments
A step voltage test applies a series of rising DC voltages and records insulation resistance at the end of each stage. A widely used procedure divides the range into five equal steps and holds each one for one minute. Equal durations keep absorption effects similar from stage to stage, so differences between readings reflect the voltage rather than timing.
A Significant Change in Resistance as Test Voltage Increases Can Reveal Behavior Hidden at a Single Voltage
Sound, clean insulation tends to show similar resistance across the steps. A noticeable drop as voltage rises suggests that weak spots begin to conduct under the stronger field. Typical causes include:
- Moisture or contamination within or on the insulation.
- Cracks, pinholes, or other physical damage.
- Aged, brittle material that loses strength under stress.
Manufacturer guidance treats a resistance deviation of about 25% between successive voltage levels as a meaningful indication of such defects. At a lower voltage, the field may be too weak to drive measurable current through them.
The Meaning of the Response Depends on the Equipment and Test Conditions
A falling resistance points to the insulation when the test itself is consistent. Before drawing conclusions, confirm that:
- The maximum voltage suits the asset’s rating and the applicable standard.
- Each step lasted the same time.
- Residual charge from an earlier test was discharged.
- Surface leakage was controlled through the guard terminal.
Instrument capability also shapes what can be tested. The AEMC 5070 megohmmeter, for instance, supports up to five programmable steps with configurable voltage and duration, test voltages up to 5,000 V, and an on-screen plot of results. As a graphical insulation tester, it displays the step response on site.
What Happens After the Test Voltage Is Removed
Switching off the test voltage does not end the useful part of the test. The discharge that follows carries information of its own.
The Discharge Period Contains Information About Stored Polarization
Once the source is removed, the insulation releases its accumulated energy. The capacitive charge drains within seconds, while the charge stored through polarization returns more slowly, over minutes. The rate of that slower release reflects how the dielectric polarized, which a reading under voltage does not show.
Dielectric Discharge Can Add Insight Into Heterogeneous or Multilayer Insulation
The dielectric discharge test (DD) suits insulation built from several layers, such as the winding insulation of rotating machines. The procedure has three steps:
- Charge the insulation at test voltage, typically for 10 to 30 minutes, until absorption is complete.
- Discharge it and measure the remaining current after 60 seconds.
- Divide that current in nanoamperes by the product of test voltage in volts and capacitance in microfarads.
When one layer of multilayer insulation is damaged or contaminated, its leakage resistance falls while its capacitance changes little. The healthy layers dominate a standard measurement, so the defect may not appear.
During discharge, the faulty layer’s time constant no longer matches its neighbors, and the DD value rises. AEMC rates results below 2 as acceptable, 2 to 4 as questionable, 4 to 7 as poor, and above 7 as bad.
DD Does Not Replace IR, DAR, or PI With a Single Better Result
DD addresses a narrower question: whether the layers of an insulation system release their charge in balance. It says little about overall leakage or the response to rising stress.
For homogeneous insulation, AEMC lists a typical DD value of 0, so the test adds little there. It also requires an accurate capacitance value and a long charging period. DD works as a complement to the other methods rather than a replacement.
Why the Most Useful Answer May Come From the Pattern, Not One Test
Each method covered so far answers one question well. The pattern across methods, conditions, and years is what turns those answers into insulation diagnostics.
Temperature and Test Conditions Can Change the Apparent Trend
An insulation resistance trend can drift for reasons unrelated to insulation health. Temperature is a major factor: as a rule of thumb, resistance halves with each 10 °C increase, so 100 GΩ at 20 °C becomes about 25 GΩ at 40 °C.
To keep results comparable:
- Correct readings to a common base temperature; IEEE Std 43 uses 40 °C for rotating machines.
- Test above the dew point to avoid surface condensation.
- Repeat the same voltage, duration, and connection method.
Graphs Make Changes Across Time and Voltage Easier to Compare
A plotted result shows shape, and shape often holds the diagnostic information. Three graphs carry most of it:
- Resistance against time — whether the curve keeps rising or flattens early.
- Resistance against voltage — whether the value holds or falls.
- Corrected readings across years — the rate of change.
A gradual decline over several inspections can warrant attention even when each value passes its acceptance limit.
Different Tests Can Support Different Parts of the Same Insulation Assessment
Methods work best as a planned set matched to the purpose of the inspection:
- Routine maintenance — a spot reading and PI, trended over time, form a common basis for scheduled checks.
- A flattening curve or falling trend — a step voltage test helps show whether the weakness depends on voltage.
- Machines with layered insulation — a DD test adds a layer-level view when other results look normal.
The final choice also depends on the asset’s criticality and the outage time available.
An Unusual Result Is a Reason for Context and Further Investigation, Not an Automatic Diagnosis
An unexpected value signals a departure from the expected pattern without naming its cause. A practical sequence for working through it:
- Verify the setup and repeat the measurement once to check stability.
- Compare the value with earlier results and with records of repairs or cleaning.
- Run a complementary test that addresses a different question.
- Inspect the asset physically if the evidence still points to a defect.
A result that holds up after these steps justifies targeted cleaning, drying, or repair.
From One Number to a Fuller Picture of Insulation Health
A single insulation resistance value remains a useful checkpoint. Its limits come from what it leaves out: how the insulation changes over time, under rising stress, after the voltage is removed, and from one year to the next.
Reliable assessment depends less on one decisive test and more on a consistent record. When methods match the asset and are repeated under comparable conditions, each result strengthens a picture that supports maintenance decisions with evidence instead of assumption.
