Motor insulation resistance testing checks the health of a winding's insulation by applying a DC voltage from the windings to the grounded frame with a megohmmeter and measuring the tiny current that leaks through: high resistance means healthy insulation, a low or falling number means the winding is breaking down. It is the first, cheapest electrical test in any motor predictive-maintenance program, and it catches the slow killers, moisture and contamination, long before the motor faults to ground.
The test is simple to run and easy to misread. The value is not in one megohm reading but in comparing it to the right limit, correcting it for temperature, and trending it over time. This guide covers what the megohmmeter measures, how the polarization index works, what minimums to compare against from IEEE Std 43 and how to run the test so the numbers mean something.
What does the megohmmeter actually measure?
A megohmmeter, commonly called a megger, is a high-resistance meter with a built-in DC voltage source. Connected from a motor winding to the grounded frame, it forces a steady DC test voltage across the ground-wall insulation and measures the small leakage current that flows. Ohm's law does the rest: resistance equals voltage divided by current. Because good insulation leaks almost nothing, the resistances are enormous, read in megohms (millions of ohms) or gigohms.
The test voltage scales with the motor. IEEE 43 guidance puts it near 500 V DC for windings rated below 1000 V, 500 to 1000 V DC for 1000 to 2500 V windings, and higher for larger machines. Note this is a DC test at modest voltage, it stresses the ground-wall insulation but does not probe turn-to-turn weaknesses the way a surge test does. That is the test's main blind spot, and the reason it lives alongside other checks in a motor test set.
What is the polarization index and how do you read it?
A single resistance reading is a snapshot that swings with temperature and humidity. The polarization index (PI) turns the test into something more diagnostic by watching how resistance changes during the test. PI is the insulation resistance at 10 minutes divided by the resistance at 1 minute.
The physics is the useful part. When you first apply voltage, clean dry insulation keeps absorbing charge, so leakage current keeps falling and resistance keeps climbing over several minutes, the 10-minute reading ends up well above the 1-minute reading, giving a PI comfortably above 1. Wet or contaminated insulation conducts freely, the current does not tail off, resistance flattens almost immediately, and PI lands near 1. IEEE 43 recommends a minimum PI of 2.0 for Class B, F, and H insulation (the vast majority of motors in service) and 1.5 for older Class A.
One important exception: IEEE 43 warns the PI can be meaningless when the insulation is very clean and the 1-minute resistance is already above roughly 5 gigohms. At that point leakage is down in the microamp range, tiny factors swamp the ratio, and a low PI is a measurement artifact, not a fault. When absolute resistance is that high, trust it and ignore the PI. When you cannot spare ten minutes, the dielectric absorption ratio (DAR), the 60-second reading over the 30-second reading, with roughly 1.4 as an acceptable floor, is a quicker screen.
What minimum values should you compare against?
Absolute resistance is judged against IEEE 43 construction-based minimums, corrected to a standard temperature (usually 40 °C), because resistance roughly halves for every 10 °C the winding warms. The floors:
| Winding type | IEEE 43 minimum insulation resistance |
|---|---|
| Most windings built before 1970 | 1 MΩ per kV of rating, plus 1 MΩ |
| Form-wound windings built after 1970 | 100 MΩ |
| Random-wound and form-wound below 1 kV | 5 MΩ |
Treat these as the line below which a motor should not go back into service, not as a goal. A good winding reads in the hundreds or thousands of megohms. The number that should worry you is not a single low reading, it is a motor that measured 2000 MΩ last year, 400 this spring, and 90 today. That slope says moisture or contamination is winning, and it is the whole reason to test on a schedule rather than only after a failure.
Why do temperature and humidity throw off the reading?
Insulation resistance is wildly sensitive to temperature, and ignoring that is the most common way plants misread the test. As a rule of thumb, resistance roughly halves for every 10 °C the winding warms. A motor pulled straight off the line at 70 °C can read a fraction of what the same healthy winding reads cold, so comparing a hot reading to a cold minimum condemns good motors and clearing a cold reading against a hot baseline passes bad ones. Always correct to a standard temperature, 40 °C is the usual reference, before you compare to a limit or to last year's number.
Humidity and surface contamination act on the reading too. A film of moisture, oil, or conductive dust across the winding ends gives leakage current an easy path, dropping resistance and flattening the PI even when the ground-wall insulation underneath is fine. That is useful information, a dirty, damp motor genuinely is at higher risk, but it means a low reading is a prompt to clean, dry, and retest, not an automatic sentence of rewind. Test under repeatable conditions, note the winding temperature every time, and the trend stays honest.
How do you run an insulation resistance test?
- De-energize and lock out. The motor must be off, isolated, and locked out. Confirm zero voltage before you connect anything, this is a test on equipment that is normally at line voltage.
- Discharge and ground the winding. Windings and capacitors can hold a charge. Ground the winding to bleed it off before and, critically, after the test, since the megohmmeter leaves the insulation charged.
- Connect and set the test voltage. Lead from the megohmmeter to the winding, the other side to the grounded frame, with the remaining phases grounded. Pick the DC voltage from the motor's rating per IEEE 43.
- Apply voltage and log the readings. Hold the voltage steady and record resistance at 1 minute and at 10 minutes for a PI, or at 30 and 60 seconds for a DAR.
- Correct for temperature. Normalize the absolute reading to 40 °C before comparing it to the IEEE 43 minimum, or you will condemn a warm-but-healthy winding.
- Compute, compare, and trend. Calculate PI, check the corrected resistance against the floor, and, most important, file the result against the motor so you can watch the trend across tests.
What the numbers say
- Insulation failure is a leading cause of motor loss, and moisture is a prime driver, which is exactly what this test catches early. Condition-based electrical testing pays off in the same range as other predictive work: the U.S. Department of Energy's FEMP O&M guidance, maintained by PNNL, documents 8–12% savings over preventive-only maintenance and an opportunity versus reactive operation that can exceed 30–40% (PNNL, O&M Best Practices: Maintenance Approaches).
- The published minimums and the PI thresholds themselves come from a maintained standard, not folklore: IEEE Std 43, Recommended Practice for Testing Insulation Resistance of Electric Machinery which is the reference to cite when someone asks where the 2.0 PI or the 100 MΩ floor comes from.
Where does insulation testing fit with the other motor tests?
Insulation resistance and PI testing own one failure mode extremely well: ground-wall insulation breaking down from moisture, dirt, and age. They will not reliably find turn-to-turn shorts or weak spots that only fail under AC or high-voltage stress, which is the job of surge testing. And they say nothing about the rotor, that story belongs to motor current signature analysis which reads broken rotor bars from the current spectrum. A complete electrical condition-based maintenance program layers these tests so each covers the others' blind spots, which is the broader argument in our predictive maintenance and equipment reliability guides, and the failure modes are cataloged in electric motor maintenance.
As with every predictive test, the payoff is in the trend, and the trend only exists if the readings are captured against the right asset and kept over time. Plants that log test results into one operational layer next to production and downtime data, the way Harmony pulls floor information together, can see a motor's insulation sliding toward the floor months out and schedule the rewind on their terms. For how one plant built that data foundation, see the CLS case study.