Megger Test Methods: IR, PI, DAR, Step Voltage, and Ramp Testing

Insulation resistance testing is one of the most portable, non-destructive diagnostics a field crew can run on substation equipment. It requires no energization, no specialized interface to the equipment under test, and can be completed in minutes to an hour depending on the test method. The limitation is interpretation: a raw megohm reading without context is nearly meaningless. Equipment size, temperature, humidity, test voltage, and test duration all affect the number. The test methods described here — IR, PI, DAR, step voltage, and ramp — address those limitations in different ways. Knowing which method fits the situation is as important as running the test.

Southern Switch crews use the Megger MIT 525, a 5 kV insulation resistance tester that supports all five test modes across test voltages of 50 V, 100 V, 250 V, 500 V, 1000 V, 2500 V, and 5000 V DC. The instrument is used across our transformer field testing, circuit breaker inspection, and maintenance testing programs.

Test voltage selection

Before running any test mode, voltage must be matched to the equipment's rated system voltage. Applying too high a test voltage to low-voltage equipment can damage healthy insulation; applying too low a voltage to high-voltage equipment produces results that don't reflect the stress the insulation actually experiences in service. IEEE 43 and NETA MTS provide voltage selection tables. The general rule for rotating machines per IEEE 43 is 500 V DC for equipment rated below 1 kV, 1000 V DC for 1–2.5 kV equipment, 2500 V DC for 2.5–5 kV, and 5000 V DC for equipment rated 5–12 kV. Power transformers are typically tested at 2500 V or 5000 V DC depending on voltage class. Cable systems and circuit breaker insulation follow similar guidelines. The MIT 525's 5 kV ceiling covers the vast majority of distribution-class equipment a substation crew encounters.

Insulation resistance (IR) — the baseline test

The standard IR test applies a fixed DC voltage for a fixed duration — most commonly 1 minute — and records the resistance in megohms or gigohms. It is the fastest test mode and the starting point for every insulation evaluation. The result is compared against a minimum threshold (IEEE 43 sets 100 MΩ minimum for rotating machines as a rough floor, though most utilities apply higher standards) and against the equipment's prior test history. A value that has dropped 50% or more from the last measurement warrants attention even if it is still above the minimum threshold.

The primary limitation is sensitivity to temperature and equipment size. A large oil-filled transformer at 10°C will read substantially higher than the same transformer at 40°C, with identical insulation condition. Single-reading IR tests must always be accompanied by the insulation temperature at test time so that values can be temperature-corrected to a standard reference (20°C for power transformers per IEEE C57.12.90) before trending. Without temperature correction, year-over-year comparisons are unreliable.

Polarization index (PI) — 10-minute trending test

The PI is the ratio of the 10-minute IR reading to the 1-minute IR reading under continuous applied voltage. Because both values are taken at the same temperature on the same equipment, the ratio cancels size and temperature effects — making it a more reliable condition indicator than either reading alone. Good, dry insulation produces a PI well above 1.0 as absorption current decays; wet or contaminated insulation stays flat, producing a PI near or below 1.0. The test takes 10 minutes of uninterrupted applied voltage with the equipment isolated. For threshold values, temperature correction procedure, and transformer-specific interpretation, see Transformer Insulation Resistance: PI Ratio and DAR.

Dielectric absorption ratio (DAR) — quick screening

The DAR is the ratio of the 60-second IR reading to the 30-second reading. The entire test completes in one minute and serves as a rapid screening tool when time is limited. A DAR above 1.25 is generally acceptable; below 1.0 indicates a problem. Because the time window is short, the DAR is less sensitive to subtle degradation than the PI — a borderline DAR should be followed by a full PI before any maintenance decision. Detailed thresholds and comparison with PI are covered in the PI Ratio and DAR article.

Step voltage test — finding voltage-dependent problems

The step voltage test applies a series of increasing DC voltages — typically four to six steps — and records the IR reading at each level. On good insulation, the resistance remains relatively constant as voltage increases: the insulation behaves ohmmically, with leakage current proportional to applied voltage, and the ratio (resistance) stays flat. On insulation with localized defects — voids, cracks, delamination, or tracking paths — the leakage current increases disproportionately at higher voltages because the defect becomes conductive under higher field stress. The result is a resistance that drops noticeably as voltage steps up.

The step voltage test is particularly useful for cable systems and motor windings where physical damage or aging creates localized weak spots that are invisible to a single-voltage IR or PI test. Standard practice per IEEE 95 is to use four voltage steps, each applied for one minute, with the steps spaced logarithmically across the test voltage range. A ratio of greater than 25% resistance drop between the lowest and highest steps is a flag for localized insulation weakness. The test is run after baseline IR and PI testing, not as a substitute for them.

Ramp test — continuous voltage sweep

The ramp test continuously increases the applied voltage from zero to a set maximum at a controlled rate — typically 500 V/minute or 1000 V/minute — while monitoring current. Rather than discrete steps, the ramp produces a continuous current-versus-voltage curve. In healthy insulation, current increases linearly with voltage (constant resistance). A non-linear current increase — current rising faster than voltage — indicates a breakdown process beginning, a partial discharge site activating, or a conduction path that becomes more conductive as field stress increases.

The ramp test is more sensitive than the step voltage test for detecting early-stage breakdown in cable insulation and is commonly used in medium-voltage cable acceptance testing as a complement to or substitute for step voltage. It is faster than running multiple discrete steps and provides a continuous picture of insulation behavior across the full voltage range. The MIT 525 supports ramp testing and logs the full current-voltage curve for documentation. Interpretation requires comparing the curve shape against the expected linear profile; the test result is the curve, not a single number.

Discharge between tests

Regardless of test mode, the insulation under test must be fully discharged between measurements. Residual charge left from a prior test artificially elevates the apparent resistance in the next reading because the charged insulation opposes the flow of test current. The standard practice is to short-circuit the test leads to the equipment terminals for at least four times the preceding test duration — or a minimum of 10 minutes — before beginning the next test. On large, high-capacitance equipment like power transformers, 30 minutes of discharge time may be required. Skipping discharge is one of the most common sources of misleading insulation resistance results in the field.

Choosing the right test for the job

For a transformer at scheduled maintenance with a test history on file, the standard program is IR at 1 minute (for trending against prior corrected values), followed by PI for a condition ratio. For a transformer at acceptance testing with no prior history, the PI is the primary measure and the single IR reading provides the baseline. For cables at acceptance, step voltage or ramp testing supplements the IR and PI. For rotating machines, IEEE 43 specifies the PI as the primary acceptance criterion. For circuit breaker insulation — contact assembly, bushing, and tank insulation — a single-voltage IR test at 1 minute is standard NETA practice, with values compared against NETA MTS Table 100.1 acceptance criteria.

The ramp and step voltage tests add diagnostic depth when there is a specific concern — suspect insulation condition, a failure history on similar equipment, or acceptance testing on older cable where in-service history is unknown. They are not routine for every job. Applying them indiscriminately adds time without adding proportional information; applying them when there is a real question can catch a defect that a simple IR or PI would miss.