Motor Megger Value: What Reading Is Acceptable?

You meggered a motor. The display says 12 MΩ. Now what?

Almost everyone in the field carries the same number in their head: one megohm. Anything above 1 MΩ, energise it. That rule is older than most of the motors it gets applied to, and for a modern machine it is far too generous. IEEE 43-2013 — the standard that actually governs this test on rotating machines — sets the floor at 5 MΩ for a small random-wound motor and 100 MΩ for a modern medium-voltage form-wound winding.

This page is the answer key. What value you should get, why the 1 MΩ habit persists, and how to correct your reading before you compare it to anything.

For the test procedure itself, see the megger a motor guide. This page is about the number, not the method.

The short answer

Your motor	Minimum megger value (IR₁ at 40 °C)
Random-wound stator, form-wound below 1 kV, DC armature — most LV motors	5 MΩ
AC form-wound winding built after roughly 1970 (epoxy-mica)	100 MΩ
Windings built before roughly 1970, and all field windings	kV + 1 MΩ

Three conditions apply to every one of those numbers, and skipping them is how people arrive at wrong answers:

1. The reading is taken 60 seconds after voltage is applied. Not five seconds. Not when the needle looks settled.
2. The reading is corrected to 40 °C before you compare it. A raw number off the display is not a megger value in the sense the standard means.
3. kV is the rated line-to-line voltage, in kilovolts, for a three-phase machine.

That is the whole acceptance test. The rest of this page is why each line reads the way it does.

Where the 1 MΩ rule comes from — and why it fails now

The kV + 1 formula is the ancestor. A 400 V motor gives 0.4 + 1 = 1.4 MΩ. Round it down, pass it around a workshop for forty years, and it becomes “one megohm.”

It was calibrated for asphaltic-mica and shellac mica-folium insulation — the thermoplastic systems used until roughly the early 1960s. Those materials absorb moisture readily and have genuinely lower bulk resistivity. On that class of insulation, single-digit megohms was a defensible pass.

Modern insulation is not that material. Epoxy and polyester systems have essentially infinite bulk resistivity when clean and dry. A healthy modern form-wound winding reads in the hundreds of megohms, often into gigohms. If you apply kV + 1 to a 4160 V epoxy-mica motor, you get 5.16 MΩ — a threshold roughly twenty times below where IEEE 43 actually puts it. A winding at 20 MΩ would sail through your acceptance check while being soaked in moisture.

That is the whole point of the 100 MΩ row. It is not a stricter version of the old rule. It is a different rule for a different material, and IEEE 43-2013 made the split explicit.

Where kV + 1 still lives: pre-1970 windings, and all field windings regardless of age.

You cannot read a value at the wrong voltage

The megger value only means something if the machine was tested at the right voltage. Too low and you fail to stress the insulation. Too high and you can puncture a winding that was merely damp.

Winding rated voltage	DC test voltage
Below 1000 V	500 V
1000–2500 V	500–1000 V
2501–5000 V	1000–2500 V
5001–12 000 V	2500–5000 V
Above 12 000 V	5000–10 000 V

Two details from the standard that field guides usually drop:

• Test with negative polarity DC. On older, moist thermoplastic windings, reversing the leads can change the reading — the electroendosmosis effect. Negative polarity is the reference condition.
• If the endwinding carries a stress control coating, the grading material distorts the current at lower voltages and drags the apparent resistance down. On those machines, 5 kV and above gives the more trustworthy value.

And record the voltage you used. Two readings taken at different voltages are not comparable, even after temperature correction.

Temperature: your display is not your value

Insulation resistance falls exponentially as the winding warms. A motor pulled off load at 60 °C will read a fraction of what the same motor reads cold. Nothing changed in the insulation. Only the temperature did.

IEEE 43 fixes the reference at 40 °C:

R₄₀ = K_T × R_measured

K_T depends on which insulation family you have, and the two families behave very differently:

Winding temp (°C)	K_T thermoplastic (asphaltic, pre-1960s)	K_T thermosetting (epoxy/polyester)
10	0.125	0.7
20	0.25	0.8
30	0.5	0.9
40	1.0	1.0
50	2.0	1.5
60	4.0	2.3
70	8.0	3.3
80	16.0	4.6

Use the wrong column and you will over-correct badly. At 60 °C the thermoplastic factor is 4.0 and the thermosetting factor is 2.3 — nearly a 2:1 error on the corrected value.

One boundary the standard is firm about: below the dew point, correction breaks down. The factors were derived on clean dry bars, not on windings with condensation forming on them. If the winding is below dew point, lean on the machine’s own history under similar conditions rather than trusting a corrected absolute number.

Three worked examples

A 400 V random-wound pump motor. Measured 12 MΩ at 25 °C, epoxy varnish. K_T ≈ 0.85 (between the 20 °C and 30 °C thermosetting rows). Corrected: 12 × 0.85 = 10.2 MΩ. Minimum is 5 MΩ. It passes — but the raw 12 MΩ flattered it, and the margin is smaller than it looked.

A 6.6 kV form-wound motor, epoxy-mica, installed 2015. Measured 300 MΩ at 60 °C. K_T = 2.3. Corrected: 300 × 2.3 = 690 MΩ. Minimum is 100 MΩ. Comfortable pass — and note the hot reading of 300 MΩ, judged raw, would have looked far closer to the line than it really was.

A 3.3 kV motor rewound in 1968, asphaltic-mica. Measured 6 MΩ at 30 °C. K_T = 0.5 (thermoplastic). Corrected: 6 × 0.5 = 3.0 MΩ. Minimum is kV + 1 = 4.3 MΩ. It fails. The correction pushed it below the line. Anyone reading the raw display would have energised it.

That third case is the reason temperature correction is not optional paperwork.

The second number: polarization index

A single megger value tells you the level. It does not tell you the trend inside the test. Run for ten minutes, take the reading at one minute and at ten, and divide:

PI = IR₁₀ ÷ IR₁

Insulation thermal class	Minimum PI
Class 105 (A)	1.5
Class 130 (B) and above	2.0

Most industrial motors are Class B or F, so 2.0 is the number that matters in practice.

The PI is close to temperature-independent — both readings sit at nearly the same winding temperature, so K_T cancels in the ratio. That is its main advantage over a bare IR reading.

IEEE 43 requires both to clear their minimums. Good IR with a PI near 1.0 means surface leakage is swamping the absorption current — usually moisture or contamination. Good PI with an IR below the floor still fails.

One trap worth knowing: if IR₁ is above 5000 MΩ, the current involved drops into the sub-microamp range, and supply ripple, humidity and lead condition start dominating the ratio. At that point IEEE 43 says the PI may be meaningless. A motor reading 8 GΩ with a PI of 0.9 is a healthy motor with a noisy ratio. Trust the absolute value. More on the split in DAR vs PI.

Reading the result

Corrected IR₁	PI	What it usually means	Action
Above minimum	Above minimum	Clean, dry insulation	Return to service; log it
Below minimum	Near 1.0	Moisture or surface contamination	Clean and dry, retest — do not hipot it
Below minimum	Above minimum	Surface condition dragging it down	Clean, dry, retest
Below minimum after drying	Low	Bulk degradation, cracking, deep contamination	Investigate: PF, PD, visual — not a return-to-service
Very high, above 5000 MΩ	Any value	PI is unreliable at this level	Judge on the absolute value
High but rising year over year on an old asphaltic winding	Above 8	Bonding materials decomposing; dry and brittle	Do not hipot; treat as elevated risk

That last row surprises people. On varnished cambric, shellac or asphaltic windings, a very high PI is not a gold star. It can mean the insulation has thermally aged, lost its bonding, and is now dry and brittle — high resistance, low mechanical integrity. Tap it. If it sounds hollow, do not clean it and do not overvoltage-test it.

What a good megger value cannot tell you

The number has a ceiling on what it can prove, and IEEE 43 is direct about it.

• It is not dielectric strength. No megger value predicts the voltage at which the winding breaks down.
• It cannot locate a defect. One number covers the whole winding under test.
• It cannot see internal voids. This is the big one on modern form-wound coils. Mica blocks DC almost completely — a single layer is enough. Delamination, voids from poor impregnation, thermal cycling damage: all invisible to a megger. Only a crack forming a continuous conductive path to ground will show. Finding internal defects needs AC methods — power factor, tip-up, partial discharge.
• It cannot see anything that needs rotation. Loose coils, endwinding movement, slot vibration. The machine is standing still.

A perfect megger value on a modern motor means the surfaces are clean and dry. It does not mean the groundwall is sound.

Before you walk away

Discharge the winding. The minimum discharge time is four times the voltage application time — after a 10-minute PI test, that is 40 minutes of grounding. Large windings hold a genuine shock hazard long after the leads come off. The test is not finished until discharge current is essentially zero and return voltage is under about 20 V.

And write it down: winding temperature, ambient, humidity, dew point, test voltage, connection arrangement, IR₁, the corrected value, and the PI. A single megger value is a snapshot. Ten years of corrected values under consistent conditions is the only version of this test that actually predicts failure.

FAQ