Short answer: DAR (dielectric absorption ratio) and PI (polarization index) are both ratios taken from a timed insulation resistance test with a megohmmeter. DAR is the resistance at 60 seconds divided by the resistance at 30 seconds — a one-minute check. PI is the resistance at 10 minutes divided by the resistance at 1 minute — a ten-minute check. Both work on the same principle: healthy, dry insulation shows resistance rising over time, so the ratio comes out well above 1. Wet or contaminated insulation stays flat, and the ratio sits near 1. DAR is the quick screen; PI is the fuller diagnostic.
Neither number means much without knowing what it’s measuring and when it lies — so here’s how to read both.
Table of Contents
What both ratios actually measure
When you apply DC voltage to insulation, three currents flow:
- Capacitive charging current — large at first, gone in seconds.
- Absorption (polarization) current — molecules in the insulation slowly align to the field; this current decays over minutes.
- Leakage (conduction) current — the steady current through and across the insulation. This is the one that reveals contamination and moisture.
In clean, dry insulation, the absorption current decays smoothly and leakage is tiny. As the fast currents die away, the measured resistance rises — so a later reading is higher than an earlier one, and the ratio of the two exceeds 1.
In wet or dirty insulation, leakage current dominates and stays high. Resistance barely changes over time, so the later reading is about the same as the earlier one, and the ratio sits close to 1.
That’s the whole idea behind both DAR and PI: the shape of the resistance-vs-time curve tells you about moisture and contamination, independent of the raw megohm value.
DAR: the one-minute check
DAR = IR at 60 s ÷ IR at 30 s
You apply the test voltage, read the resistance at 30 seconds, read it again at 60 seconds, and divide. It captures the early, fast part of the absorption decay — which is why it only needs a minute.
Typical interpretation:
| DAR value | Insulation condition |
|---|---|
| < 1.0 | Bad — investigate before energizing |
| 1.0 – 1.25 | Questionable |
| 1.25 – 1.6 | Good |
| > 1.6 | Excellent |
DAR is the go-to when you’re time-limited, or on low-voltage random-wound motors where a full PI often stabilizes so fast that the ten-minute ratio flattens toward 1 even on healthy windings. The short window catches the early decay that PI can miss.
PI: the ten-minute check
PI = IR at 10 min ÷ IR at 1 min
Same idea, longer window. PI watches the full absorption decay, so it’s the more complete picture — the standard choice for form-wound motors, generators, and transformers.
Typical interpretation (IEEE 43 style):
| PI value | Insulation condition |
|---|---|
| < 1.0 | Unacceptable |
| 1.0 – 2.0 | Questionable |
| 2.0 – 4.0 | Good |
| > 4.0 | Excellent |
For a deeper, clause-level treatment of PI on rotating machines, see our IEEE 43 clause-by-clause guide and the dedicated polarization index testing of motor windings page.
DAR vs PI: which to use
They’re not rivals — they’re the same measurement at two time scales. Pick by situation:
| DAR | PI | |
|---|---|---|
| Test time | 1 minute | 10 minutes |
| Reads | Early absorption decay | Full absorption decay |
| Best for | LV / random-wound motors, quick screening | Form-wound machines, transformers, acceptance testing |
| Weakness | Less information than PI | Can flatten to ~1 on very high-IR modern insulation |
A common field workflow: run DAR first as a fast go/no-go. If it’s marginal, or the equipment is critical, run the full PI.
The trap: when the ratio lies
This is where experience separates from rote reading. Both ratios can come out low on perfectly healthy insulation, and panicking over it is a classic mistake.
Modern epoxy-mica insulation has very fast-decaying absorption current. By one minute, most of the decay is already over, so the 10-minute reading isn’t much higher than the 1-minute one — and PI lands near 1 on a clean, dry, excellent winding.
IEEE 43 addresses this directly: when the 1-minute IR (corrected to 40 °C) exceeds 5000 MΩ, PI should not be used as a diagnostic. At that resistance the total current is in the sub-microamp range, where tiny changes in humidity, lead condition, or supply voltage swamp the real absorption signal. The right response to a 1.3 PI on a machine reading 15,000 MΩ is to record the very high IR as a good result and disregard the PI — not to condemn the winding.
The lesson: read the ratio and the raw IR together. A low ratio on very high resistance is fine. A low ratio on low resistance is the real warning.
What counts as a good megger reading?
This is the question behind a lot of field tests — “what should a 480 V motor read?” A few anchors:
The rule of thumb. Historically, minimum acceptable IR ≈ (rated kV + 1) MΩ at 40 °C. For a 480 V motor that’s about 1.5 MΩ — a floor, not a target. IEEE 43’s minimum for a random-wound LV stator is on the order of 5 MΩ at 40 °C.
The reality. A genuinely healthy 480 V motor reads far above the minimum — hundreds of MΩ to several GΩ. A reading of a few MΩ technically clears the old floor but should worry you; it usually means moisture or contamination.
Correct for temperature — always. IR roughly halves for every 10 °C rise. A reading only means something at a known temperature, corrected to a 40 °C reference:
R₄₀ = R_measured × 2^((40 − T) / 10)
For example, a 480 V motor reading 180 MΩ at 25 °C corrects to about 500 MΩ at 40 °C — and its PI works out cleanly around 2.5. This is also why DAR and PI are so useful: both readings are taken at the same winding temperature within minutes, so temperature largely cancels out of the ratio. The ratio is more portable than the raw megohm number.
Match the test voltage to the equipment:
| Equipment rated voltage | DC test voltage |
|---|---|
| < 1 kV | 500 – 1000 V |
| 1 – 2.5 kV | 1000 – 2500 V |
| 2.5 – 5 kV | 2500 V |
| 5 – 12 kV | 5000 V |
FAQ
What is the difference between DAR and PI?
DAR is the ratio of insulation resistance at 60 seconds to 30 seconds — a one-minute test. PI is the ratio at 10 minutes to 1 minute — a ten-minute test. Both measure how much resistance rises over time; DAR is the quick screen, PI the fuller diagnostic.
What is a good DAR value?
Above 1.25 is generally good and above 1.6 excellent. Between 1.0 and 1.25 is questionable, and below 1.0 suggests wet or contaminated insulation that needs investigation.
What is a good PI value?
Above 2 is good and above 4 excellent, with 1 to 2 questionable and below 1 unacceptable — but only when the 1-minute IR is below 5000 MΩ. Above that, IEEE 43 says PI is unreliable and should be disregarded.
What is a good megger reading for a 480 V motor?
The old rule of thumb is a minimum of about 1.5 MΩ (rated kV + 1) at 40 °C, and IEEE 43 puts the minimum around 5 MΩ for a random-wound LV stator. A healthy motor reads far higher — hundreds of MΩ to several GΩ. Always correct to 40 °C.
Why can PI be low on good insulation?
Modern epoxy-mica insulation decays its absorption current quickly, so by one minute the decay is nearly done and PI lands near 1 even on excellent windings. When 1-minute IR is above 5000 MΩ, treat the high IR as the good result and ignore the PI.
Key takeaways
- DAR = 60 s / 30 s (one minute); PI = 10 min / 1 min (ten minutes). Both measure the rise in resistance over time.
- Good insulation gives ratios above 1 because absorption current decays; wet/dirty insulation stays flat near 1.
- DAR suits quick checks and LV motors; PI suits form-wound machines and transformers.
- Read the ratio with the raw IR. A low ratio on very high IR (> 5000 MΩ) is fine per IEEE 43 — a low ratio on low IR is the real problem.
- Always correct to 40 °C — IR halves roughly every 10 °C.