Technical Articles

Creepage and Clearance: Material Group, Pollution Degree and the Groove Rule

2026-11-14

Creepage and clearance are measured on the same drawing, with the same calliper, in the same afternoon, and they answer two different questions. Clearance is the shortest path through air. Creepage is the shortest path along the surface of the insulation. Swap them and a design that looks compliant on one table fails on the other, which is why a measurement dispute usually turns out to be a definition dispute.

This article covers the two definitions, the three inputs that select the limits, the groove rule that catches most first-time measurements, and a measurement routine that survives a second opinion.

The two paths, and why they diverge

Clearance is decided by the dielectric strength of air, so it depends on the electric field, the shape of the electrodes and the air pressure. Creepage is decided by the ability of a contaminated surface to carry a tracking current, so it depends on how wet and how dirty the surface gets and on how resistant the material is to tracking. Air does not care what the housing is made of. The surface does, which is why one table is indexed by material group and the other is not.

A practical consequence: raising the pollution degree moves the creepage requirement and leaves the clearance requirement alone, while moving to a higher altitude moves the clearance requirement and leaves creepage alone. A design change aimed at one often does nothing for the other.

Material group comes from CTI, not from the datasheet headline

IEC 60664-1 sorts solid insulation into four groups by comparative tracking index, measured to IEC 60112:

GroupCTIWhere it usually lands
I600 V and aboveGlass-filled PBT and PA, melamine with a mineral filler
II400 V to below 600 VMineral-filled PA66, glass-filled PET, some PC blends
IIIa175 V to below 400 VUnfilled ABS, PC/ABS, unfilled PA6 and PA66
IIIb100 V to below 175 VGeneral purpose polystyrene, some soft PVC compounds

The groups are not evenly spaced, and the gap between II and IIIa is where most real materials sit. A glass-filled PBT at 425 V is group II; an unfilled ABS at 350 V is group IIIa. The difference between those two groups can be more than a millimetre of creepage at mains voltage, which is more than the wall thickness of a small outlet.

Two traps sit behind that table. The first is that CTI is measured on a specimen, and a specimen is not a moulding: flame retardant that blooms to the surface, release agent, and glass fibre orientation can all move the result by a group. The second is that CTI is quoted at a thickness in the test report, and a 0.8 mm wall may not reproduce the value obtained on a 3 mm plaque.

Pollution degree decides how much surface you have

Pollution degree 1 means no pollution or only dry, non-conductive pollution. Degree 2 means non-conductive pollution that can become conductive under condensation. Degree 3 means conductive pollution or dry non-conductive pollution that becomes conductive because condensation is expected. Degree 4 means pollution that persists, such as conductive dust or rain.

An enclosed outlet is normally degree 2. The same outlet with a ventilation slot that lets shop dust settle on a live part is degree 3, and the creepage requirement at 250 V can jump by several millimetres for that one slot. The rating is a property of the application, not of the product, which is why the same moulding can be compliant in a bedroom and non-compliant on a factory wall.

The groove rule: narrower than X and it does not exist

Not every millimetre of surface counts. A groove narrower than the dimension X is treated as a bridge: the path jumps across it rather than following it down and up. X is 1.0 mm at pollution degree 2 and 1.5 mm at pollution degree 3.

Take a ribbed housing with five ribs of 0.8 mm width between two live parts at pollution degree 2. The ribs add 1.6 mm of surface each, so the drawing shows 8 mm of extra path. The rule says the path crosses each groove at its mouth, so the extra path is zero and the creepage is the straight distance between the two conductors. This is the single measurement that most often reverses a pass in a second-opinion review, and it is invisible to anyone who measures with a calliper and a straight edge.

The complementary rule matters for clearance: an internal air gap narrower than 1 mm at pollution degree 2 is also treated as bridged, because the field does not follow the narrow channel.

How to measure so the number holds up

Use a steel strip about 0.5 mm thick and 2 mm wide, or a piece of shim cut to a known width, and lay it along the surface rather than pressing a calliper jaw onto it. The strip follows the contour without compressing it, and a soft thermoplastic compresses by more than the tolerance you are trying to hold. Where the path runs around a curve, a thin feeler or a piece of solder bent to shape does the job and can then be straightened and measured.

  • Measure the shortest path, not the most accessible one. If two routes exist, both must be measured and the shorter one governs.
  • Record the reading to 0.1 mm and take three measurements at the tightest point, not three at convenient points.
  • Note the ambient temperature and humidity. A nylon part absorbs moisture and grows; a measurement taken in a dry room in winter and one taken in a humid room in summer can differ by more than the drawing tolerance.
  • Mark the measured path on a photograph. A second reviewer can then disagree about a line instead of about a number.

Altitude, and the one correction people forget

Clearance is a function of air density, so the table is written for altitudes up to 2000 m. Above that, the required clearance is multiplied by a correction factor that rises with height, and a product specified for 3000 m needs noticeably more air gap than the same product at sea level. Creepage does not change, because the surface is at the same pressure either way.

Two exceptions are worth knowing. A hermetically sealed enclosure that is sealed at the factory keeps its internal pressure and does not need the altitude correction for internal clearances. And a conformal coating that is qualified as a solid insulation changes the creepage problem entirely, because the surface path is no longer in contact with polluted air.

Five checks before the drawing is signed

  • Group from a test report, not a datasheet. Ask which thickness and which specimen the CTI value came from.
  • Pollution degree stated per cavity. A sealed cavity and a vented one in the same product are different environments.
  • Grooves measured, not estimated. A 0.9 mm groove at degree 2 is a bridge, and 0.1 mm is the whole difference.
  • Clearance checked at the electrode, not the terminal. The tightest air gap is usually between a bent conductor and a screw head, not at the nominal gap in the drawing.
  • Both distances measured on the worst pole. On a three-phase device the worst case is almost never the one that is easiest to reach.

The instruments used for the electrical tests that sit either side of this measurement are on our electrical safety test equipment page; the full range starts at the product index.

Contact Us

Contact: Eason Wang

Phone: +86-13751010017

Tel: +86-755-33168386

Add: 1F Junfeng Building, Gongle, Xixiang, Baoan District, Shenzhen, Guangdong, China

close
Scan the qr codeClose
the qr code