Comparative Tracking Index (CTI): IEC 60112 Test Method Explained

Insulating material does not have to break down through its bulk to fail. It can fail across its face. A thin film of dirty water, a few hundred volts, and enough time will burn a conductive path into a plastic surface that measured hundreds of megohms when it was clean.

That path is called a track. IEC 60112 is the test that tells you how well a material resists forming one.

The output is a number: the comparative tracking index, or CTI. It shows up on material datasheets, in UL yellow cards, and in the creepage tables of every low-voltage equipment standard you have ever worked to. This page covers what the number means, how the test produces it, and where it stops being useful.

The failure mechanism first

Tracking needs three things at once. Take one away and it does not happen.

An electrolyte on the surface. Pure water does almost nothing. Contaminated water conducts. Salt spray, condensation carrying airborne dust, hydraulic mist, agricultural ammonia — anything that puts ions on the surface.

Electric stress across that surface. Two conductive parts at different potential, separated by insulation.

Time and repetition. A single wetting event rarely does damage. Repeated wet-dry cycling does.

The sequence runs like this. The contaminated film carries leakage current between the two conductors. Current heats the film. Heat dries the film unevenly, and a dry band forms somewhere along the path. All the voltage now drops across that narrow dry band. It breaks down. You get a small surface discharge — a scintillation.

The scintillation is what does the damage. It pyrolyzes the polymer underneath it. Organic materials that carbonize leave conductive residue behind. That residue extends the effective electrode a fraction of a millimeter, so the next dry band forms slightly further along, and the next scintillation lands there.

The path grows. Once it bridges the gap, you have a short circuit through solid insulation that used to be intact.

Two things follow from this that matter for material selection:

• Materials that carbonize track badly. Materials that erode or volatilize cleanly, without leaving carbon, track well. This is why unfilled PTFE and silicone perform very differently from phenolic laminate.
• Inorganic fillers change the answer. Glass and mineral fillers can raise or lower tracking performance depending on how they sit at the surface, so you cannot predict CTI from base resin chemistry alone.

CTI and PTI are not the same number

Both come out of the same apparatus. They answer different questions.

Proof tracking index (PTI) is a pass/fail check at one voltage you specify in advance. Five specimens, 50 drops, one voltage. They pass or they do not. PTI is an acceptance criterion — incoming inspection, quality control on molded parts, verifying that a supplier’s batch still behaves like the qualified batch.

Comparative tracking index (CTI) is the maximum voltage the material survives. You search for it by testing at progressively different voltages until you find the ceiling. CTI is a characterization value — you use it to compare materials, and to place a material into a group for creepage design.

	PTI	CTI
Question asked	Does it pass at this voltage?	What is the highest voltage it passes?
Voltage	Single, specified beforehand	Found by iteration
Drops	50	100 first, then 50
Typical use	Batch acceptance, QC	Material characterization, creepage design
Reported as	Pass/Fail PTI 175	CTI 175

If a datasheet quotes “PTI 250,” the material was proven at 250 V. It may or may not survive 275 V — nobody checked. If it quotes “CTI 250,” 250 V is the ceiling and 275 V failed.

What the apparatus actually does

The geometry is deliberately harsh and deliberately simple.

Two platinum electrodes sit on the flat top face of a specimen, tips 4 mm apart, pressing down with about the weight of a 100 g mass each. AC voltage between 100 V and 600 V is applied across them. A dropper releases a drop of electrolyte roughly every 30 seconds onto the surface between the tips. The whole thing sits inside a fume-vented enclosure.

Then you wait. A full test runs under an hour.

Fixed test conditions

Parameter	Value
Electrode material	Platinum, minimum 99 % purity
Electrode cross-section	5 mm × 2 mm
Chisel angle	30° ± 2°
Flattened tip width	0.01 mm to 0.1 mm
Angle between electrodes	60° ± 5°
Gap at surface	4.0 mm ± 0.1 mm
Contact force, each electrode	1.00 N ± 0.05 N
Test voltage range	100 V to 600 V, integer multiples of 25 V
Frequency	48 Hz to 62 Hz
Source rating	≥ 0.6 kVA
Short-circuit current setting	1.0 A ± 0.1 A
Over-current trip	0.50 A sustained for 2.00 s
Drop interval	30 s ± 5 s
Drop height	35 mm ± 5 mm
Drop mass	≈ 20 mg
Time for 50 drops	24.5 min ± 2 min
Specimen thickness	≥ 3 mm (stacking allowed)
Recommended specimen area	≥ 20 mm × 20 mm
Glass support thickness	≥ 4 mm
Conditioning	24 h at 23 °C ± 5 K, 50 % ± 10 % RH
Ambient during test	23 °C ± 5 °C

Platinum is used because it is inert. It barely reacts with the electrolyte or with the pyrolysis products, so the material under test stays the dominant variable. Brass, copper, steel, silver and gold electrodes are all used in practice for combination testing — insulating material plus the actual electrode metal in the product — but those results are not CTI or PTI values. They cannot be quoted as such.

The two electrolytes

	Solution A	Solution B
Ammonium chloride	≈ 0.1 % by mass	≈ 0.1 % by mass
Wetting agent	none	0.5 % sodium di-butyl naphthalene sulfonate
Resistivity at 23 °C	3.95 Ω·m ± 0.05	1.98 Ω·m ± 0.05
Base water conductivity	≤ 1 mS/m	≤ 1 mS/m
Marking on result	none	suffix M

Solution A is the default. Solution B is more aggressive on two counts: it is roughly twice as conductive, and the surfactant makes it wet hydrophobic surfaces instead of beading up on them.

That second point matters more than the conductivity. Silicones, fluoropolymers and some filled thermoplastics score very well with Solution A simply because the drop never spreads. Solution B removes that advantage. A large gap between a material’s A and B results is a signal that its performance depends on surface wettability, which is exactly the property that degrades in service as the surface weathers or picks up detergent residue.

Always read the suffix. CTI 400 and CTI 400 M describe very different materials.

How a test ends

Three outcomes, and one of them is not a result at all.

Tracking failure. Current across the surface reaches 0.5 A and holds for 2 s. The over-current device trips. The specimen failed at that voltage.

Persistent flame. The specimen ignites and burns for more than 2 seconds. Also a failure. Brief flashes are tolerated, but a material that never flames is preferred regardless.

Withstand. At least 25 s pass after the last drop with no trip and no persistent flame. The specimen passed.

Air arc — invalid. Sometimes the arc jumps between the electrode tips through the air rather than tracking across the surface. The over-current device trips, but nothing happened to the material. This is not a failure. Clean the apparatus, take a fresh site, repeat at the same voltage.

Air arcing is the practical ceiling of the method. Some materials go straight from clean withstand at one voltage to persistent air arcing at the next step up. For those, no valid failure can be produced, and a CTI simply cannot be determined. Report it as such rather than quoting the last passing voltage as if it were a ceiling.

Note the asymmetry: for a PTI test, an air-arc trip does not count as a tracking failure. For a CTI determination, it invalidates the test entirely.

Finding the CTI value

The current edition runs the 100-drop determination first. Earlier editions implied the opposite order, and running 100 drops first is cheaper, so a lot of older lab procedures are still backwards on this point.

Step 1 — Find the 100-drop withstand voltage. If the material is unknown, start at 350 V. Test to 100 drops. Move up or down in 25 V steps depending on the outcome. Above 400 V, limit increases to 50 V per step. The result is the highest voltage at which five consecutive specimens survive 100 drops.

Step 2 — Find the 50-drop withstand voltage. Infer a starting point from the 100-drop data. Same iteration, 50 drops this time. The result is the highest voltage at which five consecutive specimens survive 50 drops.

The CTI is the 50-drop figure. The 100-drop figure is carried along as a qualifier.

Five specimens, or five separate sites on one plaque, provided the sites are far enough apart that splash and fumes from one do not contaminate the next.

Reading the notation

Reported as	Meaning
CTI 175	50-drop ceiling 175 V, Solution A, 100-drop result within 25 V
CTI 175 M	Same, tested with Solution B
CTI 400(350)	50-drop ceiling 400 V, but 100-drop ceiling only 350 V
CTI 400(350) M	Same, Solution B
CTI 275 – 1,2	CTI 275 V with maximum erosion depth 1.2 mm
CTI 400(350) M – 3,4	All of the above combined

The parenthetical value only appears when the 100-drop result falls more than 25 V below the 50-drop result. When you see it, treat it as the real number for anything involving prolonged or repeated wetting. A wide spread means the material holds up to a short exposure but degrades under a sustained one — which is the condition most outdoor and washdown equipment actually sees.

Erosion is a separate measurement

Tracking is not the only way this test damages a specimen. Some materials never form a conductive path but get eaten away instead. The scintillations volatilize material, the surface hollows out, and the electrodes sink into the pit.

When erosion is required, specimens that survived the 50-drop test are cleaned of loose debris and measured with a depth gauge — 1.0 mm nominal diameter probe, hemispherical tip, read to 0.1 mm. Take the deepest of the five. Anything under 1 mm is simply reported as < 1 mm.

Two related observations get recorded whether or not erosion is being measured:

• If a hole forms all the way through the specimen, report it along with the depth — which is the specimen or stack thickness. The test result still stands. Retests on thicker specimens, up to 10 mm, can be run for more information.
• If the specimen flamed badly enough that erosion cannot be measured, say so.

A material with a high CTI and 3 mm of erosion is not a good material. It is a material that failed in a different way than the one the headline number tracks.

Turning CTI into creepage distance

This is where the number earns its keep. Low-voltage equipment standards do not ask for CTI directly. They ask which material group a part belongs to, and the group comes from the CTI value measured with Solution A.

Material group	CTI range
I	CTI ≥ 600
II	400 ≤ CTI < 600
IIIa	175 ≤ CTI < 400
IIIb	100 ≤ CTI < 175

The group, the working voltage, and the pollution degree of the environment together set the minimum creepage distance. A Group I material in a clean enclosure needs far less surface distance than a Group IIIb material in a polluted one. On a crowded PCB or a compact terminal block, moving from Group IIIa to Group I can be the difference between a design that fits and one that does not.

North American work uses a parallel scheme, the performance level categories of UL 746A:

PLC	CTI range
0	≥ 600
1	400 – 599
2	250 – 399
3	175 – 249
4	100 – 174
5	< 100

Two cautions before you lean on any of this.

CTI is not a creepage distance. The test result cannot be used directly to calculate a safe creepage distance. It feeds a table. The table was built with margin that the raw test number does not contain.

The test is an indoor-severity test. It discriminates between poor, moderate and good tracking resistance for equipment that may operate under moist conditions. It does not qualify material for outdoor service. Outdoor assessment uses longer, more severe methods at higher voltages on larger specimens — the inclined plane test being the usual reference. Those methods sometimes rank materials in a completely different order than the drop test does. A material that wins on CTI can lose badly on the inclined plane.

Where results go wrong

Most CTI disputes trace back to a handful of causes.

Contaminated electrolyte. Residue on the dropper from the previous test contaminates the solution. Evaporation between tests raises the concentration. Both push results low. Flushing 10 to 20 drops through before each test clears non-conforming liquid.

Worn electrode tips. The 0.01–0.1 mm flattened edge does not stay that way. Tips need mechanical refurbishing between tests, and the edges and corners are what matter. A rounded tip changes the field concentration and the result with it.

Electrolyte running off. If liquid escapes over the specimen edge, the surface never accumulates the contamination the test depends on, and the material scores high for the wrong reason. This is why 20 × 20 mm is the recommended minimum. Thin metal foil around the edge of the glass support makes escaping electrolyte easy to detect.

Specimens thinner than 3 mm. Thin specimens lose heat to the glass support faster, which changes the drying behavior of the film. Values are not comparable to those from thicker specimens. Stack thin material rather than testing it as-is.

Surface condition. Scratches, mold release, fingerprints, and texture all move the result. Test surfaces are supposed to be smooth, untextured and clean. Where they are not, the report has to describe them.

Anisotropy. Laminates, fiber-filled compounds and parts with a strong flow direction behave differently along and across the feature. Test both orientations. Report the lower value unless something specifies otherwise.

Different apparatus, different answer. Electrode pivot geometry varies between machines. When electrodes sink into a softening specimen, the tips swing through an arc and the gap changes — by how much depends on where the pivots sit relative to the contact points. Round-robin scatter between labs is real, and it is worst on materials that soften.

Comparing molded parts to molded plaques. A plaque made by a different process, or molded in a different flow direction, from the same resin can give a different CTI than the finished part. When you have to test plaques instead of the part, match the fabrication process as closely as you can.

Where CTI sits among insulation tests

Tracking resistance is a surface property under contamination. It tells you nothing about the bulk. A material can have an excellent CTI and poor dielectric strength, or vice versa.

• Insulation resistance testing measures bulk and surface leakage at DC, on a clean, dry, assembled system. It finds moisture and contamination in service. It does not predict tracking.
• Dielectric withstand testing stresses the insulation through its thickness. Tracking failures happen across the face, at voltages far below the puncture level.
• Partial discharge testing detects discharge inside voids and at internal interfaces. Scintillation during a tracking test is a surface discharge phenomenon, driven by the electrolyte film rather than by internal defects.

CTI belongs to material qualification. The others belong to system verification. You use CTI to choose the plastic and set the creepage distance; you use the others to confirm the assembly built from it is sound.

FAQ