Run an IEC 60112 test and the specimen can go bad in two completely different ways. One leaves a conductive path. The other leaves a hole. The headline number — the CTI or PTI — only tracks the first. The second gets measured separately, if at all, and it is the one people forget to ask for.
Same apparatus, same electrodes, same drops of electrolyte. Two failure modes that mean different things for the material you are about to specify.
Table of Contents
The fork
Both start the same way. A contaminated film carries leakage current between the electrodes. The film dries unevenly, a dry band forms, it breaks down, and a small surface discharge — a scintillation — fires across it. That scintillation is the agent of damage in both cases.
What happens next depends on the material.
Tracking is the electrical outcome. The scintillation pyrolyzes the polymer under it. If the material carbonizes, it leaves conductive residue behind. That residue acts as an extended electrode, so the next dry band forms slightly further along, the next scintillation lands there, and a conductive path creeps across the surface. When it bridges the gap, current runs freely. The over-current device trips. The insulation has been defeated across its face while its bulk is untouched.
Erosion is the mechanical outcome. The scintillation removes material without leaving a conductive path. The polymer volatilizes or decomposes and is carried off. No carbon track forms. Instead the surface hollows out. The electrodes sink into the pit they are digging. Given enough drops, the hole goes clean through.
The standard names them separately: tracking is the progressive formation of conductive paths; electrical erosion is the wearing away of material by electrical discharges.
| Tracking | Erosion | |
|---|---|---|
| What forms | Conductive path across the surface | A pit or hole in the surface |
| Residue | Carbon, conductive | Little or none |
| How the test ends | Over-current trip (0.5 A for 2 s) | May not trip at all — material just disappears |
| What the CTI/PTI captures | This | Not this |
| How it is measured | Voltage at failure | Depth gauge, in mm |
Why a material does one and not the other
The dividing line is what the polymer leaves behind when a discharge hits it.
Materials that carbonize track. The decomposition products are conductive, so every scintillation builds the path. Phenolic laminates and many filled thermosets sit here.
Materials that volatilize cleanly erode instead. There is no conductive residue to build a path, so the surface is removed rather than bridged. Unfilled PTFE and silicones lean this way — they can post excellent tracking numbers precisely because they refuse to carbonize, then lose material to erosion under the same discharges.
Fillers move the line in both directions. Inorganic fillers can suppress carbonization, improving tracking resistance, while changing how much material a discharge removes. This is why you cannot predict either failure mode from the base resin alone.
The practical consequence: a high CTI does not mean a material is durable under contamination. It means the material does not form a conductive track. It may still be eroding away the whole time, and the CTI number will not tell you.
How erosion gets measured
Erosion is not measured automatically. It is measured only when the specification asks for it — and if nobody asks, a badly eroding material can pass a tracking test with a clean-looking result.
When erosion is required, the procedure is separate and comes after the tracking test:
- Take the specimens that survived the 50-drop test without a tracking failure.
- Clean off loose debris and degradation products.
- Set each on the platform of a depth gauge.
- Measure the deepest point with a 1.0 mm nominal probe, hemispherical tip, to 0.1 mm.
- The result is the deepest of the five specimens.
Depths under 1 mm are reported as < 1 mm. If a specimen flamed too badly to measure, that gets reported instead of a number.
For a CTI determination, erosion is measured on the five specimens tested at the maximum 50-drop voltage. For a PTI test, it is measured on the specimens that withstood the specified voltage.
When a hole goes all the way through
Sometimes erosion does not stop at a pit. The discharge burns a hole clean through the specimen.
The standard is explicit that this does not void the result. A test where a hole formed is still valid, regardless of specimen thickness. But the hole has to be reported, along with its depth — which is simply the specimen or stack thickness, since it went all the way through.
You have an option when this happens: retest on thicker material. Specimens can be stacked or replaced up to a 10 mm maximum to gain more information about how deep the damage would have gone if the specimen had not run out.
This is worth doing when erosion is the real concern. A 3 mm specimen that holes through tells you the material lost at least 3 mm — but not how much more it would have lost. A thicker retest gives you the actual depth.
Reading both in the notation
The reported result folds erosion in when it was measured. The erosion depth follows the voltage, in millimeters.
| Reported as | Tracking result | Erosion |
|---|---|---|
| CTI 275 | 275 V | not measured |
| CTI 275 – 1,2 | 275 V | 1.2 mm max |
| CTI 400(350) M – 3,4 | 400 V, 100-drop 350 V, Solution B | 3.4 mm max |
| Pass PTI 250 – 3 | passed at 250 V | 3 mm max |
Read the two numbers as independent verdicts. CTI 275 – 1,2 and CTI 275 – 0,3 have identical tracking performance and very different durability. On a part that sees years of wet-dry cycling, the erosion figure can matter more than the CTI, because a track is a sudden failure while erosion is a slow one that eventually becomes a puncture or a hole.
A material with a strong CTI and heavy erosion has not passed. It has failed in the mode nobody put in the headline.
What this means for selection
Two rules fall out of the split.
Always ask for erosion when the part sees prolonged or repeated wetting. The default test does not measure it. If your creepage table only cares about the tracking group, you can specify a material with a great CTI that erodes badly, and the acceptance paperwork will look clean. Add the erosion requirement so the failure mode you actually face is on the report.
Do not read a clean CTI as a durability rating. CTI answers one question — does a conductive path form. Erosion answers a different one — does the material survive the discharges intact. A full picture of contamination performance needs both.
Both of these live in material qualification, upstream of the assembled product. For how the tracking side produces the CTI and PTI numbers, see the comparative tracking index pillar and the proof tracking index page. And when the part is bound for severe or outdoor service, the IEC 60587 inclined plane test evaluates both failure modes together at kilovolt levels — a hole through the specimen is an automatic failure there, which puts erosion on equal footing with tracking rather than treating it as an optional measurement.
FAQ
What is the difference between tracking and erosion?
Tracking is the formation of a conductive path across the insulating surface, driven by carbonized decomposition products. Erosion is the wearing away of material by the same discharges, leaving a pit or hole but no conductive path. Both happen in an IEC 60112 test; only tracking is captured by the CTI or PTI value.
Does the CTI value include erosion?
No. CTI measures the voltage at which tracking failure occurs. Erosion is measured separately, and only when the specification requires it. When measured, it is appended to the result as a depth in millimeters.
Why does a material erode instead of track?
Because its decomposition products are not conductive. Instead of leaving carbon that builds a conductive path, the material volatilizes and is removed. Materials that resist carbonization tend to erode rather than track.
Is a hole through the specimen a failed test?
Not automatically. The result is still valid, but the hole and its depth must be reported. You can retest on thicker specimens, up to 10 mm, to find out how deep the erosion would have gone.
How is erosion depth measured?
With a depth gauge using a 1.0 mm probe with a hemispherical tip, read to 0.1 mm, on cleaned specimens that survived the tracking test. The result is the deepest of the specimens tested. Depths below 1 mm are reported as < 1 mm.
Can a material pass the tracking test but still be a poor choice?
Yes. A material can post a high CTI and erode heavily under the same discharges. If the erosion requirement was not specified, that damage never appears on the report. For parts exposed to sustained contamination, always request the erosion measurement.