Ask an outdoor instrument transformer why it failed and the answer is rarely the winding. It is the surface: a polluted, wetted shed that flashed over at dawn, or a silicone housing that eroded after ten summers of UV. Pollution performance is the design axis that separates an outdoor unit from an indoor one — and it is the axis most often under-specified.
This note covers what pollution class actually means for a 24–35 kV unit, how creepage distance is counted, why silicone-rubber sheds behave differently from epoxy after contamination, and which environmental numbers belong on the datasheet.
- Pollution severity is a site property (ESDD/NSDD), not a product property. A coastal or industrial site is class III–IV even if the transformer is silicone.
- Creepage is quoted in mm/kV, and the number is meaningless without its reference voltage: IEC 60815-1 unified specific creepage distances run from about 22 mm/kV for very light sites to 43 mm/kV and above for heavy pollution.
- Silicone-rubber sheds transfer hydrophobicity to the pollution layer, so water beads instead of filming — that is the mechanism, not the colour of the shed.
- UV, ozone and thermal cycling age the housing; −40 °C to +55 °C and UV stability are datasheet items, not marketing.
1. Pollution is measured at the site, not chosen from a catalogue
Two measurements describe a polluted site: ESDD (equivalent salt deposit density, mg/cm² of NaCl equivalent) and NSDD (non-soluble deposit density, mg/cm² of kaolin or similar inert material). They are taken from unenergised reference insulators exposed at the site, usually over one to three years.
IEC 60815-1 then maps the pair onto a pollution severity class, from very light to very heavy. A coastal site with onshore wind, a cement plant, a chemical works and a dry inland site all land in different classes — and a site can move class when a new industry opens next door.
This is why a specification that says "suitable for polluted areas" without a class is not a specification. The supplier cannot size the sheds, and the buyer has no way to compare offers.
2. Creepage distance, and why the reference voltage matters
Creepage distance is the shortest path along the external surface between the high-voltage terminal and earth. It is quoted per kilovolt, and the kilovolt in question changes the number by a factor of √3:
- Referred to the highest system voltage Um (line-to-line), a 36 kV unit with 1,000 mm of creepage has 27.8 mm/kV.
- Referred to the phase-to-earth voltage (36/√3 = 20.8 kV), the same 1,000 mm is 48 mm/kV.
Both conventions appear in tender documents. The IEC 60815-1 unified specific creepage distance (USCD) is referred to Um: about 22 mm/kV for very light, 25 for light, 31 for medium, 43 for heavy and 53 for very heavy pollution. Older national tables and some customer standards quote 25–31 mm/kV for what they call class III–IV, which corresponds to the lighter end of that range when referred to phase-to-earth voltage. Always ask which reference voltage the number uses — the difference is not cosmetic.
| Site class (IEC 60815-1) | USCD, mm/kV referred to Um | Typical site |
|---|---|---|
| Very light | ~22 | Clean rural, low dust |
| Light | ~25 | Agricultural, light industry |
| Medium (III) | ~31 | Industrial, moderate coastal |
| Heavy (IV) | ~43 | Coastal with onshore wind, cement, chemical |
| Very heavy | ~53 | Desert with salt, severe industrial plume |
3. Shed profile: creepage is only half the story
Creepage can be increased by adding sheds, but sheds that are too close together do not help: the arc can bridge them, and pollution collects in the narrow gaps. IEC 60815-3 sets rules for shed overhang, spacing and the ratio of creepage to strike distance, which is why a "high creepage" unit with a cramped shed stack can perform worse than a unit with fewer, better-proportioned sheds.
For 24–35 kV outdoor units, the practical shape is a stack of 6–10 sheds with an overhang of 25–40 mm and a spacing that keeps the creepage-to-strike-distance ratio above about 3.5–4. Below that ratio, a pollution flashover is more likely to establish across the air gap than along the surface — and the sheds stop mattering.
4. Silicone rubber versus epoxy outdoors
Both materials are used for outdoor instrument transformers, and the difference is not mechanical strength but surface behaviour after contamination.
Epoxy is hard, dimensionally stable and resistant to tracking, but its surface is hydrophilic once contaminated. Pollution plus light rain forms a continuous wet film, the film carries leakage current, and the current dries it into dry bands where the voltage concentrates — the classic pollution flashover mechanism.
Silicone rubber is hydrophobic, and — more importantly — it is restoring. Low-molecular-weight silicone chains migrate from the bulk to the surface, diffuse into the pollution layer, and make the contamination itself water-repellent. Water beads on a dirty silicone shed instead of filming. This transfer takes hours to days after contamination and is the reason silicone units recover performance between pollution events.
The same mechanism is why handling matters: wiping a silicone shed with a solvent, or painting it, can remove or seal the low-molecular-weight fraction and destroy the effect.
5. UV, ozone and the temperature range
An outdoor housing sees more than pollution. Ultraviolet radiation breaks polymer chains at the surface; ozone attacks unsaturated bonds; thermal cycling between a cold night and a sunlit afternoon creates mechanical stress at the interface between the shed and the core.
Good practice on the datasheet:
- Ambient range: −40 °C to +55 °C covers most export destinations; a unit rated only to −25 °C will crack its sheds in a cold snap, not in the factory.
- UV stability: the compound, not the colour, determines it — a UV-stabilised silicone with carbon-black filler is standard for outdoor service.
- Tracking and erosion resistance: tested per IEC 60587 (inclined plane) for the material, and demonstrated in the design's type tests.
6. Salt fog, coastal sites and washing
Coastal sites combine high ESDD with humidity that never lets the surface dry. Salt fog tests (IEC 60507 artesian fog, or the site-specific equivalent) are the usual qualification, and the result is a withstand voltage at a given salinity — not a guarantee about a particular site.
Two operating rules for coastal installations:
- Do not wash silicone with a high-pressure jet. The water can lift the pollution layer and damage the shed surface; low-pressure washing, or none, is better.
- Inspect after the first wet season. If the sheds show erosion, chalking or a permanent water film, the class was under-specified — a replacement is cheaper than a flashover.
7. Mounting and clearances outdoors
Outdoor units are mounted on poles, brackets and ring-main boxes, and the mounting hardware is part of the insulation coordination: a floating bracket or a rusty bolt at the base can create a spark gap that no amount of creepage will fix. Earth the base directly, keep the high-voltage lead short and rigid, and respect the phase-to-phase clearance the switchgear designer specified — the transformer's own sheds do not compensate for a lead routed too close to the next phase.
8. A datasheet line that can be built
VT, outdoor, single-phase, 35/√3 kV / 0.1/√3 kV / 0.1/3 kV, class 0.5 at 30 VA, voltage factor 1.2/1.9/8 h, insulation 36/70/170 kV, silicone-rubber sheds, creepage ≥ 1,100 mm (≈31 mm/kV referred to Um), UV-stabilised, ambient −40 °C to +55 °C, IEC 61869-3.
9. Frequently asked questions
Can I use an indoor cast-resin unit outdoors under a canopy?
Only if the site is genuinely sheltered from rain, sun and pollution. A canopy stops direct rain but not humidity, and an indoor unit's surface is not designed for a wet pollution layer. If the site is class II or above, use an outdoor housing.
Does more creepage always mean better pollution performance?
No. Beyond a point, additional creepage comes from cramped sheds that collect pollution and are bridged by arcs. Profile quality and surface hydrophobicity matter as much as the millimetre count.
How often should outdoor units be cleaned?
It depends on the site class and on what the first wet season shows. Many utilities clean class III–IV sites every two to five years; silicone units often need it less because the surface self-restores hydrophobicity. Inspect before scheduling.
Is silicone suitable for very heavy pollution?
It is the better choice where erosion and wetting are the main threats, but very heavy pollution may still require a larger unit, a higher creepage class, or RTV-coated porcelain. Silicone improves the surface; it does not repeal the pollution class.
10. References and further reading
- IEC 60815-1, Selection and dimensioning of high-voltage insulators intended for use in polluted conditions — Part 1: Definitions, information and general principles.
- IEC 60815-3, Part 3: Polymer insulators for a.c. systems.
- Clean fog evaluation of RTV silicone rubber coatings with different thicknesses and pollution levels — laboratory pollution performance of silicone surfaces. Google Scholar
- Super-hydrophobic silicone rubber for outdoor electrical insulation, Elsevier — surface engineering of silicone housings. ScienceDirect
- IEC 60587, Electrical insulating materials used under severe ambient conditions — Test methods for evaluating resistance to tracking and erosion.
Choosing between an epoxy and a silicone outdoor unit for a 24 kV or 35 kV site? Our voltage transformer range lists the shed count, creepage and ambient rating per model — and the pollution class is usually the number that decides which model fits.