Matching Polybutylene Terephthalate CTI to Creepage and Clearance in a Relay

Polybutylene terephthalate (PBT) is specified for relay bases, coil formers, terminal barriers, and arc isolation walls because its semicrystalline structure provides the required heat deflection temperature, stiffness, and flow length for thin-wall insulation components. In insulation coordination, the decisive material property is not short-term dielectric strength but comparative tracking index (CTI), determined under IEC 60112 or ASTM D3638 using 0.1 % ammonium chloride solution and 50 drops per test specimen. The measured CTI places the PBT grade into a material group defined by IEC 60664-1. That group controls minimum creepage distance along the molded surface of the relay while clearance through air remains a function of impulse withstand voltage, altitude, and field geometry. When a relay terminal pitch is fixed by a standardized socket or PCB footprint, the difference between a PBT grade at 225 V CTI and a grade at 450 V CTI can determine whether the design passes without additional ribs, slots, or barrier walls.

Why Does a PBT Grade’s CTI Reclassify Creepage in a Relay Base?

Under IEC 60664-1 Annex F, creepage is assigned from the RMS working voltage, the pollution degree, and the material group. Material group IIIa covers CTI values from 175 V to 399 V, while material group IIIb covers 100 V to 174 V. Many 30 % glass-reinforced, brominated flame-retardant PBT compounds fall in group IIIa or IIIb. This classification is not a linear safety margin. A shift from group II to group IIIa at 250 V rms and pollution degree 2 raises the representative basic-insulation creepage distance from 2.5 mm to 3.2 mm for the same working voltage. For reinforced insulation, the standard doubles the basic-insulation value, so the design delta becomes 1.4 mm—from 5.0 mm to 6.4 mm—when the material group shifts from II to IIIa. The consequence in a miniature relay with a 3.0 mm terminal pitch is direct: a group II PBT may pass with a straight flat surface, while a group IIIa PBT requires a barrier or formed groove to satisfy the same creepage.

The CTI value is an electrical surface test, not a bulk property. It evaluates the ability of a molded surface to resist ionization of a contaminated liquid film under electrical stress. PBT is hygroscopic relative to polyolefins but far less than polyamide; moisture absorption under ISO 62 at 23 °C and 50 % RH is typically 0.15 %–0.25 % for unfilled grades. The CTI test imposes a more aggressive aqueous electrolyte, so the result is highly sensitive to surface polarity, filler exposure, and additive migration. Glass fiber ends exposed on the molded surface create local field enhancement and wicking paths, which is why glass reinforcement commonly depresses CTI by one or two material groups relative to an unfilled polymer. A PBT compound with 30 % short glass fiber may show CTI in the 200 V–250 V range, while a mineral-filled high-tracking grade may reach 400 V–600 V.

Clearance, Impulse Voltage, and the Non-Tracking Side of a Relay Terminal Gap

Clearance in a relay is the shortest distance through air between conductive parts of different potential. It is not derived from CTI. The assigned clearance is selected from the overvoltage category and the impulse withstand voltage, then corrected for altitude and field homogeneity. A relay coil-to-contact separation rated for a 4 kV impulse voltage under overvoltage category III and pollution degree 2 may require a clearance near 3.0 mm in an inhomogeneous field, but the same terminal gap may need 3.2 mm creepage if the selected PBT is group IIIa at 250 V rms. In that case creepage governs the design, and the CTI classification of PBT is the limiting material variable. If the relay operates in an unsealed enclosure with dust, condensation, or salt mist, pollution degree 3 increases the creepage requirement more steeply for lower-CTI PBT than for high-CTI PBT. Clearance is still adjusted only for impulse withstand and altitude; material tracking resistance does not enter the clearance calculation. For altitudes above 2000 m, clearance correction factors from IEC 60664-1 apply, while creepage remains unchanged.

On an injection molding line, the CTI value reported on a supplier datasheet is produced from a standardized 3 mm or 4 mm plaque, not from the relay base geometry. Transfer of that value to the relay component requires verification because the surface that carries creepage is formed under different shear, packing, and cooling conditions. Molded-in stresses and glass-fiber orientation at weld lines can create locally reduced tracking resistance that is not captured by plaque testing. In a relay base with an insert-molded terminal, the weld line commonly forms at the terminal shank or between the coil former and the terminal barrier. At that weld line, glass fibers orient parallel to flow, and the resin-rich region may exhibit lower CTI than the bulk material. Production experience with glass-reinforced PBT relay housings shows that shrinkage-induced microcracks and exposed fiber ends can appear after thermal cycling from -40 °C to 85 °C. Those surface features act as moisture condensation points and can initiate tracking. The insulation coordination review should therefore include CTI testing on specimens cut from actual molded relay bases at the terminal weld line, not solely the supplier’s plaque value.

Before molding, PBT must be dried to below 0.02 % residual moisture by Karl Fischer or equivalent. A desiccant-bed dryer with a dew point of -40 °C and a set point of 120 °C for 3 h–4 h is typical. Melt temperature should remain below 270 °C; higher temperatures or residence times above 120 s promote hydrolytic chain scission and the generation of low-molecular-weight oligomers that bloom to the surface. These oligomers can attract moisture and ionic contamination, reducing surface insulation and potentially lowering CTI. In flame-retardant PBT, antimony trioxide and brominated additive packages are thermodynamically stable when processed correctly, but overmolding with hot runner temperatures above 260 °C can cause additive segregation at the gate. Gate blush from high-speed injection, typically above 300 mm/s linear injection velocity, has been observed to create a surface layer with measurable reduction in tracking resistance. For this reason, relay molders may limit injection velocity to 200 mm/s–250 mm/s and use sequential valve gating to move weld lines away from terminal gaps.

Regrind content in glass-filled PBT is another CTI variable. Repeated exposure to melt processing shortens glass fibers and broadens fiber length distribution, reducing both mechanical strength and tracking resistance. Many relay molding operations limit regrind to 20 % or less by weight because higher percentages can shift CTI from 250 V toward 200 V or below in some flame-retardant grades. Fiber attrition increases the number of exposed fiber tips on the surface, and polycondensation during reprocessing changes surface polarity. Published data for regrind-specific CTI changes in PBT relay housings is limited, so full qualification testing under IEC 60112 is required when introducing regrind into a relay application.

If a Relay Terminal Pitch Falls Between Two Material Groups

Consider a printed-circuit relay with a 3.0 mm fixed terminal pitch and a working voltage of 250 V rms across coil-to-contact insulation in pollution degree 2. For basic insulation, material group II requires 2.5 mm creepage, so a flat molded PBT surface between terminals is sufficient if the grade is group II. A group IIIa PBT requires 3.2 mm, failing the 3.0 mm direct path. The designer has three options: widen the terminal pitch, add a molded rib or slot that increases effective surface path without changing pitch, or select a higher-CTI PBT grade. Widening the terminal pitch may not be possible because the relay must fit an industry-standard footprint. Adding a rib increases tool complexity but can be validated by creepage measurement on the molded part. Selecting a high-CTI PBT may require giving up a specific flame-retardant package, because many brominated flame-retardant systems depress CTI. This is the core matching problem: the electrical clearance may be entirely adequate at 3.0 mm, but the CTI of a lower-cost glass-filled PBT drives the creepage requirement above the available pitch.

Molded Ribs, Slots, and Surface Geometry in CTI-Limited Relay Bases

A rib of height 1.5 mm or a slot of width 1.0 mm may interrupt or extend the surface path under IEC 60664-1 geometry rules. The rib does not reduce the required creepage; it increases the measured surface path by forcing the path over the rib. In a relay base with a 3.0 mm terminal pitch, a rib 1.0 mm high can add roughly 2.0 mm to the creepage path if placed between terminals. The effective path then becomes 5.0 mm, exceeding the 3.2 mm group IIIa requirement. The air clearance remains 3.0 mm. This geometry solves the CTI-to-creepage mismatch without changing the footprint. However, the rib must be molded without sink marks, weld lines, or glass fiber orientation defects; otherwise the extended path may still contain a weak tracking zone. Molders may use gas counterpressure or valve-gated hot runners to maintain packing integrity in tall ribs.

Relay safety standards such as IEC 61810-1 and IEC 60730-1 reference the insulation coordination rules of IEC 60664-1. For a relay with a rated impulse withstand voltage of 4 kV, the clearance is selected from the standard’s table for that impulse level and the applicable pollution degree. A PBT with CTI 200 V does not change that clearance. It changes only the creepage side. For a 250 V rms working voltage in pollution degree 2, the required basic creepage for group IIIa is 3.2 mm; for group IIIb it is 4.0 mm. If the PCB footprint has a 3.5 mm terminal gap, the group IIIa material passes with a flat surface, while the group IIIb material needs a barrier or grade change. The design review should therefore start from the lowest expected CTI of the molded part, not from the datasheet typical.

Tracking failures in PBT relays frequently occur under condensation and dust exposure. A sealed relay with the coil and contacts in an enclosure may have a lower internal pollution degree than an unsealed PCB relay. In open construction, pollution degree 2 assumes non-conductive pollution but occasional condensation; pollution degree 3 assumes conductive pollution or dry non-conductive pollution that becomes conductive with condensation. The creepage dimension for a group IIIa PBT at 250 V rms in pollution degree 3 is larger than the 3.2 mm required for pollution degree 2. If the relay operates near a fan, in a humid environment, or under salt spray, the higher pollution degree should be selected, and the CTI-based material group becomes even more restrictive.

Qualification of a PBT relay base requires four data streams: the supplier CTI under IEC 60112, the UL 746A performance category in the PLC system, the molded-part creepage measurement under IEC 60664-1 Annex F, and production lot-to-lot CTI reproducibility. A PLC 3 rating corresponds to CTI 175 V–249 V, which is group IIIa or IIIb depending on the precise value. A PLC 2 rating maps to 250 V–399 V. If the relay safety approval requires UL 840 or IEC 60730-1 insulation coordination, the CTI source should be the actual molded compound, not a generic PBT literature value. Lot-to-lot variance of ±25 V in CTI is meaningful because it can shift a grade from group II into group IIIa. Incoming quality control should therefore include CTI tests at a frequency sufficient to detect a one-group shift before parts are molded.