In switchgear connector enclosures, the glow wire test specified in IEC 60695-2-11:2021 subjects the moulded insulating part to a localized thermal insult at a selected tip temperature between 550°C and 960°C, applied with a contact force of 1.0 N ± 0.2 N for 30 s ± 1 s. The acceptance criteria require that flames and glowing do not persist for more than 30 s after the glow wire is removed and that the specified wrapping tissue is not ignited. For low-voltage switchgear and controlgear assemblies, IEC 61439-1:2020 includes verification of resistance to abnormal heat and fire, with the applied glow wire temperature selected according to the function of the insulating part and the current-carrying or earthing role. For switchgear connector enclosures, wall thicknesses are commonly in the range of 0.8 mm to 2.0 mm, and materials such as glass-fibre-reinforced polyamide 66, polybutylene terephthalate, and polycarbonate blends are selected to provide adequate mechanical strength together with a defined glow wire ignition temperature. The actual performance of these materials under glow wire conditions is not determined solely by the total flame retardant loading or by the base polymer chemistry; it is strongly influenced by the degree of dispersive and distributive mixing achieved during compounding and by the local concentration of flame retardant at the moulded surface and in thin ribs. A compound that passes the glow wire test on a homogeneous plaque may fail on a connector enclosure because of agglomerates, surface depletion, weld-line segregation, or thermal degradation of the flame retardant during processing. Dispersion quality therefore operates as a hidden variable within the entire manufacturing chain, from feed port sequencing and screw geometry to injection moulding back pressure and melt temperature.

Does Poorly Dispersed Magnesium Hydroxide Create an Ignition Pathway That Survives the 30 s Glow Wire Removal Interval?

Magnesium hydroxide decomposes endothermically between approximately 330°C and 450°C, releasing water vapour and forming magnesium oxide; in a polyamide or polyolefin connector enclosure, the effectiveness of this mechanism under glow wire conditions depends on interparticle spacing and the presence of agglomerates. When magnesium hydroxide is present as agglomerates with a median diameter above 10 µm, the endothermic decomposition becomes localized, and the surrounding polymer matrix remains unprotected. During the 30 s contact period of IEC 60695-2-11:2021, the glow wire tip can raise the local temperature above the polymer ignition threshold before the decomposition front reaches the critical volume. Agglomerates also reduce the surface area available for heat absorption per unit mass and can initiate microcracks at the moulded surface, exposing fresh polymer to oxygen and promoting flaming after the glow wire is withdrawn. A well-dispersed magnesium hydroxide grade with a particle size D50 near 1.0 µm to 2.0 µm and a D90 below 5.0 µm distributes the endothermic reaction across the heat-affected zone more uniformly, lowering the local peak temperature and reducing the probability of persistent flaming. Compounding such formulations on a co-rotating twin-screw extruder with a 40:1 L/D ratio and a screw profile containing at least two kneading block sections followed by a vacuum vent produces a specific mechanical energy input of 0.22 kWh/kg to 0.28 kWh/kg for a 55 wt% magnesium hydroxide loading in polyamide 66. Below this energy input the agglomerate count measured by image analysis remains above the acceptance threshold, while above this range the melt temperature exceeds 245°C and viscosity degradation of the polymer is observed. The lower processing limit is not merely a mixing recommendation; it is an ignition-relevant boundary because insufficient dispersive mixing preserves large particles that function as localized defects during the glow wire test. Published data for this specific configuration is limited because industrial compounders rarely report the simultaneous agglomerate size distribution and glow wire failure rate, but the causal link between large endothermic filler particles and char discontinuity is well established.

On a production line equipped with a 25 mm co-rotating twin-screw extruder, a side-feeder for glass fibre, and a melt pump, the pressure differential across a 100 mesh screen pack provides a real-time indication of dispersion quality for brominated and mineral-based flame retardant systems. A pressure rise greater than 1.0 MPa/min during a compounding run typically indicates agglomerate accumulation or premature crosslinking; if the pressure rise accelerates, the resulting moulded connector housings show an increase in glow wire failures at 750°C even though the formulation and the total flame retardant content remain unchanged. The practical response is not to increase barrel temperature indiscriminately because that can degrade temperature-sensitive synergists such as antimony trioxide, melamine polyphosphate, or phosphinate salts. Instead the screw configuration is adjusted to replace a narrow distributive mixing element with a wider dispersive kneading block, or the feed sequence is changed so that the flame retardant is introduced after the polymer has reached a fully molten state. These interventions alter the residence time distribution and the maximum shear stress experienced by the additive, and they must be evaluated by both melt filtration and glow wire testing because a formulation optimized only for low pressure rise may still contain local concentration gradients that affect ignition. The use of a 150 mesh screen pack reduces the threshold at which agglomerates are detected but can increase melt temperature by 5°C to 10°C, which may be unacceptable for a halogen-free formulation with a narrow processing window.

Agglomerate Size and Char Coherence in 0.8 mm Switchgear Connector Rib Sections

In thin connector rib sections of 0.8 mm wall thickness, the thermal mass available to dissipate the glow wire heat input is reduced relative to a 2.0 mm plaque, and any localized flame-retardant depletion becomes substantially more dangerous. The test criteria of IEC 60695-2-12:2021 and IEC 60695-2-13:2021 for glow wire flammability index and glow wire ignition temperature use standardized specimen thicknesses; connector manufacturers often qualify materials at one thickness but mould ribs at another, which can invalidate the comparative dispersion-performance relationship. A well-dispersed halogen-free system based on aluminium diethylphosphinate and melamine polyphosphate forms a coherent intumescent char that insulates the underlying polymer and limits oxygen diffusion. If the phosphinate is located in agglomerates larger than 20 µm, the char layer develops discontinuously, with visible pinholes and cracks at the interface between the char and the molten polymer. These defects permit volatile degradation products to escape and allow air to reach the unmodified polymer beneath, sustaining flaming after the glow wire is removed. Image analysis of polished cross-sections according to ISO 18553:2002 shows that agglomerate counts above 3 per 6 mm² in the 0.8 mm rib correspond to a measurable increase in flaming persistence time from below 5 s to above 20 s, converting a pass into a fail in end-product assessment. The mechanical strength of the char is further reduced by large agglomerates because the decomposition of the flame retardant leaves behind voids that interrupt the continuity of the carbonaceous layer; therefore dispersion control is also a char morphology control. In a 25 wt% glass-fibre-reinforced polyamide 66 compound, the glass fibres themselves create additional char discontinuities if they are not fully wetted by the polymer matrix, and the presence of flame retardant agglomerates at fibre ends amplifies this effect.

When Glass Fibre Is Side-Fed After the Flame Retardant, Dispersive Mixing and Glass Length Retention Compete

Side-feeding of glass fibre into a flame-retardant polyamide 66 compound after the first mixing zone is preferred to preserve fibre length; however, this sequence can create a low-shear region in which the flame retardant has already passed through the high-stress kneading block without being fully distributed. The glass fibre then enters the melt and increases viscosity, which decreases the ability of downstream mixing elements to disperse fine flame retardant particles. The result is a moulding compound that shows acceptable fibre length but poor flame retardant distribution at the surface of connector housings, where the glow wire tip contacts the part. On a 32 mm co-rotating twin-screw extruder with a 44:1 L/D ratio, a screw profile using three kneading blocks of 30°, 60°, and 90° staggering can produce a melt temperature of 248°C at 400 rpm. If the flame retardant is fed at the main throat and glass fibre is side-fed at 10:1 L/D, the specific energy input is 0.18 kWh/kg and the glass fibre length is 0.35 mm, but the dispersion index is poor and glow wire failures are intermittent. If the same flame retardant is split and a portion is side-fed with the glass fibre, the local concentration in the outer layer of the connector is improved, but the side feeder requires a reinforced screw liner because the modified flame retardant can bridge at low feed rates. The processing window for this approach is narrow: a barrel temperature reduction of more than 5°C below the set point increases melt viscosity and overloads the side feeder, while an increase above 5°C can volatilize low-molecular-weight fractions of the flame retardant and reduce the final glow wire performance. These observations from production-scale runs demonstrate that feeder sequencing and screw geometry cannot be separated from the ignition behaviour of the moulded part.

Compounding of a brominated epoxy oligomer with antimony trioxide in polybutylene terephthalate requires the antimony trioxide to be predispersed to a D90 below 2.0 µm; otherwise the synergistic reaction with liberated hydrogen bromide is incomplete during the brief glow wire heating interval. On a production injection moulding machine with a 35 mm screw diameter, a 900 kN clamp force, and a screw back pressure of 8 MPa, the shear heating during plastication can reduce the melt viscosity just enough to mask the presence of agglomerates, but the moulded part still contains undispersed antimony trioxide particles. The glow wire test performed at 850°C on a 1.5 mm wall section can then produce a flame persistence time above 10 s because the bromine release profile is not uniform across the contact zone; antimony-rich regions volatilize early and leave bromine-lean domains that ignite. A melt filtration trial using a 150 mesh screen pack and a pressure limit of 12 MPa at 250°C shows that the undispersed material produces a filter cake with antimony trioxide agglomerates larger than 50 µm, while the fully dispersed compound produces a filter cake consisting primarily of glass fibre fragments. The standard ISO 3451-1:2019 ash method is not sufficient to detect this dispersion problem because total flame retardant content may remain within specification even when the distribution is inadequate; only a quantitative dispersion index or a pressure-rise test under controlled shear conditions reveals the inconsistency. Avoid combining brominated epoxy oligomers with low-molecular-weight amine-based lubricants or secondary amine stabilizers because the amine reacts with the brominated species and forms a crosslinked gel that raises melt viscosity and prevents adequate dispersion; the resulting material may pass on a plaque but fail on a thin connector rib due to surface defects and poor char formation.

Thermal Degradation of Aluminium Diethylphosphinate Is Initiated by Local Shear Heating Above 260°C

Aluminium diethylphosphinate is a halogen-free flame retardant used in glass-fibre-reinforced polyamide 66 and high-temperature polyamide connectors requiring glow wire ignition temperatures above 775°C. The additive is thermally stable up to approximately 350°C in its pure state, but in the presence of polyamide and residual moisture, hydrolytic and oxidative degradation can occur at extruder melt temperatures above 260°C. The degradation products include diethylphosphinic acid and related species that can corrode processing equipment and reduce the phosphorus content available for char formation. In a 40:1 L/D extruder run with a barrel profile of 220°C to 255°C, the specific energy input is typically 0.24 kWh/kg; if the screw speed is increased from 350 rpm to 500 rpm without adjusting the barrel temperature, melt temperature can exceed 270°C, and the resulting moulded parts fail the glow wire test at 750°C despite retaining the same total phosphorus content. The failure is often attributed to the loss of the char-forming nitrogen synergist, melamine polyphosphate, which decomposes at a lower temperature and leaves the phosphinate to act without the intumescent network. The processing window is therefore defined not only by the melting point of the polymer, but also by the decomposition boundary of the least thermally stable flame retardant in the formulation; for phosphinate/melamine polyphosphate systems in polyamide 66, the melt temperature must be maintained within 245°C to 255°C, and any excursion above 260°C for more than 2 min is unacceptable. Production-scale compounding with a gear pump and a melt cooler can reduce this thermal damage, but the capital cost is justified only when the connector enclosure is specified for glow wire testing at 850°C or above. Over-dispersion is also not beneficial because excessive shear after the additive has reached its domain size can fracture the encapsulated flame retardant and expose reactive surfaces to moisture and oxygen, further reducing the char yield.

Polyamide 66 and polyamide 6 absorb moisture; if the polymer is not dried to below 0.20 wt% moisture, the melt viscosity is reduced by hydrolysis and the dispersive mixing intensity falls. A moisture level of 0.35 wt% in a 30 wt% glass-fibre-reinforced polyamide 66 containing a halogen-free flame retardant package can reduce the specific energy input by 0.03 kWh/kg to 0.05 kWh/kg, which is sufficient to leave aluminium diethylphosphinate agglomerates intact. The subsequent glow wire test at 750°C on a 1.0 mm connector wall then shows ignition of the wrapping tissue because the char layer is discontinuous. Pre-drying in a desiccant dryer at 80°C for 4 h with a dew point below -30°C is standard for polyamide 66; for polyamide 6 the residence time is extended to 6 h at 80°C. Polybutylene terephthalate must be dried at 120°C for 4 h, because hydrolysis of the ester linkages becomes significant above 0.02 wt% moisture during melt processing at 250°C. These moisture limits are not simply handling recommendations; they are part of the dispersion control envelope because water-induced viscosity reduction changes the maximum shear stress transmitted to agglomerates. A compound produced from insufficiently dried material may exhibit a melt flow index within the specification range according to ISO 1133-1:2022 and still fail the glow wire test because the flame retardant is present as isolated particles rather than as a network. Pre-drying is required whenever ambient relative humidity exceeds 60%, and the material residence time in the dryer should not exceed 8 h to avoid thermal yellowing of polyamide.

At weld lines and sharp corners of switchgear connector enclosures, the flow front carries partially agglomerated flame retardant particles along the skin region; fountain flow at the melt front deposits them on the surface in a pattern that is not uniform. During the glow wire test at 750°C, the contact zone may coincide with a weld line where the local concentration of the char-forming additive is reduced by 15% to 25% relative to the bulk, producing premature ignition. This effect is observed in glass-fibre-reinforced polyamides and in PC/ABS blends where phosphoric ester flame retardants exude to the surface over time. The surface segregation is especially problematic for phosphoric ester additives with limited compatibility, because the glow wire test heats the surface directly, and a flame retardant that has migrated into a surface film can be volatilized before it can protect the underlying polymer. A small amount of an epoxy-based chain extender can rebalance the melt strength and reduce the migration rate, but the addition changes the processing viscosity and must be evaluated under the same screw and barrel conditions as the production run. Published data for the exact combination of brominated polystyrene, antimony trioxide, and glass-fibre-reinforced PBT at 0.8 mm thickness is limited, but the surface depletion mechanism has been documented for other thick-thin transitions in connector housings.

Can Surface Migration of Phosphoric Ester Flame Retardants Deplete the Glow Wire Contact Zone?

Phosphoric ester flame retardants such as resorcinol bis(diphenyl phosphate) and bisphenol A bis(diphenyl phosphate) are used in polycarbonate and PC/ABS connector enclosures to achieve glow wire compliance without halogenated additives. These esters have limited solubility in the polymer matrix and can migrate to the surface during storage or after repeated thermal cycling, creating a surface layer that is either enriched or depleted depending on the diffusion coefficient, the mould temperature, and the degree of post-mould crystallization. If the flame retardant migrates to the surface and is subsequently volatilized or abraded, the remaining surface becomes depleted and the glow wire test at 750°C on a 1.0 mm wall can fail even though the bulk material retains the original phosphorus content. Conversely, a thick surface film can ignite more readily under direct contact with the glow wire before the intumescent mechanism begins. The dispersion issue here is not agglomeration but distribution across the thickness; injection moulding with a cold mould below 80°C freezes a depleted skin before the additive can redistribute, while a mould temperature above 110°C allows more surface migration and can create a sticky surface that affects connector assembly. The processing window for this class of flame retardant is therefore defined by both the mould temperature and the residence time in the barrel; a barrel residence time exceeding 5 min at 270°C can cause hydrolysis of the phosphate ester and reduce the char-forming phosphorus content. A compliance verification matrix for this material should include surface phosphorus measurement by X-ray photoelectron spectroscopy or infrared spectroscopy, because bulk phosphorus content measured by ISO 3451-1:2019 does not identify skin depletion.

When switchgear connector enclosures contain metallic inserts such as brass terminals or steel retaining clips, the glow wire test is affected by the thermal conductivity of the insert and the interface between the polymer and the metal. Flame retardant particles often orient along the melt flow around the insert, creating a skin layer that can be enriched or depleted. If the polymer around the insert is under shear, the flame retardant may migrate away from the immediate metal surface because of shear-induced particle migration, leaving a thin layer of unprotected polymer. The contact of the glow wire at a point near the insert may produce a different failure mode than the same material away from the insert. This behaviour is observed in production parts with insert overmoulding, where the local wall thickness may be below 0.8 mm and the thermal mass of the metal insert can delay the temperature rise but also quench the char formation. The injection moulding process must be set with sufficient hold pressure and a melt temperature high enough to allow the flame retardant to wet the insert surface; if the melt is too cool, the flame retardant remains as a compacted layer and the connector fails the glow wire test at 750°C even though the bulk compound passes. Published data for this specific configuration is limited, but insert-bearing connector enclosures require a separate glow wire qualification because the dispersion state at the insert interface cannot be inferred from plaque data.

On a manufacturing line producing 2.0 mm glass-reinforced polyamide 66 connector enclosures, batch-to-batch variation in the particle size distribution of the same flame retardant grade can produce glow wire failures that are not correlated with the total flame retardant assay. A batch with a D90 of 35 µm instead of the specified 15 µm may pass incoming quality control if only the sieve residue is measured; however, after compounding and injection moulding, the larger particles remain as agglomerates and create localized char defects at the connector boss and snap-fit. The corrective action is not to increase the flame retardant loading, because that can reduce the comparative tracking index and the mechanical strength below the minimum required for the application, but to tighten the incoming particle size specification and to use the same dispersive screw configuration for the replacement batch. In one production-scale failure mode, a filter pack pressure rise of 1.3 MPa/min was recorded before the run was stopped, and the moulded parts from that batch showed a 40% increase in glow wire failures at 750°C; after replacing the screen pack and adjusting the screw profile, the pressure rise stabilized below 0.15 MPa/min and the glow wire failure rate returned to the expected value. The operational boundary is that screen pack changes and screw adjustments must be logged against the dispersion index and the glow wire result, because the relationship between pressure rise and ignition behaviour is specific to the formulation and to the test temperature.