Spunbond polyester carrier web is unwound from a braked or servo-driven station and fed through a polymer-modified bitumen coating and calendering line. The web enters a molten bitumen bath or roll coater at line speeds between 10 m/min and 40 m/min, where the compound is typically held at 160 °C to 190 °C depending on SBS or APP content. The carrier web is exposed to hot bitumen on one or both sides, then passed through a set of water-cooled or chilled steel drums. Dimensional changes in the PET nonwoven during this thermal cycle are the main cause of width loss, edge waviness, and post-roll camber. Incoming carrier shrinkage is controlled by fiber drawing, bonding conditions, and heat-setting, while process-induced shrinkage is governed by web tension, preheating, melt temperature, and cooling drum differential speed. International product standards such as EN 13707 and ASTM D6222 require dimensional stability verification at elevated temperature, making carrier shrinkage a direct compliance risk.

During melt spinning and calender bonding, polyethylene terephthalate nonwoven is produced as a semi-crystalline material whose fibers retain orientation from the spunbond drawing step. Exposure to molten bitumen at temperatures above the PET glass transition temperature, approximately 78 °C, allows amorphous chain segments to relax. Shrinkage becomes measurable when the web reaches 110 °C to 130 °C, where relaxed oriented amorphous chains can undergo cold crystallization and release frozen-in stress. The rate of thermal shrinkage is a function of draw ratio, bonding temperature, and the thermal history applied during calender bonding. Spunbond PET carrier of conventional grades may exhibit machine-direction shrinkage of 3 % to 6 % when tested at 180 °C for 5 min in a circulating air oven according to ASTM D1204; low-shrinkage grades are typically below 1.5 % under the same conditions. Cross-machine shrinkage is usually lower but becomes critical in wide-web coating because edge restraint is reduced.

Why Does Spunbond PET Carrier Shrinkage Destabilize Dimensional Compliance of Polymer-Modified Bitumen Membranes?

Because the carrier is embedded between bitumen layers, shrinkage does not occur uniformly. In the machine direction, the web under tension may contract after leaving the coating nip when the pull rolls release tensile stress. In the cross-machine direction, selvedge areas that are not fully impregnated contract at a different rate than the center, producing edge curl and width loss. A membrane roll produced with 1.0 m nominal width may lose 8 mm to 15 mm of effective width when the carrier shrinks 1 % in the cross-machine direction. This width loss must be compensated by slitting or results in nonconforming roll dimensions under EN 1848-1. Dimensional stability tests on the finished membrane, such as EN 1107-2, measure shrinkage at elevated temperature; a failed result is often traced to retained carrier stress rather than to bitumen flow. The carrier's shrinkage force also tears the bitumen matrix when restrained, creating internal voids that reduce low-temperature flexibility tested under EN 1109.

At the unwind station, tension control is the first process variable that determines whether carrier shrinkage is expressed immediately as web necking or later as residual stress. A typical production unwind station uses a magnetic particle brake or a servo motor with a dancer roller and load cells having a resolution of ±1 N. Machine-direction tension is normally set between 80 N/m and 250 N/m of web width; excessive tension pulls the web forward through the hot zone, allowing shrinkage to occur as width loss rather than length retraction. S-wrap pull rolls with rubber-coated steel nipples isolate tension zones between the unwind, coater, and cooling sections. When tension isolation is absent, the molten bitumen adhesive drag introduces a transient tension spike estimated from line speed and viscosity that can exceed 400 N/m at speeds above 30 m/min, causing instantaneous neck-in of 1 % to 2 %.

When Inline Thermal Relaxation Is Applied Ahead of the Impregnation Nip

If the carrier web is preheated through an infrared tunnel or over a heated roller stack before entering the bitumen bath, a portion of the frozen-in stress can be released under controlled conditions. On a pilot line equipped with medium-wave IR panels rated at 40 kW/m², preheating a 150 g/m² spunbond PET web to 140 °C for 2 s reduces subsequent shrinkage at 180 °C from 4.5 % to 1.8 % when the web is allowed to relax freely between two driven rolls. This reduction is attributed to chain relaxation and partial crystallization before bitumen contact. The preheat step is not universally beneficial; excessive preheating above 160 °C initiates rapid crystallite growth and may reduce the web's tear resistance, impairing its ability to bridge cracks in the finished membrane. The correct preheat temperature depends on the carrier's initial crystallinity, which can be determined by differential scanning calorimetry at a heating rate of 10 K/min.

Because carrier basis weight and bonding pattern interact with shrinkage, heat transfer into the web depends on mass per unit area and bonded points. Point-bonded spunbond PET with a bond area of 15 % to 25 % conducts heat through the fused regions faster than through the lofted fiber matrix. The impulse heat seal at the calender creates localized crystallinity gradients; the bonded points shrink less than the surrounding unbonded fibers, producing a crenellated surface when the web is heated above 140 °C. A heavier web of 200 g/m² requires longer thermal penetration than a 100 g/m² web, so line speed must be reduced or preheat temperature increased. Air permeability measured according to ISO 9237 at 200 Pa is used as an indirect indicator of bitumen penetration and heat transfer; carriers with air permeability below 200 L/m²/s may show incomplete saturation and high residual shrinkage.

Carrier Web Specification Limits for SBS-Modified Bitumen Impregnation

SBS-modified bitumen membranes typically use a polyester carrier with mass per unit area from 80 g/m² to 180 g/m². The lower limit is set by the need to withstand tensile forces during installation and by the crack-bridging requirement in standards such as EN 13707. The upper limit is set by bitumen saturation and line economics. Table 1 lists typical incoming carrier acceptance limits used on a high-speed SBS membrane line.

In SBS-modified compounds, the processing temperature of the polymer-modified bitumen is not a single value but a window defined by viscosity and thermal stability. For SBS-modified compounds, viscosity at 180 °C measured by rotational viscometer per ASTM D4402 typically falls between 0.6 Pa·s and 2.5 Pa·s; for APP-modified compounds, the range is 0.4 Pa·s to 2.0 Pa·s. Lower viscosity improves carrier wetting but increases heat transfer coefficient, accelerating web heating and shrinkage. A compound containing 25 % to 35 % mineral filler has higher thermal conductivity than unfilled bitumen, further shifting the heat flux into the web. Therefore, shrinkage control cannot be separated from formulation: a reduction in filler content from 35 % to 25 % may allow the same carrier to pass the dimensional stability test at a lower preheat temperature.

Controlled Cooling Drum Differential Speed and Its Effect on Residual Shrinkage

After the bitumen impregnation nip, the composite membrane is cooled on a series of chilled drums. The first cooling drum is typically held at 15 °C to 25 °C, and the second at 10 °C to 15 °C, with differential speed between the pull roll and the winder used to control final tension. Applying a slight overdrive of 0.5 % to 1.5 % on the cooling section draws the membrane forward and compensates for contraction during bitumen solidification. The cooling rate must be controlled because quenching below the PET glass transition while the bitumen matrix is still soft can lock in shrinkage stress. Cooling drums with independent servo drives allow the operator to program a speed profile that reduces machine-direction tension to below 100 N/m before winding, preventing the roll from becoming telescoped due to core migration.

To achieve low shrinkage, PET grades are produced by increasing the degree of crystallinity and by heat-setting the spunbond web. Heat-setting involves heating the web under restrained conditions to a temperature between 180 °C and 220 °C for 10 s to 60 s while the web is held on a tenter frame. This process reduces free shrinkage but may also reduce elongation at break, which is undesirable for membranes that must accommodate building movement. When a low-shrinkage grade with free shrinkage below 1.5 % at 180 °C is used, the process window for molten bitumen temperature can be widened by approximately 10 °C without exceeding the finished membrane dimensional stability limit. However, published data for the interaction between specific tenter heat-setting profiles and SBS bitumen adhesion is limited; factory trials are required to verify that the higher crystallite surface does not reduce peel adhesion.

On a high-speed membrane line, online monitoring is necessary because carrier shrinkage develops within seconds and cannot be corrected after the roll is wound. A typical line uses an infrared pyrometer aimed at the incoming web just before the bitumen coater, with a temperature setpoint of 70 °C to 110 °C. A second pyrometer after the cooling drums measures the membrane surface temperature before winding, which should be below 35 °C to prevent blocking. Machine-direction tension load cells provide a closed-loop signal to the unwind brake or servo; alarm limits are set at ±15 % of the nominal tension. Width cameras at the winder monitor edge position; a width loss of more than 0.5 % triggers an immediate line slowdown and notifies the quality control station. These systems record data at 1 Hz and allow root-cause correlation between shrinkage failures and process excursions.

Residual shrinkage measurement on the finished membrane is performed according to EN 1107-2, which measures the change in length and width after exposure to elevated temperature. For reinforced bitumen membranes, the test temperature is selected from the product standard and is commonly 80 °C or 100 °C depending on the product class; specimens are cut from the roll and conditioned at 23 °C and 50 % RH before measurement. A residual shrinkage value greater than 0.5 % in either direction is often considered nonconforming for high-quality polyester-reinforced products, although declared values vary by manufacturer. The test does not distinguish between carrier shrinkage and bitumen flow, so a complementary carrier test at 180 °C is necessary to isolate the source of failure.

If relative humidity during carrier storage exceeds 60 %, moisture in the PET carrier becomes a critical accelerator of heat-induced degradation and shrinkage variability. PET is hygroscopic and can absorb up to 0.4 % moisture at 23 °C and 65 % RH. When the web enters the molten bitumen, trapped moisture undergoes rapid vaporization, causing localized fiber hydrolysis and reducing molecular weight. This degradation lowers the carrier's tensile strength and can increase shrinkage by releasing chain entanglements. Therefore, carrier rolls stored in warehouses with relative humidity above 60 % for more than 24 h should be pre-dried at 120 °C for 4 h or until residual moisture is below 0.05 %. Incompatibility with highly alkaline fillers should be assessed; exposure to calcium hydroxide at high pH accelerates PET surface hydrolysis.

Table 2 summarizes the critical control points, the measurement method, and the corresponding acceptance criteria for a polymer-modified bitumen membrane line running a spunbond PET carrier.

At the end of the process, the operational limits of spunbond PET carrier shrinkage control are defined by the compromise between thermal relaxation, mechanical strength, and bitumen adhesion. Excessive heat-setting reduces shrinkage but also reduces elongation and may compromise crack-bridging capability under EN 13707 or low-temperature flexibility under EN 1109. The use of an inline annealing tunnel adds capital cost and energy load, with medium-wave IR panels consuming up to 40 kW/m². In addition, PET carrier webs that contain recycled polymer may exhibit batch-to-batch variation in intrinsic viscosity, causing shrinkage values to vary by up to 1 % between lots. Incoming inspection therefore requires a combination of thermal shrinkage testing, differential scanning calorimetry, and tensile testing. Without closed-loop tension control, online temperature monitoring, and a defined carrier specification, a polymer-modified bitumen membrane line cannot maintain the dimensional stability required by harmonized product standards.