BASF 3D Ultrafuse PET CF15 Fused Filament is a chopped-carbon-fiber-filled semi-crystalline polyethylene terephthalate feedstock for fused filament fabrication. The product designation identifies 15% carbon fiber loading by weight, with the balance a PET matrix; the filament is black and is supplied in 1.75 mm and 2.85 mm diameters with a nominal net weight of 750 g per spool. Published density is 1.31 g/cm³ to ISO 1183-1. The matrix is not glycol-modified PETG, so the drying and extrusion temperatures are higher than those of many unfilled PETG grades, but moisture control is still required to prevent hydrolysis. The carbon fiber raises melt viscosity, raises tensile and flexural modulus, lowers elongation at break, and converts the feedstock into an abrasive medium that is incompatible with brass and aluminum nozzles over sustained use. Supplier documentation does not assign a direct food-contact approval under EU 10/2011 or FDA 21 CFR to this carbon-filled grade; the intended use is industrial tooling, fixtures, vacuum plates, and low-draft structural parts.

What Mechanical Property Envelope Is Reported for Printed XY Specimens?

Supplier-reported tensile and flexural results for Ultrafuse PET CF15 are generated on vertically built XY specimens conditioned for 24 h before testing. Tensile modulus is reported between 5,200 MPa and 6,200 MPa under ISO 527-2, while tensile strength is reported at 50–65 MPa in the same orientation. Flexural modulus tested to ISO 178 is typically 4,800–5,500 MPa, and flexural strength is cited at 80–95 MPa. Charpy unnotched impact strength tested to ISO 179-1/1eU is approximately 20–30 kJ/m², which is lower than many unfilled PETG compounds. Heat deflection temperature under 0.45 MPa load to ISO 75-2/B is reported near 120–135 °C, while the 1.80 MPa deflection temperature to ISO 75-2/A is lower, typically 70–85 °C. This distinction is relevant in heated fixture services: a 85 °C build chamber or operating environment approaches the 1.80 MPa HDT threshold, whereas the 0.45 MPa value permits more margin. Published Z-axis tensile data for this specific carbon-filled filament is limited, but filled and unfilled PET systems commonly show 30–50% lower interlayer bond strength than their XY tensile strength when tested to ISO 527-2 on vertical coupons.

On an unenclosed Cartesian FFF machine with a hardened steel 0.4 mm nozzle and a polyetherimide build surface, a reproducible starting profile uses an extruder set point of 250 °C, a build-platform set point of 85 °C, a line width of 0.45 mm, and a volumetric flow rate below 4.5 mm³/s. The extrusion multiplier is typically reduced by 2–5% relative to an unfilled PET profile because fiber-filled melt is less compressible and the road cross-section is more stable. Linear speed should remain below 60 mm/s for the first layer and below 45 mm/s on tall vertical walls if corner lifting appears. Retraction start settings of 1–2 mm on direct-drive and 4–6 mm on Bowden toolheads are common, but the carbon-filled melt is shear-sensitive. Direct-drive retraction distances above 3 mm can create pressure spikes that delay waypoint starts and disturb top-layer fill. A 0.4 mm brass nozzle can be visibly enlarged after fewer than 500 g of this feedstock; hardened steel or ruby nozzles are required for dimensional control. Bed adhesion is maintained with an 85 °C build platform and an adhesive suited to PET; textured polyetherimide surfaces are preferred over smooth glass to limit edge lifting on parts longer than 120 mm. An enclosure is not mandatory, but a draft shield or chamber air temperature above 25 °C reduces curl on long straight walls.

Production-line failure records for carbon-filled PET in FFF systems identify three recurring modes: nozzle inlet bridging from fiber bundles during aggressive retraction, first-layer delamination when the build platform drops below 75 °C, and interlayer splitting at abrupt cross-section changes. The fiber bundles can temporarily lodge at the nozzle entrance if the filament is retracted too far and re-fed; a polished steel heat break and constrained filament path reduce this frequency. Batch-to-batch variation is generally low in sealed spools, but re-spooling or rough handling generates carbon dust that can pack the extruder hobb teeth and reduce feed consistency. In a 3:1 gear-ratio direct-drive extruder, idler tension should be set lower than for unfilled PET; excessive radial compression can crush the filament, producing longitudinal splits that open during retraction. The processing thermal window is narrow enough to require control: below 240 °C, melt viscosity rises and the extruder force increases, while above 270 °C, oxidative and thermal degradation produce black speck and loss of mechanical strength. This 30 °C window is usable on production machines only when drying, flow rate, and retraction are controlled together.

When the Spool Is Unsealed in a High-Humidity Production Area

Absorbed water in the melt phase hydrolyzes the ester linkages of the PET backbone. At a moisture content above 0.04%, the melt can exhibit nozzle bubbles, splay, reduced surface gloss, and a lower effective viscosity from chain scission. A circulating-air dryer at 60 °C for 4–6 h is the standard pre-drying cycle for spools stored in sealed containers. Spools exposed to >60% RH for more than 48 h may require 8–12 h at 60–80 °C. Drying above 80 °C is not recommended because the spool body may soften and release residual stress. Once dry, the filament should be fed from a sealed desiccant box with a dew point below -20 °C on prints longer than 12 h. In a machine-shop environment, the outer windings of a spool left mounted overnight can re-absorb enough moisture within 8–12 h to create defects in the next build; the spool should be returned to dry storage or placed in a heated dry box after use.

Because the fiber loading preferentially orients in the raster direction, the XY stiffness gain is not replicated in the Z direction. Tall thin fixtures fail by interlayer delamination or cleavage, not by tensile fracture. Unfilled PETG grades normally display yielding and whitening before rupture; PET CF15 has an XY elongation at break below 5% and fails more abruptly. Snap fits, thread-forming screw bosses, and living hinges designed for unfilled PETG should not be transferred to PET CF15 without revising radii, wall thickness, and insertion strain. The material is better placed in compression-loaded locating nests, vacuum-tooling plates, inspection gauge frames, and stiff end-of-arm brackets where bending deflection is the controlling rejection criterion. The carbon fiber reduces the coefficient of thermal expansion along the raster direction, improving dimensional agreement with steel reference blocks in precision assembly stations, but the through-thickness expansion remains PET-dominated. For parts longer than 150 mm, published part-specific distortion data is limited; build trials with a draft shield or heated enclosure are required because shrinkage reduction is not isotropic.

Mechanical post-processing of PET CF15 parts uses carbide tooling. High-speed steel drills and taps wear rapidly because the carbon fiber acts as a hard phase at the cut zone. Heat-stake inserts are preferred over directly tapped threads in load-bearing bosses. The printed surface carries fine axial ridges that reduce paint and epoxy primer adhesion; solvent polishing with dichloromethane or other solvent systems is not effective because the semi-crystalline PET matrix resists uniform solvent penetration. Painting trials require cleaning with isopropanol to remove build-surface adhesives and then light abrasion with 240–320 grit aluminium oxide paper. Drilling should be performed at low spindle speed with coolant or compressed air to reduce matrix melting at the hole wall. Published guidance for machining feed and speed combinations specific to 15% carbon-filled PET filament is limited; initial machining parameters should be derived from glass-filled PET or established carbon-filled PET injection-molding practice and then reduced to account for the lower density of layered polymer structure.

Material Selection Trade-Offs Against Unfilled PETG and Carbon-Filled Polyamide

For parts that require continuous contact with high-humidity air or polar process fluids, PET CF15 differs from carbon-filled polyamide. PAHT CF15 often provides higher dry heat deflection and greater damage tolerance, but its dimensions and strength shift with absorbed moisture. PET CF15 has lower moisture uptake and a more stable ambient envelope after conditioning. Compared with unfilled PETG, the tensile modulus is more than doubled, but elongation at break is reduced and the feedstock is abrasive. Compared with ABS CF, PET CF15 generally produces lower styrene emission during printing, higher modulus, and a different chemical resistance profile; side-by-side published data for these exact formulations is limited, particularly for long-term chemical exposure evaluated to ISO 175. The product is not a reliable electrical-conductivity compound: the discontinuous carbon fiber is not controlled for surface resistivity under ESD packaging standards, so additional antistatic handling measures are required. For direct food-contact applications, the grade is not indicated for compliance; a validated unfilled PETG or PET with explicit food-contact certification should be used when that constraint exists.

Selection should not be made on tensile modulus alone. A fixture near a machining-centre coolant spray has a moisture-tolerant requirement that favors PET CF15 over PAHT CF15, but the same fixture at 150 °C dry heat would exceed the 0.45 MPa HDT of PET CF15 and require a polyamide-based carbon-filled grade or a different polymer family. In low-volume prototyping without a hardened nozzle, unfilled PETG may be selected even though its modulus is lower, because the production cost of nozzle replacement and filament abrasion is removed. For inspection equipment with tight thermal flatness, PET CF15 is appropriate when the ambient temperature remains below 70 °C under load; above that boundary, published data for long-term creep and flatness retention under sustained fixture load is limited.