Stringing in Engineering Filaments: Complete Troubleshooting

Printing regular PLA or PETG? Start with the general guide Stringing: Why Your Printer Pulls Strings — this article covers engineering materials (CF/GF composites, nylon, ABS).

Stringing is one of the most common and most frustrating defects in FDM printing. Thin filament strands that connect different parts of a model or fill the space between towers can ruin an otherwise successful print, and in engineering filaments such as PETG, ABS, ASA or nylon they are significantly more stubborn than in regular PLA. This guide systematically walks through every lever you can pull: retraction, temperatures, speeds, slicer settings and hardware. The focus is on materials with higher print temperatures and higher hygroscopicity, where standard recipes for PLA fail and a different approach is required.

What stringing is and why it occurs more often in engineering filaments

Stringing, sometimes called oozing or hairy prints, is a defect in which molten filament escapes from the nozzle during travel moves, that is, at the moment when no material should be extruded. The result is thin web-like strands connecting individual parts of the model, or a fuzzy surface on towers and pillars. The defect usually does not affect the structural integrity of the part, but it worsens aesthetics, complicates post-processing, and in functional prototypes with moving elements it can even cause mechanical issues.

Engineering filaments are more prone to stringing for several reasons. PETG has lower viscosity in the molten state and high adhesion, so it pulls strands easily. ABS and ASA print at temperatures around 240 to 260 degrees Celsius, which means a longer time during which the material remains liquid after retraction. Nylon and PA-CF composites additionally combine higher print temperatures with strong hygroscopicity, so absorbed moisture in the material turns into steam inside the nozzle and literally shoots material out. With PLA it is often enough to lower the temperature by 5 degrees and the problem disappears; with engineering filaments you have to address several factors at once.

The material science behind this phenomenon is relatively straightforward. Polymers lose viscosity non-linearly when heated, so small changes in temperature can significantly influence the behavior of the melt. Polymaker’s documentation emphasizes that understanding the behavior of a specific polymer is the key to solving stringing, and this approach works particularly well with engineering materials.

Moisture in the filament as the primary suspect

With engineering filaments, drying should be the first thing a printer operator checks. Hygroscopic materials such as nylon, PA-CF, ABS, ASA and PETG absorb airborne moisture quickly, sometimes within a few dozen hours in a normal room. When wet filament passes through a nozzle heated to 240 degrees Celsius or more, the water contained in the material instantly turns into steam. The steam shoots small droplets of plastic outward and at the same time creates microbubbles, which then literally sputter material to the sides during travel moves.

The characteristic symptom of wet filament is crackling or popping in the nozzle, visible steam around the hotend, and unstable extrusion. Polymaker’s documentation states that if you hear popping or cracking during extrusion, the filament most likely has excessive moisture content and must either be dried or replaced with a dry spool. SUNLU, as a filament manufacturer, lists material quality among the main causes of stringing, where quality also includes moisture content at delivery.

Drying should be done in a dedicated dryer or in an oven at a temperature below the glass transition of the material. For PETG, 65 degrees Celsius for 4 to 6 hours is typically sufficient; ABS and ASA tolerate 70 to 80 degrees Celsius for the same duration; nylon requires 70 to 80 degrees Celsius for 8 to 12 hours. PA-CF and other carbon composites usually need even longer drying, often 12 hours or more. After drying, ideally print directly from a drybox system with active desiccant, otherwise the material will start absorbing moisture again within a few hours.

• PETG: 65 °C for 4 to 6 hours
• ABS and ASA: 70 to 80 °C for 4 to 6 hours
• Nylon: 70 to 80 °C for 8 to 12 hours
• PA-CF and carbon composites: 80 °C for 12 hours or more
• After drying, ideally print from a drybox with active desiccant

Retraction settings: distance and speed for different extruders

Retraction is the process in which the extruder pulls a certain amount of filament back into the hotend before the nozzle moves to the next location. The reduced pressure in the nozzle limits the leakage of material during the travel move. Optimal values differ depending on the type of extruder and the material, so it is not possible to apply a single universal recipe.

For a direct drive extruder, which is dominant in most modern printers such as the Bambu Lab X1C, Prusa MK4 or Creality K2, a retraction distance in the range of 0.5 to 2 mm works well. MatterHackers recommends starting at 0.8 to 1 mm and gradually increasing. With a Bowden setup, where the filament is guided through a long tube between the extruder and the hotend, the required values are significantly higher, typically 2 to 6 mm. The Prusa MINI with a Bowden setup has 3.2 mm as the default.

Retraction speed determines how quickly the motor pulls the filament back. Too slow a retraction cannot reduce the pressure in the nozzle quickly enough; too fast a retraction causes the motor to skip or wears down the filament. The standard range is 25 to 45 mm per second, and MatterHackers confirms this range as a safe default zone. For engineering filaments, a higher retraction speed usually helps, but you must always check whether the extruder is starting to grind the filament.

For flexible materials such as TPU, the situation is specific. The material stretches like rubber during retraction, so the Prusa Knowledge Base recommends a longer retraction than with rigid materials. With flex filaments, tuning retraction is more difficult and often needs to be combined with lower print speeds overall.

Temperature tower as the basic diagnostic tool

If this article were to recommend a single technique for tuning stringing, it would be the temperature tower. This test print contains several vertical sections, each printed at a different temperature, usually in steps of 5 degrees Celsius. After printing, you visually compare which section shows the least stringing while still maintaining sufficient layer adhesion.

The procedure is straightforward. You take the recommended temperature range from the filament’s manufacturer specs, for example 220 to 245 degrees Celsius for PETG, and the tower is printed in 5-degree increments. SUNLU suggests printing the temperature tower and the stringing test separately, because a tower usually also contains elements for evaluating stringing, but a standalone retraction test provides cleaner results. Loohney’s Basic Stringing Test, available on Thingiverse under number 2080224, has been used in the community for years and provides two pillars spaced so that the travel move between them reliably triggers stringing on poorly tuned profiles.

The typical optimal temperatures for the most common engineering filaments are in a narrower range than what the manufacturer advertises on the spool. MatterHackers lists 220 to 240 degrees Celsius as the baseline for PETG, 230 to 250 for ABS, and 190 to 210 for PLA. For PETG-CF and PETG-GF composites, you usually need to go 5 to 10 degrees higher due to the higher viscosity of fiber-reinforced material. For ASA the values are similar to ABS; for nylon and PA-CF the temperatures fall in the range of 250 to 280 degrees Celsius depending on the specific formulation.

Walk your specific problem through the wizard below.

Travel speed, wipe and coasting as secondary levers

Once retraction and temperature are dialed in, there are additional parameters that can eliminate residual stringing. Travel speed is the simplest optimization. Polymaker’s documentation states that on modern CoreXY printers such as Bambu Lab or Voron, you can set travel speed to 200 mm per second or higher, with travel acceleration around 2500 mm per second squared. The faster the nozzle moves between two objects, the less time the molten material has to ooze out. The limit is usually set by the machine’s mechanics and input shaper calibration.

Wipe is a function in which, after retraction, the nozzle moves back over the wall it just printed and wipes any leftover material against it. The Prusa Knowledge Base recommends leaving wipe enabled, and the same applies to retract on layer change. Wipe is especially effective with PETG, where the material’s characteristics cause a droplet to remain on the nozzle after retraction, which without wipe creates a visible blob at the start of the next segment.

Coasting is a technique in which the slicer replaces the last short stretch of an extruded path with a travel move, so the residual pressure in the nozzle completes the extrusion on its own. Polymaker’s documentation recommends coasting mainly for Bowden printers and for PETG. Be careful: coasting enabled without need can cause holes in walls at the end of segments, so it makes sense to apply it only when retraction alone is not enough.

Avoid crossing perimeters and only retract when crossing perimeters are two slicer settings that reduce the number of retraction cycles. The first optimizes the travel path so that it crosses outer perimeters as little as possible. The second disables retraction during travel moves within a single perimeter, so minor oozing is hidden inside the walls. For PrusaSlicer, OrcaSlicer and Bambu Studio, these options are available as standard in the Quality or Layers and perimeters section.

Hardware: nozzle, hotend and maintenance

Stringing can also originate in hardware, not just in slicer settings. The Prusa Knowledge Base emphasizes that if the same material is printed for a long time, for example PETG, a dried-on layer of polymer can build up on the nozzle. This layer then causes new stringing strands to stick to the residue on the nozzle during subsequent prints, creating visible smears. Thorough cleaning of the nozzle before each new type of filament, ideally with the cold pull method, is therefore a basic step.

The choice of nozzle is another factor. For abrasive materials such as PETG-CF, PA-CF or glass-filled composites, a hardened nozzle or ruby nozzle is a necessity, not a choice. A copper or brass nozzle wears out within a few dozen hours of printing with carbon-filled filaments, which leads to a widened orifice and worse flow control. A widened nozzle then contributes to stringing, because it is harder to achieve precise control over extrusion.

Hotend design influences how effectively temperature and pressure can be controlled. All-metal hotends are necessary for materials above 240 degrees Celsius, because a PTFE liner would degrade. For engineering filaments with high print temperatures, such as ABS, ASA, PC blends or PA-CF, an all-metal hotend is mandatory. A short distance between the melt zone and the nozzle reduces the amount of material that can escape after retraction. Hemera, Revo and similar modern hotends have this distance optimized.

Maintenance of the PTFE coupler, calibration of e-steps on the extruder, and checking that the extruder gear is not grinding the filament are additional points worth verifying with stubborn stringing. Incorrect e-step calibration means that the configured retraction distance does not match the actual filament movement, so even theoretically correct values do not produce the expected result.

• Clean the nozzle using the cold pull method before every material change
• Hardened nozzle for CF and GF composites, mandatory with abrasive filaments
• All-metal hotend for temperatures above 240 °C
• Calibrate e-steps and check extruder gear for wear
• Check the PTFE coupler on Bowden setups

Material-specific notes: PETG, ABS, ASA, nylon

PETG is the most common engineering filament and at the same time a material that can never be fully tuned to a complete absence of stringing. Polymaker’s documentation points this out directly, noting that with PETG some post-processing using a heat gun or a razor blade is usually unavoidable. The optimal temperature falls in the range of 230 to 245 degrees Celsius, retraction 1.5 to 2 mm for direct drive, wipe enabled, travel speed around 150 to 200 mm per second.

ABS tolerates higher temperatures and has a lower tendency to string than PETG, but it requires a heated chamber to eliminate warping. Optimal temperature is 240 to 255 degrees Celsius, retraction similar to PETG, but travel speed can be even higher. The Bambu Lab X1C with its enclosed chamber is well configured for ABS out of the box. ASA behaves similarly to ABS, except that it tolerates UV radiation better and is by default a bit less prone to warping.

Nylon is the most demanding of the common engineering filaments. Its high hygroscopicity means that even short exposure to airborne moisture dramatically increases stringing. Printing from an active drybox is practically mandatory. Depending on the type of nylon, the optimal temperature ranges from 240 to 280 degrees Celsius, with retraction usually longer than for other materials, around 2 to 3 mm for direct drive. The community reports that with nylon composites containing carbon or glass fibers, stringing can be reduced even more significantly by raising travel speed above 250 mm per second, if the printer’s mechanics can handle it.

TPU and other flexible materials are outside the main engineering filament category, but they are worth mentioning because their retraction profile is radically different. The Prusa Knowledge Base recommends a significantly longer retraction for flex due to the elastic stretching of the material, but at the same time warns that too long a retraction causes the filament to be torn out of the extruder gear. For TPU it usually pays off to combine a retraction of around 3 to 5 mm with slower print speeds overall.

A systematic tuning procedure: what to do when nothing helps

If the standard recipes fail, it pays to proceed methodically. The first step is always to verify that the filament is dry. Dried material, by various estimates, solves 30 to 50 percent of stringing problems with engineering filaments. Without this step, there is no point in tuning slicer settings, because every change may be masked by the influence of moisture.

The second step is a temperature tower with a range typical for the given material. The goal is to find the lowest temperature at which the material still adheres well to the previous layer and at the same time does not pull strands. With PETG, the optimum is often unexpectedly low, around 225 to 230 degrees Celsius, compared with the commonly recommended 240. The third step is a retraction tower with a step of 0.2 or 0.5 mm, which is used to find the boundary value for a specific extruder.

The fourth step is the optimization of other parameters: travel speed, wipe, coasting, avoid crossing perimeters. These settings usually bring the final 10 to 20 percent of improvement. The fifth step is checking the hardware if the problem persists: nozzle, hotend, extruder calibration. The last step is replacing the filament with a different brand, because some spools have quality issues that no setting will resolve.

An important principle of the entire process is to change only one variable at a time. Simultaneously changing temperature, retraction and travel speed makes it impossible to determine which factor actually produced the improvement. Keeping notes on what has been tested and with what result is standard practice for all serious 3D printer operators. MatterHackers recommends this approach explicitly as a pro tip.

Stringing in engineering filaments is not a single defect with a single solution, but a complex of phenomena with several parallel causes. A systematic procedure that starts with drying the filament, continues with a temperature tower, retraction calibration and optimization of slicer settings will eliminate stringing in most cases. For PETG and similarly prone materials, it holds that minor residual stringing is acceptable and post-processing with a heat gun will finish what settings cannot resolve. The key is patience, one variable per test, and careful notes.