3D Printing with Fused Filament Fabrication and Fused Granulate Fabrication (Pellet Printing)

By Greg Costantino, Covestro Additive Manufacturing, and Zac DiVencenzo, JuggerBot 3D

Fused Granulate Fabrication (FGF) — also known as pellet extrusion printing, pellet printing, fused pellet fabrication or fused particle fabrication — offers production speeds that are up to 200x faster than Fused Filament Fabrication (FFF). Undoubtedly, some large applications that were previously not economical with FFF (i.e., too slow to print at large scale) will become economical with FGF.

That said, FGF printing isn’t appropriate for all designs. This white paper helps explain when a product design is best suited for either FFF or FGF.

We will first review the fundamentals of 3D printing for both FFF and FGF, their polymer portfolios, material handling and drying. We will also discuss the ways in which FGF differs from FFF and how to decide between the two technologies for an application design. Next, we will review critical processing variables for FFF and FGF printing and the key operator decisions needed for successfully printing parts.

Finally we will discuss testing FFF and FGF parts vs. injection molded parts — both how the tests were run as well as reviewing test results.

Fundamentals of FFF and FGF Additive Manufacturing

Additive manufacturing, or 3D printing, is a method of creating objects that offers new possibilities for part designs with complex geometries — designs that are beyond the capabilities of traditional processes such as machining, casting and molding. As the name implies, additive manufacturing creates an object by adding material — polymers and metal most commonly though also concrete and medicine for example —layer by layer. In addition to final part production, 3D printing is particularly useful for rapid prototyping and tooling.

There are multiple 3D printing technologies, including Fused Filament Fabrication and Fused Granulate Fabrication (sometimes better known as “pellet printing.”)

For FGF printing, thermoplastic pellets are heated in a barrel and screw extrusion system and then deposited through a nozzle to print the layers of a part.

Feedstock Both filament and pellet 3D printing thermoplastics start out as granules. FFF materials are made by heating granules and forming them into a filament. FGF pellets skip this step — one less heating cycle potentially means the pellet thermoplastics will process easier, perform better and cost less than comparable filaments.

Throughput — 3D printing throughput is measured by the rate of flow of the extruded thermoplastics. Throughput is usually measured in grams/kilograms or pounds per hour. FFF printers have smaller heater zones that limit their throughput, with typical throughput being 2.27 grams/hr to 113 grams/hr (0.005 lbs/hr to 0.25 lbs/hr). FGF printers, on the other hand, are limited by the size of their extruder and screw design, with typical throughputs being 227 grams/hr to 9 kg/hr (0.50 lbs/hr. to 20 lbs/hr).

Resolution — 3D printing resolution refers to the height of the layer and the width of its bead. With FFF printing, thin filaments and small nozzles mean higher resolutions. Typical layer heights range from 0.15 mm (0.006 in) to 0.4 mm (0.016 in) and bead widths from 0.3mm (0.12 in) to 1.00 mm (0.04 in). Notably, FGF printing has lower resolutions, with typical layer heights ranging from 1mm (0.04 in) to 5mm (0.2 in) and bead widths 2mm (0.08 in) to 10mm (0.4 in).

Print dimensions — Because of its slower throughput and higher resolution, filament printing tends to be used for smaller parts: a 91 x 91cm (36 x 36 in) part is possible using FFF, but it may not be economical or efficient. Pellet printing tends to be used for larger parts that don’t need high resolution and benefit from FGF’s faster throughput; a 91 x 91cm (36 x 36 in) part could print 200 times faster using FGF instead of FFF, depending on the printer and the material.

Print complexity — FFF printing allows for more detailed parts and complicated features due to its smaller nozzles and bead width. Conversely, FGF-printed parts have fewer features and lower complexity due to FGF’s lower resolution and wider extrusion paths.

Tool path For both types of 3D printing, the extruder’s programmed route is called the “tool path.” FFF printing allows for complex paths to build complex parts — however, one must be careful to limit excessive movements since that will reduce throughput. The tool path for FGF printing is simpler but requires more focus on part design and print orientation to avoid a tool path that crosses over itself.

Processing temperatures — Getting the material temperature just right is critical for successful additive manufacturing. For both technologies, a controlled, consistent temperature throughout the printer for the entire process is needed. Interlaminar bonding is crucial — it’s the biggest weakness within material extrusion printing — so it is important to make sure the deposited material bonds well with subsequent layers. Using a controlled and heated chamber allows the operator to prevent the parts from cooling too rapidly — which increases the strength of the printed part in the z direction. A heated build chamber ensures sufficient bonding, reduces warping and ultimately helps ensure accurate part dimensions.

For filament printing, the goal is getting more heat into the material to print faster and have each layer stick to the previous by both layers being in an equivalent temperature state. Pellet printing, on the other hand, may require cooling due to the printing speed and amount of thermal energy being deposited with each layer, potentially leading to sagging layers — causing poor quality or a failed print.

Nozzle sizes for FGF printing — The  design of the part, the 3D printing material, and the surface appearance will all factor into what size nozzle used. Using highly-filled, high viscosity material usually means small FGF nozzle sizes.

Materials for FFF and FGF Polymers

Thermoplastic materials used for fused filament fabrication and fused granulate fabrication are based on nylon, polyester or polypropylene chemistries. As manufacturers use 3D printing technology for functional and end-use parts, manufacturers of 3D printing material place more emphasis on higher performance polymers with higher mechanical and thermal properties. These polymers are better suited for more stringent industrial, transportation and electronics applications.

Covestro's filament portfolio currently features four nylon-based and three polyester-based materials.

Highlighted nylon-based materials include:

  • Novamid® ID1030 CF10: Carbon fiber-filled PA666 with 3.5 times the stiffness and 2.5 times the strength of Covestro’s standard PA666.
  • Novamid® AM1030 FR: Flame-retardant PA666, which has a UL® Blue Card Listing.

Highlighted PET polyester-based materials include:

  • Arnitel® ID2060 HT: High-temperature material, tested at 190°C for 500 hours, suitable for the most stringent under-the-hood applications.

Covestro’s pellet portfolio currently includes Arnite® AM8527 (G), a highly filled PET polyester, Arnilene® AM6001 GF (G), a glass-filled polypropylene and EcoPaXX® AM4001 GF (G), a bio-based polyamide PA410. Additional Covestro pellet grades will be launched in the near future.

Covestro material solutions — Thermoplastics

Some pellet materials will be better suited for FGF than others. With thousands of formulations available, it takes time to vet materials. The priority for process development is being set by applications. As material and printer companies work with manufacturers on new projects, more will be learned about which materials run well, why they run well and what makes viable solutions for additive manufacturing.

Soft and flexible parts can be printed with FGF, although there may be challenges with materials having a lower durometer (being softer) than 70 Shore D. Soft and flexible FGF parts will be subject to complications associated with using support material and post-processing, as discussed later. FFF printing may alleviate some of the challenges associated with printing soft and flexible materials. JuggerBot 3D’s filament systems, for example, incorporate a proprietary push-pull system that ensures control over the flow of material beyond other methods, and mitigates clogging issues common with soft and flexible filaments.  

Material Handling and Drying of FFF and FGF Materials

Feeding mechanisms — FFF printing uses 1.75mm or 2.85mm (0.07 or 0.11 in) diameter filament strands wound onto spools. When loaded into the 3D printer, the spool is unwound as the filament is extruded. Conversely, FGF pellets are first loaded into a drying hopper, typically outside of the 3D printer. The printer’s sensors then “call” for pellets, which are conveyed through corrugated tubing to a smaller secondary hopper.

The importance of drying — 3D printing polymers are either hygroscopic or non-hygroscopic. Hygroscopic materials seek to absorb moisture internally; they include nylon, ABS, PET and polycarbonate. Non-hygroscopic materials, on the other hand, can collect moisture on their surfaces; they include PVC, polypropylene, polystyrene and polyethylene.

Moisture in 3D printing polymers affects extrusion, part appearance and part performance. Moisture can affect surface finish and cause discoloration, lower mechanical properties (strength and elongation to break), change electrical properties and lower the material’s viscosity.

Moisture is easily noticed with some 3D printing materials: excessive drooling from the barrel, bubbles on the surface of extruded material and off-gassing of trapped water vapor. However, some materials don’t display moisture problems until printing, with delamination being a major issue. Both FFF and FGF materials must be properly stored and dried before 3D printing. Filament comes in vacuum-sealed packages, but the operator still must dry the spool per the polymer’s specific drying procedure. Pellets also come in vacuum-sealed packages of various sizes but, again, the operator must properly dry the material before printing.

Which 3D Printing Technology to Use: FFF or FGF?

  1. Determine the polymer material suitable for the application.
  2. Is that polymer available as filament or pellet, both or neither?

If both filament and pellet materials are available, consider the part design (see below).

Process considerations for FFF — Appropriate for applications that:

a) Are small to medium sized, typically less than 3’ x 3’ (90 x 90 cm)

b) Have thin walls

c) Have small features

d) Have small holes

e) Have complex designs needing FFF’s smaller diameter nozzles, smaller bead widths and smaller layer height

Process Considerations for FGF — Appropriate for applications that:

a) Are medium to large, typically greater than 2’ x 2’ (60 x 60 cm) in size

b) Have thick walls

c) Are low complexity

d) Have no overhangs (don’t require support materials)

If printing larger-size parts, the faster printing time of FGF (up to 200x faster) could be the deciding factor.

FGF printing also lends itself to net shape printing: do fast pellet printing, then machine the part down to its final, higher-tolerance dimensions.

Successful Applications in FFF and FGF

Both filament and pellet extrusion technologies work well for tooling, specifically for patterns and sand casting, as well as molds for thermal forming. In addition, FGF is suitable for production of parts with higher-performing thermoplastics — like dies for sheet metal forming. FFF printing works well for jigs and fixtures, specifically with routing fixtures and masking fixtures and robotic tooling like end-of-arm tooling.

As FGF printers like the JuggerBot 3D P3-44 have the ability to print up to 200x faster, production printing is growing fast, especially in the aerospace, automotive, medical and energy industries. Plus, 3D printing provides a competitive edge: design freedom and flexibility, the ability to produce parts on demand and mass customization.

Critical Processing Variables for FFF and FGF

The critical variables that dictate flow rate, throughput, part strength, and quality are:

  • Nozzle diameter
  • Bead width
  • Layer height
  • Temperatures
  • Speed

After determining these variables, use slicing software to simulate the tool path, looking for any problems in each layer to ensure the features can be efficiently produced. Select 3D printers with robust gantry systems for high precision and repeatability; then “dial in” the critical variables to assure the part’s dimensional tolerances.

Bead width is considered optimized if the wall thickness can be evenly divided by the bead width. Bead width must be smaller than the minimum wall thickness: if a minimum wall thickness is 0.4mm (0.016 in) and the nozzle is 1mm (0.04 in), the tool path could completely miss that section when printing. Current software is not intelligent enough to alert the operator to this mistake; operators must catch this mistake during software simulations. Attempting to print beads smaller than the nozzle is called “under-extruding” and it is not consistent enough to hold any particular tolerance.

Layer height is considered optimized if the total part height can be evenly divided by the layer height. Layer height is derived from nozzle diameter. A good start for layer height is 60% to 80% of the nozzle diameter; for example, a 1mm (0.04 in) nozzle will deliver a layer height between 0.6mm (0.02 in) and 0.8mm (0.03 in).

Extrusion/flow rate After determining bead width and layer height, establish a flow rate that ensures accurate part dimensions. The goal is a consistent material flow rate with a temperature that melts and bonds the material to the previous layers but is rigid enough to hold itself up during printing. For the desired bead width, the RPMs of the FFF rollers or FGF screw feeder must be matched to the speed of the gantry.

Temperature — Filament and pellet manufacturers provide a temperature range for operator guidance. Providing exact temperatures is not possible, since different printer manufacturers use different equipment, and temperature sensors can provide different readings for the same material. Start at the low end of the material manufacturer’s temperature range: the goal is consistent flow and bead diameter. 

If flow is interrupted, first check the feed mechanism, then increase the temperature to reduce high pressures in the extruders. Thin-looking or drooling extruded material could mean the temperature is too hot.

Selecting gantry speeds After establishing extrusion flow, select the gantry speed to determine the accuracy of the bead width. A gantry speed that is too fast for the RPMs for the extruder will cause the bead to thin out and be smaller than needed (aka under-extrude). A gantry speed that is too slow will cause the bead to enlarge (aka over-extrude).

Carbonizing thermoplastic 3D printing material — Improperly heated material, or material that sits too long at too high of a temperature, degrades the polymer. A filament 3D printer’s small heating section means it is less likely to degrade much of the material; if it does, a change in nozzle or extruder may be needed. Pellets are much more vulnerable to degradation due to heat from friction in the screw feeder; repairs are much more costly and take longer than repairs for FFF degradation.

Jumping to a new start point without oozing of material — When jumping to a new start point in filament printing, it’s natural to set an open-source machine to do a retraction: retract the filament and reduce the pressure so it doesn’t ooze, then over-extrude at the start. In pellet printing, retracting isn’t possible (cannot turn the screw backwards), however there are a few techniques to address this within the slicing software. One technique is to appropriately coast into the starts and stops. This is essentially reducing RPMs of the screw or turning off the screw 6 to 10 mm before the end of segment. Then to rebuild pressure, slow down the gantry to allow pressure to build up to allow a clean start, or increase RPMs for a short time. Using these techniques are band-aids: oozing is typically connected to temperature — it’s important to match the temperature with the extrusion process itself and the material selected.

Supports and Post-Printing Options

Supports In FGF printing The use of supports in FGF printing can be complicated for the following reasons:

  • Pellet printing favors larger parts using larger format printers. For example, the JuggerBot 3D’s Tradesman Series™ P3-44 FGF printer has a printing space of 3’ x 3’ x 4’ (91 x 91 x 122cm) — larger than most chemical bath basins required to remove dissolvable supports.
  • Because FGF beads are higher volume and printed at higher temperatures, removing supports can be quite difficult and require a tool (e.g. channel locks).
  • To avoid the use of supports with pellet printing, try to eliminate any overhangs through optimum orientation of the printer.
  • If supports are used with pellet printing, it’s best to machine them away during post-processing.

Dissolvable supports —Water-soluble supports can also be an option; they dissolve after sitting in a bath of warm water for several hours. In applications where water-soluble support materials cannot be used, materials like high impact polystyrene (HIPS) are used to dissolve the supports.

Final finishing of 3D printed parts When it comes to achieving an immaculate finish with high dimensional tolerance, light facing or machining of the printed part is an option — though more expensive and time consuming. Other traditional methods to improve surface finish — to make it look like an injection molded part — could be abrasive blasting, matte finishing (like tumbling and polishing), buffing and sanding. The final finishing could be applying coatings or paint.

Testing FFF vs. Injection Molding

Novamid® ID1030 CF10: FFF tensile strength results — Tests were run at room temperature using various bead widths, or extrusion widths, while keeping the layer height constant. On-edge specimens had higher proprieties than upright specimens:

  • Modulus values were 3 to 3.5x greater
  • Stress values were 2 to 3.5x greater
  • As the extrusion width changed for the on-edge specimens, stress values were very similar (varying only 5.4 MPa) and strain values were virtually unchanged
  • As the extrusion width increased for upright specimens, stress values deviated by 26 MPa and strain values by 50%

Preparation of FFF test specimens

  • “Upright” samples start as 3D printed hexagonal shapes, then test bars are milled from each face of the hexagon. Upright samples indicate the z-axis strength (the interlayer adhesion) since the pull direction is perpendicular to the print direction.
  • “On-edge” samples start as 3D printed rectangular shapes, then test bars are milled from each face of the rectangle. On-edge samples exhibit greater fiber orientation since the filament layers are in direct alignment with the pull direction.

Explaining the test differences: on-edge vs. upright specimens — On-edge specimens had much greater fiber orientation since the filament layers were in the same direction as the pull.

Upright specimens, on the other hand, were pulled perpendicular to the direction of the filament layers; micro-voids between layers could create areas of weakness. For upright specimens, wider beads embodied higher levels of energy, which resulted in better bonding and higher stress values.

Arnitel® ID2045: FFF vs. injection molding — Makers of additive manufacturing materials are focused on creating polymers that, when 3D printed, have properties that are similar to injection molded parts made from similar materials.

The following graph shows tensile strength testing of bars printed in four orientations with Arnitel® ID2045, as well as the strength of injection molded bars using a similar grade of material. The key takeaway is that Covestro continues to improve the correlation of properties between its 3D and injection molding polymers.

Arnitel® ID2060 HT: heat aging tests on high temperature FFF samples — Unlike some polymers, Arnitel® ID2060 HT has higher thermal performance and mechanical properties after heating aging up to 190°C/375°F. The following graph shows the change in modulus during heat aging at 175°C/347°F for 1000 hours compared to a similar injection molded grade; note the correlation of performance across most of the heat age.

Testing FGF vs. Injection Molding

Preparation of FGF test specimens — Four-sided cubes were 3D printed and then machined into tensile strength specimen bars. The layer thickness varied depending on the goal of the tests, and the size of the cube was altered depending on the number of samples needed.

FGF stress-strain test data on Arnite® AM8527 (G) was generated from bars milled in the x and z axis, as shown the previous graphic. Bars in the print direction show significantly higher maximum stress values. The 3D printed stress-strain results are highly correlated to the injection molded results. The slightly higher properties of the 3D printed bars are most likely due to the better overall alignment of fibers in the printed beads versus those of the injection molded bars.

Summary of Technical Learnings

  • Operators for either FFF or FGF machines must be sure to adequately dry materials before printing.
  • Consider the distinct process advantages/disadvantages for both FFF and FGF when evaluating which technology to utilize.
  • Be sure to optimize critical process variables, including: (1) bead width, (2) layer height and (3) flow rate.
  • Understand how each process parameter impacts part performance, including:
  • Part orientation – “On-edge” performs better than “Upright”.
  • Extrusion (bead) width – Larger beads may correspond to higher stress values.
  • 3D printed parts can exhibit either decreased or increased performance (vs. injection molding) depending on the combination of materials, print parameters and process optimization.

How can we help?

Contact our experts to discuss how Covestro can help you tap into the full potential of additive manufacturing.