Machining PEEK and Other Plastics: How They Differ from Metals, Cutting Conditions and Tools, Annealing, Dimensions and Plastic Gears
Plastics grow and shrink with heat and moisture. Measure right after cutting, and the size might be different tomorrow!


Plastics such as PEEK, POM and nylon (PA) can be cut on the same lathes and machining centers as metals. Yet people still run into trouble: the size changes after machining, holes come out small, parts crack, burrs won't come off. The reason is that plastics are materials that trap heat, spring back elastically, and expand and shrink with temperature and moisture. This article covers how plastics differ from metals, types of plastics and where to use them, cutting conditions and tools, cooling, annealing and dimensional stability, a calculator for dimensional change with temperature, workholding and burrs, tool wear with glass- and carbon-fiber-filled grades, and plastic gears and how they are used in humanoid and collaborative robot joints, based on technical documents from plastic manufacturers, JIS standards and research papers (as of October 2026).
How plastics differ from metals
Even though it is all "cutting," plastics behave differently from metals in the following five ways. Each one directly affects dimensions and surface finish.
| Difference | What happens | What to do on the shop floor |
|---|---|---|
| Heat doesn't escape | Low thermal conductivity and a low melting point mean heat builds up at the cutting edge and surface. The material melts, gets gummy, or keeps residual stress[1][3] | Clear chips quickly. Cool with air. Use a sharp edge that cuts rather than rubs |
| Soft and springs back | The elastic modulus is lower than metal's, so gripping hard distorts the part. The material springs back after the edge passes, so holes tend to shrink and ODs tend to grow[2][1] | Grip lightly and remove stock symmetrically. Aim ODs at the low side and holes at the high side |
| Expands and shrinks a lot | The coefficient of linear expansion is about 4–17 times that of steel (chart below). Hand heat or cutting heat changes the measured value | Measure after the part has cooled, close to 20 °C. Use the Calculators to estimate the temperature effect |
| Absorbs water | Nylon can absorb up to about 7% of its weight in water when immersed, changing dimensions by up to about 2%[5]. PEEK and POM absorb far less | For materials that absorb moisture, keep storage and measuring conditions the same. Use low-absorption plastics for precision parts |
| Cracks from chemicals | Amorphous plastics such as PC and PEI can stress-crack from mineral-oil-based cutting oils[1][4] | Use air or water-soluble coolant. If coolant is used, wash with IPA and deionized water |
Watch out for slow deformation, tooUnder sustained load, plastics keep deforming over time (creep). Seating faces under tightly torqued bolts, press fits, and clamp marks from long fixturing may not recover, so take measures such as adding metal collars at fastening points or widening the gripping surfaces.
types of plastics and where to use them
Here are plastics commonly machined from stock, compared by service temperature, expansion, water absorption and machining notes. The values are typical figures from plastic manufacturers and vary by manufacturer and grade. Detailed values and sources for each material are also in the Materials database.
| Plastic | Continuous service temperature (guide) | Coefficient of linear expansion (×10⁻⁵/K) | Water absorption (23 °C, immersed) | Machining note |
|---|---|---|---|---|
| POM-C (acetal) | 100 ℃ | 14 | 24 h 0.05%, 96 h 0.1% | The standard choice: no voids and easy to machine. Gears and sliding parts[16] |
| PA6 (cast nylon) | 100 ℃ | 12 | Saturated 6.5% | Dimensions move with water absorption. Large parts, rollers[17][18] |
| UHMW-PE | 80 ℃ | 18 | 24 h 0.01% | Soft and deflects, prone to welding onto the tool. Raise feed and speed to break chips away[22][1] |
| PTFE | 260 ℃ | 10–13 | 0.01% | Low strength and large expansion, so discuss tight tolerances at the design stage[20][21] |
| PC (polycarbonate) | 120 ℃ | 8 | 24 h 0.03–0.06% | Amorphous. Can crack from mineral-oil-based cutting oils[19][1] |
| PPS | 230 ℃ | About 6 (about 11 at 100–150 °C) | 0.01% or less | Expansion nearly doubles above the 97 °C glass transition. PCD for reinforced grades[11][12] |
| PEI | 170 ℃ | 5 | 24 h 0.05%, 96 h 0.1% | Amorphous (glass transition 216 °C). Watch for stress cracking from cutting oil[10][1] |
| PEEK | 260 ℃ | 5 (12 above Tg 143 °C) | 24 h 0.02–0.03%, saturated 0.45% | Expensive, so plan roughing setups carefully. Can be annealed at up to 250 °C[6][7][1] |
| PEEK GF30, CF30 | 260 ℃ | 4 (extrusion direction) | 24 h 0.02–0.03% | The fibers wear the tool. PCD or CVD diamond tools[8][9][1] |
| PAI (polyamide-imide) | 210 ℃ | 4.3 | Dimensions change with moisture absorption | Hard and strong. PCD for reinforced grades[13][14][1] |
| PI (polyimide) | 290 ℃ | 4.3–4.4 (50–200 °C) | 24 h 0.47% | Low cutting speeds of 25–100 m/min. PCD is best[15][1] |
Continuous service temperature, coefficient of linear expansion and water absorption are typical values from Ensinger product pages (POM-C: TECAFORM AH, PA6: TECAST T, PI: TECASINT 2011, PAI: TECAPAI CM XP403, etc.). Saturated water absorption of PA6 from MCAM; saturated water absorption and Tg of PEEK from Victrex.
How to choosePOM is cheap and easy to machine. Avoid PA where dimensions must hold around water. For areas above 100 °C or exposed to chemicals and steam, use PPS or PEEK; for even hotter areas or where strength is needed, PAI or PI. PEEK is often chosen for precision parts that need both dimensional stability and heat resistance because its linear expansion is small for a plastic, it absorbs little water, and it can be annealed at 250 °C[6][1].
Starting cutting conditions and tools
Below are values from plastic manufacturers' machining guides. They are all starting points; adjust them for the machine, workholding and part shape. The table from Mitsubishi Chemical Group (MCAM, formerly Quadrant) divides plastics into several groups[1].
| Plastic group (MCAM table) | Turning: speed, feed | Milling: speed, feed | Drilling: speed, feed | Tools |
|---|---|---|---|---|
| POM, PA, UHMW-PE, etc. | 200–500 m/min, 0.05–0.5 mm/rev | 200–500 m/min (POM up to 400), 0.5 mm/tooth or less | 50–100 m/min, 0.1–0.3 mm/rev | HSS or carbide |
| PC, PEI, PSU, PPSU | 200–400, 0.05–0.4 | 200–400, 0.4 or less | 50–100, 0.1–0.3 | Carbide |
| PEEK (unfilled), PAI (unfilled), PET | 200–400, 0.05–0.5 | 150–300, 0.4 or less | 50–80, 0.1–0.3 | Carbide |
| GF/CF-filled PEEK, PAI, PPS; PA66 GF30 | 100–200, 0.05–0.3 | 75–150, 0.3 or less | 50–80, 0.1–0.3 | Carbide or PCD/CVD |
| PI, PBI | 25–100, 0.05–0.2 | 25–100, 0.15 or less | 25–50, 0.1–0.3 | Carbide or PCD/CVD |
Tool angles for both turning and milling: relief angle 5–15°, rake angle 0–10° (0–15° for milling, 3–5° for turning PI and PBI). For drills: relief angle 5–10°, rake angle 3–5°, point angle 90–120°[1].
For PEEK, the plastic manufacturer Victrex also publishes conditions. For unfilled PEEK: turning at a cutting speed of 100–300 m/min, feed 0.1–0.7 mm/rev, relief angle 5–15°, top rake angle 5–15° and depth of cut 1–6.5 mm; milling with carbide at 180–230 m/min; drilling at 120 m/min, feed 0.05–0.20 mm/rev and a 118° point angle[2]. Victrex's lower speed limit is lower and its upper feed limit is higher than MCAM's, so a safe start is where the two sources overlap (turning at around 200–300 m/min and 0.1–0.5 mm/rev).
- Sharp edges, positive or 0° rake: When plastic is rubbed, the heat makes it melt or tear. HSS works, but carbide lasts longer for large quantities[1]
- Generous relief angle: The material springs back after the edge passes, so with too little relief the flank rubs and generates heat
- End mills: 2–3 flutes, high helix: Give the chips plenty of room and prevent them from wrapping around the tool[1]
- Don't drop the feed too low: Too slow, and the tool rubs the same spot and heat builds up. Find conditions where chips come off continuously[1]
- Even diamond tools wear: According to a report from a tool manufacturer's research lab, cutting PEEK with single-crystal diamond tools wears and damages the cutting edge, and the type of damage changes with edge preparation[26]. Even for mirror-like finishes, tool life must be managed
Cooling: air, water-soluble coolant and cutting oil
Except for drilling and parting off, the basic rule for plastics is to machine them without coolant. When surface finish and tolerances matter, water-soluble coolants free of aromatics, mist, or compressed air are effective[1].
Don't use mineral-oil-based cutting oils on amorphous plasticsAmorphous plastics such as PC, PPSU, PEI and PSU can stress-crack (chemical cracking) from mineral-oil-based cutting oils. If you use coolant, wash the parts with isopropyl alcohol (IPA) and deionized water right after machining[1][4]. Even if no cracks appear right after cutting, you can't relax while oil remains on the part.
On the other hand, for PEEK, which is semi-crystalline and chemically resistant, Victrex says both water-based and oil-based coolants can be used[2]. Deep holes (deeper than 2× the diameter) tend to heat up, so use coolant and peck drill to clear the chips[1].
Cooling also reduces burrs. In a study of micro-milling (fine slot cutting) of a biomedical polymer, burrs 185 µm wide blocked the slot with 16 °C air, but blowing air chilled to −2.2 °C with a vortex tube reduced them to 70 µm. The cold plastic becomes stiffer, which changes how it tears[28].
annealing and dimensional stability
Plastic stock (plate and rod) contains residual stress from extrusion or molding. Machining off the surface upsets the stress balance and causes warping, and cutting heat also introduces new stress. Relieving this stress is the job of annealing (stress-relief heat treatment)[2][1].
| Plastic | Holding temperature (MCAM) | Atmosphere |
|---|---|---|
| PE-UHMW, PC | 130 ℃ | Air, nitrogen (oil also for PE) |
| POM-C, POM-H, PET | 155 ℃ | Air, nitrogen, oil |
| PA | 160 °C (170 °C for PA4.6 and PA66 GF30) | Air, nitrogen, oil |
| PSU | 170 ℃ | Air, nitrogen |
| PPS, PEI, PPSU | 200 ℃ | Air, nitrogen |
| PEEK | 250 ℃ | Air, nitrogen |
Common procedure: heat at 10–20 °C/h, hold for 10 minutes per mm of wall thickness, and cool at 5–10 °C/h. Allow a further 3 minutes per mm of wall thickness to return to room temperature. Keep the oven temperature within ±3 °C, and keep the part held in a fixture until it has cooled[1].
Victrex publishes a different procedure for PEEK: heat evenly to 200 °C, hold for 1 hour per mm of wall thickness, then cool at 10 °C/h to below 140 °C before switching off the oven. To stabilize the dimensions of parts used at high temperature, hold at a temperature at least 20 °C above the maximum service temperature instead[2]. Both the temperature and the time differ from MCAM's (250 °C, 10 min/mm), so record in the process sheet which source you followed.
- Roughing: Leave the finished size plus the amount annealing will move it plus finishing stock[1]
- Annealing: At the temperatures in the table above. For parts with heavy stock removal, anneal once more before final finishing[2]
- Light finishing cut: Small depth of cut and a fresh edge. Even just letting parts rest for 24 hours after roughing helps the dimensions settle[1]
If you want to skip annealingPlate and rod from stock manufacturers are normally annealed before shipping. For parts with little stock to remove, a symmetrical shape and generous tolerances, roughing and finishing in one go often causes no problems. If warping occurs, first check whether you removed a lot of material from one side only (machine both sides evenly and symmetrically about the extrusion centerline)[1].
Dimensional targets and measurement (with calculator)
Three things matter when measuring plastic parts: Temperature and moisture and whether the part was just machined. JIS dimensions are values at 20 °C, so for materials with large expansion like plastics, pass/fail judgments go wrong unless you allow for the measuring temperature (see Measurement error basics).
The coefficient of linear expansion is an average over a temperature range and varies with grade, direction (reinforced grades differ between the extrusion direction and perpendicular to it) and degree of crystallinity. Plastic manufacturers' documents give no formula for moisture growth, so this field takes a value you already know (for nylon, up to about 2% when saturated in water[5]).
- Don't measure right after cutting: The part is swollen from cutting heat. Measure after it cools, ideally after it has settled to the measuring room temperature
- Don't hold parts in your hand for long: If hand heat raises the temperature a few °C, a 100 mm POM part moves by more than 0.01 mm (you can check with the calculator)
- Aim ODs at the low side of the tolerance and holes at the high side: In turning, ODs tend to grow and holes tend to shrink. Also check how tap drill holes actually cut[1]
- How tight can you go?: MCAM's "best tolerances" table gives, for sizes of 10, 50, 100 and 200 mm: 0.02, 0.04, 0.06 and 0.10 mm for PEEK, PEI, PPS and PC; 0.03, 0.05, 0.09 and 0.15 mm for POM and PA; 0.05, 0.08, 0.15 and 0.26 mm for HDPE; and 0.02, 0.03, 0.05 and 0.08 mm for filled PEEK. It notes that tighter tolerances need special tools and processes and may increase cost[1]
Workholding, burrs and surface finish
Plastics have a low elastic modulus, so if you grip them with the same force as metal, they are machined in their clamped shape, then spring back and lose their form after unclamping. Victrex notes that strong clamping easily causes warping and distortion, so stock should be removed evenly and symmetrically[2].
- Chucks: soft jaws, wide contact: For thin-walled rings, form the jaws to support the full circumference, or insert a plug in the bore before gripping
- Plates: vacuum, double-sided tape or fixtures rather than a vise: Spread out the clamping force. Removing a lot from one side only causes warping, so machine both sides evenly[1]
- Burrs: prevent them first: Burrs depend on the material, tool condition, feed, speed and cooling[28]. Keep the edge sharp, and use a backing plate or down milling on the exit side
- Deburr without heat: Trim lightly with a blade, or cool first. Rubbing hard with a file melts the plastic again and raises a lip
Glass- and carbon-fiber-filled grades and tool wear
With PEEK GF30 (30% glass fiber), CF30 (30% carbon fiber) and reinforced grades of PAI and PPS, the plastic itself may be soft, but the fibers wear the tool like abrasive grains. MCAM states that for glass- or carbon-fiber-filled and graphite-filled plastics, polycrystalline diamond (PCD) or CVD diamond tools are essential. Cutting speeds are also lowered to about half those for unfilled grades (100–200 m/min for turning)[1]. Victrex also says abrasive reinforced grades require carbide, coated carbide or diamond tools[2].
- Cut stock with a band saw with 4–6 mm tooth pitch: An ordinary circular saw can cause cracking[1]
- Expansion depends on direction: The coefficient of linear expansion of GF30 and CF30 is 4×10⁻⁵/K in the extrusion direction. Since the value varies with fiber orientation, match the measuring direction to the drawing[8][9]
- Don't breathe the dust: Chips from filled grades turn into fine dust. Use dust extraction and protective equipment
When fibers are cut while the plastic is still soft, heat can also weld the plastic onto the tool. In a study of trimming a thermoplastic CFRP with a PA66 matrix (45% carbon fiber), the chips were continuous, and when grinding with an electroplated diamond tool, the higher the speed and feed, the more molten plastic stuck to the tool, leaving fuzz, and the machining force rose from 49 N to over 100 N. On the other hand, cold air about 15 °C below ambient showed no clear effect under these conditions[27]. With reinforced grades, before "cooling," what works is to avoid generating heat through tool grade and speed.
Plastic gears: molding vs. cutting, strength and wear
Plastic gears are injection-molded for mass production, and machined from stock (hobbing or milling) for prototypes, small lots, large gears, and expensive materials like PEEK. Their advantages are light weight, quiet running, operation without lubrication, and no rust[35]. For gear cutting tools and gear basics, see Gear calculations and Gear study.
| Standard | What it specifies | How it applies to machined gears |
|---|---|---|
| JIS B 1702-3:2020 | Accuracy grades for injection-molded plastic cylindrical gears (flank and double-flank composite deviations). Module 0.1–2.0 mm, reference diameter 1–280 mm, 3 or more teeth, helix angle 45° or less, face width 0.2–40 mm[29] | A standard for injection-molded gears. For machined gears, confirm on the drawing and purchase order which standard and grade apply |
| JIS B 1759:2019 | Method for evaluating the tooth root bending strength of plastic cylindrical gears. The material's allowable bending stress is found by running tests, and safety against the nominal tangential force is evaluated. There is no corresponding international standard[30] | Intended for injection-molded gears, but also applicable to machined gears. However, only where the failure mode is tooth root breakage; it cannot be used when wear is extremely advanced[30] |
Strength standards for plastic gears were also developed by the Technical Committee on Molded Plastic Gears of the Japan Society for Precision Engineering. Because there were many cases of "designing plastic gears without understanding them and causing accidents" and "plastic gears not even being considered, even though they could cut costs if used well," the committee created a strength calculation framework modeled on ISO 6336 for metal gears, and set values starting with the tooth root bending strength of POM, based mainly on endurance tests running against a steel master gear without lubrication[31][32].
Strength and wear: temperature affects everything
- Life is often set by wear rather than bending: In tests where gears machined from extruded POM rod were run without lubrication against steel and cast iron gears, wear rather than breakage set the usable limit under low-speed, long-life conditions, and wear was lower against steel than against cast iron[33]
- Heat reduces backlash: A follow-up report from the same study found that when expansion from friction heat exceeded wear, backlash during running became smaller than just after start-up. The wear rate was determined by load rather than speed[34]. Set the center distance and backlash of plastic gears allowing for expansion when they warm up
- Higher speed doesn't mean longer life: The tooth root breakage life of POM gears (oil-lubricated at 50 °C, constant torque) peaked at about 750 min⁻¹; at higher speeds, thermal failure took over and life became shorter. Shaping the root fillet to reduce stress concentration extended life at every speed[35]. With machined gears, the root shape can be built into the cutting tool
- Raise the load and they melt: For PVDF gears (module 3, 17 teeth, steel pinion, unlubricated at 1000 rpm), B10 life was 6.3×10⁶ cycles at 4 Nm; at 6 Nm the tooth root temperature rose to 123 °C, and at 8 Nm the gears melted and failed[36]
- Adding oil doesn't always help: In PEEK-on-steel sliding, depending on the conditions, lubricating oil can make wear worse than running dry[37]. Choose lubricants by testing
Victrex, the plastic manufacturer, makes PEEK gears by injection molding as "VICTREX HPG," claiming 70% less weight, 80% less moment of inertia, and up to 50% (3 dB) less powertrain noise and vibration compared with metal gears. It also cites one-step injection molding that eliminates secondary machining, hardening and finishing as a cost advantage[38]. The division of labor is molding for mass production and machining for prototypes and small lots.
Humanoid and collaborative robot joints and plastic gears
In robots that work alongside people (collaborative robots, service robots and humanoid robots), there is strong demand for lighter, quieter joint reducers, and there is a move toward plastic gears. Here we list only cases confirmed by company announcements and technical documents (as of October 2026).
| Company | What they are doing | How plastic gears are used |
|---|---|---|
| Riken NPR × Harmonic Drive Systems | On September 18, 2026, announced an agreement to jointly develop a "plastic strain wave gear." For applications that share space with people, such as service robots, wearable devices and small mobility vehicles, the aim is safety on contact with people and lower power consumption through weight reduction. The development has future commercialization in view[39] | Riken NPR has already developed a strain wave reducer with a flexspline (thin external gear) and ring gear (internal gear) made of super engineering plastics, and lists applications such as the tips of collaborative and SCARA robots, precision equipment, and each joint and the hands of humanoid robots as targets[40] |
| igus | At Hannover Messe in June 2022, unveiled a prototype humanoid service robot. Its arms are the company's ReBeL robot arm, each able to carry 2 kg[41] | At the heart of each ReBeL joint is a lubrication-free plastic (tribo-polymer) strain wave gear, integrated with the motor, encoder, force control and controller[41] |
| Victrex | Handles PEEK gears (VICTREX HPG) from design to manufacturing for automotive, robotics and e-mobility[38] | Says they can replace metal planetary gear sets in collaborative robots. Has not disclosed specific robot customers[38] |
What we left outNews reports and analyst views that a particular humanoid robot has adopted PEEK gears or parts could not be confirmed by company announcements, so they are not included. How far plastic gears will be used in robot joints depends on the results of the developments above.
From a machining standpoint, plastic gears for joints are expected to be mostly molded in mass production, but prototype and evaluation gears, molds, and the metal parts that work with plastic gears (housings, shafts, cams) are machining jobs. If you cut thin plastic gears, the "workholding," "annealing" and "temperature and dimensions" sections of this article apply directly.
8 common mistakes
- Passed when measured right after cutting, failed the next day: Cutting heat and stress recovery moved the dimensions. Measure after cooling; for precision parts, rough → anneal → finish[1][2]
- The hole won't take the gauge: Holes in plastic shrink after machining. Set the pilot hole and reamer sizes based on how the material actually cuts[1]
- PC or PEI cracked later: Stress cracking from mineral-oil-based cutting oil. Switch to air or water-soluble coolant, and wash with IPA and deionized water after use[1]
- The plate warped: A lot was removed from one side only. Machine both sides evenly and symmetrically, and anneal if needed[1]
- The ring went oval: The chuck gripped too hard. Support the full circumference with soft jaws or a plug[2]
- Tools wear out fast on GF30: The fibers wear the tool. Switch to PCD or CVD diamond tools and lower the speed[1]
- UHMW-PE or PA wraps around the tool: Feed too slow or edge dull. Raise feed and speed, and use 2–3 flute, high-helix end mills[1]
- A plastic gear jammed during operation: It warmed up, expanded and lost its backlash. Set center distance and tooth thickness allowing for expansion[34]
FAQ
- Do I need special tools to machine plastics?
- Unfilled POM, PA and PE can be cut with sharp HSS or carbide tools. For large quantities, carbide lasts longer. Glass- or carbon-fiber-filled grades and PI need PCD or CVD diamond tools[1]. In every case, the common requirements are "a sharp edge, positive or 0° rake, and a generous relief angle."
- Can I use the same tools I used on metal?
- A tool whose edge has been rounded on metal will rub plastic and generate heat, causing tearing, burrs and dimensional errors. It is safer to use a fresh edge or tools set aside for plastics only.
- Does PEEK always need annealing?
- Stock is often already annealed by the manufacturer, and if there is little stock to remove and the tolerances are generous, annealing can sometimes be skipped. For parts with heavy stock removal, thin walls, tight tolerances or high-temperature service, anneal after roughing and then finish. The conditions differ between MCAM (250 °C, 10 min/mm) and Victrex (200 °C, 1 hour/mm), so decide which one you will follow[1][2].
- At what temperature should plastic parts be measured?
- Drawing dimensions are referenced to 20 °C. If the measuring room is not at 20 °C, estimate the temperature effect with the calculator above and check whether it is negligible relative to the tolerance. For 100 mm of PEEK, an 8 °C difference moves it by about 0.04 mm, which is not negligible for a ±0.05 mm tolerance.
- My nylon parts changed size during the rainy season.
- Nylon absorbs water and grows. When saturated in water it can absorb up to about 7% of its weight, and its dimensions can change by up to about 2%[5]. Where dimensional stability is needed, switch to a low-absorption plastic such as POM or PEEK, or condition the parts to the service environment before measuring.
- Are machined plastic gears as strong as molded ones?
- No. Molded and machined parts differ in material state, such as the surface skin, fiber orientation and residual stress, so their strength can differ too. JIS B 1759 says it can be applied to machined gears, but the material's allowable stress is to be found by running tests[30].
Values for each material are in the Materials database, measurement principles in Measurement error basics, gears in Gear study and Gear calculations, and cutting condition calculations in Calculators.
Sources
- Mitsubishi Chemical Group (Mitsubishi Chemical Advanced Materials), "Guide to machining plastic parts (Machinist Toolkit)" (PDF)
- Victrex, "VICTREX Finishing Brochure" (2022, PDF)
- Ensinger, "CNC turning of plastics"
- Ensinger, "FAQ"
- Quadrant Engineering Plastic Products, design and fabrication brochure (hosted by Polymershapes, PDF)
- Ensinger, "TECAPEEK natural"
- Victrex, "VICTREX PEEK 450G" datasheet
- Ensinger, "TECAPEEK GF30 natural"
- Ensinger, "TECAPEEK CF30 black"
- Ensinger, "TECAPEI natural"
- Ensinger, "PPS plastic"
- Ensinger, "TECATRON SX natural"
- Ensinger, "TECAPAI CM XP403 green"
- Ensinger, "Torlon PAI (polyamide-imide)"
- Ensinger, "TECASINT 2011 natural"
- Ensinger, "TECAFORM AH natural"
- Ensinger, "TECAST T natural"
- MCAM, "Ertalon 6PLA FG" product data (PDF)
- Ensinger, "TECANAT natural"
- Ensinger, "TECAFLON PTFE natural"
- Nichias, "NAFLON PTFE" catalog (includes the Japan Fluoropolymers Industry Association table, PDF)
- Ensinger, "TECAFINE PE 1000 natural"
- Ovako, "Steel Navigator: C45"
- Nippon Steel, "Stainless Steel Shapes" catalog S006 (PDF)
- Kaiser Aluminum, "Alloy 6061 Rod and Bar"
- F. Sakano, N. Tatsumi, Y. Kobayashi, H. Sumiya, "Damage analysis of diamond tools in cutting PEEK," Proceedings of the JSPE Spring Meeting 2019, pp.642-643 (2019, abstract)
- H. Tsuge, T. Kaga, M. Kayaoka, "Establishing cutting and grinding techniques for thermoplastic CFRP (Report 1)," Research Reports of the Gifu Prefectural Research Institute of Industrial Technology No.2 (2014, PDF)
- M. Miranda-Giraldo et al., "Burr formation and control for polymers micro-milling: A case study with vortex tube cooling," DYNA 84(203), pp.150-159 (2017, PDF)
- JIS B 1702-3:2020, "Cylindrical gears — Accuracy grades — Part 3: Definitions and allowable values of deviations relevant to flanks and double-flank composite engagement of injection-molded plastic gears" (Japanese Standards Association preview, PDF)
- JIS B 1759:2019, "Methods for evaluating the bending strength of plastic cylindrical gears" (Japanese Standards Association preview, PDF)
- T. Koide, A. Ueda, "[Technical Committee on Molded Plastic Gears] Toward wider use of plastic gears and established strength evaluation methods," Journal of the Japan Society for Precision Engineering Vol.79 No.10, pp.904-907 (2013)
- I. Moriwaki, "Developing strength design standards for plastic gears," Proceedings of the JSPE Semestrial Meeting 2007S (2007, PDF)
- M. Iwai, S. Bushimata, "Study on the strength of polyacetal gears for power transmission (1st report): Bending strength and wear strength," Journal of the Japan Society of Precision Engineering (Seimitsu Kikai) Vol.46 No.10, pp.1285-1290 (1980)
- S. Bushimata, M. Iwai, "Study on the strength of polyacetal gears for power transmission (2nd report): Tracking wear progression by the capacitance method," Journal of the Japan Society of Precision Engineering (Seimitsu Kikai) Vol.47 No.8, pp.956-961 (1981)
- R. Eberlein, S. Düzel, "Fatigue Lifetime Analysis of POM Gears for Generalized Tooth Root Shapes," PPS-39 Proceedings (2024)
- E. Muratović et al., "An Analysis of the Tribological and Thermal Performance of PVDF Gears in Correlation with Wear Mechanisms and Failure Modes Under Different Load Conditions," Coatings 15(7), 800 (2025)
- ENEOS Technical Review Vol.62 No.1, "Effect of lubrication on friction and wear properties of PEEK with steel counterparts" (Tatsumi, Shitara, Sakamoto; award report, 2020, PDF)
- Victrex, "PEEK Gears VICTREX HPG – High-Precision Gear Solutions"
- Riken NPR, "Riken NPR and Harmonic Drive Systems agree to jointly develop next-generation 'plastic strain wave gearing'" (September 18, 2026, PDF)
- Riken NPR, "Electric unit products"
- igus, "Everyday assistants of the future: igus accelerates humanoid robotics with Low Cost Automation" (June 2022)
More articles
IndustryMachining EV Parts: Parts That Shrink and Grow, eAxles and Battery Cases, Quiet Gears, Major Japanese Companies, Machine and Cutting Tools, New Technology, How to Enter
NC11 Machining Simulators That Run in Your Browser: CAM, Machining Center, CNC Lathe, 5-Axis, Gears, Grinding
Gauges