
Dewey Wu General Manager & senior mechanical engineer at EPOC CRAFTER, 15 years in design engineering, quality, and metallurgy. Hands-on across CNC machining, metalwork, sheet metal, and prototyping (subtractive + 3D printing).
Nylon (polyamide) mechanical properties span a wide range depending on grade and moisture state: PA66 dry-as-molded (DAM) tensile strength reaches 83 MPa at 23°C per ASTM D638, while the same grade conditioned to 50% RH drops to 77 MPa, a 7% reduction that compounds into dimensional shift of +0.6% on linear dimensions per ASTM D570. Add glass fiber reinforcement (PA66 GF30, e.g. Zytel 70G33L) and tensile strength climbs to 170–190 MPa DAM with heat deflection temperature (HDT) rising from 90°C to 210–230°C at 1.82 MPa per ASTM D648.
The machining consequence is direct: a PA66 rod stock at 23°C with a coefficient of thermal expansion (CTE) of 81×10⁻⁶/°C expands approximately 0.007 in. per inch of diameter across a 70°F temperature rise during cutting, enough to invalidate mid-cut measurements and push a finished bore outside ±0.05 mm tolerance without pre-drying and thermal stabilization.
This guide covers tensile strength, flexural modulus, and impact data across PA6, PA66, PA6 GF30, and PA66 GF30 in both DAM and conditioned states; HDT interpretation under 0.455 MPa and 1.82 MPa load conditions per ASTM D648; moisture absorption rates and dimensional change calculations per ASTM D570; and CNC turning, milling, and drilling parameters including tool selection, cutting speeds, pre-drying protocols, and achievable tolerances for each grade.
1. The Five-State Property Framework
Nylon does not have a single set of mechanical properties. It has five distinct performance states, each corresponding to a physical condition your part passes through from raw stock to final service. Selecting a grade based on DAM datasheet values alone and ignoring moisture conditioning is the single most common cause of tolerance failures in precision nylon components.
The five states, in order of a part’s lifecycle, are: dry-as-molded (DAM) stock as received from the supplier; conditioned to 50% relative humidity equilibrium; fully saturated in water immersion; glass-fiber-reinforced (any of the above states); and the cutting zone state during active CNC machining, where thermal expansion from frictional heat temporarily shifts dimensions independently of moisture content.
Each state produces measurably different tensile strength, flexural modulus, and dimensional envelope. The table below maps all five states to their key property ranges so you can anchor every downstream machining and tolerance decision to the correct baseline.
| 1. DAM | 23°C, dry stock as received | PA66: 83 / PA6: 75–80 | PA66: 2,800–3,500 / PA6: 2,000–2,500 | Baseline (0%) | Machine to this state; dimensions are smallest |
| 2. Conditioned 50% RH | ~2.5% moisture by weight | PA66: 77 / PA6: ~65 | PA66: 1,800–2,300 / PA6: 1,300–1,700 | +0.6% linear | Service fit dimensions; design clearances here |
| 3. Saturated | ~8.5% moisture (PA66) | PA66: ~45–55 | PA66: 1,100–1,500 | +2.6% linear | Wet environment worst-case envelope |
| 4. GF30 Reinforced | DAM, 23°C | PA66 GF30: 170–190 | PA66 GF30: 9,000–11,000 | +0.15–0.25% (reduced by fiber) | High-stiffness, high-HDT applications |
| 5. Cutting Zone | Active machining, no coolant | Locally softened at tool tip | N/A | +0.007 in./in. per 70°F rise (CTE 81×10⁻⁶/°C) | Finish-cut only after thermal stabilization |
Data sources: States 1–3 per DuPont Zytel Design Guide (Module II) and ASTM D638/D570; State 4 per CAMPUS Plastics Database typical values (Zytel 70G33L); State 5 CTE per DuPont Zytel 101 at 23°C dry per Minlon/Zytel Design Info Module 2.
Engineering Note The most common misapplication is using State 1 (DAM) tensile and modulus values for service load calculations on parts that operate in ambient indoor environments. Indoor equilibrium for PA66 in a factory or lab at 50% RH corresponds to State 2, where flexural modulus is 35–45% lower than the DAM datasheet value. Size to State 2 for fit-critical features; use State 1 only for as-machined inspection dimensions.
For a full comparison of nylon grades including PA12 and MoS₂-filled variants, see PA6 vs PA66 vs PA12: Nylon Grades Compared.
2. Tensile Strength, Modulus, and Impact Grade-by-Grade Data
PA66 tensile strength at 23°C DAM per ASTM D638 is 83 MPa for general-purpose unfilled grade (Zytel 101L). Conditioning to 50% RH reduces this to 77 MPa. At 121°C DAM, tensile strength falls to 43 MPa, a 48% reduction from room-temperature DAM that defines the practical upper load limit for unfilled PA66 structural components without safety factor adjustment.
These numbers shift significantly across grades and reinforcement levels. The four sections below give grade-specific data for the conditions engineers encounter most: DAM at 23°C, conditioned at 50% RH, and elevated temperature.
2.1 PA66 Unfilled
PA66 unfilled (Zytel 101L, Ultramid A) delivers the highest stiffness and strength among unreinforced nylon grades at room temperature. Tensile strength drops 48% from 40°C to 121°C in DAM condition per DuPont Module II data, and moisture conditioning at 50% RH shifts the entire strength curve downward by approximately 7–15% depending on temperature.
| Tensile Strength | 23°C, DAM | 83 MPa | ASTM D638 |
| Tensile Strength | 23°C, 50% RH | 77 MPa | ASTM D638 |
| Tensile Strength | 40°C, DAM | 114 MPa | ASTM D638 |
| Tensile Strength | 77°C, DAM | 62 MPa | ASTM D638 |
| Tensile Strength | 121°C, DAM | 43 MPa | ASTM D638 |
| Elongation at Break | 23°C, DAM | 40–60% | ISO 527 |
| Flexural Modulus | 23°C, DAM | 2,800–3,500 MPa | ISO 178 |
| Flexural Modulus | 23°C, 50% RH | 1,800–2,300 MPa | ISO 178 |
| Notched Charpy Impact | 23°C, DAM | ~5 kJ/m² | ISO 179 |
| Melting Point | — | 262°C | ASTM D3418 |
| HDT | 1.82 MPa | 90°C | ASTM D648 |
| HDT | 0.455 MPa | 235°C | ASTM D648 |
Source: DuPont Zytel Design Guide Module II; CAMPUS Plastics Database (Zytel 101L / Ultramid A typical values).
2.2 PA6 Unfilled
PA6 unfilled (Ultramid B, Durethan B40F) runs 5–10 MPa lower in tensile strength than PA66 at equivalent conditions and absorbs more moisture at equilibrium, which amplifies dimensional instability. PA6 is the more forgiving grade to machine due to its higher elongation at break and lower stiffness, but it requires the same pre-drying discipline as PA66 before tight-tolerance work.
| Tensile Strength | 23°C, DAM | 70–85 MPa | ISO 527 |
| Tensile Strength | 23°C, 50% RH | 55–65 MPa | ISO 527 |
| Elongation at Break | 23°C, DAM | 40–70% | ISO 527 |
| Flexural Modulus | 23°C, DAM | 2,000–2,500 MPa | ISO 178 |
| Flexural Modulus | 23°C, 50% RH | 1,300–1,700 MPa | ISO 178 |
| Notched Charpy Impact | 23°C, DAM | 4–8 kJ/m² | ISO 179 |
| Melting Point | — | 218–222°C | ISO 11357 |
| HDT | 1.82 MPa | 65–80°C | ISO 75 |
| HDT | 0.455 MPa | 180–200°C | ISO 75 |
Source: CAMPUS Plastics Database typical values (Ultramid B / Durethan B40F).
2.3 PA66 GF30
PA66 GF30 (Zytel 70G33L, Durethan AKV30) delivers tensile strength of 170–190 MPa DAM at 23°C, approximately 2.1× the unfilled PA66 value, and flexural modulus of 9,000–11,000 MPa, roughly 3–3.5× higher. HDT at 1.82 MPa jumps from 90°C to 210–230°C, which is the primary reason GF30 is specified for under-hood automotive and industrial thermal environments where unfilled grades fail.
The glass fiber content also reduces CTE in the flow direction from 81×10⁻⁶/°C to 23×10⁻⁶/°C, cutting thermal expansion during machining to less than one-third of the unfilled value. This is the single most important dimensional stability benefit of GF reinforcement for precision CNC work.
| Tensile Strength | 23°C, DAM | 170–190 MPa | ISO 527 |
| Elongation at Break | 23°C, DAM | 2.5–3.5% | ISO 527 |
| Flexural Modulus | 23°C, DAM | 9,000–11,000 MPa | ISO 178 |
| Notched Charpy Impact | 23°C, DAM | 9–13 kJ/m² | ISO 179 |
| HDT | 1.82 MPa | 210–230°C | ISO 75 |
| HDT | 0.455 MPa | 250°C+ | ISO 75 |
| Water Absorption 24h | 23°C | 0.4–0.8% | ISO 62 |
| Saturation Absorption | 23°C | ~5.4% (Zytel 70G33L) | ASTM D570 |
| CTE (flow direction) | 23°C, dry | 23×10⁻⁶/°C | Minlon/Zytel Module 2 |
| Density | — | 1.36–1.40 g/cm³ | ISO 1183 |
Source: DuPont Zytel Design Guide Module II (Zytel 70G33L); CAMPUS Plastics Database.
2.4 PA6 GF30
PA6 GF30 (Durethan BKV30) performs comparably to PA66 GF30 in tensile strength (150–180 MPa DAM) but carries a lower melting point (218–222°C vs 258–265°C) and slightly higher moisture absorption. For applications below 150°C service temperature where cost matters more than peak thermal performance, PA6 GF30 is a viable alternative. Above 150°C or in continuous thermal cycling, PA66 GF30 is the correct choice.
| Tensile Strength | 23°C, DAM | 150–180 MPa | ISO 527 |
| Elongation at Break | 23°C, DAM | 3–4% | ISO 527 |
| Flexural Modulus | 23°C, DAM | 8,000–10,000 MPa | ISO 178 |
| Notched Charpy Impact | 23°C, DAM | 8–12 kJ/m² | ISO 179 |
| HDT | 1.82 MPa | 200–215°C | ISO 75 |
| Water Absorption 24h | 23°C | 0.5–1.0% | ISO 62 |
| Density | — | 1.34–1.38 g/cm³ | ISO 1183 |
Source: CAMPUS Plastics Database typical values (Durethan BKV30).
Engineering Note When comparing TDS values from different suppliers, confirm whether HDT was measured at 0.455 MPa or 1.82 MPa per ASTM D648. The same PA66 unfilled grade reports 235°C at 0.455 MPa and 90°C at 1.82 MPa, a 145°C gap from a single material under two load conditions per ASTM D648-01 Section 13.1.6. Specifying “HDT” without the load condition in a procurement document gives your supplier a 145°C window of ambiguity.
According to Dewey Wu, General Manager and senior mechanical engineer at EPOC CRAFTER: “On precision gear and bushing projects we request the 1.82 MPa HDT figure, not the 0.455 MPa number. The low-load figure is cited in marketing materials because it looks better, but for a load-bearing part running at 80°C, only the 1.82 MPa value tells you whether the part holds its geometry.”
For CNC machining capability across these grades including achievable tolerances and surface finish, see CNC Machining Services at EPOC CRAFTER.
3. Heat Deflection Temperature Reading HDT Without Getting Burned
HDT measured per ASTM D648 is the temperature at which a standard test bar deflects 0.25 mm under a specified bending stress, raised at 2.0 ±0.2°C/min in a liquid bath per ASTM D648-01 Section 4.1. It is not the melting point, not the continuous use temperature, and not a design safety limit. ASTM D648-01 Section 5.1 explicitly states the data “shall not be used to predict the behavior of plastics at elevated temperatures” outside the specific test conditions.
The 0.455 MPa and 1.82 MPa load conditions produce results that differ by over 140°C on the same material. Both numbers appear on supplier datasheets. Using the wrong one in a thermal specification produces a part that fails in service at temperatures the datasheet appeared to clear.
3.1 What the Two Load Conditions Actually Mean
At 0.455 MPa, the test bar carries a light bending stress. At 1.82 MPa, the stress is four times higher and more representative of a structural part under moderate mechanical load at temperature. For any load-bearing nylon component, gears, bushings, brackets, or housings, the 1.82 MPa figure is the operationally relevant number.
The gap between the two figures is not fixed. For unfilled PA66 the gap is approximately 145°C (235°C vs 90°C per Zytel 101L data from DuPont Module II). For PA66 GF30 the gap narrows to roughly 40–50°C because glass fiber carries load at elevated temperature, reducing deflection under both stress levels. This narrowing is itself an engineering signal: the more the two HDT figures converge, the more thermally stable the grade under real structural loading.
| PA6 unfilled | 180–200 | 65–80 | ~120 | ISO 75 / ASTM D648 |
| PA66 unfilled (Zytel 101L) | 235 | 90 | 145 | ASTM D648 |
| PA6 GF30 | 210–220 | 200–215 | 10–15 | ISO 75 |
| PA66 GF30 (Zytel 70G33L) | 250+ | 210–230 | ~30 | ASTM D648 / ISO 75 |
Source: DuPont Zytel Design Guide Module II; Minlon Zytel Design Info Module 2; CAMPUS Plastics Database.
3.2 HDT Is Not Continuous Use Temperature
ASTM D648-01 Section 5.1 places four explicit restrictions on HDT data: it applies only when time, temperature, loading mode, and surface fiber stress in service are similar to test conditions; it does not predict long-term durability; it does not account for creep under sustained load; and it does not account for thermal aging or chemical environment effects.
For practical CNC part specification, a conservative working rule is to derate the 1.82 MPa HDT by 20–30°C as an estimated upper continuous use temperature for lightly loaded structural parts, and by 40–50°C for parts under sustained mechanical load. PA66 GF30 at HDT 210–230°C supports approximately 160–190°C continuous service in lightly loaded enclosure applications, and 160–170°C under sustained structural load, not 210°C.
3.3 Why GF30 HDT Jumps at 1.82 MPa
Unfilled PA66 softens progressively above its glass transition region, and the amorphous phase begins to lose load-bearing capacity well below the melting point of 262°C. At 90°C under 1.82 MPa, the amorphous chains have enough thermal energy to allow the 0.25 mm deflection criterion to be met, triggering the HDT endpoint per ASTM D648-01 Section 4.1.
Glass fibers at 30% by weight create a load-bearing skeleton within the polymer matrix. At temperatures where the unfilled matrix deflects, the fiber network continues to carry the bending stress, delaying the 0.25 mm deflection to 210–230°C. PA66 GF30 HDT at 1.82 MPa is 2.3–2.6× higher than unfilled PA66 at the same load. The test is measuring fiber-network stiffness retention, not polymer softening. Specify PA66 GF30 any time your part must hold dimensional tolerances above 80°C under mechanical load.
Engineering Note ASTM D648-01 Section 9.1, Note 8 specifies that specimens near 13 mm width can read 2–4°C higher than specimens at 4 mm width due to slower internal heat transfer. Two datasheets from different suppliers showing HDT values 3°C apart may reflect specimen geometry differences rather than actual material differences. When comparing suppliers, request confirmation that HDT was measured on the same specimen geometry and load condition before treating the numbers as directly comparable. Also note per ASTM D648-01 Section 7.1.1 that Method A (101.6 mm span) and Method B (100 mm span) can differ by 1.0–4.5°C on the same material.
For material selection across nylon grades and alternative engineering plastics including POM and PEEK, see Materials & Properties.
4. Moisture Absorption What Happens to Your Part After It Leaves the Machine
PA66 absorbs 1.20% of its weight in water after 24 hours at 23°C per ASTM D570 (Zytel 101L measured value from DuPont Module II), and reaches 8.50% at full saturation. A PA66 part machined to print dimensions in dry stock condition and shipped without conditioning will continue absorbing moisture from ambient air until it reaches equilibrium at approximately 2.5% moisture content in a 50% RH environment, at which point its linear dimensions have grown by +0.6% and its flexural modulus has dropped 35–45% from the DAM datasheet value.
For a 50 mm bore machined to H7 tolerance (±0.025 mm), a +0.6% dimensional growth equals +0.30 mm on diameter, twelve times the tolerance band. This is not a worst-case scenario. It is the expected behavior of an unmoisturized PA66 part in a standard indoor environment.
4.1 Absorption Rates by Grade
Glass fiber reinforcement reduces moisture absorption by displacing polymer volume with inert fiber, but does not eliminate it. Zytel 70G33L (PA66 GF30) absorbs 0.70% at 24h and 5.40% at saturation per ASTM D570 from DuPont Module II data, still enough to produce measurable dimensional shift in precision assemblies. PA6 grades absorb more than PA66 at equivalent conditions due to higher amide group density per unit of polymer chain length.
| PA6 unfilled | 1.0–1.5 | 7–9 | ~0.7 | ~2.8 | ISO 62 / ASTM D570 |
| PA66 unfilled (Zytel 101L) | 1.20 | 8.50 | +0.6 | +2.6 | ASTM D570 |
| PA6 GF30 | 0.5–1.0 | 3–4 | ~0.3 | ~1.0 | ISO 62 |
| PA66 GF30 (Zytel 70G33L) | 0.70 | 5.40 | ~0.2–0.3 | ~0.8 | ASTM D570 |
Source: DuPont Zytel Design Guide Module II; CAMPUS Plastics Database typical values.
4.2 How Moisture Changes Mechanical Properties
Water absorbed by nylon acts as a plasticizer, disrupting the hydrogen bonding network between amide groups (–CONH–) and reducing chain-to-chain cohesion. The mechanical result is lower stiffness and higher toughness: the material becomes more impact-resistant but less capable of holding load without creep.
For Zytel 101L (PA66 unfilled), the shift from DAM to 50% RH equilibrium reduces tensile strength from 83 MPa to 77 MPa at 23°C per ASTM D638. Flexural modulus falls from approximately 3,000 MPa to 1,800–2,300 MPa, a reduction of 25–40% that directly affects part stiffness under load in service. At full saturation (~8.5% moisture), tensile strength drops further to approximately 45–55 MPa and modulus to 1,100–1,500 MPa.
| DAM (dry-as-molded) | ~0.2 | 83 | 2,800–3,500 | Lower, more brittle |
| 50% RH equilibrium | ~2.5 | 77 | 1,800–2,300 | Moderate |
| Fully saturated | ~8.5 | 45–55 | 1,100–1,500 | Higher, tougher |
All values: PA66 unfilled (Zytel 101L), 23°C, per ASTM D638 / DuPont Zytel Design Guide Module II.
4.3 Dimensional Change Calculations
DuPont Module II data establishes two anchor points for PA66 dimensional change: +0.6% linear at 50% RH equilibrium (2.5% moisture content) and +2.6% linear at full saturation (8.5% moisture content). A working engineering estimate of approximately 0.24% dimensional change per 1% moisture content gain applies across the range from DAM to saturation for unfilled PA66.
For PA66 GF30, the fiber skeleton constrains swelling and reduces the dimensional change rate to approximately 0.09–0.12% per 1% moisture content, roughly half the unfilled value. The fiber constraint is anisotropic: swelling in the transverse direction (perpendicular to fiber alignment) remains higher than in the flow direction.
To calculate expected dimensional shift for a specific feature:
ΔL = L₀ × (Δmoisture% × 0.0024)
Where L₀ is the machined dimension in mm, Δmoisture% is the change in moisture content from machining state to service state, and 0.0024 is the linear strain coefficient per 1% moisture for unfilled PA66 DAM stock.
For a 100 mm shaft diameter machined dry, equilibrating to 50% RH (Δmoisture = 2.5%): ΔL = 100 × (2.5 × 0.0024) = +0.60 mm
This growth must be accounted for in clearance fits, bore diameters, and any assembly feature where nylon mates with metal or another polymer with negligible moisture uptake.
4.4 Pre-Conditioning and Design Compensation
Two engineering approaches handle moisture-induced dimensional change. Pre-conditioning exposes the machined part to service-equivalent humidity before final inspection and assembly, so dimensions are measured and accepted in their equilibrium state. Design compensation builds the expected dimensional growth into the machined target: a bore intended to be 20.000 mm in service at 50% RH is machined to 19.880 mm dry, with the 0.120 mm growth (0.6% of 20 mm) closing the gap during conditioning.
Pre-conditioning is the more reliable approach for tight tolerance work. Boiling water conditioning for 3 hours (3.2 mm section) or 18 hours (6.4 mm section) per DuPont Zytel specifications brings PA66 to near-equilibrium faster than ambient humidity exposure, which requires approximately 125 days for a 1.5 mm dry specimen at 50% RH per DuPont Module II data. For production parts, potassium acetate solution conditioning (125 parts potassium acetate per 100 parts water, boiling point 121°C) achieves up to 2.5% moisture uptake in 20 hours for 3.2 mm sections. This is the fastest practical method for reaching 50% RH equilibrium moisture content without full water saturation.
Engineering Note Flood coolant during CNC machining introduces uncontrolled surface moisture absorption in real time. A PA66 part being machined with water-based coolant is absorbing moisture and expanding at the cutting zone simultaneously, producing dimensional drift that cannot be corrected by post-machining inspection alone. Use compressed air for chip clearance and thermal management on all unfilled nylon grades. Water-based coolant is only acceptable on PA66 GF30 for tool temperature management, applied as mist rather than flood, and only when final dimensions are verified after a minimum 2-hour stabilization period.
For design guidelines on tolerancing nylon parts in assemblies with metal components, see DFM Design Guidelines.
5. CNC Machining Parameters Grade-Specific Numbers
Nylon machines cleanly under the right conditions and produces scrap under the wrong ones. The three variables that determine which outcome you get are tool sharpness, thermal management, and moisture state of the stock before the first cut. Get all three right and unfilled PA66 holds ±0.05 mm routinely. Get any one wrong and you are chasing dimensional drift caused by heat expansion, moisture uptake from coolant, or residual stress release, none of which show up until after the part has left the spindle.
5.1 Tool Selection
Unfilled PA6 and PA66 machine with HSS tooling provided the cutting edge is sharp and geometry is correct. A worn HSS tool on nylon does not cut, it rubs, generates heat, and produces a smeared surface finish with subsurface stress that causes warping hours after machining. Replace or resharpen HSS tooling at the first sign of edge rounding.
PA6 GF30 and PA66 GF30 require carbide tooling without exception. Glass fiber at 30% by weight has a hardness that abrades HSS cutting edges within minutes of contact. Uncoated carbide is acceptable for short runs; TiAlN-coated carbide extends tool life by 40–60% on GF30 grades and is the preferred choice for production volumes above 50 pieces.
| PA6 unfilled | HSS or uncoated carbide | 10–15° positive | 8–12° | 0.4–0.8 mm | None required |
| PA66 unfilled | HSS or uncoated carbide | 10–15° positive | 8–12° | 0.4–0.8 mm | None required |
| PA6 GF30 | Carbide only | 10–15° positive | 8–12° | 0.4–0.8 mm | TiAlN preferred |
| PA66 GF30 | Carbide only | 10–15° positive | 8–12° | 0.4–0.8 mm | TiAlN preferred |
Source: Ensinger Engineering Plastics Machining Guide; Davantech machining guide for PA6/PA66.
5.2 Turning Parameters
Use positive rake angle geometry throughout. Negative rake angles increase cutting forces and heat generation on nylon, producing smeared surfaces and accelerated tool wear. Chip breaker geometry is required on all turning inserts to manage nylon’s tendency to produce long continuous chips that wrap around the spindle and drag across finished surfaces.
The roughing-to-finishing sequence for tight-tolerance turned parts follows three steps: rough at maximum depth and feed to remove bulk material; semi-finish at 1.0–1.5 mm depth, then stop and allow the part to return to ambient temperature for 10–15 minutes before measuring; finish at 0.1–0.4 mm depth with high surface speed and light feed, measuring only after the part has stabilized at ambient temperature.
| PA6 / PA66 unfilled | HSS | 60–180 | 0.1–0.4 | 1.0–4.0 | 0.1–0.4 |
| PA6 / PA66 unfilled | Carbide | 180–350 | 0.1–0.4 | 1.0–4.0 | 0.1–0.3 |
| PA6 / PA66 GF30 | Carbide | 100–200 | 0.05–0.2 | 0.5–2.0 | 0.1–0.2 |
Source: Davantech machining guide for PA6/PA66; Ready Plastics nylon machining guide.
Engineering Note GF30 turning speeds are set 40–50% below unfilled PA66 carbide speeds. The glass fiber generates abrasive heat at the tool-chip interface that accumulates faster than thermal conductivity (0.25 W/m·K for PA66 per DuPont Zytel data) can dissipate. Exceeding 200 m/min on GF30 with carbide accelerates tool wear exponentially. A doubling of cutting speed can reduce tool life by 70–80% on abrasive-filled grades.
5.3 Milling Parameters
Climb milling is preferred over conventional milling on all nylon grades. Conventional milling lifts and pulls the workpiece upward at the entry point, producing chatter on thin-walled features and delamination on GF30 grades at the surface layer. Climb milling pushes the part down into the fixture, produces cleaner entry cuts, and reduces the tendency for long chip strands to re-enter the cut zone.
| PA6 / PA66 unfilled | HSS | 50–150 | 0.05–0.15 | Up to 1.5× D | 0.3–0.5× D |
| PA6 / PA66 unfilled | Carbide | 150–300 | 0.05–0.15 | Up to 1.5× D | 0.3–0.5× D |
| PA6 / PA66 GF30 | Carbide | 80–150 | 0.03–0.10 | Up to 1.0× D | 0.2–0.4× D |
Source: Davantech machining guide; Ready Plastics nylon machining guide.
5.4 Drilling
Drill bits must be sharp and correctly ground for plastics. A dull drill on nylon does not cut the material. It displaces it radially, building compressive hoop stress in the bore wall that causes splitting during or after drilling, particularly in PA66 GF30 where the fiber network offers no stress relief through plastic deformation. Standard twist drills reground with a 118° point angle and polished flutes work on unfilled grades. For GF30, solid carbide drills with through-coolant capability (air, not water) reduce heat buildup and extend tool life.
For holes deeper than 3× diameter, use peck drilling cycles to clear chips and prevent heat accumulation. Chip packing in deep bores generates enough heat to locally melt nylon at the drill tip, producing a fused plug that damages both the tool and the bore wall on retraction.
5.5 Chip Control and Cooling
Nylon produces long, continuous, stringy chips that wrap around tooling, score finished surfaces, and generate friction heat if allowed to accumulate in the cut zone. Chip control on production runs determines surface finish quality as much as cutting parameters do.
Compressed air directed at the cut zone is the correct cooling and chip clearance method for all unfilled nylon grades. It removes chips before they re-enter the cut, provides sufficient thermal management for moderate cutting speeds, and introduces zero moisture. Flood coolant is prohibited on unfilled nylon: water-based coolants cause surface moisture absorption during machining, produce dimensional drift that invalidates in-process measurements, and can cause stress cracking in some PA66 grades exposed to certain cutting fluid chemistries.
For PA66 GF30 at high cutting speeds where compressed air is insufficient for thermal management, a light mist of water-soluble coolant is acceptable, applied as mist rather than flood, with a mandatory 2-hour dimensional stabilization period before final inspection.
5.6 Pre-Drying and Annealing
Stock nylon rod and plate absorbs moisture during storage. PA66 rod stored in open warehouse conditions can reach 1–2% moisture content, enough to cause dimensional instability during precision machining. Pre-drying to below 0.2% moisture content before machining is required for any tolerance tighter than ±0.10 mm.
Dry PA6 and PA66 stock at 80–90°C for 8–12 hours in a circulating air oven. Seal in moisture-barrier bags immediately after drying and machine within 4 hours of removal from the bag. For PA66 GF30, the same drying protocol applies; the lower absorption rate (0.70% at 24h vs 1.20% for unfilled per ASTM D570) does not eliminate the need for pre-drying on tight-tolerance work.
Annealing relieves residual stress in stock material before precision machining and reduces post-machining warping on complex geometries. Anneal at 90–100°C for 2–4 hours with slow furnace cooling (less than 20°C/hour) before finish machining. For Zytel grades, DuPont specifies annealing at 149°C (300°F) for 15 minutes per 3.2 mm of section thickness, approximately 28°C above the maximum intended use temperature per DuPont Module II data. A second annealing pass after roughing and before finishing is justified on parts with wall thickness variation greater than 2:1 or deep pockets where material removal is asymmetric.
| Pre-drying (unfilled PA6/PA66) | 80–90°C | 8–12 h | Reduce moisture to <0.2% before machining |
| Pre-drying (PA66 GF30) | 80–90°C | 6–10 h | Reduce moisture to <0.2% before machining |
| Annealing before finish machining | 90–100°C | 2–4 h + slow cool | Relieve residual stress, reduce post-machining warp |
| Zytel annealing (DuPont spec) | 149°C | 15 min per 3.2 mm section | Stress relief to DuPont specification |
| Boiling water conditioning | 100°C | 3 h (3.2 mm) / 18 h (6.4 mm) | Accelerated moisture equilibration to near-50% RH |
Source: DuPont Zytel Design Guide Module II; Davantech PA66 machining guide.
5.7 Achievable Tolerances
Unfilled PA6 and PA66 machined from pre-dried, stress-relieved stock in a temperature-controlled environment (±2°C) holds ±0.05 mm on linear dimensions and ±0.08 mm on bores without post-machining conditioning. Tightening to ±0.02 mm is achievable but requires pre-dried and annealed stock, light finishing passes with thermal stabilization between rough and finish cuts, controlled environment machining, and dimensional verification after a minimum 24-hour post-machining stabilization period.
PA66 GF30 holds ±0.05 mm more consistently than unfilled grades due to lower CTE and reduced moisture uptake, but the abrasive fiber content accelerates tool wear, which shifts dimensions progressively on long production runs. Tool offset compensation every 10–20 pieces is standard practice on GF30 production runs where tolerances are tighter than ±0.08 mm.
| PA6 / PA66 unfilled | ±0.08–0.10 mm | ±0.05 mm | Pre-dried stock, annealed, temperature-controlled machining, 24h stabilization |
| PA6 / PA66 unfilled | ±0.05 mm | ±0.02 mm | All above + light finishing pass + post-machining conditioning |
| PA66 GF30 | ±0.05–0.08 mm | ±0.05 mm | Pre-dried stock, carbide tooling, tool offset compensation every 10–20 pcs |
For EPOC CRAFTER’s CNC machining capability on nylon and engineering plastics, see CNC Machining.
6. Thermal Expansion and Tolerance Stacking
PA66 dry-as-molded has a CTE of 81×10⁻⁶/°C at 23°C per DuPont Zytel 101 data from Minlon/Zytel Design Info Module 2, approximately 4.5× the CTE of aluminum (18×10⁻⁶/°C) and 9× the CTE of steel (11×10⁻⁶/°C). When nylon is assembled with metal components and the assembly experiences temperature variation, nylon expands and contracts at a rate that metal cannot match, producing interference at elevated temperature or clearance at low temperature depending on fit configuration.
Moisture absorption compounds the problem independently of temperature. A PA66 part at full saturation has a CTE of 117×10⁻⁶/°C, 45% higher than the dry value, meaning a wet part in a hot environment expands at nearly double the rate of a dry part at room temperature.
6.1 CTE Values by Grade and Condition
Glass fiber reinforcement in the flow direction reduces CTE substantially. Zytel 70G30HSL (PA66 GF30) measures 23×10⁻⁶/°C in the flow direction per DuPont Minlon/Zytel Design Module 2 data, less than one-third of the unfilled PA66 dry value. The transverse direction CTE remains higher, at 60–80×10⁻⁶/°C, creating anisotropic expansion behavior that must be accounted for in part orientation relative to the mold or extrusion direction.
| PA66 unfilled (Zytel 101) | 63 | Dry, 40°C | — | DuPont Minlon/Zytel Module 2 |
| PA66 unfilled (Zytel 101) | 81 | Dry, 23°C | — | DuPont Minlon/Zytel Module 2 |
| PA66 unfilled (Zytel 101) | 90 | Dry, 77°C | — | DuPont Minlon/Zytel Module 2 |
| PA66 unfilled (Zytel 101) | 117 | Saturated, 23°C | — | DuPont Minlon/Zytel Module 2 |
| PA66 GF30 (Zytel 70G30HSL) | 23 | Dry | Flow direction | DuPont Minlon/Zytel Module 2 |
| PA66 GF30 (Zytel 70G30HSL) | 60–80 | Dry | Transverse | Typical value |
| PA6 unfilled | 80–100 | Dry, 23°C | — | CAMPUS Plastics typical |
6.2 Thermal Expansion During Machining
The CTE of 81×10⁻⁶/°C means a PA66 rod at 23°C that heats to 93°C during aggressive machining (a 70°C rise) expands by:
ΔL = L₀ × CTE × ΔT = 1 mm × 81×10⁻⁶ × 70 = 0.0057 mm per mm of diameter
For a 50 mm diameter bore, this equals +0.28 mm dimensional growth at the cutting zone, eleven times a ±0.025 mm H7 tolerance band. A dimension measured mid-cut on a hot part and accepted as in-tolerance will be 0.28 mm undersize after the part cools to ambient temperature.
Finish cuts on nylon must be taken at low speed and light depth to minimize heat generation, and dimensions must be measured only after the part has returned to ambient temperature. At EPOC CRAFTER, the standard protocol for nylon parts with tolerances tighter than ±0.05 mm includes a mandatory thermal stabilization pause between roughing and finishing, with part temperature verified by contact thermometer before final measurement.
6.3 Tolerance Stacking in Nylon-Metal Assemblies
When a nylon component fits into a metal housing or onto a metal shaft, the dimensional relationship between the two changes with temperature and humidity. The calculation below covers a common case: PA66 unfilled bushing pressed into an aluminum housing, operating between 20°C and 80°C in a 50% RH environment.
At 20°C dry (as-machined state): clearance or interference is set at assembly.
At 80°C in service (60°C rise above assembly temperature):
- PA66 bushing expands: 81×10⁻⁶ × 60°C = 0.0049 mm/mm
- Aluminum housing expands: 23×10⁻⁶ × 60°C = 0.0014 mm/mm
- Net differential expansion: 0.0035 mm/mm. Nylon grows 0.35 mm per 100 mm of diameter relative to aluminum.
At 50% RH equilibrium (moisture contribution):
- Additional +0.6% linear growth on nylon = +0.60 mm per 100 mm diameter
- Aluminum: zero moisture-related growth
Total dimensional shift relative to aluminum over the full temperature and humidity range: up to +0.95 mm per 100 mm of nylon diameter. For a 20 mm bushing bore, this is +0.19 mm, enough to convert a designed clearance fit into an interference condition or to close a designed interference fit to the point of stress cracking.
Engineering Note The standard engineering response to nylon-metal differential expansion is to design the nylon feature with a clearance that accommodates the maximum expected dimensional growth at the highest service temperature and moisture condition, then verify the minimum clearance at the lowest expected temperature and moisture state does not produce excessive play. For PA66 GF30 in the flow direction (CTE 23×10⁻⁶/°C), differential expansion relative to aluminum is reduced to approximately 0.0000 mm/mm at moderate temperatures. GF30 flow-direction CTE is nearly identical to aluminum, making it the preferred grade when thermal stability in metal assemblies is the primary design constraint.
For prototyping nylon components and validating fit in metal assemblies before production, see Rapid Prototyping.
7. Chemical Resistance Boundaries
PA66 and PA6 resist hydrocarbons, oils, lubricants, and most aliphatic solvents at room temperature, which is why nylon dominates gear, bushing, and bearing applications in automotive and industrial environments where oil contact is continuous. The failure modes are specific and predictable: strong acids hydrolyze the amide bond (–CONH–) directly, zinc chloride and calcium chloride cause stress cracking under load, and strong oxidizing agents attack the polymer backbone at elevated temperature.
| Aliphatic hydrocarbons | Gasoline, diesel, mineral oil, kerosene | Good to excellent | None under normal conditions | Primary reason nylon is used in fuel system components |
| Aromatic hydrocarbons | Toluene, xylene | Good at room temperature | Stress cracking under load at elevated temperature | Test under actual stress and temperature conditions |
| Alcohols (short-chain) | Ethanol, isopropanol | Good | None under normal conditions | Compatible at room temperature |
| Lubricating oils and greases | Mineral oil, hydraulic fluid, cutting oil | Good to excellent | None | Suitable for oil-bath lubricated components |
| Dilute inorganic acids | Dilute HCl, dilute H₂SO₄ | Poor to moderate | Amide bond hydrolysis, mass loss, strength reduction | Avoid even brief contact; damage is cumulative |
| Concentrated inorganic acids | Concentrated H₂SO₄, concentrated HNO₃ | Not resistant | Rapid hydrolysis and dissolution | Prohibited contact |
| Dilute alkalis | Dilute NaOH, KOH | Good to moderate at room temperature | Chain scission at elevated concentration and temperature | Monitor concentration and temperature |
| Concentrated alkalis | High-concentration NaOH/KOH, hot alkali | Moderate to poor | Polymer chain degradation over time | Not recommended for continuous exposure |
| Zinc chloride | ZnCl₂ solutions | Not resistant | Stress cracking under load | Critical failure mode; prohibited in stressed parts |
| Calcium chloride | CaCl₂ solutions | Not resistant | Stress cracking under load | Same mechanism as ZnCl₂ |
| Phenols | Phenol, cresol | Not resistant | Dissolution and severe stress cracking | Prohibited contact |
| Oxidizing agents | Hydrogen peroxide, bleach, concentrated HNO₃ | Not resistant at elevated temperature | Backbone oxidation and degradation | Prohibited at elevated temperature |
| Esters and ketones | Ethyl acetate, acetone, MEK | Moderate to poor | Swelling, dimensional change, potential cracking | Test before specifying |
| Water and steam | — | Moderate (see Section 4) | Dimensional change, property reduction | Managed by design compensation; long-term steam exposure not recommended |
Source: DuPont Zytel Design Guide Module II (Sheet 6: boundary conditions and failure modes); BASF Ultramid chemical resistance guide.
7.1 Stress Cracking The Hidden Failure Mode
Stress cracking from zinc chloride, calcium chloride, and phenol-family solvents is the failure mode most frequently missed in chemical compatibility screening because the material shows no visible degradation in unstressed immersion testing. The crack initiates only when the part carries mechanical load simultaneously with chemical exposure.
DuPont Zytel Design Guide data confirms PA66 GF30 (Zytel 70G33L) passed a stress cracking screen at 93.2 MPa (13,500 psi) for 5 minutes at 23°C across 12 pure chemical environments with no cracking observed per DuPont Module II Sheet 7. Any chemical environment outside these tested conditions requires prototype testing under actual service stress before production release.
7.2 Thermal Degradation Boundary
PA66 begins oxidative degradation above approximately 120°C under prolonged load-bearing service, well below its melting point of 262°C and below the HDT values on datasheets. DuPont UL temperature index data defines the temperature at which a material retains 50% of its initial property value after 60,000 hours as the UL thermal index per DuPont Module II Sheet 7. For Zytel grades at 3.0 mm thickness, the UL mechanical index (without impact) is the conservative design limit for continuous structural service.
Long-term steam exposure is a specific degradation pathway for PA66: DuPont Module II data notes that internal stress accelerates deterioration in steam environments, and chlorinated water at 8 ppm chlorine concentration reduces service life by 20–30% compared to unchlorinated water at the same temperature. Parts for steam sterilization or hot water immersion above 80°C require specific grade selection and prototype validation.
Engineering Note Chemical resistance data in supplier datasheets is generated on unstressed specimens in ambient-temperature immersion. Real service conditions combine chemical exposure with mechanical stress, elevated temperature, and cyclic loading simultaneously. A material that shows “good” resistance in a compatibility table can fail under combined conditions. For any application where nylon contacts aggressive chemicals under load above 60°C, request prototype testing under combined stress and chemical exposure before committing to the grade.
For nylon parts in automotive and industrial chemical environments, see Automotive Applications and CNC Machining.
8. Frequently Asked Questions
What is the tensile strength of nylon 66?
PA66 unfilled tensile strength is 83 MPa at 23°C dry-as-molded (DAM) per ASTM D638, measured on Zytel 101L general-purpose grade. At 40°C DAM the value rises to 114 MPa as polymer chain mobility decreases. At 121°C DAM it falls to 43 MPa. Conditioning to 50% relative humidity reduces room-temperature tensile strength to 77 MPa. Glass fiber reinforcement at 30% by weight (PA66 GF30, e.g. Zytel 70G33L) raises tensile strength to 170–190 MPa DAM at 23°C, approximately 2.1× the unfilled value. All tensile strength values cited here are per ASTM D638 or ISO 527 at the stated condition; values from datasheets without stated test conditions and moisture state are not directly comparable.
How does moisture absorption affect nylon’s mechanical properties?
Water absorbed by PA66 acts as a plasticizer, reducing flexural modulus by 35–45% and tensile strength by approximately 7% from DAM to 50% RH equilibrium at 23°C. At full saturation (8.5% moisture content for PA66 unfilled per ASTM D570), flexural modulus drops from 2,800–3,500 MPa to 1,100–1,500 MPa and tensile strength falls to approximately 45–55 MPa. Toughness and elongation at break increase at the same time — the material becomes less stiff but more impact-resistant. For structural applications where stiffness under load is the design constraint, use DAM datasheet modulus values only for as-machined inspection; use 50% RH conditioned values for service load calculations.
Can nylon be CNC machined to tight tolerances?
PA66 and PA6 unfilled hold ±0.05 mm on linear dimensions when machined from pre-dried stock (moisture below 0.2%, dried at 80–90°C for 8–12 hours), annealed before finish machining (90–100°C for 2–4 hours with slow furnace cooling), and measured after a minimum 24-hour post-machining stabilization period at ambient temperature. Tightening to ±0.02 mm requires all of the above plus light finishing passes with thermal stabilization between rough and finish cuts, and dimensional verification after post-machining conditioning to service moisture state. PA66 GF30 holds ±0.05 mm more consistently than unfilled grades due to lower CTE (23×10⁻⁶/°C flow direction vs 81×10⁻⁶/°C for unfilled) and lower moisture uptake, but requires tool offset compensation every 10–20 pieces on production runs.
What cutting tools work best for machining nylon?
Unfilled PA6 and PA66 machine with HSS tooling with positive rake angle 10–15° and relief angle 8–12°, provided the edge is sharp. Worn HSS tooling generates heat by rubbing rather than cutting, producing dimensional errors and surface smearing. PA6 GF30 and PA66 GF30 require carbide tooling without exception. Glass fiber abrades HSS cutting edges within minutes. TiAlN-coated carbide is preferred for GF30 grades on runs above 50 pieces, extending tool life by 40–60% compared to uncoated carbide.
Does nylon need to be dried before CNC machining?
Yes. Pre-dry PA66 and PA6 stock at 80–90°C for 8–12 hours in a circulating air oven to bring moisture content below 0.2%. Machine within 4 hours of removal from the oven. Without pre-drying, tolerances tighter than ±0.10 mm are unreliable due to moisture-related dimensional expansion during cutting. PA66 GF30 requires the same pre-drying protocol despite its lower absorption rate of 0.70% at 24h vs 1.20% for unfilled PA66 per ASTM D570.
What is the HDT of nylon 6 vs nylon 66?
PA6 unfilled HDT is 65–80°C at 1.82 MPa and 180–200°C at 0.455 MPa per ISO 75. PA66 unfilled HDT is 90°C at 1.82 MPa and 235°C at 0.455 MPa per ASTM D648. Both values must be specified with load condition. Citing HDT without load condition introduces up to 145°C of ambiguity on PA66 alone per ASTM D648-01 Section 13.1.6. PA66 GF30 HDT at 1.82 MPa is 210–230°C, approximately 2.5× the unfilled PA66 value at the same load.
Why does nylon deform after machining?
Post-machining deformation in nylon parts has three independent causes. First, residual stress release: cutting removes material asymmetrically, unbalancing internal stresses built into the stock during extrusion or molding. Annealing at 90–100°C for 2–4 hours before finish machining reduces this effect. Second, thermal expansion contraction as the part cools from cutting temperature. A 70°C rise on PA66 produces 0.28 mm expansion on a 50 mm feature. Third, moisture re-absorption from ambient air causing dimensional growth of approximately 0.24% per 1% moisture content gain per DuPont Module II data. All three must be controlled independently through pre-drying, annealing, thermal stabilization pauses, and post-machining conditioning before final inspection.
How does glass fiber reinforcement change nylon CNC machinability?
GF30 reinforcement changes nylon machinability in four ways. Tool wear rate increases significantly, requiring carbide tooling only. Cutting speeds must be reduced 40–50% below unfilled grade parameters. Surface finish is rougher at equivalent parameters due to exposed fiber ends. Dimensional stability improves: CTE drops from 81×10⁻⁶/°C to 23×10⁻⁶/°C in the flow direction and 24-hour moisture absorption drops from 1.20% to 0.70% per ASTM D570, both of which reduce dimensional drift during and after machining. The trade-off is higher tooling cost and shorter tool life in exchange for better part stability on precision features.
Related Resources
Related Capability: CNC Machining at EPOC CRAFTER — covers achievable tolerances, surface finishes, and material options for precision CNC machined parts in nylon, metals, and engineering plastics.
Related Article: PA6 vs PA66 vs PA12: Nylon Grades Compared — Properties, Modifications, and When to Switch Materials— covers grade selection across the full polyamide family including GF reinforcement, MoS₂-filled variants, and substitution decisions against POM and PEEK.
Related Material Guide: Materials & Properties — covers material data, grade selection criteria, and processing characteristics across metals and engineering plastics available for CNC machining at EPOC CRAFTER.
Related Standards Reference: Tolerances & Standards — covers ISO 2768, IT grade tolerancing, and how to specify dimensional requirements for machined plastic and metal components.
Related Article: Nylon Parts for Engineering Applications: Industry Use Cases, Tolerances, and Procurement Checklist covers industry-specific nylon component requirements, tolerance specification for procurement, and supplier qualification checklist for machined nylon parts.
If you need EPOC CRAFTER to quote machined nylon components, upload your drawing to CNC Machining.
Ready to Quote Your CNC Prototype?
Upload your STEP file and get a DFM review + quote within 12 hours.
No tooling cost. Minimum 1 part.