What tolerance can be held on a machined plastic part?

On a dimensionally stable plastic such as PEEK, PPS or acetal, a rigid machine with sharp tooling holds about ±0.075 mm (±0.003 in) in normal work and ±0.025 mm (±0.001 in) with care and temperature control. On PTFE, nylon or UHMW-PE the same machine holds no better than ±0.25 mm in practice, because the material itself moves more than that with temperature and humidity.

Why does my plastic part measure differently after a few days?

Almost always moisture or stress relaxation rather than a machining error. Nylon 6/6 absorbs 1.35% of its weight in water in 24 hours and swells roughly 0.5% linearly; PTFE, by contrast, absorbs almost nothing but moves with temperature. Machined-in stresses also relax over hours to days, bending thin sections. Condition the stock before final machining, or re-measure at the same temperature and humidity as the original measurement.

Should I anneal plastic stock before machining?

Yes, for tight-tolerance work in PEEK, PEI, PPS and other high-performance plastics. Extruded and moulded stock carries residual stress from cooling, and machining releases it unevenly, so the part distorts after the cut. A controlled anneal cycle — ramp up, hold below Tg or Tm, cool slowly — relaxes that stress and is the single most effective step for holding tolerance. Acetal and nylon benefit too, though less.

Do I need coolant when machining plastics?

Rarely, and sometimes it makes things worse. Most plastics machine cleanly dry, and dry machining avoids contaminating a material often chosen for chemical purity and avoids the stress-cracking that solvent-based coolants can cause in polycarbonate and PEI. Use air blast for chip clearance. Coolant helps on PEEK and other high-modulus materials where the cut generates real heat, and on any material where chip welding is a problem.

Plastic Machining Tolerances

Practical turning and milling tolerances for twelve engineering plastics, paired with thermal expansion and 24-hour water absorption — the two effects that move a machined plastic part after the cutter has left it.

Data verified 2026-09-29 · based on ASTM D696-16

Quick Answer

A well-set-up machine holds about ±0.075 mm in a dimensionally stable plastic such as PEEK, and roughly ±0.10 mm in acetal. That is not the limit that matters, though: PTFE, nylon and UHMW-PE move more than that from temperature and moisture alone, so the tolerance you can hold is set by the environment, not by the machine.

Find a plastic

Achievable Tolerance, Thermal Expansion and Water Absorption by Plastic

Material^[3]	CLTE µm/m·°C^[2]	Growth of a 100 mm part per 10 °C µm^[1]	Typical tolerance ± mm^[3]	Precision tolerance ± mm^[3]	Water absorption, 24 h %^[2]
PEEK #	47	47	0.075	0.025	0.15
PPS #	55	55	0.075	0.025	0.05
PEI (Ultem) #	56	56	0.075	0.025	0.25
PET-P (Ertalyte) #	60	60	0.10	0.025	0.10
Polycarbonate #	68	68	0.10	0.05	0.15
Acrylic (PMMA) #	70	70	0.10	0.05	0.30
Rigid PVC #	70	70	0.13	0.05	0.05
Nylon 6/6 #	80	80	0.25	0.10	1.35
POM-C (Acetal copolymer) #	110	110	0.10	0.025	0.22
POM-H (Delrin) #	122	122	0.10	0.025	0.25
PTFE #	135	135	0.25	0.10	0.01
UHMW-PE #	200	200	0.50	0.25	0.01

Tolerances are shop practice for stock-shape material on a rigid, well-set-up machine with sharp tooling — there is no governing standard for them, and a light machine or a dull cutter will do worse. The thermal expansion column is the part that matters: a 100 mm PTFE part grows 135 µm for a 10 °C temperature change, which is already larger than the precision tolerance in the same row. Dimensions are meaningless on a plastic part unless the measurement temperature is stated.

Why Plastic Tolerances Are Not Metal Tolerances

Three effects make a plastic part harder to hold to size than a metal one, and only the first of them is the machinist's problem.

Elastic deflection. Plastics have moduli between 0.7 and 3.8 GPa, against 200 GPa for steel. A workpiece held in a vise, or a thin wall pushed by the cutter, deforms elastically and springs back after the cut. The dimension you measure while the part is clamped is not the dimension you get when you release it. This is the reason soft jaws, light clamping and light finishing passes matter far more on plastic than on metal.

Thermal expansion. Coefficients of linear thermal expansion run from 47 to 200 µm/m·°C, against about 12 for steel and 23 for aluminium — five to sixteen times higher. A 100 mm acetal part grows 110 µm per 10 °C. Cutting fluid, a warm spindle and a warm shop can easily put the part 5–10 °C away from the 20 °C the drawing assumes. Measure at a controlled temperature, or accept the movement.

Moisture absorption. Hygroscopic plastics take up water from the air, and absorbed water pushes the chain segments apart, so the part swells. Nylon 6/6 absorbs 1.35% in 24 hours and can grow over 0.5% linearly between dry and equilibrium. PTFE and UHMW-PE absorb essentially nothing, which is why their printed tolerances look tighter relative to how soft they are.

The practical rule: for a dimensionally stable plastic, tolerance is a machine and tooling question. For nylon, PTFE or UHMW-PE, tolerance is an environment question and must be specified together with the temperature and moisture condition of measurement.

Machining Practice That Actually Holds Tolerance

Tooling. Sharp, polished, high-rake carbide with generous clearance. A dull edge rubs, generates heat, and on a low-melting plastic such as acetal or nylon melts the surface and produces a smear that no measurement can rescue.

Feeds and speeds. Higher feed per tooth than you would use on metal, at moderate surface speed. The objective is to cut the material cleanly, not to rub it away. Climb milling on the finish pass gives a better surface on most plastics.

Workholding. Soft jaws, minimum clamping pressure, and support under the cut. Where possible, clamp against a feature that will not be measured, and never measure a dimension while the part is still held.

Finishing. A final light pass — 0.1 to 0.2 mm — removes the work-hardened or heat-affected surface left by roughing and releases the internal stresses that would otherwise move the part after machining. On stress-relievable materials such as PEEK and PEI, annealing the stock before machining removes the residual stress from extrusion or moulding and is the single most effective step for holding a tight tolerance.

Deburring. Plastics burr by smearing rather than by cutting, so an abrasive belt tends to fold the burr over instead of removing it. Use a sharp scraper, a deburring blade, or a quick flame pass on the materials that tolerate it.

Frequently Asked Questions

What tolerance can be held on a machined plastic part?: On a dimensionally stable plastic such as PEEK, PPS or acetal, a rigid machine with sharp tooling holds about ±0.075 mm (±0.003 in) in normal work and ±0.025 mm (±0.001 in) with care and temperature control. On PTFE, nylon or UHMW-PE the same machine holds no better than ±0.25 mm in practice, because the material itself moves more than that with temperature and humidity.
Why does my plastic part measure differently after a few days?: Almost always moisture or stress relaxation rather than a machining error. Nylon 6/6 absorbs 1.35% of its weight in water in 24 hours and swells roughly 0.5% linearly; PTFE, by contrast, absorbs almost nothing but moves with temperature. Machined-in stresses also relax over hours to days, bending thin sections. Condition the stock before final machining, or re-measure at the same temperature and humidity as the original measurement.
Should I anneal plastic stock before machining?: Yes, for tight-tolerance work in PEEK, PEI, PPS and other high-performance plastics. Extruded and moulded stock carries residual stress from cooling, and machining releases it unevenly, so the part distorts after the cut. A controlled anneal cycle — ramp up, hold below Tg or Tm, cool slowly — relaxes that stress and is the single most effective step for holding tolerance. Acetal and nylon benefit too, though less.
Do I need coolant when machining plastics?: Rarely, and sometimes it makes things worse. Most plastics machine cleanly dry, and dry machining avoids contaminating a material often chosen for chemical purity and avoids the stress-cracking that solvent-based coolants can cause in polycarbonate and PEI. Use air blast for chip clearance. Coolant helps on PEEK and other high-modulus materials where the cut generates real heat, and on any material where chip welding is a problem.

Value Sources

Each data column on this page is tied to the source it came from. The numbers in square brackets correspond to the table headers above.

#	Source	Type	Revision / method
[1]	Thermal expansion of a 100 mm part per 10 °C	derived	ΔL = L × α × ΔT, with L = 0.1 m and ΔT = 10 °C, so ΔL in micrometres is numerically equal to the coefficient of linear thermal expansion in µm/m·°C. The column is a restatement of the CLTE column in shop-floor units, not an independent measurement.
[2]	Published unfilled-grade property data sheets (thermal section)	standard	supplier data sheets as published 2024–2026
[3]	Machining tolerances as practised by plastics stock-shape distributors	standard	n/a — shop practice, not a governing standard

Data Sources

Standard	Revision	What it covers on this page
ASTM D696 — Coefficient of Linear Thermal Expansion of Plastics	ASTM D696-16	the CLTE column and everything derived from it
ASTM D570 — Water Absorption of Plastics	ASTM D570-22	the 24-hour water absorption column
ISO 2768-1 — General tolerances for linear and angular dimensions	ISO 2768-1:1989	the general-tolerance convention the shop values are compared against

Cross-checked against:

CLTE and water absorption values checked between two independent suppliers' data sheets per material
Achievable tolerances compared against published machining guidance from plastics stock-shape distributors
Thermal expansion cross-checked against the arithmetic in ΔL = L × α × ΔT at several sizes

Derived values — the following values on this page are calculated, not taken directly from the standard:

Value	How it is derived
Growth of a 100 mm part per 10 °C	Restatement of the CLTE column: ΔL = L × α × ΔT with L = 0.1 m and ΔT = 10 °C, so ΔL in micrometres equals the CLTE in µm/m·°C. No independent measurement is involved.

Tolerances are practical shop values, not a specification. They assume sharp tooling, a rigid machine, controlled clamping and a stable shop temperature; they are not achievable on a light machine or with a dull cutter. For hygroscopic materials the tolerance must be quoted together with the moisture condition, and for all materials together with the measurement temperature — otherwise the number is not reproducible.

Accuracy and use. The values on this page are compiled from the published standards and cross-checked sources listed above. Where values are derived, the derivation is stated. No warranty, express or implied, is made as to the accuracy or completeness of this information, and no liability is accepted for any loss or damage arising from its use. Engineering reference data is provided for guidance in preliminary work — before a value is used for design, fabrication or acceptance testing, verify it against the current revision of the governing standard and against your own inspection. The user assumes all risk and responsibility in connection with the use of this information.

Cite This Page

Every value on this page is traceable to the sources listed above. If you use the data in a document, paper or report, cite it as:

Plain text

Engineering Reference. "Plastic Machining Tolerances." https://refmanual.com/charts/plastic-tolerance-machining/ (accessed 2026-09-29). Data from ASTM D696 — Coefficient of Linear Thermal Expansion of Plastics, ASTM D696-16.

BibTeX

@misc{plastic-tolerance-machining,
  title        = {Plastic Machining Tolerances},
  author       = {{Engineering Reference}},
  howpublished = {\url{https://refmanual.com/charts/plastic-tolerance-machining/}},
  note         = {Accessed: 2026-09-29},
  source       = {ASTM D696 — Coefficient of Linear Thermal Expansion of Plastics, ASTM D696-16}
}

RIS (EndNote, Zotero, Mendeley)

TY  - ELEC
TI  - Plastic Machining Tolerances
AU  - Engineering Reference
PB  - Engineering Reference
UR  - https://refmanual.com/charts/plastic-tolerance-machining/
PY  - 2026
DA  - 2026-09-29
N1  - Source: ASTM D696 — Coefficient of Linear Thermal Expansion of Plastics, ASTM D696-16
ER  -

Each row in the tables above also has a permanent link — hover a row and use the # link to cite a single value rather than the whole page.