Mon–Sat 08:30–18:00 (GMT+8) Founded 2009 Dongguan, China

TECAPEEK natural data sheet: reading it like a machinist

Designers read a data sheet to choose a grade. Machinists read it to set up a job. The same forty lines answer two different questions, and only one of them is printed.

Material Selection 7 min read
TECAPEEK natural data sheet: reading it like a machinist

There are two ways to read a PEEK data sheet. A designer reads it to decide whether the grade can survive the application. A process engineer reads the same sheet to decide how to hold the part, how fast to cut it, when to anneal it and what to measure. The document serves both, but only the first reading is obvious, which is why so many tight-tolerance PEEK jobs go wrong at the setup stage rather than at the design stage.

This is the second reading, using TECAPEEK natural as the example. It covers which property maps onto which process decision, how the family around it differs, and the three assumptions that sit behind every number on the page.

One property, one process decision

The conversion is mechanical once you know the mapping. Each line on the sheet answers a question the machinist is going to face on the shop floor.

Sheet value Process decision it drives
Tensile modulus, about 3.7 GPa Clamping force and support. PEEK deflects under chuck pressure; a three-jaw chuck on a thin ring will ovalise it before the tool touches it
Elongation at break, 20 to 40 percent Whether a burr folds or breaks off, and whether a threaded feature can be cut or must be an insert
Thermal conductivity, about 0.25 W/mK Heat stays at the cut. No coolant means the chip carries it, so feed rate and chip load matter more than spindle speed
CTE, about 47 µm/mK below Tg Measurement temperature and the tolerance you can honestly promise. A 100 mm part grows about 0.1 mm over a 20 C shop swing
Water absorption, 0.1 to 0.5 percent Whether the part must be conditioned before final measurement, and how long it must sit
HDT, about 152 C Whether an anneal below that temperature is safe under the part’s own weight
Volume resistivity, 10^14 to 10^16 ohm·cm Whether the part needs ESD handling, and whether swarf must be kept out of electronics assembly

Two of these are routinely ignored and cause most of the scrap. Modulus decides clamping, and clamping decides roundness. CTE decides measurement, and measurement decides whether a good part gets rejected.

The family, and why the sheet looks the same across it

TECAPEEK is a family, not a single product, and the data sheets are laid out identically because the tests are the same tests. That uniformity is a trap: the numbers differ by factors, not percentages.

  • natural — unfilled virgin PEEK, the grade in this article. Highest elongation, lowest modulus, the most forgiving to machine.
  • GF30 — 30 percent glass. Modulus roughly triples, elongation drops to 2 to 3 percent, and tool life becomes the dominant cost line. Abrasive, and it chips at thread crests.
  • CF30 — 30 percent carbon. Stiffer and stronger than GF30, conductive, and the swarf is electrically conductive dust that must be extracted, not blown.
  • TF / HPV / bearing variants — PTFE, graphite and carbon combinations for low friction. Softer matrix, cleaner cutting, but lower strength and a much narrower load window.
  • MT / medical grades — same polymer, different documentation, dedicated lines and a different supply chain. Not a machining variant; a paperwork variant.

The practical consequence: a process card written for natural will produce scrap on GF30 and a poor surface on TF. The cutting data, the clamping, the tool and the dust handling all change. Reading the family column first is not optional.

Turning the sheet into a process card

Four settings matter more than the rest of the card combined.

Annealing, before and after. Stress-relieved stock is the starting point, but rough machining reintroduces stress. For anything with a wall under 10 mm or a flatness call under 0.1 mm, rough, anneal, then finish. The anneal ramp should be slow enough that the part does not see a gradient across its section, and the cool-down should be slower than the ramp. Rushing the cool is the usual reason a part moves a week after dispatch.

Clamping and support. Use soft jaws, a collet, or a fixture that supports the whole face. Point contact from a jaw deforms the part elastically, the cut is made in the deformed state, and the part springs back oval. For thin rings, machine in a sacrificial bore or use a vacuum plate.

Cutting data. Sharp, high-rake tooling, positive geometry, generous chip clearance. PEEK cuts like a soft metal with a tendency to smear if the tool rubs. A dull edge generates heat, the surface melts locally, and the result is a smeared finish that will not clean up. Carbide is fine for natural; diamond-coated pays back on GF30 and CF30.

Cooling and cleaning. Air blast rather than flood coolant for most work, because PEEK does not need lubrication and wet chips are harder to reclaim. If coolant is used, it must be compatible and the part must be dried before measurement, or the water absorption line on the sheet becomes a measurement error.

The three assumptions behind every number

A data sheet is a measurement report, and the measurement has conditions.

  • The specimen is not your part. Values come from standard test bars, moulded and cooled under controlled conditions. The middle of a 60 mm block has a different cooling history, and therefore a different crystallinity and a different modulus. A thick section will not match the sheet.
  • Dry as moulded. Mechanical values are quoted on conditioned-dry specimens. A part that has been in a humid shop for a month is not in that state, and its dimensions are not the ones in the inspection plan.
  • 23 C. Every dimensional and mechanical value is at room temperature. A part that will run at 120 C needs a different tolerance on the drawing, not a different grade.

From data sheet to first-article inspection

The sheet also tells you what to measure, and in what order.

Measure after the part has stabilised, not straight off the machine. For thin sections that means a settling period in the inspection environment. Measure the features that the sheet’s properties actually threaten: flatness on a thin plate, roundness on a bored ring, and centre distance on anything long, because those are the three that CTE and stress move most. Record the temperature at which the measurement was taken; without it, the number is not comparable to the next one.

Where a drawing carries a tolerance tighter than the CTE can justify over the service temperature range, raise it before machining rather than after. A 0.02 mm tolerance on a 150 mm part that will see a 100 C excursion is not a machining problem, it is a drawing problem.

Questions that come up on the shop floor

Can I machine TECAPEEK and KETRON PEEK-1000 with the same program? In practice yes for unfilled grades of both, since the polymer is the same and the machining behaviour follows the polymer. The differences are in stock consistency and documentation, not in the cutting data.

Why do two suppliers show different values for the same grade name? Different test standards, different specimen preparation, different dates. Compare the test method column, not just the number, and ask for a batch certificate when the difference matters.

Is the MT grade harder to machine? No. It is the same polymer with different documentation and process controls. The cost difference is in the paperwork and the traceability, not in the cutting.

How long should a part sit before final inspection? Until it has stopped moving, which for thin sections in a stable shop is usually overnight. Measuring at machine temperature is the single most common cause of a part that passes inspection and fails at the customer.

Top Plastic supplies unfilled PEEK in stress-relieved sheet and rod, with batch certificates, annealing on request and machining to drawing. Send the drawing with the service temperature on it and we will return the achievable tolerance and the process route, rather than a data sheet value that assumes 23 C and dry.

Need a material recommendation?

Send us the application and we will reply with two or three candidate grades and a price.

Ask an engineer
Share this article

Send us your drawing, get a price in 24 hours

Tell us the material, the size and how many pieces. If you are not sure about the material, describe the application and we will recommend one.

Request a quote Browse stock shapes