These three plastics dominate machined-plastic parts for mechanical applications. They look similar on a data sheet but serve very different purposes. PEEK is the aerospace/medical premium. Delrin is the mechanical-part workhorse. UHMW is the sliding-surface specialist. Picking the wrong one wastes 3–10× in cost or delivers a part that fails in service.
The short version:
They overlap — you could machine a gear in all three. But you'd pay 10× for a PEEK gear vs Delrin with no benefit, and a UHMW gear would wear out in weeks. Picking right is about matching the material to the service conditions.
| Property | PEEK | Delrin (POM-C) | UHMW-PE |
|---|---|---|---|
| Tensile strength | 95 MPa | 68 MPa | 21 MPa |
| Flexural modulus | 3,700 MPa | 2,800 MPa | 760 MPa |
| Izod impact (notched) | 87 J/m | 53 J/m | No break |
| Continuous service temp | 250 °C | 100 °C | 80 °C |
| Short-term peak temp | 310 °C | 130 °C | 100 °C |
| Coefficient of friction | 0.3 | 0.2 | 0.1 (lowest) |
| Wear resistance | Excellent | Excellent | Excellent (sliding only) |
| Moisture absorption | 0.1% | 0.8% | 0.01% |
| Density | 1.32 g/cm³ | 1.41 g/cm³ | 0.93 g/cm³ (floats) |
| Machinability | Good (slow, expensive) | Excellent (like brass) | Fair (rubbery) |
| Cost (rod, per kg) | ~$300 | ~$25 | ~$15 |
| Lead time stocked? | Yes, common sizes | Yes, very common | Yes, very common |
PEEK costs 10–20× what Delrin costs. It's worth the premium only when one of these is true:
Delrin degrades above 100 °C. Nothing else machined in the engineering-plastic class handles 250 °C like PEEK. Automotive under-hood, turbine components, dryer internals — PEEK territory.
PEEK is USP Class VI and ISO 10993-certified. Used in spinal cages, bone screws, dental. X-ray translucent. Delrin has no such pedigree.
PEEK resists almost every cleaning chemistry used in semiconductor fabs. Wafer handling, plasma-chamber windows, chemical-bath parts. Delrin dissolves in strong acids and bases.
PEEK meets FAR 25.853 flame and smoke requirements. Used in engine bracketry, cabin interior parts. Certified traceable lots available.
PEEK tolerates 1,000 kGy gamma before significant degradation. Delrin starts breaking down at 20 kGy. Nuclear, medical sterilization, space applications.
PEEK is the WRONG answer when: the part sees normal indoor temperatures, isn't medical/aerospace/chemical-exposed, and doesn't need certifications. In 90% of "engineering plastic part" applications, PEEK is massive overkill.
Delrin is the default engineering plastic for mechanical parts. It's chosen when all of these are true:
Applications where Delrin shines:
UHMW (ultra-high-molecular-weight polyethylene) has a specific sweet spot: sliding surfaces where low friction and wear resistance matter more than load-carrying capacity.
UHMW fails when:
Four questions, in order:
Edge cases that trip people up:
Each has quirks on the mill:
PEEK: slow speeds (SFM ~150–300), sharp tools, adequate coolant. Chip breaks cleanly. Thin walls can warp under machining stress — anneal if going under 1.5 mm wall. Tool costs dominate — PEEK eats edges.
Delrin: machines like brass. Fast speeds (SFM 500–1000), almost any tool works, chips eject cleanly. Holds tight tolerances easily. Thermal expansion is moderate — no special fixture considerations for most parts. The easiest plastic to machine.
UHMW: the hardest to machine of the three, counterintuitively. Very low stiffness causes workpiece flex. Rubbery chips stick to tools. Low thermal conductivity builds heat under the cutter. Sharp high-rake tools, adequate coolant, light chip loads. Hold tight tolerances is difficult — ±0.15 mm is standard.
Email [email protected] with your drawing and service conditions. We quote all three and will push back if the spec looks wrong for the application.
Start a quote →