Valve Seat Seal Material Selection: PTFE vs PEEK vs Metal Seat — Leakage Class and Pressure Rating

Overview #

When sourcing valve seat seals from China, the specification that procurement teams most consistently get wrong is not the material grade — it’s the leakage class. Buyers specify PTFE or PEEK by name, approve a sample that passes initial dimensional checks, and then discover at commissioning that the seat-to-disc contact geometry was never verified against ISO 5208 leakage rate requirements. The material is correct. The sealing performance is not. That gap between material compliance and functional compliance is where most China-sourced valve seat failures originate.

Seat seal material selection — PTFE, PEEK, or metal — is ultimately a pressure-temperature-media decision, but the sourcing decision is a qualification and consistency decision. A PTFE seat that passes leakage class A at 16 bar in the first shipment and drifts to class D by the third shipment is not a material problem. It is a supplier process control problem, and it is far more common in the Chinese valve seat supply chain than most buyers expect.

Material Performance Parameters: What the Specification Must Define #

The three seat materials — PTFE, PEEK, and metal (typically 316 SS, Stellite 6, or hardened 17-4 PH) — occupy distinct performance envelopes. The comparison below is drawn from published material data and ASTM International test standards, not from supplier marketing sheets.

PTFE’s compressive creep behavior is the parameter most buyers underweight. At sustained contact stress above 7 MPa, virgin PTFE will cold-flow, and the seat geometry will shift over time. Filled PTFE grades — 25% glass-filled or 15% carbon-filled — reduce creep significantly, but the filler content must be specified on the drawing and verified on the COA. A supplier who delivers unfilled PTFE when glass-filled was specified will pass a hardness check and fail a 72-hour compression set test.

PEEK seats are the correct choice when operating temperature exceeds 180°C continuously or when system pressure exceeds 40 bar with a soft-seat requirement. The compressive strength of 100 MPa means PEEK maintains contact geometry under load conditions that would permanently deform PTFE. The trade-off is machinability: PEEK requires tighter CNC tolerancing to achieve the same sealing geometry, and Chinese suppliers who machine PEEK seats on equipment calibrated for PTFE will produce parts that are dimensionally within drawing tolerance but functionally out of spec on contact angle.

For pump valve seals in high-pressure or high-temperature service, metal seats are the only technically defensible choice above PN 100. The leakage class achievable with metal-to-metal contact depends almost entirely on surface finish — Ra ≤ 0.4 µm on the seating face is the threshold for Class B leakage per ISO 5208. We have seen Chinese suppliers deliver metal seats with Ra 1.6 µm finish, which is acceptable for many machined surfaces but will not achieve Class B leakage regardless of material grade.

Leakage Class Verification: The Test Protocol That Separates Qualified Suppliers #

ISO 5208 defines leakage classes A through G for industrial valves, with Class A representing zero measurable leakage (bubble-tight) and Class D permitting up to 0.006 × DN mm³/s. Most buyers specify “Class A” on the purchase order without specifying the test medium, test pressure, or test duration — and Chinese suppliers will test to whatever conditions produce a passing result.

The correct specification for incoming inspection is:

• Test medium: water (for Class A–D) or air at 6 bar (for Class A bubble test)
• Test pressure: 1.1 × rated pressure (not nominal pressure)
• Test duration: minimum 60 seconds sustained at test pressure
• Acceptance criterion: zero bubbles for Class A; measured leakage rate ≤ class limit for B–D

In our supplier qualification program, we require three consecutive production batch test reports before recommending volume approval. The reason is not that a single test is unreliable — it is that leakage class performance in Chinese valve seat production is highly sensitive to tooling wear. A supplier running 500 seats per week on a single CNC fixture will produce excellent geometry in weeks 1–2 and degraded contact angles by week 4 as the fixture wears. Three consecutive batch reports spanning at least 6 weeks of production will reveal this pattern. A single PPAP sample will not.

For PTFE seats specifically, we apply ASTM D395 Method B compression set testing as a qualification gate: compression set must be ≤ 25% after 22 hours at 23°C for standard service, or ≤ 30% after 70 hours at 150°C for elevated-temperature service. Suppliers who cannot provide this data — not just hardness, but compression set — are not qualified for critical sealing applications regardless of price.

Most procurement teams over-specify tensile strength on PTFE seat COAs and under-specify the parameter that actually determines long-term sealing integrity: compression set after sustained load at operating temperature. Tensile strength for PTFE is easy to test and easy to pass. Compression set at temperature requires a 70-hour test cycle that many Chinese compounders simply do not run unless a buyer explicitly requires it.

Supplier Qualification Protocol: Certifications, COA Requirements, and Red Flags #

Certifications to Request #

For valve seats used in industrial process applications, the minimum certification baseline from a Chinese supplier should be:

• Material traceability certificate — linking the seat material to a named raw material supplier (e.g., Daikin, Solvay, or Victrex for PTFE/PEEK resin). Generic “domestic resin” declarations are a red flag.
• Dimensional inspection report — 100% or AQL 1.0 sampling per ISO 2859-1, with actual measured values (not just pass/fail) for seat bore diameter, contact angle, and surface finish Ra.
• Leakage test certificate — per ISO 5208, with test conditions and actual measured leakage rate recorded.
• Material hardness certificate — Shore D for PTFE, Rockwell M for PEEK, HRC for metal seats, with instrument calibration date.
• RoHS/REACH declaration — required for EU-destined product; verify against ECHA REACH SVHC candidate list.

For seats used in food, pharmaceutical, or potable water service, additionally require FDA 21 CFR 177.1550 compliance declaration for PTFE, or NSF International NSF/ANSI 61 certification for wetted components.

Minimum COA Requirements Checklist #

A COA that does not include all of the following should be rejected at incoming inspection:

• [ ] Material designation (grade, filler content if applicable)
• [ ] Raw material lot number and resin supplier name
• [ ] Shore D / Rockwell hardness — actual measured value, not range
• [ ] Compression set result — test method, conditions, and % value
• [ ] Dimensional report reference number (linked to inspection record)
• [ ] Surface finish Ra — seating face only
• [ ] Leakage test result — class achieved, test pressure, test medium
• [ ] Batch/lot number with production date
• [ ] Inspector signature and QC stamp
• [ ] Calibration certificate reference for test instruments used

A COA that lists only hardness and a generic “passed QC inspection” statement is not a COA — it is a shipping document with a header. We see this from approximately 40% of first-time Chinese valve seat suppliers we evaluate.

Red Flags for Substandard Suppliers #

In our qualification program, the following patterns have consistently predicted production-volume quality failures:

Red Flag 1 — Sample approval, production failure. The supplier passes PPAP with excellent dimensional and leakage data, then delivers out-of-spec material at volume. The trigger is almost always a raw material substitution at the compounder level — switching from named-brand PTFE resin to a domestic equivalent without buyer notification. A standard COA will not catch this without incoming hardness spot-testing and compression set verification on production lots.

Red Flag 2 — Tolerance compression on drawings. The supplier returns a signed drawing with tolerances tightened on non-critical dimensions (OD, length) and loosened on critical dimensions (contact angle, bore concentricity). This is a deliberate quality-gaming strategy. Always verify that the supplier’s internal inspection drawing matches your approved drawing exactly.

Red Flag 3 — No lot-to-lot consistency data. Three out of five Chinese valve seat suppliers we evaluated in a recent qualification round could not produce six months of consecutive batch COAs showing compression set and leakage class data. The absence of historical data is not a documentation problem — it means the tests were never run.

Red Flag 4 — Surface finish reported as a range. Ra “0.4–0.8 µm” on a COA means the supplier is not measuring individual parts — they are reporting a process capability estimate. For metal seats targeting Class B leakage, Ra must be measured per part or per batch sample, not estimated from process parameters.

The English technical content available for Chinese valve seat suppliers is almost entirely produced by Western valve brand owners and engineering standards bodies. Chinese suppliers rarely publish English-language technical data sheets with actual test values. That documentation gap is precisely why buyers approve suppliers based on price and sample appearance rather than on verified performance data — and why leakage failures at commissioning are so common in China-sourced valve assemblies.

Dimensional Verification: Incoming Inspection Protocol #

For o-rings and static seals and valve seat components, dimensional verification at incoming inspection should follow a structured protocol, not a visual check.

Critical dimensions to measure on every incoming lot (AQL 1.0 per ISO 2859-1):

• Seat bore diameter: tolerance ±0.05 mm for PN ≤ 40; ±0.025 mm for PN > 40
• Contact angle: ±0.5° from nominal (verify with optical comparator or CMM, not a protractor)
• Seating face surface finish: Ra ≤ 0.4 µm for metal seats (Class B target); Ra ≤ 0.8 µm for PTFE/PEEK
• Seat OD concentricity to bore: TIR ≤ 0.05 mm
• Seat height/thickness: ±0.1 mm

The contact angle is the dimension that most incoming inspection teams skip because it requires an optical comparator or CMM rather than a caliper. It is also the dimension most likely to drift in Chinese production when tooling wears. A 1° deviation in contact angle on a PTFE seat will not be visible, will not affect Shore D hardness, and will not be caught by a bore diameter check — but it will produce a Class D leakage result in a valve rated for Class A. The difference sounds marginal. In production, it accumulates.

Practical Guidance for Buyers #

When sourcing valve seat seals from China, the first document to request is not a product catalog or a price list — it is three consecutive batch COAs showing compression set data, leakage class test results, and raw material lot traceability. Most Chinese suppliers will provide a single sample COA on request. The ability to provide six months of production batch data is the actual qualification gate, and fewer than half of suppliers we evaluate can meet it.

The most common sourcing mistake is approving a supplier based on PPAP sample performance without establishing a production monitoring protocol. PTFE seat geometry is sensitive to raw material lot variation and tooling wear. A supplier who delivers Class A leakage performance on a 50-piece sample can deliver Class D performance on a 500-piece production run if the resin lot changes or the CNC fixture is not recalibrated. The consequence is not a warranty claim — it is a valve that passes factory acceptance testing and fails at process startup, after installation costs have been committed.

Before committing to volume order, require a witnessed leakage test per ISO 5208 at 1.1 × rated pressure on a production-representative sample of minimum 10 pieces from a production lot (not a pre-selected sample). Require compression set data per ASTM D395 Method B for PTFE and PEEK seats. These two tests, run on production material rather than qualification samples, will identify 90% of the quality risks before they reach your facility.

Frequently Asked Questions #

Q1: What is the most critical specification to verify on a COA for PTFE valve seats?

A: Compression set — not hardness. Hardness is easy to pass and easy to fake; compression set per ASTM D395 Method B at operating temperature is what determines whether the seat will maintain contact geometry and leakage class over service life.

Q2: When should I specify PEEK instead of PTFE for valve seat seals?

A: Specify PEEK when continuous service temperature exceeds 180°C or when system pressure exceeds 40 bar with a soft-seat requirement. PEEK’s compressive strength of 100 MPa versus PTFE’s ~12 MPa means it maintains contact geometry under load conditions that will cause PTFE to cold-flow and lose leakage class. See the comparison table above for the full performance envelope by material.

Q3: What is the most common quality failure mode in China-sourced valve seats?

A: Raw material substitution at the compounder level after initial qualification approval. The supplier passes PPAP with named-brand PTFE resin, then switches to a domestic equivalent at production volume without notification. The substituted material passes hardness checks but fails compression set testing — and the leakage class degrades from Class A to Class C or D in service. The only reliable catch is incoming compression set spot-testing on production lots, not just on qualification samples.

Q4: What certifications should I require for valve seats going into EU process applications?

A: At minimum: ISO 5208 leakage test certificate with actual measured values, ECHA REACH SVHC declaration, and dimensional inspection report per ISO 2859-1 AQL 1.0. For food or pharmaceutical contact, additionally require FDA 21 CFR 177.1550 compliance for PTFE or NSF International NSF/ANSI 61 certification. Request the actual test reports, not just a declaration letter.

Q5: Is a metal seat always better than PTFE for high-pressure applications?

A: Not automatically — it depends on the leakage class requirement and the surface finish the supplier can actually achieve. A metal seat with Ra 1.6 µm finish will not achieve Class B leakage regardless of material grade. A properly specified PTFE seat at PN 40 will achieve Class A. The material is not the variable; the manufacturing process control is.

Published by sinoraw.com Technical Team | Request a sourcing consultation