What Causes Ball Valve Seat Extrusion at High Pressure? | Polymer Clearance, Differential Pressure, Temperature

Find the Gap Where the Polymer Moved

Before deciding whether PTFE, RPTFE or PEEK is the problem, look at where the material actually went.

Common extrusion paths include:

• between the seat outside diameter and the seat pocket;
• between the polymer seat and metal retainer;
• behind the seat;
• between a polymer insert and metal carrier;
• beside the seat edge near the valve bore.

Follow the pressure direction. The first polymer edge that loses full metal support is usually the area worth checking most closely.

A thin lip pushed into that gap is not the same as ordinary wear. Extrusion moves material into a space. Erosion removes material from the seat.

Check the Hot Gap, Not the Drawing Gap

The clearance shown on a drawing is only the starting point. The seat may see a different gap once the valve is assembled, pressurized and hot.

The operating gap can change because of:

• seat OD and ID tolerance;
• seat thickness tolerance;
• seat-pocket diameter;
• retainer dimensions;
• ball diameter;
• seat-carrier dimensions;
• pressure-induced movement of metal parts;
• different thermal expansion of polymer and metal;
• assembly preload.

Small changes are easy to ignore when looking at one part, but several changes can stack together.

For example, if three dimensions each move only 0.02 mm in the unfavorable direction, the combined dimensional change can reach 0.06 mm. That is still a small number, but it can matter when the unsupported edge beside the seat is only a fraction of a millimetre wide.

The same issue applies to a nominal gap. If an unsupported gap starts at 0.05 mm and tolerance, heat or pressure movement adds another 0.05 mm, the available extrusion path has doubled.

These figures are illustrative, not universal valve tolerances. The value that matters is the worst operating clearance for the actual valve.

Use Differential Pressure, Not Line Pressure

For a closed valve, upstream pressure by itself does not tell you how hard the seat is being loaded.

The upstream gauge reads 200 bar in both cases. The second condition, however, has about 10 times the end-to-end differential pressure.

Pressure-generated force follows:

Force = Differential Pressure × Effective Pressure Area

The effective pressure area comes from the valve and seat geometry. It should not automatically be treated as the valve bore area.

For illustration, a 10 MPa differential acting over a circular effective diameter of 25 mm gives an area of about 491 mm² and a force of about 4.91 kN.

These are pressure-area examples, not valve ratings. The useful point is the change in force: when effective diameter doubles, area and pressure force increase by about four times.

This is also why a ball valve leaking at low differential pressure needs a different diagnosis from a valve damaged after a large ΔP event. One gauge reading does not show the full seat-loading condition.

Do Not Read Pressure Class as a Seat Limit

Pressure class, shell strength and polymer-seat life are different checks.

ASME B16.34 covers pressure-temperature ratings, materials, dimensions, tolerances, examination, testing and marking for valves within its scope.^[1]

For a soft-seated ball valve, check three separate limits:

• Pressure boundary: whether the body and pressure-containing parts can safely hold the pressure.
• Shutoff: whether the seat can achieve the required leakage performance at the actual differential pressure.
• Seat life: whether the polymer can keep its shape after pressure, temperature, time and cycling.

ISO 17292:2015 also covers metal ball valves as complete valve assemblies and includes material, inspection and testing requirements.^[2]

A high pressure class does not mean every polymer seat can run at the class maximum pressure and the seat’s highest temperature at the same time. For elevated-temperature service, use the rating for the actual body material, pressure class and seat configuration. The site’s ball valve pressure-class selection at elevated temperature explains why Class 150, 300 or 600 is not a fixed psi value.

A Floating Ball Usually Loads the Downstream Seat Harder

A floating ball can move slightly between the two seats. Once the valve closes and differential pressure builds, the ball is pushed toward the downstream seat.

That means the downstream seat can carry far more load than the upstream seat.

If the downstream seat is heavily damaged while the upstream seat still looks relatively normal, check:

• actual pressure direction;
• maximum ΔP;
• ball movement;
• downstream seat support;
• ball diameter;
• seat interference;
• contact-band width.

A soft-seated floating ball valve uses this pressure-assisted downstream sealing action.

The contact mark left on the removed seat is useful. A narrow, deeply compressed band suggests the load was concentrated over a small area. If the band is uneven around the circumference, check misalignment, dimensional variation, ball runout or debris before blaming the polymer alone.

Larger Floating Valves Can Create Much Higher Seat Force

Two valves can run at the same pressure and still place very different loads on their seats.

In otherwise similar floating-ball designs, a larger valve usually has a larger projected pressure area. The earlier example shows how quickly the force changes: at the same 10 MPa differential, increasing the illustrative effective diameter from 25 mm to 50 mm raises the force from about 4.91 kN to 19.63 kN.

Those numbers are not direct ratings for 25 mm or 50 mm valves. The real effective area depends on the actual design. They do show why a 1-inch and 6-inch floating valve should not be treated as the same seat-loading problem just because both run at 100 bar.

The mechanical difference is easier to see when comparing floating and trunnion ball valves at increasing size and pressure.

For Trunnion Valves, Check SPE and DPE

In a trunnion-mounted valve, the ball is mechanically supported instead of being pushed downstream like a floating ball.

The seat assembly normally uses a metal carrier, springs and a polymer insert. Pressure acting over different seat diameters creates an axial force on that carrier.

Too much seat force can:

• increase contact stress;
• raise operating torque;
• compress the polymer insert;
• push the insert toward a retaining gap;
• leave permanent deformation.

A typical soft-seated trunnion ball valve therefore needs to be checked as a pressure-energized seat system, not simply as a fixed polymer ring.

SPE: line pressure normally pushes the seat toward the ball. In a self-relieving design, sufficiently high cavity pressure can change the force balance and allow the seat to move away from the ball.

DPE: pressure from either side can push the seat toward the ball. Because of that, a DPE seat should not automatically be expected to vent excess cavity pressure by itself.

Also check whether the valve uses SPE/SPE, DPE/DPE or a mixed arrangement. The cavity-pressure behavior is different.

Separate Heat Damage from Creep

Heat does more than make a polymer “softer.” It can change the seat in several different ways.

NASA testing of PTFE compressive creep specifically examined the effect of applied stress and temperature, which is why time and temperature should be checked together instead of relying only on a short pressure test.^[3]

A five-minute pressure test and 1,000 hours of continuous loaded service are not the same creep condition. One thousand hours is about 42 days under load.

The 1,000-hour figure is only an illustrative example, not a universal seat-life requirement. Its purpose is to show why time under load matters when a valve fails after a long isolation period.

Thermal Cycles Add Up Quickly

A valve that heats and cools once per day sees about 365 thermal cycles per year. At two complete cycles per day, that becomes about 730 cycles per year.

Each cycle can change:

• seat interference;
• polymer compression;
• extrusion clearance;
• ball-to-seat contact;
• operating torque.

If the polymer does not fully recover after each cycle, those small changes can build up over time.

A valve that leaks only when hot, only when cold, or only after repeated startup and shutdown should therefore be checked for dimensional change and permanent set, not just chemical compatibility.

PTFE, RPTFE and PEEK Solve Different Problems

PTFE works well where low friction, chemical resistance and good contact with the ball are important. Under long-term mechanical load, however, creep and cold flow can become the weak point.

RPTFE or filled PTFE uses fillers such as glass, carbon, graphite or bronze to change stiffness, wear and compression behavior. The label “RPTFE” by itself is not enough to tell you how the seat will behave because the filler and percentage matter.

PEEK is normally stiffer and more resistant to long-term deformation than virgin PTFE. Published research on non-reinforced PEEK still shows measurable creep under sustained loading, so it should not be treated as a material that cannot move under load.^[4]

The site’s PTFE, PEEK and metal-seat comparison gives more detail on these differences.

A metal-seated ball valve removes soft polymer from the main ball-to-seat sealing interface, although polymer or elastomer seals may still be used elsewhere in the valve.

Cavity Pressure Can Reverse the Seat Load

A closed ball valve can trap fluid between its seats. If that trapped liquid heats up, body-cavity pressure can rise and load the seat from a direction that is different from normal line pressure.

API 6D distinguishes different isolation and cavity-pressure arrangements, including DBB and DIB configurations, and includes requirements related to valve cavity behavior.^[5]

If both seats are damaged, or the extrusion appears on the side you did not expect, check:

• upstream pressure;
• downstream pressure;
• body-cavity pressure;
• trapped-liquid heating;
• SPE/DPE arrangement;
• cavity vent or relief path;
• shutdown sequence.

The upstream gauge is not always showing the highest pressure acting on the seat.

Do Not Confuse Extrusion with Erosion or Cutting

A partly open ball valve can create high local flow velocity around the seat edge. With solids in the fluid, the seat can lose material very quickly, but the damage may have nothing to do with extrusion clearance.

Rapid gas decompression is another separate failure mode. Internal cracking or blistering should not be called extrusion unless the material has actually been pushed into an unsupported mechanical gap.

Read the Failure Pattern Before Replacing the Seat

Measure the Failed Parts Before Cleaning Them

Do not clean, trim or replace the failed seat before the useful evidence has been recorded.

Before any inspection or maintenance, the valve must first be isolated from pressure and other stored hazardous energy. OSHA 29 CFR 1910.147 covers control of hazardous energy during servicing and maintenance.^[6]

Once the valve is safe, record:

• seat OD and ID;
• seat thickness;
• seat-pocket diameter;
• ball diameter;
• retainer overlap;
• extrusion-lip width and direction;
• ball scratches or coating damage;
• debris in the body cavity.

For uneven deformation, record seat thickness at 0°, 90°, 180° and 270° instead of relying on one measurement.

For example, if seat thickness changes from 5.00 mm to 4.85 mm after testing, the reduction is 0.15 mm, or 3%, at that measurement point. That number does not prove the root cause by itself, but it gives you something useful to compare with an unused seat and with other positions around the same seat.

Similar checks are useful during trunnion ball valve seat inspection and maintenance.

If It Failed After Maintenance, Check Assembly First

If the valve worked before overhaul and failed soon after the seat was replaced, start with the replacement parts and assembly.

Check for:

• wrong replacement-seat dimensions;
• wrong polymer grade;
• seat installed in the wrong direction;
• incorrect preload;
• incorrect retainer position;
• debris trapped behind the seat;
• a nicked or folded seat edge;
• body assembly creating excessive seat compression.

A correct PEEK or PTFE grade can still fail if the installed gap or preload is wrong.

Use the Actual Pressure and Temperature History

“Normal operating pressure” is not enough for a failure investigation.

Record:

• upstream pressure;
• downstream pressure;
• maximum ΔP;
• startup pressure;
• shutdown pressure;
• reverse-pressure events;
• body-cavity pressure;
• hydrotest pressure;
• rapid venting;
• short pressure spikes.

A system operating steadily at 100 bar can, for example, briefly see 120 bar during a transient. That is a 20% pressure increase. If downstream pressure falls at the same time, the actual increase in seat ΔP can be even larger.

The 100/120 bar case is illustrative. Use the highest credible pressure condition from the real process, not just the normal gauge value.

For temperature, record:

• normal valve temperature;
• short hot periods;
• steam or cleaning cycles;
• startup and shutdown temperatures;
• heat tracing;
• number of thermal cycles.

Torque Can Warn You Before the Valve Leaks

A seat can start changing shape before leakage becomes obvious. Torque may show the problem first.

Watch for:

• higher breakaway torque;
• higher running torque;
• sticking at the start of travel;
• slower actuator movement;
• actuator torque alarms;
• incomplete travel;
• torque that rises sharply with ΔP.

For example, if breakaway torque rises from 80 N·m to 110 N·m after pressure and thermal cycling, the increase is 37.5%. The valve may still pass a leakage test, but a change of that size is worth investigating.

The example is illustrative; acceptable torque change depends on the valve design and qualification criteria.

Breakaway, running and reseating torque are different values and should not be mixed together. The site’s valve torque curve guide explains the difference.

If the valve becomes hard to turn mainly after a long shutdown, also check trapped pressure, deposits and temperature effects. These failure paths are covered in the article on ball valves that jam after shutdown.

A Passed Factory Test Does Not Prove Long-Term Seat Life

ISO 5208:2015 covers pressure testing used to assess valve pressure-boundary integrity, closure tightness and related test requirements.^[7]

Factory testing cannot reproduce every condition a valve may see during years of service.

Actual service may add:

• thousands of hours under load;
• hot operating temperature;
• hundreds of thermal cycles;
• thousands of valve operations;
• contamination;
• reverse pressure;
• cavity-pressure events;
• short pressure spikes.

A valve can therefore pass leakage testing when new and still develop creep or extrusion later.

Match the Repair to the Failure Pattern

Do These Five Checks Before Switching to PEEK

• Measure the extrusion gap. PEEK cannot correct poor mechanical support.
• Check torque. A stiffer seat can increase breakaway and running torque.
• Check low-pressure shutoff. A harder seat may not conform to the ball as easily as PTFE.
• Inspect ball finish. Scratches and dimensional errors become more important with a less flexible seat.
• Confirm the exact valve rating. Raw PEEK material data are not the pressure-temperature rating of the finished valve.

Verify the Repair with More Than a Leak Test

After changing the seat material, support or clearance, check the valve under conditions that are as close as practical to the conditions that caused the original failure.

Useful checks include:

• required pressure and temperature together;
• pressure hold where creep is important;
• pressure cycling;
• thermal cycling;
• reverse-pressure testing where applicable;
• cavity-relief verification;
• torque before and after cycling;
• leakage before and after cycling;
• seat thickness before and after testing;
• inspection for small extrusion lips.

ISO 23632:2021 covers design-validation type testing of metallic ball and butterfly valves used for isolation and includes seat-performance, torque and durability evaluation through mechanical and thermal cycling. Its validation program includes 205 mechanical cycles under the specified test framework.^[8]

Do not stop at “the valve still seals.” If torque has risen sharply or the seat has already lost measurable thickness, the design may already be moving toward another failure even before visible leakage appears.

Ask for These Numbers Before Approving the Valve

• exact PTFE, RPTFE or PEEK grade;
• filler type and formulation where applicable;
• pressure-temperature rating for the exact valve size;
• maximum allowable differential pressure;
• reverse-pressure capability;
• SPE/DPE seat arrangement;
• cavity-relief behavior;
• hot operating torque;
• cycle-test or design-validation information;
• replacement-seat dimensions and material specification.

“PEEK seat, 300 bar rated” is not enough information for a severe application. The rating needs to apply to the actual valve size, seat geometry, temperature and pressure direction.

FAQ

What normally starts ball valve seat extrusion?

Pressure pushes the polymer toward an unsupported clearance. Extrusion starts when the load is high enough, the operating gap is large enough, or the material deforms enough for part of the seat to enter that gap.

Why can the downstream seat fail first?

In a floating ball valve, differential pressure pushes the ball toward the downstream seat. That seat can therefore carry much more load than the upstream seat.

Can PTFE pass a pressure test and still creep later?

Yes. A short pressure test and thousands of hours under hot continuous load are different conditions. Creep depends on time, so the seat can continue changing shape even when the operating pressure stays the same.

Does PEEK eliminate extrusion?

No. PEEK usually resists deformation better than virgin PTFE, but a large unsupported gap can still cause trouble. PEEK can also change torque, seat contact and low-pressure sealing.

How can I tell extrusion from erosion?

Extrusion normally pushes material into a nearby mechanical gap. Erosion normally removes material from an area exposed to high flow velocity or particles.

Why can both seats fail?

Check body-cavity pressure, reverse pressure and the SPE/DPE arrangement. The pressure acting on each seat may not match the normal upstream-to-downstream flow direction.

What should be measured on a failed seat?

Record seat OD, ID and thickness, the extrusion-lip position, seat-pocket and retainer dimensions, ball diameter, contact marks and ball-surface damage. Measuring thickness at several positions around the seat is more useful than taking only one reading.

Finally

A useful seat-extrusion investigation needs numbers, not just a photo of the damaged seat. Record the real differential pressure, worst operating gap, seat temperature, time under load, ball and seat dimensions, and torque before replacing parts. A 10 MPa differential acting over only a 25 mm effective diameter already produces about 4.91 kN; doubling that diameter raises the illustrative force to 19.63 kN. If PTFE is changed to PEEK or the clearance is reduced, verify the result with pressure-temperature testing, torque checks and post-test seat measurements. Leakage alone can miss early extrusion or permanent deformation.