Guide
Valve Seat Leakage: Classification & Standards
Valve seat leakage is classified by international standards. This guide explains the FCI 70.2 and API 598 leakage classes, what each class means, and how to test and specify the right leakage class for your application.
7 min readUpdated July 2026
Leakage Classification Standards
No industrial valve is perfectly leak-tight when closed. The amount of acceptable leakage depends on the application. Two standards define leakage classes:
- FCI 70-2 (formerly ANSI B16.104): The most widely referenced standard for control valve seat leakage. Defines six classes (Class I through Class VI) with progressively tighter leakage limits.
- API 598: The standard for industrial valve testing (gate, globe, check, ball, butterfly). Defines allowable leakage rates based on valve size and type, tested with water or air.
FCI 70-2 Leakage Classes
| Class | Leakage Rate | Typical Application |
|---|---|---|
| Class I | No test required | Non-critical isolation, low-pressure systems |
| Class II | 0.5% of valve capacity | General service, non-hazardous fluids |
| Class III | 0.1% of valve capacity | Moderate isolation requirements |
| Class IV | 0.01% of valve capacity | Standard for most control valves |
| Class V | 0.0005 ml/min per mm of port diameter per bar differential | Tight shutoff for high-pressure or hazardous service |
| Class VI (bubble-tight) | No visible bubbles (tested with air/water) | Critical isolation, hazardous/toxic service, food/pharma |
Causes of Excessive Seat Leakage
- Wire-drawing (erosion): High-velocity fluid through a nearly-closed valve cuts grooves in the seat surface. Once a groove is cut, the seat can't seal. Fix: replace or lap the seat. Prevent by ensuring the valve is fully open or fully closed, not throttled near closed.
- Particulate damage: Solid particles (scale, rust, weld slag) get trapped between the seat and the closing member, creating a leak path. Fix: clean the internals. Prevent with an upstream strainer.
- Chemical attack: The fluid is incompatible with the seat material, causing it to swell, harden, or dissolve. Fix: select a compatible seat material.
- Thermal distortion: Rapid temperature changes distort the valve body and seat, breaking the seal. Fix: use a valve rated for the temperature range and ramp temperatures gradually.
- Mechanical wear: Normal operation wears the seat over time. The rate depends on the cycle frequency, pressure, and fluid properties. Fix: replace the seat at scheduled intervals.
- Improper closure: The actuator doesn't have enough force to push the closing member fully against the seat, or the valve is installed backwards. Fix: verify actuator sizing and installation direction.
Testing for Seat Leakage
- Bubble test (Class VI): Connect a tube from the valve outlet to a water container. Pressurise the inlet with air. Count the bubbles per minute — Class VI allows zero bubbles for most sizes.
- Hydrostatic test: Pressurise the valve with water at 1.1× the rated pressure. Measure the leakage flow rate through the outlet.
- Ultrasonic test: Use an ultrasonic leak detector on the downstream side. Sensitive enough to detect very small leaks in pressurised gas systems.
- Pressure decay test: Pressurise the upstream side, close the valve, and monitor the upstream pressure. Any pressure decay indicates leakage through the seat.
Selecting the Right Leakage Class
- Class IV is sufficient for most general process control applications
- Class V for high-pressure steam and hydrocarbon service where leakage is costly or hazardous
- Class VI for toxic, hazardous, or food/pharmaceutical applications where any leakage is unacceptable
- Specify metal seats for high-temperature or abrasive service; soft seats for tight shutoff at moderate temperatures
- Consider the cost — higher leakage classes require more expensive seat materials and tighter manufacturing tolerances
- Verify that the actuator has sufficient force to achieve the specified leakage class — tight shutoff requires more actuator force