Clean soft-seated ball valves are often suitable for rough and medium vacuum. High vacuum, repeated bakeout, frequent cycling, particles or contamination-sensitive processes may require a specially built vacuum valve. Selection should be based on absolute pressure, effective pumping speed, gas type, temperature, seat design and measured leak rate—not only on the words “vacuum rated” or “bubble-tight.”
Vacuum Range
Vacuum is measured as absolute pressure. A lower absolute pressure means a deeper vacuum. The ranges below are common engineering descriptions, although the exact boundaries may differ slightly between industries.
| Vacuum range | Approximate absolute pressure | Ball valve use |
|---|---|---|
| Rough vacuum | 1,000 to 1 mbar | Many industrial ball valves can work after leakage and material checks |
| Medium vacuum | 1 to 10−3 mbar | Clean, vacuum-tested soft-seated valves are often suitable |
| High vacuum | 10−3 to 10−7 mbar | Low-gas-load materials, controlled cleaning and helium testing are normally required |
| Ultra-high vacuum | Below 10−7 mbar | Ordinary industrial ball valves are rarely the first choice |
These ranges are not fixed valve limits. A clean, purpose-built valve may work at a lower pressure than expected. A contaminated valve with weak stem packing may cause problems even in rough vacuum. ISO 3529-1 provides the general vocabulary used in vacuum technology.[1]
With one port near atmospheric pressure and the other near zero absolute pressure, the differential pressure is about 1.013 bar at standard atmospheric conditions. Most industrial valve bodies can withstand this load. The harder task is controlling very small gas flows through the seats, stem seals, body joints and non-metallic materials.
Pressure Rating
A Class 150, Class 600, PN16 or PN40 rating mainly describes the valve’s positive-pressure capability at a stated temperature. It does not prove that the valve can maintain a low vacuum pressure.
A normal pressure rating does not confirm:
- Leakage across the closed ball
- Outside air entering around the stem
- Leakage through body gaskets
- Gas passing through elastomer seals
- Outgassing from seats, grease or cleaning residue
- Leakage after heating or repeated operation
ISO 5208 covers pressure-boundary and closure-tightness testing for industrial metallic valves.[2] These tests remain useful, but a water or compressed-gas pressure test is not the same as a vacuum helium test.
ASME B16.34 covers pressure-temperature ratings, materials, dimensions, testing and marking for several forms of industrial valves.[3] Vacuum testing should be added to these mechanical requirements, not used instead of them.
Positive-pressure strength still matters. A vacuum line may see nitrogen purging, cleaning pressure, pump backpressure or an upstream process fault. Liquid trapped inside a closed valve cavity can also produce high pressure when heated.
Carilo’s API 6D ball valve factory acceptance test guide explains common shell, seat and functional checks. API Specification 6D defines manufacturing and testing requirements for pipeline valves, but it is not a complete high-vacuum qualification standard.[4]
Sealing Direction
The direction of the pressure difference can change how tightly a ball valve seals. This is especially important in floating ball valves.
When one port is at atmospheric pressure and the other is under vacuum, pressure pushes a floating ball toward the vacuum-side seat. That seat normally carries most of the pressure-assisted sealing load.
The two seats may use the same material but still perform differently because of:
- Different initial compression
- Machining tolerances
- Relief grooves behind one seat
- A vent hole in the ball
- Different cavity passages
- Unequal seat support
A valve may pass the required leakage limit in one direction but fail after the pressure direction is reversed. The test certificate must identify the vacuum port, higher-pressure port and test direction.
Before installation, confirm:
- Whether the ball is floating or trunnion mounted
- Which port faces the vacuum chamber
- Which port normally sees the higher pressure
- Whether the seats are mechanically symmetrical
- Where the ball and cavity vents are located
- Which direction was used during factory testing
- Whether pressure can reverse during operation
Do not select the vacuum direction only from an external flow arrow. Ask for a sectional drawing that shows the ball, seats and vent passages in the closed position.
Low Pressure Difference
A floating ball receives strong pressure-assisted sealing when one side is at atmospheric pressure and the other side is under vacuum. It receives much less help when both sides are already at low pressure.
| Pressure condition | Approximate differential pressure | Effect on floating ball |
|---|---|---|
| Atmosphere versus 10−3 mbar | About 1,013 mbar | Strong pressure-assisted seat loading |
| 10−2 versus 10−4 mbar | 0.0099 mbar | Very little pressure-assisted seat loading |
The two examples differ in pressure differential by roughly 100,000 times. This is why a valve that seals well between atmosphere and vacuum may perform differently when both ports are already evacuated.
At low differential pressure, sealing depends more on:
- Initial seat preload
- Seat elasticity
- Ball roundness
- Ball surface finish
- Seat alignment
- Spring loading
- Correct actuator stops
Pressure Force and Valve Size
The force pushing a floating ball toward the lower-pressure seat increases with the projected area exposed to the pressure difference.
The simplified relationship is:
Force = pressure difference × projected area
The following values use a differential pressure of 1 bar and a circular projected area. They are theoretical examples, not actual actuator-torque or seat-contact values.
| Projected diameter | Projected area | Force at 1 bar differential |
|---|---|---|
| 50 mm | About 1,963 mm² | About 196 N |
| 100 mm | About 7,854 mm² | About 785 N |
| 150 mm | About 17,671 mm² | About 1,767 N |
Doubling the projected diameter increases the area and pressure force by about four times. Larger floating ball valves can therefore develop much higher seat loading, operating torque and polymer deformation than smaller valves.
The actual load depends on the ball geometry, port size, seat diameter, body clearances and friction. These figures should be used to understand the scale of the force, not to size an actuator.
Floating Ball Valves
A floating ball is supported by the stem and two seats. It can move slightly toward the lower-pressure side. This simple design can provide tight shutoff with a compliant soft seat.
Repeated pressure reversal moves the ball between both seats. A system that alternates between vacuum, atmospheric venting and positive-pressure purging should therefore be tested in both directions and after a suitable number of operating cycles.
Carilo’s forged soft-seated floating ball valve guide shows the basic floating-ball construction. Vacuum suitability still depends on the exact seat grade, stem seal, lubricant, cleaning method and completed-valve test result.
Trunnion Ball Valves
A trunnion-mounted ball is supported at the top and bottom. The ball has little sideways movement. Spring-loaded or pressure-loaded seat rings move toward the ball instead.
This design reduces operating torque in larger valves, but vacuum performance depends on the pressure path behind each seat. The terms single-piston effect and double-piston effect describe seat movement; they do not describe a helium leak class.
Carilo’s trunnion-mounted ball valve selection guide explains the basic construction, pressure classes and end connections.
SPE and DPE Seats
A single-piston-effect seat normally uses line pressure to push the seat toward the ball. If cavity pressure rises above line pressure, the seat may move away from the ball and release cavity pressure into the pipe.
In vacuum service, the engineer must check where that gas is released. A self-relieving seat may protect the valve cavity but send trapped gas toward the vacuum chamber.
A double-piston-effect seat can use pressure from either the line side or body-cavity side to increase contact with the ball. This may improve isolation, but it can also trap pressure in the cavity.
A DPE arrangement may need a separate cavity-relief or monitoring connection. SPE, DPE, double block and bleed, and double isolation and bleed do not prove that both directions have the same vacuum leakage.
Seat Materials
Soft seats can deform slightly and fill small surface marks on the ball. This often gives lower gas leakage than a normal metal seat. The trade-off is that polymers can creep, wear, release gas and allow gas to pass through the material.
| Seat material | Main advantage | Main vacuum concern |
|---|---|---|
| PTFE | Low friction and good sealing | Creep, cold flow and gas permeation |
| Filled PTFE | Better wear and shape retention | Performance depends on filler type and percentage |
| PCTFE | Lower gas permeability and better dimensional stability | Harder surface needs better ball finish and alignment |
| PEEK | High strength and improved creep resistance at elevated temperature | Less able to fill scratches or machining errors |
| Metal | High-temperature and particle resistance | Needs accurate contact, finish and lapping |
A PTFE seat can lose preload after long compression, repeated heating, particle damage or incorrect actuator adjustment. “Filled PTFE” is not a complete specification because glass, carbon and other fillers change hardness, wear and cleanliness.
PCTFE and PEEK can improve shape retention, but harder seats need better ball roundness, smoother surfaces and closer alignment. A stronger material does not automatically produce a lower leak rate.
Metal seats tolerate high temperature and abrasive material, but low gas leakage may require a matched and lapped ball-seat set. Replacing only the ball or one seat can change the result.
Temperature limits depend on the exact compound, filler, pressure and valve design. Carilo’s ball valve temperature and material guide explains why the body rating and seat rating must be checked separately.
Stem Seals
The stem seal is often the main route for outside air to enter the valve. Unlike a body gasket, it must remain tight while the stem rotates.
Packing must be compressed enough to stop air entry without creating excessive operating torque. Packing load can fall because of temperature cycling, stem wear, vibration or material relaxation. Live-loaded packing uses springs to help maintain compression.
O-rings can seal well when the elastomer, groove, compression and stem finish are correct. However, gas can slowly pass through an elastomer by permeation. Tightening the gland may close a physical gap, but it cannot stop gas moving through the seal material.
A welded metal bellows separates the moving stem from atmosphere and can greatly reduce stem leakage. It also adds cost, size, welds and a limited fatigue life. Bellows or double seals with a monitored space are often considered for toxic, reactive or high-purity gases.
Body Joints
A three-piece valve is easier to clean and repair but has two body joints. A one-piece valve has fewer joints but is harder to inspect internally.
Body-joint leakage depends on:
- Gasket material
- Sealing-surface finish
- Bolt preload
- Tightening sequence
- Temperature cycling
- Cleanliness during reassembly
After a valve is disassembled, the original factory leak result is no longer enough. The rebuilt valve should be cleaned and tested again.
Body Cavity
A closed ball valve can trap gas around the ball, behind the seats and inside the ball bore. That gas may enter the vacuum line slowly through a narrow internal path.
This is called a virtual leak. There may be no opening to atmosphere, but the pressure curve can look like a small external leak.
A cavity volume of only 20 cm³ equals 0.02 L. If it is initially filled at about 1,000 mbar, it contains approximately:
1,000 mbar × 0.02 L = 20 mbar·L of gas
If this gas can escape only through a narrow path, it may continue entering the vacuum chamber long after the main line has been pumped down.
Common trapped spaces include:
- The cavity around the ball
- The closed ball bore
- Spaces behind the seats
- Blind threaded holes
- Stem recesses
- Small gaps below gaskets
A vent hole can connect the ball bore or cavity to one port. The correct direction depends on the internal design. The vent should not automatically face the vacuum side. It may help empty the cavity in one design but release trapped gas into a clean chamber in another.
The cavity can also trap liquid. If that liquid is heated, thermal expansion can create pressure much higher than the normal vacuum differential. Confirm whether the cavity self-relieves, where the fluid is released and whether a separate relief connection is needed.
Leakage, Outgassing and Permeation
A vacuum system that will not reach its target pressure does not always have a physical leak. The gas may come from five different paths.
| Gas source | What happens | Useful check |
|---|---|---|
| External leakage | Air enters through the stem, body joint, weld or connection | Spray helium around the outside of the valve |
| Seat leakage | Gas crosses the closed ball from the higher-pressure port | Test the closed valve with pressure across the seats |
| Outgassing | Water, oil or volatile material leaves internal surfaces | Compare pressure over time and after cleaning or bakeout |
| Permeation | Gas passes through a polymer or elastomer | Use a timed test and compare seal materials |
| Virtual leakage | Gas leaves a trapped internal volume slowly | Review the sectional drawing and isolate internal cavities |
Carilo’s stem leak and seat leak troubleshooting guide explains how to separate common internal and external leak paths.
Outgassing
Outgassing is gas or vapor released from materials after pressure is reduced. Common valve sources include water on metal surfaces, machining oil, solvent residue, grease, adhesives, polymer seats and elastomer seals.
NASA’s outgassing database reports total mass loss and collected volatile condensable material using ASTM E595 test methods.[5] This data can help screen a material, but it does not prove that the complete valve is clean or leak-tight.
A seat compound may have good material-level results while the assembled valve still contains oil, excess lubricant or trapped solvent. The finished valve must therefore be cleaned and tested as a complete unit.
Gas-Load Calculation
Near a stable operating condition, pressure, gas load and effective pumping speed are related by:
Q = S × P
- Q = total gas load in mbar·L/s
- S = effective pumping speed at the chamber in L/s
- P = chamber pressure in mbar
The effective speed at the chamber can be much lower than the pump’s catalog value because pipes, bends, filters, fittings and valves restrict gas flow. NIST vacuum guidance explains that connection diameter, pipe length and bends affect system conductance and the pumping speed available at the chamber.[6]
| Target pressure | Effective pumping speed | Maximum total gas load at that pressure |
|---|---|---|
| 1 × 10−3 mbar | 10 L/s | 1 × 10−2 mbar·L/s |
| 1 × 10−5 mbar | 20 L/s | 2 × 10−4 mbar·L/s |
| 1 × 10−7 mbar | 50 L/s | 5 × 10−6 mbar·L/s |
These values show the mathematical relationship only. They are not universal valve acceptance limits.
For example, a chamber operating at 1 × 10−5 mbar with an effective pumping speed of 20 L/s has a total gas-load limit of:
20 × 1 × 10−5 = 2 × 10−4 mbar·L/s
If 20% of this example budget is assigned to four isolation valves:
2 × 10−4 × 20% ÷ 4 = 1 × 10−5 mbar·L/s per valve
The 20% allowance is only an example. A real calculation must include chamber-wall outgassing, gauges, feedthroughs, process gas, other seals, aging and measurement uncertainty.
Multiple Valves
Small leakage values add together when several valves are connected to the same vacuum space.
Four valves each contributing 1 × 10−5 mbar·L/s create a combined gas load of:
4 × 1 × 10−5 = 4 × 10−5 mbar·L/s
With an effective pumping speed of 20 L/s, this gas load alone produces a pressure contribution of:
4 × 10−5 ÷ 20 = 2 × 10−6 mbar
A system targeting 1 × 10−5 mbar would use about 20% of its pressure allowance on these four valves alone. This is why the number of valves must be included in the gas-load budget.
| Combined valve allowance | Number of valves | Example allowance per valve |
|---|---|---|
| 8 × 10−5 mbar·L/s | 2 | 4 × 10−5 mbar·L/s |
| 8 × 10−5 mbar·L/s | 4 | 2 × 10−5 mbar·L/s |
| 8 × 10−5 mbar·L/s | 8 | 1 × 10−5 mbar·L/s |
The final specification should also leave room for seal wear, test uncertainty and leakage from other components.
Cleaning and Lubricants
Vacuum cleaning should remove oil, particles, moisture and cleaning residue without damaging the seats or seals.
A practical process is:
- Disassemble the valve and remove visible contamination.
- Clean and rinse all parts with compatible fluids.
- Dry the parts fully and inspect them.
- Reassemble with clean gloves, tools and controlled lubricant.
- Leak-test and package the finished valve.
“Degreased,” “oxygen cleaned” and “vacuum cleaned” are different requirements. The supplier should state the procedure, permitted chemicals, inspection method and packaging conditions.
Many standard ball valves contain grease around the seats, stem or O-rings. An unsuitable grease can release vapor and contaminate gauges, optical surfaces or products. A dry valve is not always better because some seats and seals need a small amount of approved lubricant to prevent wear.
The purchase specification should identify the lubricant, permitted amount, application points, cleaning method and maximum bakeout temperature.
Bakeout
Heating helps remove water and volatile material from a vacuum system. The metal-body temperature limit is not enough; every seat, O-ring, gasket, packing ring, lubricant and actuator accessory must also survive the cycle.
The supplier should state:
- Maximum bakeout temperature
- Permitted valve position during heating
- Number of qualified heating cycles
- Whether the actuator must be removed
- Whether the valve must be tested while hot or after cooling
A room-temperature test does not prove that the valve will remain tight after repeated heating and cooling. Different thermal expansion rates can reduce seat preload, relax a gasket or change stem-packing compression.
End Connections
The valve may be tight while its pipe connection leaks. End connections should be selected according to temperature, cleanliness, maintenance and required pressure.
Threaded connections contain helical spaces and usually need tape or sealant. They may be suitable for rough vacuum but are harder to clean and qualify for low gas loads.
Quick-release vacuum connections use an O-ring and carrier. ISO 2861 specifies dimensions for clamped vacuum couplings, O-rings and carriers.[7] Dimensional compatibility does not remove the need for an assembled-system leak test.
Knife-edge flanges use a metal gasket and are widely used where high-temperature bakeout and low gas load are required. ISO 3669 specifies dimensions for knife-edge vacuum flanges.[8]
Welded ends remove detachable seals but can introduce oxide, particles and distortion. Clean and leak-test the valve after welding.
Port Size and Conductance
A full-port ball valve normally restricts gas flow less than a reduced-port valve. However, the smallest opening in the complete line controls the effective flow to the pump.
Restrictions may include:
- Reducers
- Small fittings
- Long narrow tubing
- Filters
- Sharp bends
- Small vent passages
Carilo’s full-bore and reduced-bore ball valve guide explains the basic bore difference. In a vacuum line, pipe length, connection size and bends must also be included in the conductance check.
Helium Leakage Testing
A vacuum ball valve normally needs two separate tests:
- Seat test: measures gas passing across the closed ball
- Envelope test: measures gas entering through the valve body, stem and joints
ISO 20485 describes leak-testing techniques that use a tracer gas and a tracer-gas-specific detector.[9] The buyer must still state the pressure, direction, temperature, helium exposure time and acceptance value.
Seat Test
- Close the valve using the specified actuator pressure or torque.
- Connect the intended vacuum-side port to the leak detector.
- Evacuate that port.
- Keep the other port at the stated higher pressure.
- Apply helium to the higher-pressure side.
- Wait for the defined response time.
- Record the background and measured leakage.
- Reverse the test if both directions must be qualified.
The report must identify the test direction. The words “seat tested” do not show which seat carried the sealing load.
Envelope Test
- Arrange the valve so the internal test space can be evacuated.
- Connect the valve to the detector.
- Measure the test-fixture background.
- Apply helium around the stem, body joints, welds and connections.
- Use a helium enclosure when total external leakage is required.
- Record the highest valid result.
Local helium spraying helps locate a defect. An enclosure gives the total external leakage but may not identify the exact leak point.
Test Background
The detector, temporary seals, hoses and test fittings create their own background signal. The background must be low enough to separate the valve leakage from the test equipment.
For example, assume the valve acceptance limit is 1 × 10−7 mbar·L/s, while the fixture background is already 7 × 10−8 mbar·L/s. Only 3 × 10−8 mbar·L/s remains between the background and the rejection limit.
This narrow difference makes the result sensitive to drift, temperature and connection quality. The fixture should be improved before the valve is accepted or rejected.
The test record should include:
- Detector background before helium exposure
- Fixture-only background where practical
- Reference-leak reading
- Valve reading
- Background recovery after testing
Test Timing
Do not simply wait for an undefined “stable reading.” Large cavities, long hoses and narrow passages can delay the detector response. Helium passing through an elastomer may cause a slower rise after the first response.
The procedure should state:
- Helium exposure time
- Expected detector response time
- Measurement window
- Background recovery requirement
- How delayed permeation is treated
A rapid detector rise after helium reaches the stem usually suggests a direct leak path. A slower increase may be caused by a long internal path, a trapped volume or gas permeating through an elastomer.
Temperature During Testing
Temperature affects seals, valve dimensions and helium reference leaks. NIST notes that some helium permeation leak artifacts can change by as much as 4% per degree Celsius.[10]
| Temperature change | Possible change at 4% per °C |
|---|---|
| 1°C | About 4% |
| 2°C | About 8% |
| 5°C | About 20% |
These percentages apply to some permeation-type reference leaks, not to every ball valve. They show why the reference-leak temperature, valve temperature and room temperature should be recorded during a sensitive test.
Other Test Methods
A pressure-rise test measures how quickly chamber pressure increases after the pump is isolated. It is useful for checking the complete system, but it cannot by itself separate external leakage, seat leakage, outgassing, permeation and trapped gas.
Bubble and pressure-decay tests can find larger leaks. They may be suitable for preliminary checks or rough-vacuum equipment, but “bubble-tight” is not a numerical high-vacuum acceptance value.
Leakage Units
Common leakage-rate units include:
- mbar·L/s
- Pa·m³/s
- Torr·L/s
- std cm³/s
Useful approximate conversions are:
- 1 mbar·L/s = 0.1 Pa·m³/s
- 1 Pa·m³/s = 10 mbar·L/s
- 1 std cm³/s is approximately 1 mbar·L/s
Always check the reference temperature and pressure used for standard volume. Do not compare two catalog values only by the exponent. The gas, unit, direction and test method must also match.
Leakage Levels
The table below helps show the difference between leakage-rate orders of magnitude. It is not a universal classification or acceptance standard.
| Leakage-rate order | Practical meaning |
|---|---|
| 10−3 mbar·L/s | A large gas load for many small vacuum systems |
| 10−5 mbar·L/s | May fit part of an industrial vacuum gas-load budget |
| 10−7 mbar·L/s | A much tighter requirement that needs controlled helium testing |
| 10−9 mbar·L/s | A range published for some purpose-built high-vacuum components |
A change from 10−5 to 10−7 mbar·L/s is not a small improvement. It is a 100-fold reduction in gas flow. Moving from 10−5 to 10−9 mbar·L/s is a 10,000-fold reduction.
Leakage Limit
There is no universal leakage limit for every vacuum ball valve. The allowable value depends on:
- Target chamber pressure
- Effective pumping speed
- Chamber volume
- Required pump-down time
- Number of valves and seals
- Process gas
- Allowable contamination
- Expected wear
- Measurement uncertainty
A leak rate that is acceptable in a large vacuum dryer may be too high for a small analytical chamber. Set the valve limit from the system gas-load budget, not from the lowest number shown on the detector.
Use separate limits for:
- Seat leakage
- External leakage
- Normal and reverse directions
- New and post-cycle conditions
- Room, hot or cold temperatures
Cycle Testing
A new valve may pass its first test and leak after repeated operation. The ball slides across the seats, while the stem packing rotates and changes compression.
Operating frequency changes the required qualification level:
| Operating frequency | Approximate cycle count |
|---|---|
| 2 operations per year | 20 operations in 10 years |
| 10 operations per day | About 3,650 operations per year |
| 1 operation per minute | 1,440 operations per day |
A maintenance isolation valve used twice per year should not have the same cycle requirement as an automated process valve operating every minute.
A practical qualification test can include:
- Initial seat and envelope tests
- A stated number of open-close cycles
- Pressure reversal where required
- Heating and cooling cycles
- Final leakage tests
- Operating torque measurement
Actuator Check
The final test should use the actual actuator, bracket and coupling. Common problems include insufficient torque, excessive torque, incorrect mechanical stops and stem side loading.
A closed position switch does not prove that the ball has reached the correct sealing position. The switch can operate before the ball reaches its final mechanical stop.
Record:
- Pneumatic supply pressure or electric actuator setting
- Closing torque where measured
- Mechanical stop position
- Open and closed position-switch settings
- Leakage after the actuator is installed
Process Conditions
Vacuum drying, distillation and resin degassing can send water vapor, solvent or process by-products into the valve. Vapor may condense in a cool body, collect behind the seats and evaporate during the next pump-down.
Particles can also be carried across the seat as the ball rotates. Hard particles may cut a channel in a soft seat, while sticky deposits may stop the ball before it fully closes.
A standard on-off ball valve should not be used for precise vacuum pressure control unless it has been designed and tested for throttling. V-port ball valves and purpose-built vacuum control valves use different port and seat designs.
For toxic, flammable, oxidizing or corrosive gas, a vacuum leak may allow air into the process or let process gas escape after the line returns to positive pressure. The system may need two valves in series, double stem seals, a monitored space, inert-gas purging, fail-closed actuation and gas detection.
When to Use a Ball Valve
A ball valve is often suitable when the system needs:
- Fast quarter-turn isolation
- Compact installation
- Low flow restriction when open
- Rough or medium vacuum
- Moderate temperature
- Moderate operating frequency
- A gas compatible with the seats and seals
A dedicated vacuum gate, angle, butterfly or diaphragm valve may be better for ultra-high vacuum, high-temperature bakeout, frequent cycling, low particle generation, bellows stem isolation or precise pressure control.
Where bore size, torque or installation space affects the choice, see Carilo’s ball valve and butterfly valve comparison.
Troubleshooting
| Observed problem | Likely causes | Useful check |
|---|---|---|
| Pressure stops above the target | External leak, seat leak, outgassing, process gas or permeation | Separate seat and envelope tests, then review the pressure-time curve |
| Pressure improves slowly for hours | Water, oil, condensate or trapped gas | Clean, heat carefully and isolate trapped volumes |
| The valve passes in one direction only | Floating-ball movement or unequal seat design | Review the sectional drawing and test both directions |
| The new valve passes but the cycled valve fails | Seat wear, packing relaxation or actuator movement | Check torque, alignment and post-cycle leakage |
| The helium reading rises slowly | Delayed detector response or elastomer permeation | Use a fixed exposure and measurement window |
| Leakage appears after heating | Thermal movement, gasket relaxation or seat distortion | Test while hot and again after cooling |
Purchase Requirements
The buyer should provide:
- Target absolute pressure
- Normal and reverse pressure directions
- Process gas and contaminants
- Operating and bakeout temperatures
- Valve size and required bore
- Connection type
- Operating frequency
- Seat and external leak limits
- Test gas and test direction
- Cleaning and lubricant restrictions
- Required actuator fail position
The test certificate should report:
- Valve model and serial number
- Seat and stem-seal materials
- Pressure on each port
- Test direction and temperature
- Helium concentration
- Detector and fixture background
- Exposure and response time
- Actual measured leakage
- Acceptance limit
- Cycle and bakeout condition
- Reference-leak or calibration identification
Test the finished valve with a calibrated helium mass-spectrometer leak detector. Measure seat leakage in the installed pressure direction and external leakage around the complete valve. Record the test pressure, temperature, direction, background, exposure time and actual measured result. For bidirectional service, report each direction separately.
Conclusion
Ball valves can work well in vacuum service when the seat direction, stem sealing, cavity design and gas load match the system. At a 1 bar differential, a 100 mm projected area can place about 785 N on a floating ball, while two ports already under vacuum may provide almost no pressure-assisted sealing. Set the allowable leak rate from Q = S × P, include every valve in the gas-load budget, and test seat and external leakage separately. Below about 10−7 mbar, during high-temperature bakeout or in very clean processes, a dedicated vacuum valve is usually easier to qualify.
