Whether a valve “leaks” is actually three different questions: shell strength, seat leakage, and leakage to atmosphere through the stem. GB/T 13927, API 598 and ISO 5208 answer the first two. When a project specifies Low-E, LDAR or VOC requirements, the standards that matter are a different pair — ISO 15848-1/-2 and API 624. Both measure the same thing (micro-leakage from the stem, gland and body joints to atmosphere), but their classification systems, cycling profiles and units differ completely, which is where quotations and acceptance criteria most often fail to line up.
This article separates the two systems in order: what each standard covers, how to read the classes, what production testing measures, and what the test bench must provide. All figures follow the official text referenced by the project; the tables below are for quick orientation during sourcing discussions.
1. What fugitive-emission testing actually measures
Fugitive emissions are the uncontrolled, very small releases from dynamic seal points — mainly the stuffing box where the stem or shaft exits the pressurized body — and from body joints. The media are usually methane, VOCs or process gas; the leak rate is expressed in ppmv (volume concentration, sniffing method) or mg·s⁻¹·m⁻¹ (volumetric leak rate per unit stem diameter, helium method) — orders of magnitude below seat-leakage criteria, and far beyond what a visual check or a routine hydrostatic test can detect.
A common scoping error: answering a fugitive-emissions clause with an ISO 5208 Class A or API 598 leakage rate. The former judges the seat (internal leak), the latter judges the stem packing (external leak); the methods, units and acceptance limits are not interchangeable. Both requirements usually apply to the same valve and are accepted separately.
2. ISO 15848-1: type testing, expressed as a class code
ISO 15848-1 is a type-testing standard: a representative sample valve is exercised under controlled temperature, pressure and cycling, and the leak performance of the stem seal and body joints is not reported as pass/fail but as a quotable class code — tightness class + endurance class + temperature class + test medium.
Tightness classes
The test medium is helium (H) or methane of at least 97% purity (M), and classes are labelled per medium:
| Medium | Class | Leak-rate limit | Typically achieved with |
|---|---|---|---|
| Helium | AH | ≤10⁻⁵ mg·s⁻¹·m⁻¹ | Bellows seals, or equivalent stem sealing on quarter-turn valves |
| Helium | BH | ≤10⁻⁴ mg·s⁻¹·m⁻¹ | PTFE-based packings or elastomeric seals |
| Helium | CH | ≤10⁻² mg·s⁻¹·m⁻¹ | Flexible graphite packings |
| Methane | AM / BM / CM | ≤50 / ≤100 / ≤500 ppmv | Corresponding to the three sealing families above |
Two points worth noting: Class A can only be measured with helium using the vacuum method — conventional sniffer hardware cannot reach that resolution; and the leak rate is normalized per unit stem diameter, so a larger valve of the same class is allowed a larger absolute leak. When comparing two manufacturers’ certificates, check that the sizes are comparable first.
Endurance and temperature classes
| Valve type | Endurance class | Mechanical cycles | Thermal cycles |
|---|---|---|---|
| Isolation | CO1 | 205 | 2 |
| Isolation | CO2 | 1500 | 3 |
| Isolation | CO3 | 2500 | 4 |
| Control | CC1 / CC2 / CC3 | 20000 / 60000 / 100000 | 2 / 3 / 4 |
Temperature classes cover -196 °C, -46 °C, room temperature (-29 to +40 °C), 200 °C and 400 °C. A complete ISO 15848-1 certificate should read as four elements, e.g. “BH–CO2–200 °C–helium”. Check a supplier’s report against all four; anything missing cannot be quoted as-is.
3. ISO 15848-2 and API 624: from type to production
Type testing proves the design; batch delivery also asks whether this particular valve is acceptable. That is the role of production testing (ISO 15848-2): for valves already type-tested under Part 1, a short verification — room temperature, a small number of cycles — is performed before shipment on a sampling or per-valve basis, with methods and limits tied to the type-test class.
API 624 is the American-system path for rising and rising-rotating stem valves equipped with graphite packing, where the packing set must previously have qualified under API 622. Key requirements:
| Item | API 624 requirement |
|---|---|
| Test medium | Methane, ≥97% purity |
| Cycling | 310 mechanical cycles + 3 thermal cycles (260 °C/500 °F); optional -29 °C low-temperature cycle |
| Test pressure | The lower of 600 psig or the ASME B16.34 rating at 260 °C, held constant |
| Measurement | Based on EPA Method 21 sniffing, foil enclosure shielding, static and dynamic leak recorded separately |
| Packing adjustment | Not permitted during the test |
| Acceptance | ≤100 ppmv throughout |
| Scope | ≤NPS 24, ≤Class 1500; stem vertical |
The companion API 641 covers quarter-turn valves (ball, butterfly) with the same methane 100 ppmv limit. A frequent specifying mistake is requesting API 624 for a quarter-turn valve — it only covers rising-stem designs; quarter-turn valves should go through API 641 or ISO 15848-1.
4. Mapping the two systems onto one valve
Projects frequently require both. A practical alignment: ISO 15848-1 states the design capability as a class; API 624/641 states the type-test conclusion of that design against the methane/100 ppmv line. A packed valve can hold both a BH-CO2 certificate and an API 624 report, and neither substitutes for the other — helium measurement cannot express the field comparability that the methane system cares about, and methane sniffing cannot reach BH resolution. List the two references separately in the technical agreement and accept each by its own criteria.
Another recurring split is helium versus methane. Helium offers the resolution needed for class assignment in toxic or severe service; methane follows the field LDAR logic of EPA Method 21. Put the medium choice into the contract, and design the two media circuits on the bench separately — deciding this at the equipment stage matters, because switching afterwards is expensive.
5. What the test bench must provide
Compared with a conventional pressure-test bench, a fugitive-emissions bench differs in four areas:
- Actuation and cycling: mechanical cycling is done by a driver that must not impose side loads on the stem or packing; running torque is recorded on the first and last cycle.
- Heating and temperature control: the valve body must reach 260 °C ±5% and hold; thermocouples are placed per the standard (stuffing box as the control point), with reproducible ramp and soak profiles.
- Measurement chain: the sniffing route needs a Method 21-grade detector, foil enclosures, probe-speed control and background compensation; the helium route uses a vacuum enclosure plus helium mass spectrometer. The two chains are not interchangeable.
- Safety envelope: methane is a pressurized flammable gas — supply circuit, venting, interlocks and personnel protection are designed for the test facility itself, which is the largest cost gap versus a hydrostatic bench.
If only production testing (ISO 15848-2 / pre-shipment sampling) is planned, the bench can be simpler — room temperature, short cycling, sniffing-based measurement — but leave headroom to upgrade to type-test conditions, because the first customer question is usually whether the API 624 profile can be reproduced.
6. Five things to check at acceptance
- The four elements of the type certificate (tightness–endurance–temperature–medium) match the contract clause by clause;
- The packing type, ring count and torque in the API 624 report match the delivered valve — the packing set is what API 622 qualified, and changing it changes the conclusion;
- Leak records separate static from dynamic readings, with background correction applied;
- Whether any packing adjustment occurred (API 624 forbids it; ISO 15848-1 allows 0/1/2/3 depending on class);
- The sampling rule and record format for production testing are written into the technical agreement.
FAQ
Is ISO 15848-1 Class A “zero leakage”?
No. Class A is a quantified upper limit (≤10⁻⁵ mg·s⁻¹·m⁻¹ with helium) and can only be measured with the vacuum method. Writing “zero leakage” into an agreement creates a criterion no instrument can demonstrate.
If a valve has an API 624 report, does it still need ISO 15848-1?
Depends on the contract. API 624 is a type-test conclusion for rising-stem graphite-packed valves at the methane 100 ppmv line; European projects or clauses that require a class code (e.g. BH-CO3) still need an ISO 15848-1 report. The media, units and cycling systems differ, and the results cannot be converted into each other.
Can a conventional pressure-test bench be adapted for fugitive-emissions testing?
Hydrostatic/pneumatic benches judge internal leakage and shell strength; the magnitudes and the measurement chain differ. Retrofitting item by item (methane circuit, heating, sniffing chain) usually costs more than a purpose-built bench; define the requirement boundary once, then size the solution.
Specific parameters, criteria and scopes of application follow the official ISO 15848-1/-2 and API 622/624/641 texts referenced by the project; the tables above are for quick orientation in sourcing discussions. How these systems divide scope from internal-leakage standards is covered in ISO 5208 leakage classes and API 598 vs GB/T 13927; the current editions and scopes of all referenced standards are summarized in the Valve Test Standards Index, and the pressure-rating check logic is in the PN vs CWP checklist.
Related test equipment
Each test above maps to a bench. Pick by valve type and by the item you need to prove:
- Pipe and connection leak test bench — pressure-decay and helium methods, judged against a leak-rate limit
- Valve life-cycle test bench — pressurised cycling with automatic counting and stop-on-failure
- Valve comprehensive test bench — pressure, tightness, flow, torque and life cycling on one clamping system
- Valve testing equipment range — browse all models by valve type and test item