“Leakage” on a gas valve is not the same thing as “leakage” on a water valve. EN 13774 requires both external and internal tightness to be tested at two pressure tiers: 6 mbar and 1.1 × pmax. The 6 mbar tier looks for micro-leakage — a slow seep in a buried valve pit can accumulate to the lower explosive limit in a few months. The 1.1× tier looks at pressure-containing tightness. A report that only shows 1.1× and says “gas-tight: pass” has not completed the test.
EN 13774:2013, “Valves for gas distribution systems with maximum operating pressure ≤ 16 bar — Performance requirements”, covers metal isolating valves: plug, ball, gate, globe and butterfly. The medium is fuel gas of the first, second and third families per EN 437, and the temperature is either T1 (−10 to +60 °C) or T2 (−20 to +60 °C). It is not a product standard but an additional-requirements standard — where it conflicts with the product standards listed in Annex B, EN 13774 prevails. This article runs: scope and exclusions, the four pressure and torque terms, the test list, sampling rules, the 6 mbar tier, the three temperature points, the boundary with EN 331, the Chinese mapping, and equipment.
1. Scope and exclusions — is your valve even in it?
| Item | EN 13774:2013 |
|---|---|
| Valve types | Metal isolating valves: plug, ball, gate, globe, butterfly |
| Medium | EN 437 first family (manufactured gas), second (natural gas), third (LPG) |
| Pressure | Maximum operating pressure MOP ≤ 16 bar |
| Temperature | T1: −10 to +60 °C; T2: −20 to +60 °C; special designs as specified by the purchaser |
| Not covered | Domestic (in-building) gas valves, safety relief valves, wellhead valves; for power-operated valves the power source is excluded |
| Relationship to other standards | Additional to the product standards in Annex B; EN 13774 prevails on conflict |
This exclusion is the one most often misapplied: a DN 15 in-building gas ball valve belongs to EN 331, not EN 13774; conversely a DN 200 buried distribution gate valve belongs to EN 13774 and cannot ship on an EN 331 report. Pressure tiers, temperature tiers and endurance counts all differ — see EN 331.
2. pmax, MOP, MOT and maximum strength torque — four terms, four uses
| Term | Definition | Used for |
|---|---|---|
| pmax | Maximum allowable pressure: the highest pressure any part may carry continuously at the specified temperature | Basis of test pressure (1.1 × pmax) |
| MOP | Maximum operating pressure: the highest pressure in a system operated continuously under normal conditions | Selection and system design; ≤ 16 bar |
| MOT | Maximum operating torque: the highest torque needed to operate the valve at pmax and any service temperature | Operability criterion, actuator sizing |
| Maximum strength torque | The highest torque applied to the stem, with the obturator fully open or closed, that causes no loss of function | Stop strength and excessive-actuating-force tests |
The pmax / MOP distinction matters in practice: MOP is a system promise, pmax is a product capability. A valve declared pmax 16 bar installed on a 4 bar branch is still tested at 1.1 × 16 = 17.6 bar. That sets the bench range — not the 4 bar.
3. The test list — type tests and production tests are not the same set
| Test | Type | Production | Notes |
|---|---|---|---|
| Shell strength | ✔ | ✔ | Per product standard and EN 12266-1; no visible leakage, no permanent deformation |
| External tightness | ✔ | ✔ | Two tiers: 6 mbar and 1.1 × pmax, air or inert gas, method per EN 12266-1 Annex A.3; type test at ta, tmax and tmin, production at ta |
| Internal tightness | ✔ | ✔ | Same two tiers, method A.4; for DN ≥ 500 the type-test hold is 15 minutes |
| Operability / MOT | ✔ | ✔ | Open and close at pmax; torque must not exceed the declared value |
| Strength of stops | ✔ | — | After maximum strength torque, stops and stem survive and the valve still seals |
| Resistance to excessive actuating forces | ✔ | — | Simulates cheater bars and impact wrenches in the field |
| Obturator resistance to static differential pressure | ✔ | — | One side pressurised, zero on the other, both directions |
| Endurance (cycling) | ✔ | — | Internal and external tightness plus torque re-tested afterwards |
| Resistance to wear | Optional | — | Only the smallest size in the range; pass criterion after wear is zero visible leakage |
| Reference flow rate | Optional | — | Air flow at 0 °C and 1 013 mbar, declared at full opening |
The two items most often dropped are “excessive actuating forces” and “wear”. The first mirrors real field abuse; the second mirrors long-term erosion by dust and condensate in gas service. A buyer who accepts “shell + tightness” has waived two type tests.
4. Sampling — why smallest, largest and one in the middle
The range-qualification logic is specific: the smallest size, the largest size (capped at DN 500) and one chosen from the middle, one sample each, run through the full type-test programme. Above DN 500, a DN 500 sample qualifies the range up to DN 1000. The precondition is that materials, design and construction do not change across the range — change the body material, the sealing arrangement or the stem connection and the qualification chain breaks.
Wear is the exception: smallest size only, because wear scales with travel and velocity, and the small bore sees the harshest erosion per unit area.
Two hard document requirements: material certificates to EN 10204 — 3.1 for body and obturator, 2.2 for the stem; and all tests certified by an independent, accepted body. An in-house report without third-party signature does not complete the type test.
5. The 6 mbar tier — the real gate on a gas valve
6 mbar is roughly 60 mm of water column. Why would a standard push the test pressure that low? Because that is the actual condition: medium- and low-pressure distribution networks run at tens of mbar, and once a valve is closed the differential across it is small. At that differential, poor seat contact, washed-out sealant grease or a speck of debris on the seat all leak — while 1.1 × pmax presses the rubber harder and hides the defect. It is the same logic as the 0.5 bar tier on water supply valves.
Instrument choice decides whether you can judge it: use pressure decay with a differential sensor, confirmed by bubble/soap film. A high-range gauge will not move for a 0.1 mbar seep. For a comparison of methods see choosing a tightness test method.
6. Three temperature points — why a type test goes into a chamber
External tightness is tested at ta (20 ± 5 °C), tmax (+60 °C) and tmin (−10 or −20 °C). The reason is plain: elastomers stiffen and lose rebound when cold, and creep and age faster when hot. Room-temperature compliance says nothing about either end. Buried valves also see annual soil-temperature cycles, so T2 (−20 °C) is not optional on northern projects.
For equipment this means either the test station goes inside a chamber or the medium is conditioned to the setpoint; the cold point also needs dew-point control and frost protection, so supply gas must be dried below the test temperature. Chinese reports routinely state only “tightness passed at ambient”, which simply does not map onto an EN 13774 type test.
7. Boundary with EN 331 and EN 1074
| Standard | Object | Pressure | Sizes | Key difference |
|---|---|---|---|---|
| EN 331:2015 | In-building manual gas ball/plug valves | MOP 20 mbar – 2 bar (five classes) | DN 6 – 100 | Indoor, not direct-buried; endurance 500 – 5 000 cycles |
| EN 13774:2013 | Gas distribution metal isolating valves | MOP ≤ 16 bar | No upper limit (qualified to DN 1000) | Buried/stations, dual-tier tightness (6 mbar + 1.1 pmax), three temperatures, range sampling |
| EN 1074-1/-2 | Water supply network isolating valves | PFA 6 – 25 bar | DN 50 – 2000 | Water, PFA/PMA/PEA, disinfection resistance |
The three cannot substitute for each other, but one bench can serve all three: dual-tier pressure circuits, a chamber, torque and cycling hardware are common — only criteria and record fields change. For a multi-standard frame see the valve comprehensive test bench.
8. The Chinese mapping
| Dimension | Chinese counterpart | Difference |
|---|---|---|
| Pressure test method | GB/T 13927-2022, GB/T 26480-2011 | 1.5× shell, 1.1× seat; no 6 mbar micro-pressure tier |
| Gas valve products | GB/T 12232 and other steel valve standards plus city-gas codes | Organised by PN, not by MOP with T1/T2 temperature classes |
| Design and installation | GB 50028 (code for city gas engineering) | System design and acceptance, not a product type test |
| Type testing | TSG D7002-2023 (pressure piping components) | Licensing-scope type test; items do not map one-to-one — see TSG D7002 |
| Material certificates | GB/T 18253 / EN 10204 equivalent | Export projects need EN 10204 3.1 / 2.2 |
For export gas valves, three things are most often missing: 6 mbar tier data, type-test records at all three temperatures, and the range-sampling statement. Supplying those is worth far more than pushing the shell pressure higher.
9. Configuring the equipment
- Dual-tier circuits: one at 6 mbar (differential sensor, 0.1 % FS, 0 – 20 mbar range), one at 1.1 × pmax (17.6 bar for a 16 bar valve), switchable, logged separately
- Temperature stations: −20, +20 and +60 °C, either in-chamber or medium-conditioned; gas dew point below the lowest test temperature
- Torque and stops: two loading levels — MOT and maximum strength torque — with torque-versus-rotation logging
- Cycling bench: cycles under pressure with automatic counting, routed straight into the tightness re-test
- Large sizes: DN 500 and above hold 15 minutes at 17.6 bar; blind flanges and clamping sized on the resulting axial force
- Helium (optional): for ppm-level requirements or leak localisation where decay sensitivity is insufficient
Individual machines include the piping tightness test bench, valve pressure test machine, valve life test bench and valve torque test machine.
FAQ
Is a 1.1 × pmax tightness test enough for EN 13774?
No. Both external and internal tightness are required at 6 mbar and at 1.1 × pmax. Without the 6 mbar tier you have not verified sealing at the small differential that actually exists across a closed distribution valve, and technical review will record the test as incomplete.
Must the type test be third-party?
Yes. EN 13774 requires all tests to be certified by an independent, accepted body. In-house data can only serve as internal pre-verification — unlike some Chinese type-test arrangements that allow the manufacturer to test in-house.
What qualifies a DN 600 valve?
A DN 500 sample, which qualifies the range up to DN 1000. That holds only while materials, design and construction stay unchanged; a different body material or sealing arrangement at DN 600 breaks the coverage.
T1 or T2?
Driven by the minimum ambient temperature at the installation. T1 is −10 °C and T2 is −20 °C, and the purchaser states the class in the technical specification. Northern buried work generally specifies T2. Declaring T1 and installing at −20 °C is outside the declared range.
Specific clauses follow EN 13774:2013; the tables above are for orientation in selection, quoting and test discussions. For in-building gas valves see EN 331; for the European water-supply system see EN 1074; for US pressure classes and test pressures see ASME B16.34; and the full map is in the valve test standards index.
Related test equipment
Each test above maps to a bench. Pick by valve type and by the item you need to prove:
- Valve comprehensive test bench — pressure, tightness, flow, torque and life cycling on one clamping system
- Valve torque test machine — break-to-open and seating torque curves for actuator sizing
- Valve pressure test machine — shell at 1.5x and seat at 1.1x, with a report per valve
- Valve testing equipment range — browse all models by valve type and test item